<!DOCTYPE html>
<html lang="en">
<head>
<meta charset="utf-8" />
<meta http-equiv="X-UA-Compatible" content="IE=edge" />
<title>FFmpeg documentation</title>
<link rel="stylesheet" href="bootstrap.min.css" />
<link rel="stylesheet" href="style.min.css" />
<meta name="description" content="FFmpeg Filters Documentation: ">
<meta name="keywords" content="FFmpeg documentation : FFmpeg Filters ">
<meta name="Generator" content="texi2html 5.0">
<!-- Created on January 13, 2016 by texi2html 5.0 -->
<!--
texi2html was written by:
Lionel Cons <Lionel.Cons@cern.ch> (original author)
Karl Berry <karl@freefriends.org>
Olaf Bachmann <obachman@mathematik.uni-kl.de>
and many others.
Maintained by: Many creative people.
Send bugs and suggestions to <texi2html-bug@nongnu.org>
-->
<meta http-equiv="Content-Type" content="text/html; charset=utf-8">
</head>
<body>
<div style="width: 95%; margin: auto">
<h1 class="titlefont">FFmpeg Filters Documentation</h1>
<hr>
<a name="SEC_Top"></a>
<a name="SEC_Contents"></a>
<h1>Table of Contents</h1>
<div class="contents">
<ul class="no-bullet">
<li><a name="toc-Description" href="#Description">1 Description</a></li>
<li><a name="toc-Filtering-Introduction" href="#Filtering-Introduction">2 Filtering Introduction</a></li>
<li><a name="toc-graph2dot" href="#graph2dot">3 graph2dot</a></li>
<li><a name="toc-Filtergraph-description" href="#Filtergraph-description">4 Filtergraph description</a>
<ul class="no-bullet">
<li><a name="toc-Filtergraph-syntax-1" href="#Filtergraph-syntax-1">4.1 Filtergraph syntax</a></li>
<li><a name="toc-Notes-on-filtergraph-escaping" href="#Notes-on-filtergraph-escaping">4.2 Notes on filtergraph escaping</a></li>
</ul></li>
<li><a name="toc-Timeline-editing" href="#Timeline-editing">5 Timeline editing</a></li>
<li><a name="toc-Audio-Filters" href="#Audio-Filters">6 Audio Filters</a>
<ul class="no-bullet">
<li><a name="toc-aconvert" href="#aconvert">6.1 aconvert</a>
<ul class="no-bullet">
<li><a name="toc-Examples-61" href="#Examples-61">6.1.1 Examples</a></li>
</ul></li>
<li><a name="toc-adelay" href="#adelay">6.2 adelay</a>
<ul class="no-bullet">
<li><a name="toc-Examples" href="#Examples">6.2.1 Examples</a></li>
</ul></li>
<li><a name="toc-aecho" href="#aecho">6.3 aecho</a>
<ul class="no-bullet">
<li><a name="toc-Examples-5" href="#Examples-5">6.3.1 Examples</a></li>
</ul></li>
<li><a name="toc-aeval" href="#aeval">6.4 aeval</a>
<ul class="no-bullet">
<li><a name="toc-Examples-15" href="#Examples-15">6.4.1 Examples</a></li>
</ul></li>
<li><a name="toc-afade" href="#afade">6.5 afade</a>
<ul class="no-bullet">
<li><a name="toc-Examples-70" href="#Examples-70">6.5.1 Examples</a></li>
</ul></li>
<li><a name="toc-aformat-1" href="#aformat-1">6.6 aformat</a></li>
<li><a name="toc-allpass" href="#allpass">6.7 allpass</a></li>
<li><a name="toc-amerge" href="#amerge">6.8 amerge</a>
<ul class="no-bullet">
<li><a name="toc-Examples-40" href="#Examples-40">6.8.1 Examples</a></li>
</ul></li>
<li><a name="toc-amix" href="#amix">6.9 amix</a></li>
<li><a name="toc-anull" href="#anull">6.10 anull</a></li>
<li><a name="toc-apad" href="#apad">6.11 apad</a>
<ul class="no-bullet">
<li><a name="toc-Examples-47" href="#Examples-47">6.11.1 Examples</a></li>
</ul></li>
<li><a name="toc-aphaser" href="#aphaser">6.12 aphaser</a></li>
<li><a name="toc-aresample-1" href="#aresample-1">6.13 aresample</a>
<ul class="no-bullet">
<li><a name="toc-Examples-28" href="#Examples-28">6.13.1 Examples</a></li>
</ul></li>
<li><a name="toc-asetnsamples" href="#asetnsamples">6.14 asetnsamples</a></li>
<li><a name="toc-asetrate" href="#asetrate">6.15 asetrate</a></li>
<li><a name="toc-ashowinfo" href="#ashowinfo">6.16 ashowinfo</a></li>
<li><a name="toc-astats" href="#astats">6.17 astats</a></li>
<li><a name="toc-astreamsync" href="#astreamsync">6.18 astreamsync</a>
<ul class="no-bullet">
<li><a name="toc-Examples-27" href="#Examples-27">6.18.1 Examples</a></li>
</ul></li>
<li><a name="toc-asyncts" href="#asyncts">6.19 asyncts</a></li>
<li><a name="toc-atempo" href="#atempo">6.20 atempo</a>
<ul class="no-bullet">
<li><a name="toc-Examples-48" href="#Examples-48">6.20.1 Examples</a></li>
</ul></li>
<li><a name="toc-atrim" href="#atrim">6.21 atrim</a></li>
<li><a name="toc-bandpass" href="#bandpass">6.22 bandpass</a></li>
<li><a name="toc-bandreject" href="#bandreject">6.23 bandreject</a></li>
<li><a name="toc-bass" href="#bass">6.24 bass</a></li>
<li><a name="toc-biquad" href="#biquad">6.25 biquad</a></li>
<li><a name="toc-bs2b" href="#bs2b">6.26 bs2b</a></li>
<li><a name="toc-channelmap" href="#channelmap">6.27 channelmap</a></li>
<li><a name="toc-channelsplit" href="#channelsplit">6.28 channelsplit</a></li>
<li><a name="toc-compand" href="#compand">6.29 compand</a>
<ul class="no-bullet">
<li><a name="toc-Examples-79" href="#Examples-79">6.29.1 Examples</a></li>
</ul></li>
<li><a name="toc-earwax" href="#earwax">6.30 earwax</a></li>
<li><a name="toc-equalizer" href="#equalizer">6.31 equalizer</a>
<ul class="no-bullet">
<li><a name="toc-Examples-14" href="#Examples-14">6.31.1 Examples</a></li>
</ul></li>
<li><a name="toc-flanger" href="#flanger">6.32 flanger</a></li>
<li><a name="toc-highpass" href="#highpass">6.33 highpass</a></li>
<li><a name="toc-join" href="#join">6.34 join</a></li>
<li><a name="toc-ladspa" href="#ladspa">6.35 ladspa</a>
<ul class="no-bullet">
<li><a name="toc-Examples-35" href="#Examples-35">6.35.1 Examples</a></li>
<li><a name="toc-Commands" href="#Commands">6.35.2 Commands</a></li>
</ul></li>
<li><a name="toc-lowpass" href="#lowpass">6.36 lowpass</a></li>
<li><a name="toc-pan" href="#pan">6.37 pan</a>
<ul class="no-bullet">
<li><a name="toc-Mixing-examples" href="#Mixing-examples">6.37.1 Mixing examples</a></li>
<li><a name="toc-Remapping-examples" href="#Remapping-examples">6.37.2 Remapping examples</a></li>
</ul></li>
<li><a name="toc-replaygain" href="#replaygain">6.38 replaygain</a></li>
<li><a name="toc-resample" href="#resample">6.39 resample</a></li>
<li><a name="toc-silencedetect" href="#silencedetect">6.40 silencedetect</a>
<ul class="no-bullet">
<li><a name="toc-Examples-39" href="#Examples-39">6.40.1 Examples</a></li>
</ul></li>
<li><a name="toc-silenceremove" href="#silenceremove">6.41 silenceremove</a>
<ul class="no-bullet">
<li><a name="toc-Examples-58" href="#Examples-58">6.41.1 Examples</a></li>
</ul></li>
<li><a name="toc-treble" href="#treble">6.42 treble</a></li>
<li><a name="toc-volume" href="#volume">6.43 volume</a>
<ul class="no-bullet">
<li><a name="toc-Commands-1" href="#Commands-1">6.43.1 Commands</a></li>
<li><a name="toc-Examples-68" href="#Examples-68">6.43.2 Examples</a></li>
</ul></li>
<li><a name="toc-volumedetect" href="#volumedetect">6.44 volumedetect</a>
<ul class="no-bullet">
<li><a name="toc-Examples-22" href="#Examples-22">6.44.1 Examples</a></li>
</ul>
</li>
</ul></li>
<li><a name="toc-Audio-Sources" href="#Audio-Sources">7 Audio Sources</a>
<ul class="no-bullet">
<li><a name="toc-abuffer" href="#abuffer">7.1 abuffer</a>
<ul class="no-bullet">
<li><a name="toc-Examples-60" href="#Examples-60">7.1.1 Examples</a></li>
</ul></li>
<li><a name="toc-aevalsrc" href="#aevalsrc">7.2 aevalsrc</a>
<ul class="no-bullet">
<li><a name="toc-Examples-11" href="#Examples-11">7.2.1 Examples</a></li>
</ul></li>
<li><a name="toc-anullsrc" href="#anullsrc">7.3 anullsrc</a>
<ul class="no-bullet">
<li><a name="toc-Examples-32" href="#Examples-32">7.3.1 Examples</a></li>
</ul></li>
<li><a name="toc-flite" href="#flite">7.4 flite</a>
<ul class="no-bullet">
<li><a name="toc-Examples-72" href="#Examples-72">7.4.1 Examples</a></li>
</ul></li>
<li><a name="toc-sine" href="#sine">7.5 sine</a>
<ul class="no-bullet">
<li><a name="toc-Examples-8" href="#Examples-8">7.5.1 Examples</a></li>
</ul>
</li>
</ul></li>
<li><a name="toc-Audio-Sinks" href="#Audio-Sinks">8 Audio Sinks</a>
<ul class="no-bullet">
<li><a name="toc-abuffersink" href="#abuffersink">8.1 abuffersink</a></li>
<li><a name="toc-anullsink" href="#anullsink">8.2 anullsink</a></li>
</ul></li>
<li><a name="toc-Video-Filters" href="#Video-Filters">9 Video Filters</a>
<ul class="no-bullet">
<li><a name="toc-alphaextract" href="#alphaextract">9.1 alphaextract</a></li>
<li><a name="toc-alphamerge" href="#alphamerge">9.2 alphamerge</a></li>
<li><a name="toc-ass" href="#ass">9.3 ass</a></li>
<li><a name="toc-bbox" href="#bbox">9.4 bbox</a></li>
<li><a name="toc-blackdetect" href="#blackdetect">9.5 blackdetect</a></li>
<li><a name="toc-blackframe" href="#blackframe">9.6 blackframe</a></li>
<li><a name="toc-blend" href="#blend">9.7 blend</a>
<ul class="no-bullet">
<li><a name="toc-Examples-78" href="#Examples-78">9.7.1 Examples</a></li>
</ul></li>
<li><a name="toc-boxblur" href="#boxblur">9.8 boxblur</a>
<ul class="no-bullet">
<li><a name="toc-Examples-81" href="#Examples-81">9.8.1 Examples</a></li>
</ul></li>
<li><a name="toc-codecview" href="#codecview">9.9 codecview</a>
<ul class="no-bullet">
<li><a name="toc-Examples-6" href="#Examples-6">9.9.1 Examples</a></li>
</ul></li>
<li><a name="toc-colorbalance" href="#colorbalance">9.10 colorbalance</a>
<ul class="no-bullet">
<li><a name="toc-Examples-77" href="#Examples-77">9.10.1 Examples</a></li>
</ul></li>
<li><a name="toc-colorchannelmixer" href="#colorchannelmixer">9.11 colorchannelmixer</a>
<ul class="no-bullet">
<li><a name="toc-Examples-56" href="#Examples-56">9.11.1 Examples</a></li>
</ul></li>
<li><a name="toc-colormatrix" href="#colormatrix">9.12 colormatrix</a></li>
<li><a name="toc-copy" href="#copy">9.13 copy</a></li>
<li><a name="toc-crop" href="#crop">9.14 crop</a>
<ul class="no-bullet">
<li><a name="toc-Examples-10" href="#Examples-10">9.14.1 Examples</a></li>
</ul></li>
<li><a name="toc-cropdetect" href="#cropdetect">9.15 cropdetect</a></li>
<li><a name="toc-curves-1" href="#curves-1">9.16 curves</a>
<ul class="no-bullet">
<li><a name="toc-Examples-20" href="#Examples-20">9.16.1 Examples</a></li>
</ul></li>
<li><a name="toc-dctdnoiz" href="#dctdnoiz">9.17 dctdnoiz</a>
<ul class="no-bullet">
<li><a name="toc-Examples-59" href="#Examples-59">9.17.1 Examples</a></li>
</ul></li>
<li><a name="toc-decimate-1" href="#decimate-1">9.18 decimate</a></li>
<li><a name="toc-dejudder" href="#dejudder">9.19 dejudder</a></li>
<li><a name="toc-delogo" href="#delogo">9.20 delogo</a>
<ul class="no-bullet">
<li><a name="toc-Examples-75" href="#Examples-75">9.20.1 Examples</a></li>
</ul></li>
<li><a name="toc-deshake" href="#deshake">9.21 deshake</a></li>
<li><a name="toc-drawbox" href="#drawbox">9.22 drawbox</a>
<ul class="no-bullet">
<li><a name="toc-Examples-16" href="#Examples-16">9.22.1 Examples</a></li>
</ul></li>
<li><a name="toc-drawgrid" href="#drawgrid">9.23 drawgrid</a>
<ul class="no-bullet">
<li><a name="toc-Examples-29" href="#Examples-29">9.23.1 Examples</a></li>
</ul></li>
<li><a name="toc-drawtext-1" href="#drawtext-1">9.24 drawtext</a>
<ul class="no-bullet">
<li><a name="toc-Syntax" href="#Syntax">9.24.1 Syntax</a></li>
<li><a name="toc-Text-expansion" href="#Text-expansion">9.24.2 Text expansion</a></li>
<li><a name="toc-Examples-46" href="#Examples-46">9.24.3 Examples</a></li>
</ul></li>
<li><a name="toc-edgedetect" href="#edgedetect">9.25 edgedetect</a>
<ul class="no-bullet">
<li><a name="toc-Examples-52" href="#Examples-52">9.25.1 Examples</a></li>
</ul></li>
<li><a name="toc-extractplanes" href="#extractplanes">9.26 extractplanes</a>
<ul class="no-bullet">
<li><a name="toc-Examples-37" href="#Examples-37">9.26.1 Examples</a></li>
</ul></li>
<li><a name="toc-elbg" href="#elbg">9.27 elbg</a></li>
<li><a name="toc-fade" href="#fade">9.28 fade</a>
<ul class="no-bullet">
<li><a name="toc-Examples-69" href="#Examples-69">9.28.1 Examples</a></li>
</ul></li>
<li><a name="toc-field" href="#field">9.29 field</a></li>
<li><a name="toc-fieldmatch" href="#fieldmatch">9.30 fieldmatch</a>
<ul class="no-bullet">
<li><a name="toc-p_002fc_002fn_002fu_002fb-meaning-1" href="#p_002fc_002fn_002fu_002fb-meaning-1">9.30.1 p/c/n/u/b meaning</a>
<ul class="no-bullet">
<li><a name="toc-p_002fc_002fn" href="#p_002fc_002fn">9.30.1.1 p/c/n</a></li>
<li><a name="toc-u_002fb" href="#u_002fb">9.30.1.2 u/b</a></li>
</ul></li>
<li><a name="toc-Examples-62" href="#Examples-62">9.30.2 Examples</a></li>
</ul></li>
<li><a name="toc-fieldorder" href="#fieldorder">9.31 fieldorder</a></li>
<li><a name="toc-fifo" href="#fifo">9.32 fifo</a></li>
<li><a name="toc-format-1" href="#format-1">9.33 format</a>
<ul class="no-bullet">
<li><a name="toc-Examples-41" href="#Examples-41">9.33.1 Examples</a></li>
</ul></li>
<li><a name="toc-fps-1" href="#fps-1">9.34 fps</a>
<ul class="no-bullet">
<li><a name="toc-Examples-34" href="#Examples-34">9.34.1 Examples</a></li>
</ul></li>
<li><a name="toc-framepack" href="#framepack">9.35 framepack</a></li>
<li><a name="toc-framestep" href="#framestep">9.36 framestep</a></li>
<li><a name="toc-frei0r-1" href="#frei0r-1">9.37 frei0r</a>
<ul class="no-bullet">
<li><a name="toc-Examples-36" href="#Examples-36">9.37.1 Examples</a></li>
</ul></li>
<li><a name="toc-geq" href="#geq">9.38 geq</a>
<ul class="no-bullet">
<li><a name="toc-Examples-19" href="#Examples-19">9.38.1 Examples</a></li>
</ul></li>
<li><a name="toc-gradfun" href="#gradfun">9.39 gradfun</a>
<ul class="no-bullet">
<li><a name="toc-Examples-44" href="#Examples-44">9.39.1 Examples</a></li>
</ul></li>
<li><a name="toc-haldclut-1" href="#haldclut-1">9.40 haldclut</a>
<ul class="no-bullet">
<li><a name="toc-Workflow-examples" href="#Workflow-examples">9.40.1 Workflow examples</a>
<ul class="no-bullet">
<li><a name="toc-Hald-CLUT-video-stream" href="#Hald-CLUT-video-stream">9.40.1.1 Hald CLUT video stream</a></li>
<li><a name="toc-Hald-CLUT-with-preview" href="#Hald-CLUT-with-preview">9.40.1.2 Hald CLUT with preview</a></li>
</ul>
</li>
</ul></li>
<li><a name="toc-hflip" href="#hflip">9.41 hflip</a></li>
<li><a name="toc-histeq" href="#histeq">9.42 histeq</a></li>
<li><a name="toc-histogram" href="#histogram">9.43 histogram</a>
<ul class="no-bullet">
<li><a name="toc-Examples-55" href="#Examples-55">9.43.1 Examples</a></li>
</ul></li>
<li><a name="toc-hqdn3d-1" href="#hqdn3d-1">9.44 hqdn3d</a></li>
<li><a name="toc-hqx" href="#hqx">9.45 hqx</a></li>
<li><a name="toc-hue" href="#hue">9.46 hue</a>
<ul class="no-bullet">
<li><a name="toc-Examples-26" href="#Examples-26">9.46.1 Examples</a></li>
<li><a name="toc-Commands-4" href="#Commands-4">9.46.2 Commands</a></li>
</ul></li>
<li><a name="toc-idet" href="#idet">9.47 idet</a></li>
<li><a name="toc-il" href="#il">9.48 il</a></li>
<li><a name="toc-interlace" href="#interlace">9.49 interlace</a></li>
<li><a name="toc-kerndeint" href="#kerndeint">9.50 kerndeint</a>
<ul class="no-bullet">
<li><a name="toc-Examples-74" href="#Examples-74">9.50.1 Examples</a></li>
</ul></li>
<li><a name="toc-lenscorrection" href="#lenscorrection">9.51 lenscorrection</a>
<ul class="no-bullet">
<li><a name="toc-Options-1" href="#Options-1">9.51.1 Options</a></li>
</ul></li>
<li><a name="toc-lut3d-1" href="#lut3d-1">9.52 lut3d</a></li>
<li><a name="toc-lut_002c-lutrgb_002c-lutyuv" href="#lut_002c-lutrgb_002c-lutyuv">9.53 lut, lutrgb, lutyuv</a>
<ul class="no-bullet">
<li><a name="toc-Examples-53" href="#Examples-53">9.53.1 Examples</a></li>
</ul></li>
<li><a name="toc-mergeplanes" href="#mergeplanes">9.54 mergeplanes</a>
<ul class="no-bullet">
<li><a name="toc-Examples-50" href="#Examples-50">9.54.1 Examples</a></li>
</ul></li>
<li><a name="toc-mcdeint" href="#mcdeint">9.55 mcdeint</a></li>
<li><a name="toc-mp" href="#mp">9.56 mp</a>
<ul class="no-bullet">
<li><a name="toc-Examples-17" href="#Examples-17">9.56.1 Examples</a></li>
</ul></li>
<li><a name="toc-mpdecimate" href="#mpdecimate">9.57 mpdecimate</a></li>
<li><a name="toc-negate" href="#negate">9.58 negate</a></li>
<li><a name="toc-noformat" href="#noformat">9.59 noformat</a>
<ul class="no-bullet">
<li><a name="toc-Examples-30" href="#Examples-30">9.59.1 Examples</a></li>
</ul></li>
<li><a name="toc-noise" href="#noise">9.60 noise</a>
<ul class="no-bullet">
<li><a name="toc-Examples-33" href="#Examples-33">9.60.1 Examples</a></li>
</ul></li>
<li><a name="toc-null" href="#null">9.61 null</a></li>
<li><a name="toc-ocv" href="#ocv">9.62 ocv</a>
<ul class="no-bullet">
<li><a name="toc-dilate-1" href="#dilate-1">9.62.1 dilate</a></li>
<li><a name="toc-erode" href="#erode">9.62.2 erode</a></li>
<li><a name="toc-smooth" href="#smooth">9.62.3 smooth</a></li>
</ul></li>
<li><a name="toc-overlay-1" href="#overlay-1">9.63 overlay</a>
<ul class="no-bullet">
<li><a name="toc-Commands-5" href="#Commands-5">9.63.1 Commands</a></li>
<li><a name="toc-Examples-76" href="#Examples-76">9.63.2 Examples</a></li>
</ul></li>
<li><a name="toc-owdenoise" href="#owdenoise">9.64 owdenoise</a></li>
<li><a name="toc-pad" href="#pad">9.65 pad</a>
<ul class="no-bullet">
<li><a name="toc-Examples-12" href="#Examples-12">9.65.1 Examples</a></li>
</ul></li>
<li><a name="toc-perspective" href="#perspective">9.66 perspective</a></li>
<li><a name="toc-phase" href="#phase">9.67 phase</a></li>
<li><a name="toc-pixdesctest" href="#pixdesctest">9.68 pixdesctest</a></li>
<li><a name="toc-pp" href="#pp">9.69 pp</a>
<ul class="no-bullet">
<li><a name="toc-Examples-18" href="#Examples-18">9.69.1 Examples</a></li>
</ul></li>
<li><a name="toc-psnr" href="#psnr">9.70 psnr</a></li>
<li><a name="toc-pullup-1" href="#pullup-1">9.71 pullup</a></li>
<li><a name="toc-removelogo" href="#removelogo">9.72 removelogo</a></li>
<li><a name="toc-rotate" href="#rotate">9.73 rotate</a>
<ul class="no-bullet">
<li><a name="toc-Examples-73" href="#Examples-73">9.73.1 Examples</a></li>
<li><a name="toc-Commands-3" href="#Commands-3">9.73.2 Commands</a></li>
</ul></li>
<li><a name="toc-sab" href="#sab">9.74 sab</a></li>
<li><a name="toc-scale-1" href="#scale-1">9.75 scale</a>
<ul class="no-bullet">
<li><a name="toc-Options" href="#Options">9.75.1 Options</a></li>
<li><a name="toc-Examples-67" href="#Examples-67">9.75.2 Examples</a></li>
</ul></li>
<li><a name="toc-separatefields" href="#separatefields">9.76 separatefields</a></li>
<li><a name="toc-setdar_002c-setsar" href="#setdar_002c-setsar">9.77 setdar, setsar</a>
<ul class="no-bullet">
<li><a name="toc-Examples-65" href="#Examples-65">9.77.1 Examples</a></li>
</ul></li>
<li><a name="toc-setfield-1" href="#setfield-1">9.78 setfield</a></li>
<li><a name="toc-showinfo" href="#showinfo">9.79 showinfo</a></li>
<li><a name="toc-shuffleplanes" href="#shuffleplanes">9.80 shuffleplanes</a></li>
<li><a name="toc-signalstats" href="#signalstats">9.81 signalstats</a>
<ul class="no-bullet">
<li><a name="toc-Examples-21" href="#Examples-21">9.81.1 Examples</a></li>
</ul></li>
<li><a name="toc-smartblur-1" href="#smartblur-1">9.82 smartblur</a></li>
<li><a name="toc-stereo3d" href="#stereo3d">9.83 stereo3d</a>
<ul class="no-bullet">
<li><a name="toc-Examples-38" href="#Examples-38">9.83.1 Examples</a></li>
</ul></li>
<li><a name="toc-spp" href="#spp">9.84 spp</a></li>
<li><a name="toc-subtitles-1" href="#subtitles-1">9.85 subtitles</a></li>
<li><a name="toc-super2xsai" href="#super2xsai">9.86 super2xsai</a></li>
<li><a name="toc-swapuv" href="#swapuv">9.87 swapuv</a></li>
<li><a name="toc-telecine" href="#telecine">9.88 telecine</a></li>
<li><a name="toc-thumbnail" href="#thumbnail">9.89 thumbnail</a>
<ul class="no-bullet">
<li><a name="toc-Examples-3" href="#Examples-3">9.89.1 Examples</a></li>
</ul></li>
<li><a name="toc-tile" href="#tile">9.90 tile</a>
<ul class="no-bullet">
<li><a name="toc-Examples-24" href="#Examples-24">9.90.1 Examples</a></li>
</ul></li>
<li><a name="toc-tinterlace" href="#tinterlace">9.91 tinterlace</a></li>
<li><a name="toc-transpose" href="#transpose">9.92 transpose</a></li>
<li><a name="toc-trim" href="#trim">9.93 trim</a></li>
<li><a name="toc-unsharp" href="#unsharp">9.94 unsharp</a>
<ul class="no-bullet">
<li><a name="toc-Examples-23" href="#Examples-23">9.94.1 Examples</a></li>
</ul></li>
<li><a name="toc-vidstabdetect-1" href="#vidstabdetect-1">9.95 vidstabdetect</a>
<ul class="no-bullet">
<li><a name="toc-Examples-25" href="#Examples-25">9.95.1 Examples</a></li>
</ul></li>
<li><a name="toc-vidstabtransform-1" href="#vidstabtransform-1">9.96 vidstabtransform</a>
<ul class="no-bullet">
<li><a name="toc-Options-2" href="#Options-2">9.96.1 Options</a></li>
<li><a name="toc-Examples-54" href="#Examples-54">9.96.2 Examples</a></li>
</ul></li>
<li><a name="toc-vflip" href="#vflip">9.97 vflip</a></li>
<li><a name="toc-vignette-1" href="#vignette-1">9.98 vignette</a>
<ul class="no-bullet">
<li><a name="toc-Expressions" href="#Expressions">9.98.1 Expressions</a></li>
<li><a name="toc-Examples-42" href="#Examples-42">9.98.2 Examples</a></li>
</ul></li>
<li><a name="toc-w3fdif" href="#w3fdif">9.99 w3fdif</a></li>
<li><a name="toc-yadif-1" href="#yadif-1">9.100 yadif</a></li>
<li><a name="toc-zoompan" href="#zoompan">9.101 zoompan</a>
<ul class="no-bullet">
<li><a name="toc-Examples-13" href="#Examples-13">9.101.1 Examples</a></li>
</ul>
</li>
</ul></li>
<li><a name="toc-Video-Sources" href="#Video-Sources">10 Video Sources</a>
<ul class="no-bullet">
<li><a name="toc-buffer" href="#buffer">10.1 buffer</a></li>
<li><a name="toc-cellauto" href="#cellauto">10.2 cellauto</a>
<ul class="no-bullet">
<li><a name="toc-Examples-43" href="#Examples-43">10.2.1 Examples</a></li>
</ul></li>
<li><a name="toc-mandelbrot" href="#mandelbrot">10.3 mandelbrot</a></li>
<li><a name="toc-mptestsrc" href="#mptestsrc">10.4 mptestsrc</a></li>
<li><a name="toc-frei0r_005fsrc" href="#frei0r_005fsrc">10.5 frei0r_src</a></li>
<li><a name="toc-life" href="#life">10.6 life</a>
<ul class="no-bullet">
<li><a name="toc-Examples-7" href="#Examples-7">10.6.1 Examples</a></li>
</ul></li>
<li><a name="toc-color_002c-haldclutsrc_002c-nullsrc_002c-rgbtestsrc_002c-smptebars_002c-smptehdbars_002c-testsrc" href="#color_002c-haldclutsrc_002c-nullsrc_002c-rgbtestsrc_002c-smptebars_002c-smptehdbars_002c-testsrc">10.7 color, haldclutsrc, nullsrc, rgbtestsrc, smptebars, smptehdbars, testsrc</a>
<ul class="no-bullet">
<li><a name="toc-Commands-2" href="#Commands-2">10.7.1 Commands</a></li>
</ul>
</li>
</ul></li>
<li><a name="toc-Video-Sinks" href="#Video-Sinks">11 Video Sinks</a>
<ul class="no-bullet">
<li><a name="toc-buffersink" href="#buffersink">11.1 buffersink</a></li>
<li><a name="toc-nullsink" href="#nullsink">11.2 nullsink</a></li>
</ul></li>
<li><a name="toc-Multimedia-Filters" href="#Multimedia-Filters">12 Multimedia Filters</a>
<ul class="no-bullet">
<li><a name="toc-avectorscope" href="#avectorscope">12.1 avectorscope</a>
<ul class="no-bullet">
<li><a name="toc-Examples-45" href="#Examples-45">12.1.1 Examples</a></li>
</ul></li>
<li><a name="toc-concat" href="#concat">12.2 concat</a>
<ul class="no-bullet">
<li><a name="toc-Examples-31" href="#Examples-31">12.2.1 Examples</a></li>
</ul></li>
<li><a name="toc-ebur128" href="#ebur128">12.3 ebur128</a>
<ul class="no-bullet">
<li><a name="toc-Examples-71" href="#Examples-71">12.3.1 Examples</a></li>
</ul></li>
<li><a name="toc-interleave_002c-ainterleave" href="#interleave_002c-ainterleave">12.4 interleave, ainterleave</a>
<ul class="no-bullet">
<li><a name="toc-Examples-49" href="#Examples-49">12.4.1 Examples</a></li>
</ul></li>
<li><a name="toc-perms_002c-aperms" href="#perms_002c-aperms">12.5 perms, aperms</a></li>
<li><a name="toc-select_002c-aselect" href="#select_002c-aselect">12.6 select, aselect</a>
<ul class="no-bullet">
<li><a name="toc-Examples-1" href="#Examples-1">12.6.1 Examples</a></li>
</ul></li>
<li><a name="toc-sendcmd_002c-asendcmd" href="#sendcmd_002c-asendcmd">12.7 sendcmd, asendcmd</a>
<ul class="no-bullet">
<li><a name="toc-Commands-syntax" href="#Commands-syntax">12.7.1 Commands syntax</a></li>
<li><a name="toc-Examples-63" href="#Examples-63">12.7.2 Examples</a></li>
</ul></li>
<li><a name="toc-setpts_002c-asetpts" href="#setpts_002c-asetpts">12.8 setpts, asetpts</a>
<ul class="no-bullet">
<li><a name="toc-Examples-80" href="#Examples-80">12.8.1 Examples</a></li>
</ul></li>
<li><a name="toc-settb_002c-asettb" href="#settb_002c-asettb">12.9 settb, asettb</a>
<ul class="no-bullet">
<li><a name="toc-Examples-66" href="#Examples-66">12.9.1 Examples</a></li>
</ul></li>
<li><a name="toc-showcqt" href="#showcqt">12.10 showcqt</a>
<ul class="no-bullet">
<li><a name="toc-Examples-4" href="#Examples-4">12.10.1 Examples</a></li>
</ul></li>
<li><a name="toc-showspectrum" href="#showspectrum">12.11 showspectrum</a>
<ul class="no-bullet">
<li><a name="toc-Examples-57" href="#Examples-57">12.11.1 Examples</a></li>
</ul></li>
<li><a name="toc-showwaves" href="#showwaves">12.12 showwaves</a>
<ul class="no-bullet">
<li><a name="toc-Examples-51" href="#Examples-51">12.12.1 Examples</a></li>
</ul></li>
<li><a name="toc-split_002c-asplit" href="#split_002c-asplit">12.13 split, asplit</a>
<ul class="no-bullet">
<li><a name="toc-Examples-9" href="#Examples-9">12.13.1 Examples</a></li>
</ul></li>
<li><a name="toc-zmq_002c-azmq" href="#zmq_002c-azmq">12.14 zmq, azmq</a>
<ul class="no-bullet">
<li><a name="toc-Examples-2" href="#Examples-2">12.14.1 Examples</a></li>
</ul>
</li>
</ul></li>
<li><a name="toc-Multimedia-Sources" href="#Multimedia-Sources">13 Multimedia Sources</a>
<ul class="no-bullet">
<li><a name="toc-amovie" href="#amovie">13.1 amovie</a></li>
<li><a name="toc-movie-1" href="#movie-1">13.2 movie</a>
<ul class="no-bullet">
<li><a name="toc-Examples-64" href="#Examples-64">13.2.1 Examples</a></li>
</ul>
</li>
</ul></li>
<li><a name="toc-See-Also" href="#See-Also">14 See Also</a></li>
<li><a name="toc-Authors" href="#Authors">15 Authors</a></li>
</ul>
</div>
<hr size="6">
<a name="Description"></a>
<h1 class="chapter"><a href="ffmpeg-filters.html#toc-Description">1 Description</a></h1>
<p>This document describes filters, sources, and sinks provided by the
libavfilter library.
</p>
<a name="Filtering-Introduction"></a>
<h1 class="chapter"><a href="ffmpeg-filters.html#toc-Filtering-Introduction">2 Filtering Introduction</a></h1>
<p>Filtering in FFmpeg is enabled through the libavfilter library.
</p>
<p>In libavfilter, a filter can have multiple inputs and multiple
outputs.
To illustrate the sorts of things that are possible, we consider the
following filtergraph.
</p>
<div class="example">
<pre class="example"> [main]
input --> split ---------------------> overlay --> output
| ^
|[tmp] [flip]|
+-----> crop --> vflip -------+
</pre></div>
<p>This filtergraph splits the input stream in two streams, then sends one
stream through the crop filter and the vflip filter, before merging it
back with the other stream by overlaying it on top. You can use the
following command to achieve this:
</p>
<div class="example">
<pre class="example">ffmpeg -i INPUT -vf "split [main][tmp]; [tmp] crop=iw:ih/2:0:0, vflip [flip]; [main][flip] overlay=0:H/2" OUTPUT
</pre></div>
<p>The result will be that the top half of the video is mirrored
onto the bottom half of the output video.
</p>
<p>Filters in the same linear chain are separated by commas, and distinct
linear chains of filters are separated by semicolons. In our example,
<var>crop,vflip</var> are in one linear chain, <var>split</var> and
<var>overlay</var> are separately in another. The points where the linear
chains join are labelled by names enclosed in square brackets. In the
example, the split filter generates two outputs that are associated to
the labels <var>[main]</var> and <var>[tmp]</var>.
</p>
<p>The stream sent to the second output of <var>split</var>, labelled as
<var>[tmp]</var>, is processed through the <var>crop</var> filter, which crops
away the lower half part of the video, and then vertically flipped. The
<var>overlay</var> filter takes in input the first unchanged output of the
split filter (which was labelled as <var>[main]</var>), and overlay on its
lower half the output generated by the <var>crop,vflip</var> filterchain.
</p>
<p>Some filters take in input a list of parameters: they are specified
after the filter name and an equal sign, and are separated from each other
by a colon.
</p>
<p>There exist so-called <var>source filters</var> that do not have an
audio/video input, and <var>sink filters</var> that will not have audio/video
output.
</p>
<a name="graph2dot"></a>
<h1 class="chapter"><a href="ffmpeg-filters.html#toc-graph2dot">3 graph2dot</a></h1>
<p>The ‘<tt>graph2dot</tt>’ program included in the FFmpeg ‘<tt>tools</tt>’
directory can be used to parse a filtergraph description and issue a
corresponding textual representation in the dot language.
</p>
<p>Invoke the command:
</p><div class="example">
<pre class="example">graph2dot -h
</pre></div>
<p>to see how to use ‘<tt>graph2dot</tt>’.
</p>
<p>You can then pass the dot description to the ‘<tt>dot</tt>’ program (from
the graphviz suite of programs) and obtain a graphical representation
of the filtergraph.
</p>
<p>For example the sequence of commands:
</p><div class="example">
<pre class="example">echo <var>GRAPH_DESCRIPTION</var> | \
tools/graph2dot -o graph.tmp && \
dot -Tpng graph.tmp -o graph.png && \
display graph.png
</pre></div>
<p>can be used to create and display an image representing the graph
described by the <var>GRAPH_DESCRIPTION</var> string. Note that this string must be
a complete self-contained graph, with its inputs and outputs explicitly defined.
For example if your command line is of the form:
</p><div class="example">
<pre class="example">ffmpeg -i infile -vf scale=640:360 outfile
</pre></div>
<p>your <var>GRAPH_DESCRIPTION</var> string will need to be of the form:
</p><div class="example">
<pre class="example">nullsrc,scale=640:360,nullsink
</pre></div>
<p>you may also need to set the <var>nullsrc</var> parameters and add a <var>format</var>
filter in order to simulate a specific input file.
</p>
<a name="Filtergraph-description"></a>
<h1 class="chapter"><a href="ffmpeg-filters.html#toc-Filtergraph-description">4 Filtergraph description</a></h1>
<p>A filtergraph is a directed graph of connected filters. It can contain
cycles, and there can be multiple links between a pair of
filters. Each link has one input pad on one side connecting it to one
filter from which it takes its input, and one output pad on the other
side connecting it to one filter accepting its output.
</p>
<p>Each filter in a filtergraph is an instance of a filter class
registered in the application, which defines the features and the
number of input and output pads of the filter.
</p>
<p>A filter with no input pads is called a "source", and a filter with no
output pads is called a "sink".
</p>
<p><a name="Filtergraph-syntax"></a>
</p><a name="Filtergraph-syntax-1"></a>
<h2 class="section"><a href="ffmpeg-filters.html#toc-Filtergraph-syntax-1">4.1 Filtergraph syntax</a></h2>
<p>A filtergraph has a textual representation, which is
recognized by the ‘<samp>-filter</samp>’/‘<samp>-vf</samp>’ and ‘<samp>-filter_complex</samp>’
options in <code>ffmpeg</code> and ‘<samp>-vf</samp>’ in <code>ffplay</code>, and by the
<code>avfilter_graph_parse()</code>/<code>avfilter_graph_parse2()</code> functions defined in
‘<tt>libavfilter/avfilter.h</tt>’.
</p>
<p>A filterchain consists of a sequence of connected filters, each one
connected to the previous one in the sequence. A filterchain is
represented by a list of ","-separated filter descriptions.
</p>
<p>A filtergraph consists of a sequence of filterchains. A sequence of
filterchains is represented by a list of ";"-separated filterchain
descriptions.
</p>
<p>A filter is represented by a string of the form:
[<var>in_link_1</var>]...[<var>in_link_N</var>]<var>filter_name</var>=<var>arguments</var>[<var>out_link_1</var>]...[<var>out_link_M</var>]
</p>
<p><var>filter_name</var> is the name of the filter class of which the
described filter is an instance of, and has to be the name of one of
the filter classes registered in the program.
The name of the filter class is optionally followed by a string
"=<var>arguments</var>".
</p>
<p><var>arguments</var> is a string which contains the parameters used to
initialize the filter instance. It may have one of two forms:
</p><ul>
<li>
A ’:’-separated list of <var>key=value</var> pairs.
</li><li>
A ’:’-separated list of <var>value</var>. In this case, the keys are assumed to be
the option names in the order they are declared. E.g. the <code>fade</code> filter
declares three options in this order – ‘<samp>type</samp>’, ‘<samp>start_frame</samp>’ and
‘<samp>nb_frames</samp>’. Then the parameter list <var>in:0:30</var> means that the value
<var>in</var> is assigned to the option ‘<samp>type</samp>’, <var>0</var> to
‘<samp>start_frame</samp>’ and <var>30</var> to ‘<samp>nb_frames</samp>’.
</li><li>
A ’:’-separated list of mixed direct <var>value</var> and long <var>key=value</var>
pairs. The direct <var>value</var> must precede the <var>key=value</var> pairs, and
follow the same constraints order of the previous point. The following
<var>key=value</var> pairs can be set in any preferred order.
</li></ul>
<p>If the option value itself is a list of items (e.g. the <code>format</code> filter
takes a list of pixel formats), the items in the list are usually separated by
’|’.
</p>
<p>The list of arguments can be quoted using the character "’" as initial
and ending mark, and the character ’\’ for escaping the characters
within the quoted text; otherwise the argument string is considered
terminated when the next special character (belonging to the set
"[]=;,") is encountered.
</p>
<p>The name and arguments of the filter are optionally preceded and
followed by a list of link labels.
A link label allows one to name a link and associate it to a filter output
or input pad. The preceding labels <var>in_link_1</var>
... <var>in_link_N</var>, are associated to the filter input pads,
the following labels <var>out_link_1</var> ... <var>out_link_M</var>, are
associated to the output pads.
</p>
<p>When two link labels with the same name are found in the
filtergraph, a link between the corresponding input and output pad is
created.
</p>
<p>If an output pad is not labelled, it is linked by default to the first
unlabelled input pad of the next filter in the filterchain.
For example in the filterchain
</p><div class="example">
<pre class="example">nullsrc, split[L1], [L2]overlay, nullsink
</pre></div>
<p>the split filter instance has two output pads, and the overlay filter
instance two input pads. The first output pad of split is labelled
"L1", the first input pad of overlay is labelled "L2", and the second
output pad of split is linked to the second input pad of overlay,
which are both unlabelled.
</p>
<p>In a complete filterchain all the unlabelled filter input and output
pads must be connected. A filtergraph is considered valid if all the
filter input and output pads of all the filterchains are connected.
</p>
<p>Libavfilter will automatically insert <a href="#scale">scale</a> filters where format
conversion is required. It is possible to specify swscale flags
for those automatically inserted scalers by prepending
<code>sws_flags=<var>flags</var>;</code>
to the filtergraph description.
</p>
<p>Here is a BNF description of the filtergraph syntax:
</p><div class="example">
<pre class="example"><var>NAME</var> ::= sequence of alphanumeric characters and '_'
<var>LINKLABEL</var> ::= "[" <var>NAME</var> "]"
<var>LINKLABELS</var> ::= <var>LINKLABEL</var> [<var>LINKLABELS</var>]
<var>FILTER_ARGUMENTS</var> ::= sequence of chars (possibly quoted)
<var>FILTER</var> ::= [<var>LINKLABELS</var>] <var>NAME</var> ["=" <var>FILTER_ARGUMENTS</var>] [<var>LINKLABELS</var>]
<var>FILTERCHAIN</var> ::= <var>FILTER</var> [,<var>FILTERCHAIN</var>]
<var>FILTERGRAPH</var> ::= [sws_flags=<var>flags</var>;] <var>FILTERCHAIN</var> [;<var>FILTERGRAPH</var>]
</pre></div>
<a name="Notes-on-filtergraph-escaping"></a>
<h2 class="section"><a href="ffmpeg-filters.html#toc-Notes-on-filtergraph-escaping">4.2 Notes on filtergraph escaping</a></h2>
<p>Filtergraph description composition entails several levels of
escaping. See <a href="ffmpeg-utils.html#quoting_005fand_005fescaping">(ffmpeg-utils)the "Quoting and escaping" section in the ffmpeg-utils(1) manual</a> for more
information about the employed escaping procedure.
</p>
<p>A first level escaping affects the content of each filter option
value, which may contain the special character <code>:</code> used to
separate values, or one of the escaping characters <code>\'</code>.
</p>
<p>A second level escaping affects the whole filter description, which
may contain the escaping characters <code>\'</code> or the special
characters <code>[],;</code> used by the filtergraph description.
</p>
<p>Finally, when you specify a filtergraph on a shell commandline, you
need to perform a third level escaping for the shell special
characters contained within it.
</p>
<p>For example, consider the following string to be embedded in
the <a href="#drawtext">drawtext</a> filter description ‘<samp>text</samp>’ value:
</p><div class="example">
<pre class="example">this is a 'string': may contain one, or more, special characters
</pre></div>
<p>This string contains the <code>'</code> special escaping character, and the
<code>:</code> special character, so it needs to be escaped in this way:
</p><div class="example">
<pre class="example">text=this is a \'string\'\: may contain one, or more, special characters
</pre></div>
<p>A second level of escaping is required when embedding the filter
description in a filtergraph description, in order to escape all the
filtergraph special characters. Thus the example above becomes:
</p><div class="example">
<pre class="example">drawtext=text=this is a \\\'string\\\'\\: may contain one\, or more\, special characters
</pre></div>
<p>(note that in addition to the <code>\'</code> escaping special characters,
also <code>,</code> needs to be escaped).
</p>
<p>Finally an additional level of escaping is needed when writing the
filtergraph description in a shell command, which depends on the
escaping rules of the adopted shell. For example, assuming that
<code>\</code> is special and needs to be escaped with another <code>\</code>, the
previous string will finally result in:
</p><div class="example">
<pre class="example">-vf "drawtext=text=this is a \\\\\\'string\\\\\\'\\\\: may contain one\\, or more\\, special characters"
</pre></div>
<a name="Timeline-editing"></a>
<h1 class="chapter"><a href="ffmpeg-filters.html#toc-Timeline-editing">5 Timeline editing</a></h1>
<p>Some filters support a generic ‘<samp>enable</samp>’ option. For the filters
supporting timeline editing, this option can be set to an expression which is
evaluated before sending a frame to the filter. If the evaluation is non-zero,
the filter will be enabled, otherwise the frame will be sent unchanged to the
next filter in the filtergraph.
</p>
<p>The expression accepts the following values:
</p><dl compact="compact">
<dt>‘<samp>t</samp>’</dt>
<dd><p>timestamp expressed in seconds, NAN if the input timestamp is unknown
</p>
</dd>
<dt>‘<samp>n</samp>’</dt>
<dd><p>sequential number of the input frame, starting from 0
</p>
</dd>
<dt>‘<samp>pos</samp>’</dt>
<dd><p>the position in the file of the input frame, NAN if unknown
</p></dd>
</dl>
<p>Additionally, these filters support an ‘<samp>enable</samp>’ command that can be used
to re-define the expression.
</p>
<p>Like any other filtering option, the ‘<samp>enable</samp>’ option follows the same
rules.
</p>
<p>For example, to enable a blur filter (<a href="#smartblur">smartblur</a>) from 10 seconds to 3
minutes, and a <a href="#curves">curves</a> filter starting at 3 seconds:
</p><div class="example">
<pre class="example">smartblur = enable='between(t,10,3*60)',
curves = enable='gte(t,3)' : preset=cross_process
</pre></div>
<a name="Audio-Filters"></a>
<h1 class="chapter"><a href="ffmpeg-filters.html#toc-Audio-Filters">6 Audio Filters</a></h1>
<p>When you configure your FFmpeg build, you can disable any of the
existing filters using <code>--disable-filters</code>.
The configure output will show the audio filters included in your
build.
</p>
<p>Below is a description of the currently available audio filters.
</p>
<a name="aconvert"></a>
<h2 class="section"><a href="ffmpeg-filters.html#toc-aconvert">6.1 aconvert</a></h2>
<p>Convert the input audio format to the specified formats.
</p>
<p><em>This filter is deprecated. Use <a href="#aformat">aformat</a> instead.</em>
</p>
<p>The filter accepts a string of the form:
"<var>sample_format</var>:<var>channel_layout</var>".
</p>
<p><var>sample_format</var> specifies the sample format, and can be a string or the
corresponding numeric value defined in ‘<tt>libavutil/samplefmt.h</tt>’. Use ’p’
suffix for a planar sample format.
</p>
<p><var>channel_layout</var> specifies the channel layout, and can be a string
or the corresponding number value defined in ‘<tt>libavutil/channel_layout.h</tt>’.
</p>
<p>The special parameter "auto", signifies that the filter will
automatically select the output format depending on the output filter.
</p>
<a name="Examples-61"></a>
<h3 class="subsection"><a href="ffmpeg-filters.html#toc-Examples-61">6.1.1 Examples</a></h3>
<ul>
<li>
Convert input to float, planar, stereo:
<div class="example">
<pre class="example">aconvert=fltp:stereo
</pre></div>
</li><li>
Convert input to unsigned 8-bit, automatically select out channel layout:
<div class="example">
<pre class="example">aconvert=u8:auto
</pre></div>
</li></ul>
<a name="adelay"></a>
<h2 class="section"><a href="ffmpeg-filters.html#toc-adelay">6.2 adelay</a></h2>
<p>Delay one or more audio channels.
</p>
<p>Samples in delayed channel are filled with silence.
</p>
<p>The filter accepts the following option:
</p>
<dl compact="compact">
<dt>‘<samp>delays</samp>’</dt>
<dd><p>Set list of delays in milliseconds for each channel separated by ’|’.
At least one delay greater than 0 should be provided.
Unused delays will be silently ignored. If number of given delays is
smaller than number of channels all remaining channels will not be delayed.
</p></dd>
</dl>
<a name="Examples"></a>
<h3 class="subsection"><a href="ffmpeg-filters.html#toc-Examples">6.2.1 Examples</a></h3>
<ul>
<li>
Delay first channel by 1.5 seconds, the third channel by 0.5 seconds and leave
the second channel (and any other channels that may be present) unchanged.
<div class="example">
<pre class="example">adelay=1500|0|500
</pre></div>
</li></ul>
<a name="aecho"></a>
<h2 class="section"><a href="ffmpeg-filters.html#toc-aecho">6.3 aecho</a></h2>
<p>Apply echoing to the input audio.
</p>
<p>Echoes are reflected sound and can occur naturally amongst mountains
(and sometimes large buildings) when talking or shouting; digital echo
effects emulate this behaviour and are often used to help fill out the
sound of a single instrument or vocal. The time difference between the
original signal and the reflection is the <code>delay</code>, and the
loudness of the reflected signal is the <code>decay</code>.
Multiple echoes can have different delays and decays.
</p>
<p>A description of the accepted parameters follows.
</p>
<dl compact="compact">
<dt>‘<samp>in_gain</samp>’</dt>
<dd><p>Set input gain of reflected signal. Default is <code>0.6</code>.
</p>
</dd>
<dt>‘<samp>out_gain</samp>’</dt>
<dd><p>Set output gain of reflected signal. Default is <code>0.3</code>.
</p>
</dd>
<dt>‘<samp>delays</samp>’</dt>
<dd><p>Set list of time intervals in milliseconds between original signal and reflections
separated by ’|’. Allowed range for each <code>delay</code> is <code>(0 - 90000.0]</code>.
Default is <code>1000</code>.
</p>
</dd>
<dt>‘<samp>decays</samp>’</dt>
<dd><p>Set list of loudnesses of reflected signals separated by ’|’.
Allowed range for each <code>decay</code> is <code>(0 - 1.0]</code>.
Default is <code>0.5</code>.
</p></dd>
</dl>
<a name="Examples-5"></a>
<h3 class="subsection"><a href="ffmpeg-filters.html#toc-Examples-5">6.3.1 Examples</a></h3>
<ul>
<li>
Make it sound as if there are twice as many instruments as are actually playing:
<div class="example">
<pre class="example">aecho=0.8:0.88:60:0.4
</pre></div>
</li><li>
If delay is very short, then it sound like a (metallic) robot playing music:
<div class="example">
<pre class="example">aecho=0.8:0.88:6:0.4
</pre></div>
</li><li>
A longer delay will sound like an open air concert in the mountains:
<div class="example">
<pre class="example">aecho=0.8:0.9:1000:0.3
</pre></div>
</li><li>
Same as above but with one more mountain:
<div class="example">
<pre class="example">aecho=0.8:0.9:1000|1800:0.3|0.25
</pre></div>
</li></ul>
<a name="aeval"></a>
<h2 class="section"><a href="ffmpeg-filters.html#toc-aeval">6.4 aeval</a></h2>
<p>Modify an audio signal according to the specified expressions.
</p>
<p>This filter accepts one or more expressions (one for each channel),
which are evaluated and used to modify a corresponding audio signal.
</p>
<p>It accepts the following parameters:
</p>
<dl compact="compact">
<dt>‘<samp>exprs</samp>’</dt>
<dd><p>Set the ’|’-separated expressions list for each separate channel. If
the number of input channels is greater than the number of
expressions, the last specified expression is used for the remaining
output channels.
</p>
</dd>
<dt>‘<samp>channel_layout, c</samp>’</dt>
<dd><p>Set output channel layout. If not specified, the channel layout is
specified by the number of expressions. If set to ‘<samp>same</samp>’, it will
use by default the same input channel layout.
</p></dd>
</dl>
<p>Each expression in <var>exprs</var> can contain the following constants and functions:
</p>
<dl compact="compact">
<dt>‘<samp>ch</samp>’</dt>
<dd><p>channel number of the current expression
</p>
</dd>
<dt>‘<samp>n</samp>’</dt>
<dd><p>number of the evaluated sample, starting from 0
</p>
</dd>
<dt>‘<samp>s</samp>’</dt>
<dd><p>sample rate
</p>
</dd>
<dt>‘<samp>t</samp>’</dt>
<dd><p>time of the evaluated sample expressed in seconds
</p>
</dd>
<dt>‘<samp>nb_in_channels</samp>’</dt>
<dt>‘<samp>nb_out_channels</samp>’</dt>
<dd><p>input and output number of channels
</p>
</dd>
<dt>‘<samp>val(CH)</samp>’</dt>
<dd><p>the value of input channel with number <var>CH</var>
</p></dd>
</dl>
<p>Note: this filter is slow. For faster processing you should use a
dedicated filter.
</p>
<a name="Examples-15"></a>
<h3 class="subsection"><a href="ffmpeg-filters.html#toc-Examples-15">6.4.1 Examples</a></h3>
<ul>
<li>
Half volume:
<div class="example">
<pre class="example">aeval=val(ch)/2:c=same
</pre></div>
</li><li>
Invert phase of the second channel:
<div class="example">
<pre class="example">aeval=val(0)|-val(1)
</pre></div>
</li></ul>
<a name="afade"></a>
<h2 class="section"><a href="ffmpeg-filters.html#toc-afade">6.5 afade</a></h2>
<p>Apply fade-in/out effect to input audio.
</p>
<p>A description of the accepted parameters follows.
</p>
<dl compact="compact">
<dt>‘<samp>type, t</samp>’</dt>
<dd><p>Specify the effect type, can be either <code>in</code> for fade-in, or
<code>out</code> for a fade-out effect. Default is <code>in</code>.
</p>
</dd>
<dt>‘<samp>start_sample, ss</samp>’</dt>
<dd><p>Specify the number of the start sample for starting to apply the fade
effect. Default is 0.
</p>
</dd>
<dt>‘<samp>nb_samples, ns</samp>’</dt>
<dd><p>Specify the number of samples for which the fade effect has to last. At
the end of the fade-in effect the output audio will have the same
volume as the input audio, at the end of the fade-out transition
the output audio will be silence. Default is 44100.
</p>
</dd>
<dt>‘<samp>start_time, st</samp>’</dt>
<dd><p>Specify the start time of the fade effect. Default is 0.
The value must be specified as a time duration; see
<a href="ffmpeg-utils.html#time-duration-syntax">(ffmpeg-utils)the Time duration section in the ffmpeg-utils(1) manual</a>
for the accepted syntax.
If set this option is used instead of <var>start_sample</var>.
</p>
</dd>
<dt>‘<samp>duration, d</samp>’</dt>
<dd><p>Specify the duration of the fade effect. See
<a href="ffmpeg-utils.html#time-duration-syntax">(ffmpeg-utils)the Time duration section in the ffmpeg-utils(1) manual</a>
for the accepted syntax.
At the end of the fade-in effect the output audio will have the same
volume as the input audio, at the end of the fade-out transition
the output audio will be silence.
By default the duration is determined by <var>nb_samples</var>.
If set this option is used instead of <var>nb_samples</var>.
</p>
</dd>
<dt>‘<samp>curve</samp>’</dt>
<dd><p>Set curve for fade transition.
</p>
<p>It accepts the following values:
</p><dl compact="compact">
<dt>‘<samp>tri</samp>’</dt>
<dd><p>select triangular, linear slope (default)
</p></dd>
<dt>‘<samp>qsin</samp>’</dt>
<dd><p>select quarter of sine wave
</p></dd>
<dt>‘<samp>hsin</samp>’</dt>
<dd><p>select half of sine wave
</p></dd>
<dt>‘<samp>esin</samp>’</dt>
<dd><p>select exponential sine wave
</p></dd>
<dt>‘<samp>log</samp>’</dt>
<dd><p>select logarithmic
</p></dd>
<dt>‘<samp>par</samp>’</dt>
<dd><p>select inverted parabola
</p></dd>
<dt>‘<samp>qua</samp>’</dt>
<dd><p>select quadratic
</p></dd>
<dt>‘<samp>cub</samp>’</dt>
<dd><p>select cubic
</p></dd>
<dt>‘<samp>squ</samp>’</dt>
<dd><p>select square root
</p></dd>
<dt>‘<samp>cbr</samp>’</dt>
<dd><p>select cubic root
</p></dd>
</dl>
</dd>
</dl>
<a name="Examples-70"></a>
<h3 class="subsection"><a href="ffmpeg-filters.html#toc-Examples-70">6.5.1 Examples</a></h3>
<ul>
<li>
Fade in first 15 seconds of audio:
<div class="example">
<pre class="example">afade=t=in:ss=0:d=15
</pre></div>
</li><li>
Fade out last 25 seconds of a 900 seconds audio:
<div class="example">
<pre class="example">afade=t=out:st=875:d=25
</pre></div>
</li></ul>
<p><a name="aformat"></a>
</p><a name="aformat-1"></a>
<h2 class="section"><a href="ffmpeg-filters.html#toc-aformat-1">6.6 aformat</a></h2>
<p>Set output format constraints for the input audio. The framework will
negotiate the most appropriate format to minimize conversions.
</p>
<p>It accepts the following parameters:
</p><dl compact="compact">
<dt>‘<samp>sample_fmts</samp>’</dt>
<dd><p>A ’|’-separated list of requested sample formats.
</p>
</dd>
<dt>‘<samp>sample_rates</samp>’</dt>
<dd><p>A ’|’-separated list of requested sample rates.
</p>
</dd>
<dt>‘<samp>channel_layouts</samp>’</dt>
<dd><p>A ’|’-separated list of requested channel layouts.
</p>
<p>See <a href="ffmpeg-utils.html#channel-layout-syntax">(ffmpeg-utils)the Channel Layout section in the ffmpeg-utils(1) manual</a>
for the required syntax.
</p></dd>
</dl>
<p>If a parameter is omitted, all values are allowed.
</p>
<p>Force the output to either unsigned 8-bit or signed 16-bit stereo
</p><div class="example">
<pre class="example">aformat=sample_fmts=u8|s16:channel_layouts=stereo
</pre></div>
<a name="allpass"></a>
<h2 class="section"><a href="ffmpeg-filters.html#toc-allpass">6.7 allpass</a></h2>
<p>Apply a two-pole all-pass filter with central frequency (in Hz)
<var>frequency</var>, and filter-width <var>width</var>.
An all-pass filter changes the audio’s frequency to phase relationship
without changing its frequency to amplitude relationship.
</p>
<p>The filter accepts the following options:
</p>
<dl compact="compact">
<dt>‘<samp>frequency, f</samp>’</dt>
<dd><p>Set frequency in Hz.
</p>
</dd>
<dt>‘<samp>width_type</samp>’</dt>
<dd><p>Set method to specify band-width of filter.
</p><dl compact="compact">
<dt>‘<samp>h</samp>’</dt>
<dd><p>Hz
</p></dd>
<dt>‘<samp>q</samp>’</dt>
<dd><p>Q-Factor
</p></dd>
<dt>‘<samp>o</samp>’</dt>
<dd><p>octave
</p></dd>
<dt>‘<samp>s</samp>’</dt>
<dd><p>slope
</p></dd>
</dl>
</dd>
<dt>‘<samp>width, w</samp>’</dt>
<dd><p>Specify the band-width of a filter in width_type units.
</p></dd>
</dl>
<a name="amerge"></a>
<h2 class="section"><a href="ffmpeg-filters.html#toc-amerge">6.8 amerge</a></h2>
<p>Merge two or more audio streams into a single multi-channel stream.
</p>
<p>The filter accepts the following options:
</p>
<dl compact="compact">
<dt>‘<samp>inputs</samp>’</dt>
<dd><p>Set the number of inputs. Default is 2.
</p>
</dd>
</dl>
<p>If the channel layouts of the inputs are disjoint, and therefore compatible,
the channel layout of the output will be set accordingly and the channels
will be reordered as necessary. If the channel layouts of the inputs are not
disjoint, the output will have all the channels of the first input then all
the channels of the second input, in that order, and the channel layout of
the output will be the default value corresponding to the total number of
channels.
</p>
<p>For example, if the first input is in 2.1 (FL+FR+LF) and the second input
is FC+BL+BR, then the output will be in 5.1, with the channels in the
following order: a1, a2, b1, a3, b2, b3 (a1 is the first channel of the
first input, b1 is the first channel of the second input).
</p>
<p>On the other hand, if both input are in stereo, the output channels will be
in the default order: a1, a2, b1, b2, and the channel layout will be
arbitrarily set to 4.0, which may or may not be the expected value.
</p>
<p>All inputs must have the same sample rate, and format.
</p>
<p>If inputs do not have the same duration, the output will stop with the
shortest.
</p>
<a name="Examples-40"></a>
<h3 class="subsection"><a href="ffmpeg-filters.html#toc-Examples-40">6.8.1 Examples</a></h3>
<ul>
<li>
Merge two mono files into a stereo stream:
<div class="example">
<pre class="example">amovie=left.wav [l] ; amovie=right.mp3 [r] ; [l] [r] amerge
</pre></div>
</li><li>
Multiple merges assuming 1 video stream and 6 audio streams in ‘<tt>input.mkv</tt>’:
<div class="example">
<pre class="example">ffmpeg -i input.mkv -filter_complex "[0:1][0:2][0:3][0:4][0:5][0:6] amerge=inputs=6" -c:a pcm_s16le output.mkv
</pre></div>
</li></ul>
<a name="amix"></a>
<h2 class="section"><a href="ffmpeg-filters.html#toc-amix">6.9 amix</a></h2>
<p>Mixes multiple audio inputs into a single output.
</p>
<p>Note that this filter only supports float samples (the <var>amerge</var>
and <var>pan</var> audio filters support many formats). If the <var>amix</var>
input has integer samples then <a href="#aresample">aresample</a> will be automatically
inserted to perform the conversion to float samples.
</p>
<p>For example
</p><div class="example">
<pre class="example">ffmpeg -i INPUT1 -i INPUT2 -i INPUT3 -filter_complex amix=inputs=3:duration=first:dropout_transition=3 OUTPUT
</pre></div>
<p>will mix 3 input audio streams to a single output with the same duration as the
first input and a dropout transition time of 3 seconds.
</p>
<p>It accepts the following parameters:
</p><dl compact="compact">
<dt>‘<samp>inputs</samp>’</dt>
<dd><p>The number of inputs. If unspecified, it defaults to 2.
</p>
</dd>
<dt>‘<samp>duration</samp>’</dt>
<dd><p>How to determine the end-of-stream.
</p><dl compact="compact">
<dt>‘<samp>longest</samp>’</dt>
<dd><p>The duration of the longest input. (default)
</p>
</dd>
<dt>‘<samp>shortest</samp>’</dt>
<dd><p>The duration of the shortest input.
</p>
</dd>
<dt>‘<samp>first</samp>’</dt>
<dd><p>The duration of the first input.
</p>
</dd>
</dl>
</dd>
<dt>‘<samp>dropout_transition</samp>’</dt>
<dd><p>The transition time, in seconds, for volume renormalization when an input
stream ends. The default value is 2 seconds.
</p>
</dd>
</dl>
<a name="anull"></a>
<h2 class="section"><a href="ffmpeg-filters.html#toc-anull">6.10 anull</a></h2>
<p>Pass the audio source unchanged to the output.
</p>
<a name="apad"></a>
<h2 class="section"><a href="ffmpeg-filters.html#toc-apad">6.11 apad</a></h2>
<p>Pad the end of an audio stream with silence.
</p>
<p>This can be used together with <code>ffmpeg</code> ‘<samp>-shortest</samp>’ to
extend audio streams to the same length as the video stream.
</p>
<p>A description of the accepted options follows.
</p>
<dl compact="compact">
<dt>‘<samp>packet_size</samp>’</dt>
<dd><p>Set silence packet size. Default value is 4096.
</p>
</dd>
<dt>‘<samp>pad_len</samp>’</dt>
<dd><p>Set the number of samples of silence to add to the end. After the
value is reached, the stream is terminated. This option is mutually
exclusive with ‘<samp>whole_len</samp>’.
</p>
</dd>
<dt>‘<samp>whole_len</samp>’</dt>
<dd><p>Set the minimum total number of samples in the output audio stream. If
the value is longer than the input audio length, silence is added to
the end, until the value is reached. This option is mutually exclusive
with ‘<samp>pad_len</samp>’.
</p></dd>
</dl>
<p>If neither the ‘<samp>pad_len</samp>’ nor the ‘<samp>whole_len</samp>’ option is
set, the filter will add silence to the end of the input stream
indefinitely.
</p>
<a name="Examples-47"></a>
<h3 class="subsection"><a href="ffmpeg-filters.html#toc-Examples-47">6.11.1 Examples</a></h3>
<ul>
<li>
Add 1024 samples of silence to the end of the input:
<div class="example">
<pre class="example">apad=pad_len=1024
</pre></div>
</li><li>
Make sure the audio output will contain at least 10000 samples, pad
the input with silence if required:
<div class="example">
<pre class="example">apad=whole_len=10000
</pre></div>
</li><li>
Use <code>ffmpeg</code> to pad the audio input with silence, so that the
video stream will always result the shortest and will be converted
until the end in the output file when using the ‘<samp>shortest</samp>’
option:
<div class="example">
<pre class="example">ffmpeg -i VIDEO -i AUDIO -filter_complex "[1:0]apad" -shortest OUTPUT
</pre></div>
</li></ul>
<a name="aphaser"></a>
<h2 class="section"><a href="ffmpeg-filters.html#toc-aphaser">6.12 aphaser</a></h2>
<p>Add a phasing effect to the input audio.
</p>
<p>A phaser filter creates series of peaks and troughs in the frequency spectrum.
The position of the peaks and troughs are modulated so that they vary over time, creating a sweeping effect.
</p>
<p>A description of the accepted parameters follows.
</p>
<dl compact="compact">
<dt>‘<samp>in_gain</samp>’</dt>
<dd><p>Set input gain. Default is 0.4.
</p>
</dd>
<dt>‘<samp>out_gain</samp>’</dt>
<dd><p>Set output gain. Default is 0.74
</p>
</dd>
<dt>‘<samp>delay</samp>’</dt>
<dd><p>Set delay in milliseconds. Default is 3.0.
</p>
</dd>
<dt>‘<samp>decay</samp>’</dt>
<dd><p>Set decay. Default is 0.4.
</p>
</dd>
<dt>‘<samp>speed</samp>’</dt>
<dd><p>Set modulation speed in Hz. Default is 0.5.
</p>
</dd>
<dt>‘<samp>type</samp>’</dt>
<dd><p>Set modulation type. Default is triangular.
</p>
<p>It accepts the following values:
</p><dl compact="compact">
<dt>‘<samp>triangular, t</samp>’</dt>
<dt>‘<samp>sinusoidal, s</samp>’</dt>
</dl>
</dd>
</dl>
<p><a name="aresample"></a>
</p><a name="aresample-1"></a>
<h2 class="section"><a href="ffmpeg-filters.html#toc-aresample-1">6.13 aresample</a></h2>
<p>Resample the input audio to the specified parameters, using the
libswresample library. If none are specified then the filter will
automatically convert between its input and output.
</p>
<p>This filter is also able to stretch/squeeze the audio data to make it match
the timestamps or to inject silence / cut out audio to make it match the
timestamps, do a combination of both or do neither.
</p>
<p>The filter accepts the syntax
[<var>sample_rate</var>:]<var>resampler_options</var>, where <var>sample_rate</var>
expresses a sample rate and <var>resampler_options</var> is a list of
<var>key</var>=<var>value</var> pairs, separated by ":". See the
ffmpeg-resampler manual for the complete list of supported options.
</p>
<a name="Examples-28"></a>
<h3 class="subsection"><a href="ffmpeg-filters.html#toc-Examples-28">6.13.1 Examples</a></h3>
<ul>
<li>
Resample the input audio to 44100Hz:
<div class="example">
<pre class="example">aresample=44100
</pre></div>
</li><li>
Stretch/squeeze samples to the given timestamps, with a maximum of 1000
samples per second compensation:
<div class="example">
<pre class="example">aresample=async=1000
</pre></div>
</li></ul>
<a name="asetnsamples"></a>
<h2 class="section"><a href="ffmpeg-filters.html#toc-asetnsamples">6.14 asetnsamples</a></h2>
<p>Set the number of samples per each output audio frame.
</p>
<p>The last output packet may contain a different number of samples, as
the filter will flush all the remaining samples when the input audio
signal its end.
</p>
<p>The filter accepts the following options:
</p>
<dl compact="compact">
<dt>‘<samp>nb_out_samples, n</samp>’</dt>
<dd><p>Set the number of frames per each output audio frame. The number is
intended as the number of samples <em>per each channel</em>.
Default value is 1024.
</p>
</dd>
<dt>‘<samp>pad, p</samp>’</dt>
<dd><p>If set to 1, the filter will pad the last audio frame with zeroes, so
that the last frame will contain the same number of samples as the
previous ones. Default value is 1.
</p></dd>
</dl>
<p>For example, to set the number of per-frame samples to 1234 and
disable padding for the last frame, use:
</p><div class="example">
<pre class="example">asetnsamples=n=1234:p=0
</pre></div>
<a name="asetrate"></a>
<h2 class="section"><a href="ffmpeg-filters.html#toc-asetrate">6.15 asetrate</a></h2>
<p>Set the sample rate without altering the PCM data.
This will result in a change of speed and pitch.
</p>
<p>The filter accepts the following options:
</p>
<dl compact="compact">
<dt>‘<samp>sample_rate, r</samp>’</dt>
<dd><p>Set the output sample rate. Default is 44100 Hz.
</p></dd>
</dl>
<a name="ashowinfo"></a>
<h2 class="section"><a href="ffmpeg-filters.html#toc-ashowinfo">6.16 ashowinfo</a></h2>
<p>Show a line containing various information for each input audio frame.
The input audio is not modified.
</p>
<p>The shown line contains a sequence of key/value pairs of the form
<var>key</var>:<var>value</var>.
</p>
<p>The following values are shown in the output:
</p>
<dl compact="compact">
<dt>‘<samp>n</samp>’</dt>
<dd><p>The (sequential) number of the input frame, starting from 0.
</p>
</dd>
<dt>‘<samp>pts</samp>’</dt>
<dd><p>The presentation timestamp of the input frame, in time base units; the time base
depends on the filter input pad, and is usually 1/<var>sample_rate</var>.
</p>
</dd>
<dt>‘<samp>pts_time</samp>’</dt>
<dd><p>The presentation timestamp of the input frame in seconds.
</p>
</dd>
<dt>‘<samp>pos</samp>’</dt>
<dd><p>position of the frame in the input stream, -1 if this information in
unavailable and/or meaningless (for example in case of synthetic audio)
</p>
</dd>
<dt>‘<samp>fmt</samp>’</dt>
<dd><p>The sample format.
</p>
</dd>
<dt>‘<samp>chlayout</samp>’</dt>
<dd><p>The channel layout.
</p>
</dd>
<dt>‘<samp>rate</samp>’</dt>
<dd><p>The sample rate for the audio frame.
</p>
</dd>
<dt>‘<samp>nb_samples</samp>’</dt>
<dd><p>The number of samples (per channel) in the frame.
</p>
</dd>
<dt>‘<samp>checksum</samp>’</dt>
<dd><p>The Adler-32 checksum (printed in hexadecimal) of the audio data. For planar
audio, the data is treated as if all the planes were concatenated.
</p>
</dd>
<dt>‘<samp>plane_checksums</samp>’</dt>
<dd><p>A list of Adler-32 checksums for each data plane.
</p></dd>
</dl>
<a name="astats"></a>
<h2 class="section"><a href="ffmpeg-filters.html#toc-astats">6.17 astats</a></h2>
<p>Display time domain statistical information about the audio channels.
Statistics are calculated and displayed for each audio channel and,
where applicable, an overall figure is also given.
</p>
<p>It accepts the following option:
</p><dl compact="compact">
<dt>‘<samp>length</samp>’</dt>
<dd><p>Short window length in seconds, used for peak and trough RMS measurement.
Default is <code>0.05</code> (50 miliseconds). Allowed range is <code>[0.1 - 10]</code>.
</p></dd>
</dl>
<p>A description of each shown parameter follows:
</p>
<dl compact="compact">
<dt>‘<samp>DC offset</samp>’</dt>
<dd><p>Mean amplitude displacement from zero.
</p>
</dd>
<dt>‘<samp>Min level</samp>’</dt>
<dd><p>Minimal sample level.
</p>
</dd>
<dt>‘<samp>Max level</samp>’</dt>
<dd><p>Maximal sample level.
</p>
</dd>
<dt>‘<samp>Peak level dB</samp>’</dt>
<dt>‘<samp>RMS level dB</samp>’</dt>
<dd><p>Standard peak and RMS level measured in dBFS.
</p>
</dd>
<dt>‘<samp>RMS peak dB</samp>’</dt>
<dt>‘<samp>RMS trough dB</samp>’</dt>
<dd><p>Peak and trough values for RMS level measured over a short window.
</p>
</dd>
<dt>‘<samp>Crest factor</samp>’</dt>
<dd><p>Standard ratio of peak to RMS level (note: not in dB).
</p>
</dd>
<dt>‘<samp>Flat factor</samp>’</dt>
<dd><p>Flatness (i.e. consecutive samples with the same value) of the signal at its peak levels
(i.e. either <var>Min level</var> or <var>Max level</var>).
</p>
</dd>
<dt>‘<samp>Peak count</samp>’</dt>
<dd><p>Number of occasions (not the number of samples) that the signal attained either
<var>Min level</var> or <var>Max level</var>.
</p></dd>
</dl>
<a name="astreamsync"></a>
<h2 class="section"><a href="ffmpeg-filters.html#toc-astreamsync">6.18 astreamsync</a></h2>
<p>Forward two audio streams and control the order the buffers are forwarded.
</p>
<p>The filter accepts the following options:
</p>
<dl compact="compact">
<dt>‘<samp>expr, e</samp>’</dt>
<dd><p>Set the expression deciding which stream should be
forwarded next: if the result is negative, the first stream is forwarded; if
the result is positive or zero, the second stream is forwarded. It can use
the following variables:
</p>
<dl compact="compact">
<dt><var>b1 b2</var></dt>
<dd><p>number of buffers forwarded so far on each stream
</p></dd>
<dt><var>s1 s2</var></dt>
<dd><p>number of samples forwarded so far on each stream
</p></dd>
<dt><var>t1 t2</var></dt>
<dd><p>current timestamp of each stream
</p></dd>
</dl>
<p>The default value is <code>t1-t2</code>, which means to always forward the stream
that has a smaller timestamp.
</p></dd>
</dl>
<a name="Examples-27"></a>
<h3 class="subsection"><a href="ffmpeg-filters.html#toc-Examples-27">6.18.1 Examples</a></h3>
<p>Stress-test <code>amerge</code> by randomly sending buffers on the wrong
input, while avoiding too much of a desynchronization:
</p><div class="example">
<pre class="example">amovie=file.ogg [a] ; amovie=file.mp3 [b] ;
[a] [b] astreamsync=(2*random(1))-1+tanh(5*(t1-t2)) [a2] [b2] ;
[a2] [b2] amerge
</pre></div>
<a name="asyncts"></a>
<h2 class="section"><a href="ffmpeg-filters.html#toc-asyncts">6.19 asyncts</a></h2>
<p>Synchronize audio data with timestamps by squeezing/stretching it and/or
dropping samples/adding silence when needed.
</p>
<p>This filter is not built by default, please use <a href="#aresample">aresample</a> to do squeezing/stretching.
</p>
<p>It accepts the following parameters:
</p><dl compact="compact">
<dt>‘<samp>compensate</samp>’</dt>
<dd><p>Enable stretching/squeezing the data to make it match the timestamps. Disabled
by default. When disabled, time gaps are covered with silence.
</p>
</dd>
<dt>‘<samp>min_delta</samp>’</dt>
<dd><p>The minimum difference between timestamps and audio data (in seconds) to trigger
adding/dropping samples. The default value is 0.1. If you get an imperfect
sync with this filter, try setting this parameter to 0.
</p>
</dd>
<dt>‘<samp>max_comp</samp>’</dt>
<dd><p>The maximum compensation in samples per second. Only relevant with compensate=1.
The default value is 500.
</p>
</dd>
<dt>‘<samp>first_pts</samp>’</dt>
<dd><p>Assume that the first PTS should be this value. The time base is 1 / sample
rate. This allows for padding/trimming at the start of the stream. By default,
no assumption is made about the first frame’s expected PTS, so no padding or
trimming is done. For example, this could be set to 0 to pad the beginning with
silence if an audio stream starts after the video stream or to trim any samples
with a negative PTS due to encoder delay.
</p>
</dd>
</dl>
<a name="atempo"></a>
<h2 class="section"><a href="ffmpeg-filters.html#toc-atempo">6.20 atempo</a></h2>
<p>Adjust audio tempo.
</p>
<p>The filter accepts exactly one parameter, the audio tempo. If not
specified then the filter will assume nominal 1.0 tempo. Tempo must
be in the [0.5, 2.0] range.
</p>
<a name="Examples-48"></a>
<h3 class="subsection"><a href="ffmpeg-filters.html#toc-Examples-48">6.20.1 Examples</a></h3>
<ul>
<li>
Slow down audio to 80% tempo:
<div class="example">
<pre class="example">atempo=0.8
</pre></div>
</li><li>
To speed up audio to 125% tempo:
<div class="example">
<pre class="example">atempo=1.25
</pre></div>
</li></ul>
<a name="atrim"></a>
<h2 class="section"><a href="ffmpeg-filters.html#toc-atrim">6.21 atrim</a></h2>
<p>Trim the input so that the output contains one continuous subpart of the input.
</p>
<p>It accepts the following parameters:
</p><dl compact="compact">
<dt>‘<samp>start</samp>’</dt>
<dd><p>Timestamp (in seconds) of the start of the section to keep. I.e. the audio
sample with the timestamp <var>start</var> will be the first sample in the output.
</p>
</dd>
<dt>‘<samp>end</samp>’</dt>
<dd><p>Specify time of the first audio sample that will be dropped, i.e. the
audio sample immediately preceding the one with the timestamp <var>end</var> will be
the last sample in the output.
</p>
</dd>
<dt>‘<samp>start_pts</samp>’</dt>
<dd><p>Same as <var>start</var>, except this option sets the start timestamp in samples
instead of seconds.
</p>
</dd>
<dt>‘<samp>end_pts</samp>’</dt>
<dd><p>Same as <var>end</var>, except this option sets the end timestamp in samples instead
of seconds.
</p>
</dd>
<dt>‘<samp>duration</samp>’</dt>
<dd><p>The maximum duration of the output in seconds.
</p>
</dd>
<dt>‘<samp>start_sample</samp>’</dt>
<dd><p>The number of the first sample that should be output.
</p>
</dd>
<dt>‘<samp>end_sample</samp>’</dt>
<dd><p>The number of the first sample that should be dropped.
</p></dd>
</dl>
<p>‘<samp>start</samp>’, ‘<samp>end</samp>’, and ‘<samp>duration</samp>’ are expressed as time
duration specifications; see
<a href="ffmpeg-utils.html#time-duration-syntax">(ffmpeg-utils)the Time duration section in the ffmpeg-utils(1) manual</a>.
</p>
<p>Note that the first two sets of the start/end options and the ‘<samp>duration</samp>’
option look at the frame timestamp, while the _sample options simply count the
samples that pass through the filter. So start/end_pts and start/end_sample will
give different results when the timestamps are wrong, inexact or do not start at
zero. Also note that this filter does not modify the timestamps. If you wish
to have the output timestamps start at zero, insert the asetpts filter after the
atrim filter.
</p>
<p>If multiple start or end options are set, this filter tries to be greedy and
keep all samples that match at least one of the specified constraints. To keep
only the part that matches all the constraints at once, chain multiple atrim
filters.
</p>
<p>The defaults are such that all the input is kept. So it is possible to set e.g.
just the end values to keep everything before the specified time.
</p>
<p>Examples:
</p><ul>
<li>
Drop everything except the second minute of input:
<div class="example">
<pre class="example">ffmpeg -i INPUT -af atrim=60:120
</pre></div>
</li><li>
Keep only the first 1000 samples:
<div class="example">
<pre class="example">ffmpeg -i INPUT -af atrim=end_sample=1000
</pre></div>
</li></ul>
<a name="bandpass"></a>
<h2 class="section"><a href="ffmpeg-filters.html#toc-bandpass">6.22 bandpass</a></h2>
<p>Apply a two-pole Butterworth band-pass filter with central
frequency <var>frequency</var>, and (3dB-point) band-width width.
The <var>csg</var> option selects a constant skirt gain (peak gain = Q)
instead of the default: constant 0dB peak gain.
The filter roll off at 6dB per octave (20dB per decade).
</p>
<p>The filter accepts the following options:
</p>
<dl compact="compact">
<dt>‘<samp>frequency, f</samp>’</dt>
<dd><p>Set the filter’s central frequency. Default is <code>3000</code>.
</p>
</dd>
<dt>‘<samp>csg</samp>’</dt>
<dd><p>Constant skirt gain if set to 1. Defaults to 0.
</p>
</dd>
<dt>‘<samp>width_type</samp>’</dt>
<dd><p>Set method to specify band-width of filter.
</p><dl compact="compact">
<dt>‘<samp>h</samp>’</dt>
<dd><p>Hz
</p></dd>
<dt>‘<samp>q</samp>’</dt>
<dd><p>Q-Factor
</p></dd>
<dt>‘<samp>o</samp>’</dt>
<dd><p>octave
</p></dd>
<dt>‘<samp>s</samp>’</dt>
<dd><p>slope
</p></dd>
</dl>
</dd>
<dt>‘<samp>width, w</samp>’</dt>
<dd><p>Specify the band-width of a filter in width_type units.
</p></dd>
</dl>
<a name="bandreject"></a>
<h2 class="section"><a href="ffmpeg-filters.html#toc-bandreject">6.23 bandreject</a></h2>
<p>Apply a two-pole Butterworth band-reject filter with central
frequency <var>frequency</var>, and (3dB-point) band-width <var>width</var>.
The filter roll off at 6dB per octave (20dB per decade).
</p>
<p>The filter accepts the following options:
</p>
<dl compact="compact">
<dt>‘<samp>frequency, f</samp>’</dt>
<dd><p>Set the filter’s central frequency. Default is <code>3000</code>.
</p>
</dd>
<dt>‘<samp>width_type</samp>’</dt>
<dd><p>Set method to specify band-width of filter.
</p><dl compact="compact">
<dt>‘<samp>h</samp>’</dt>
<dd><p>Hz
</p></dd>
<dt>‘<samp>q</samp>’</dt>
<dd><p>Q-Factor
</p></dd>
<dt>‘<samp>o</samp>’</dt>
<dd><p>octave
</p></dd>
<dt>‘<samp>s</samp>’</dt>
<dd><p>slope
</p></dd>
</dl>
</dd>
<dt>‘<samp>width, w</samp>’</dt>
<dd><p>Specify the band-width of a filter in width_type units.
</p></dd>
</dl>
<a name="bass"></a>
<h2 class="section"><a href="ffmpeg-filters.html#toc-bass">6.24 bass</a></h2>
<p>Boost or cut the bass (lower) frequencies of the audio using a two-pole
shelving filter with a response similar to that of a standard
hi-fi’s tone-controls. This is also known as shelving equalisation (EQ).
</p>
<p>The filter accepts the following options:
</p>
<dl compact="compact">
<dt>‘<samp>gain, g</samp>’</dt>
<dd><p>Give the gain at 0 Hz. Its useful range is about -20
(for a large cut) to +20 (for a large boost).
Beware of clipping when using a positive gain.
</p>
</dd>
<dt>‘<samp>frequency, f</samp>’</dt>
<dd><p>Set the filter’s central frequency and so can be used
to extend or reduce the frequency range to be boosted or cut.
The default value is <code>100</code> Hz.
</p>
</dd>
<dt>‘<samp>width_type</samp>’</dt>
<dd><p>Set method to specify band-width of filter.
</p><dl compact="compact">
<dt>‘<samp>h</samp>’</dt>
<dd><p>Hz
</p></dd>
<dt>‘<samp>q</samp>’</dt>
<dd><p>Q-Factor
</p></dd>
<dt>‘<samp>o</samp>’</dt>
<dd><p>octave
</p></dd>
<dt>‘<samp>s</samp>’</dt>
<dd><p>slope
</p></dd>
</dl>
</dd>
<dt>‘<samp>width, w</samp>’</dt>
<dd><p>Determine how steep is the filter’s shelf transition.
</p></dd>
</dl>
<a name="biquad"></a>
<h2 class="section"><a href="ffmpeg-filters.html#toc-biquad">6.25 biquad</a></h2>
<p>Apply a biquad IIR filter with the given coefficients.
Where <var>b0</var>, <var>b1</var>, <var>b2</var> and <var>a0</var>, <var>a1</var>, <var>a2</var>
are the numerator and denominator coefficients respectively.
</p>
<a name="bs2b"></a>
<h2 class="section"><a href="ffmpeg-filters.html#toc-bs2b">6.26 bs2b</a></h2>
<p>Bauer stereo to binaural transformation, which improves headphone listening of
stereo audio records.
</p>
<p>It accepts the following parameters:
</p><dl compact="compact">
<dt>‘<samp>profile</samp>’</dt>
<dd><p>Pre-defined crossfeed level.
</p><dl compact="compact">
<dt>‘<samp>default</samp>’</dt>
<dd><p>Default level (fcut=700, feed=50).
</p>
</dd>
<dt>‘<samp>cmoy</samp>’</dt>
<dd><p>Chu Moy circuit (fcut=700, feed=60).
</p>
</dd>
<dt>‘<samp>jmeier</samp>’</dt>
<dd><p>Jan Meier circuit (fcut=650, feed=95).
</p>
</dd>
</dl>
</dd>
<dt>‘<samp>fcut</samp>’</dt>
<dd><p>Cut frequency (in Hz).
</p>
</dd>
<dt>‘<samp>feed</samp>’</dt>
<dd><p>Feed level (in Hz).
</p>
</dd>
</dl>
<a name="channelmap"></a>
<h2 class="section"><a href="ffmpeg-filters.html#toc-channelmap">6.27 channelmap</a></h2>
<p>Remap input channels to new locations.
</p>
<p>It accepts the following parameters:
</p><dl compact="compact">
<dt>‘<samp>channel_layout</samp>’</dt>
<dd><p>The channel layout of the output stream.
</p>
</dd>
<dt>‘<samp>map</samp>’</dt>
<dd><p>Map channels from input to output. The argument is a ’|’-separated list of
mappings, each in the <code><var>in_channel</var>-<var>out_channel</var></code> or
<var>in_channel</var> form. <var>in_channel</var> can be either the name of the input
channel (e.g. FL for front left) or its index in the input channel layout.
<var>out_channel</var> is the name of the output channel or its index in the output
channel layout. If <var>out_channel</var> is not given then it is implicitly an
index, starting with zero and increasing by one for each mapping.
</p></dd>
</dl>
<p>If no mapping is present, the filter will implicitly map input channels to
output channels, preserving indices.
</p>
<p>For example, assuming a 5.1+downmix input MOV file,
</p><div class="example">
<pre class="example">ffmpeg -i in.mov -filter 'channelmap=map=DL-FL|DR-FR' out.wav
</pre></div>
<p>will create an output WAV file tagged as stereo from the downmix channels of
the input.
</p>
<p>To fix a 5.1 WAV improperly encoded in AAC’s native channel order
</p><div class="example">
<pre class="example">ffmpeg -i in.wav -filter 'channelmap=1|2|0|5|3|4:channel_layout=5.1' out.wav
</pre></div>
<a name="channelsplit"></a>
<h2 class="section"><a href="ffmpeg-filters.html#toc-channelsplit">6.28 channelsplit</a></h2>
<p>Split each channel from an input audio stream into a separate output stream.
</p>
<p>It accepts the following parameters:
</p><dl compact="compact">
<dt>‘<samp>channel_layout</samp>’</dt>
<dd><p>The channel layout of the input stream. The default is "stereo".
</p></dd>
</dl>
<p>For example, assuming a stereo input MP3 file,
</p><div class="example">
<pre class="example">ffmpeg -i in.mp3 -filter_complex channelsplit out.mkv
</pre></div>
<p>will create an output Matroska file with two audio streams, one containing only
the left channel and the other the right channel.
</p>
<p>Split a 5.1 WAV file into per-channel files:
</p><div class="example">
<pre class="example">ffmpeg -i in.wav -filter_complex
'channelsplit=channel_layout=5.1[FL][FR][FC][LFE][SL][SR]'
-map '[FL]' front_left.wav -map '[FR]' front_right.wav -map '[FC]'
front_center.wav -map '[LFE]' lfe.wav -map '[SL]' side_left.wav -map '[SR]'
side_right.wav
</pre></div>
<a name="compand"></a>
<h2 class="section"><a href="ffmpeg-filters.html#toc-compand">6.29 compand</a></h2>
<p>Compress or expand the audio’s dynamic range.
</p>
<p>It accepts the following parameters:
</p>
<dl compact="compact">
<dt>‘<samp>attacks</samp>’</dt>
<dt>‘<samp>decays</samp>’</dt>
<dd><p>A list of times in seconds for each channel over which the instantaneous level
of the input signal is averaged to determine its volume. <var>attacks</var> refers to
increase of volume and <var>decays</var> refers to decrease of volume. For most
situations, the attack time (response to the audio getting louder) should be
shorter than the decay time, because the human ear is more sensitive to sudden
loud audio than sudden soft audio. A typical value for attack is 0.3 seconds and
a typical value for decay is 0.8 seconds.
</p>
</dd>
<dt>‘<samp>points</samp>’</dt>
<dd><p>A list of points for the transfer function, specified in dB relative to the
maximum possible signal amplitude. Each key points list must be defined using
the following syntax: <code>x0/y0|x1/y1|x2/y2|....</code> or
<code>x0/y0 x1/y1 x2/y2 ....</code>
</p>
<p>The input values must be in strictly increasing order but the transfer function
does not have to be monotonically rising. The point <code>0/0</code> is assumed but
may be overridden (by <code>0/out-dBn</code>). Typical values for the transfer
function are <code>-70/-70|-60/-20</code>.
</p>
</dd>
<dt>‘<samp>soft-knee</samp>’</dt>
<dd><p>Set the curve radius in dB for all joints. It defaults to 0.01.
</p>
</dd>
<dt>‘<samp>gain</samp>’</dt>
<dd><p>Set the additional gain in dB to be applied at all points on the transfer
function. This allows for easy adjustment of the overall gain.
It defaults to 0.
</p>
</dd>
<dt>‘<samp>volume</samp>’</dt>
<dd><p>Set an initial volume, in dB, to be assumed for each channel when filtering
starts. This permits the user to supply a nominal level initially, so that, for
example, a very large gain is not applied to initial signal levels before the
companding has begun to operate. A typical value for audio which is initially
quiet is -90 dB. It defaults to 0.
</p>
</dd>
<dt>‘<samp>delay</samp>’</dt>
<dd><p>Set a delay, in seconds. The input audio is analyzed immediately, but audio is
delayed before being fed to the volume adjuster. Specifying a delay
approximately equal to the attack/decay times allows the filter to effectively
operate in predictive rather than reactive mode. It defaults to 0.
</p>
</dd>
</dl>
<a name="Examples-79"></a>
<h3 class="subsection"><a href="ffmpeg-filters.html#toc-Examples-79">6.29.1 Examples</a></h3>
<ul>
<li>
Make music with both quiet and loud passages suitable for listening to in a
noisy environment:
<div class="example">
<pre class="example">compand=.3|.3:1|1:-90/-60|-60/-40|-40/-30|-20/-20:6:0:-90:0.2
</pre></div>
</li><li>
A noise gate for when the noise is at a lower level than the signal:
<div class="example">
<pre class="example">compand=.1|.1:.2|.2:-900/-900|-50.1/-900|-50/-50:.01:0:-90:.1
</pre></div>
</li><li>
Here is another noise gate, this time for when the noise is at a higher level
than the signal (making it, in some ways, similar to squelch):
<div class="example">
<pre class="example">compand=.1|.1:.1|.1:-45.1/-45.1|-45/-900|0/-900:.01:45:-90:.1
</pre></div>
</li></ul>
<a name="earwax"></a>
<h2 class="section"><a href="ffmpeg-filters.html#toc-earwax">6.30 earwax</a></h2>
<p>Make audio easier to listen to on headphones.
</p>
<p>This filter adds ‘cues’ to 44.1kHz stereo (i.e. audio CD format) audio
so that when listened to on headphones the stereo image is moved from
inside your head (standard for headphones) to outside and in front of
the listener (standard for speakers).
</p>
<p>Ported from SoX.
</p>
<a name="equalizer"></a>
<h2 class="section"><a href="ffmpeg-filters.html#toc-equalizer">6.31 equalizer</a></h2>
<p>Apply a two-pole peaking equalisation (EQ) filter. With this
filter, the signal-level at and around a selected frequency can
be increased or decreased, whilst (unlike bandpass and bandreject
filters) that at all other frequencies is unchanged.
</p>
<p>In order to produce complex equalisation curves, this filter can
be given several times, each with a different central frequency.
</p>
<p>The filter accepts the following options:
</p>
<dl compact="compact">
<dt>‘<samp>frequency, f</samp>’</dt>
<dd><p>Set the filter’s central frequency in Hz.
</p>
</dd>
<dt>‘<samp>width_type</samp>’</dt>
<dd><p>Set method to specify band-width of filter.
</p><dl compact="compact">
<dt>‘<samp>h</samp>’</dt>
<dd><p>Hz
</p></dd>
<dt>‘<samp>q</samp>’</dt>
<dd><p>Q-Factor
</p></dd>
<dt>‘<samp>o</samp>’</dt>
<dd><p>octave
</p></dd>
<dt>‘<samp>s</samp>’</dt>
<dd><p>slope
</p></dd>
</dl>
</dd>
<dt>‘<samp>width, w</samp>’</dt>
<dd><p>Specify the band-width of a filter in width_type units.
</p>
</dd>
<dt>‘<samp>gain, g</samp>’</dt>
<dd><p>Set the required gain or attenuation in dB.
Beware of clipping when using a positive gain.
</p></dd>
</dl>
<a name="Examples-14"></a>
<h3 class="subsection"><a href="ffmpeg-filters.html#toc-Examples-14">6.31.1 Examples</a></h3>
<ul>
<li>
Attenuate 10 dB at 1000 Hz, with a bandwidth of 200 Hz:
<div class="example">
<pre class="example">equalizer=f=1000:width_type=h:width=200:g=-10
</pre></div>
</li><li>
Apply 2 dB gain at 1000 Hz with Q 1 and attenuate 5 dB at 100 Hz with Q 2:
<div class="example">
<pre class="example">equalizer=f=1000:width_type=q:width=1:g=2,equalizer=f=100:width_type=q:width=2:g=-5
</pre></div>
</li></ul>
<a name="flanger"></a>
<h2 class="section"><a href="ffmpeg-filters.html#toc-flanger">6.32 flanger</a></h2>
<p>Apply a flanging effect to the audio.
</p>
<p>The filter accepts the following options:
</p>
<dl compact="compact">
<dt>‘<samp>delay</samp>’</dt>
<dd><p>Set base delay in milliseconds. Range from 0 to 30. Default value is 0.
</p>
</dd>
<dt>‘<samp>depth</samp>’</dt>
<dd><p>Set added swep delay in milliseconds. Range from 0 to 10. Default value is 2.
</p>
</dd>
<dt>‘<samp>regen</samp>’</dt>
<dd><p>Set percentage regeneneration (delayed signal feedback). Range from -95 to 95.
Default value is 0.
</p>
</dd>
<dt>‘<samp>width</samp>’</dt>
<dd><p>Set percentage of delayed signal mixed with original. Range from 0 to 100.
Default valu is 71.
</p>
</dd>
<dt>‘<samp>speed</samp>’</dt>
<dd><p>Set sweeps per second (Hz). Range from 0.1 to 10. Default value is 0.5.
</p>
</dd>
<dt>‘<samp>shape</samp>’</dt>
<dd><p>Set swept wave shape, can be <var>triangular</var> or <var>sinusoidal</var>.
Default value is <var>sinusoidal</var>.
</p>
</dd>
<dt>‘<samp>phase</samp>’</dt>
<dd><p>Set swept wave percentage-shift for multi channel. Range from 0 to 100.
Default value is 25.
</p>
</dd>
<dt>‘<samp>interp</samp>’</dt>
<dd><p>Set delay-line interpolation, <var>linear</var> or <var>quadratic</var>.
Default is <var>linear</var>.
</p></dd>
</dl>
<a name="highpass"></a>
<h2 class="section"><a href="ffmpeg-filters.html#toc-highpass">6.33 highpass</a></h2>
<p>Apply a high-pass filter with 3dB point frequency.
The filter can be either single-pole, or double-pole (the default).
The filter roll off at 6dB per pole per octave (20dB per pole per decade).
</p>
<p>The filter accepts the following options:
</p>
<dl compact="compact">
<dt>‘<samp>frequency, f</samp>’</dt>
<dd><p>Set frequency in Hz. Default is 3000.
</p>
</dd>
<dt>‘<samp>poles, p</samp>’</dt>
<dd><p>Set number of poles. Default is 2.
</p>
</dd>
<dt>‘<samp>width_type</samp>’</dt>
<dd><p>Set method to specify band-width of filter.
</p><dl compact="compact">
<dt>‘<samp>h</samp>’</dt>
<dd><p>Hz
</p></dd>
<dt>‘<samp>q</samp>’</dt>
<dd><p>Q-Factor
</p></dd>
<dt>‘<samp>o</samp>’</dt>
<dd><p>octave
</p></dd>
<dt>‘<samp>s</samp>’</dt>
<dd><p>slope
</p></dd>
</dl>
</dd>
<dt>‘<samp>width, w</samp>’</dt>
<dd><p>Specify the band-width of a filter in width_type units.
Applies only to double-pole filter.
The default is 0.707q and gives a Butterworth response.
</p></dd>
</dl>
<a name="join"></a>
<h2 class="section"><a href="ffmpeg-filters.html#toc-join">6.34 join</a></h2>
<p>Join multiple input streams into one multi-channel stream.
</p>
<p>It accepts the following parameters:
</p><dl compact="compact">
<dt>‘<samp>inputs</samp>’</dt>
<dd><p>The number of input streams. It defaults to 2.
</p>
</dd>
<dt>‘<samp>channel_layout</samp>’</dt>
<dd><p>The desired output channel layout. It defaults to stereo.
</p>
</dd>
<dt>‘<samp>map</samp>’</dt>
<dd><p>Map channels from inputs to output. The argument is a ’|’-separated list of
mappings, each in the <code><var>input_idx</var>.<var>in_channel</var>-<var>out_channel</var></code>
form. <var>input_idx</var> is the 0-based index of the input stream. <var>in_channel</var>
can be either the name of the input channel (e.g. FL for front left) or its
index in the specified input stream. <var>out_channel</var> is the name of the output
channel.
</p></dd>
</dl>
<p>The filter will attempt to guess the mappings when they are not specified
explicitly. It does so by first trying to find an unused matching input channel
and if that fails it picks the first unused input channel.
</p>
<p>Join 3 inputs (with properly set channel layouts):
</p><div class="example">
<pre class="example">ffmpeg -i INPUT1 -i INPUT2 -i INPUT3 -filter_complex join=inputs=3 OUTPUT
</pre></div>
<p>Build a 5.1 output from 6 single-channel streams:
</p><div class="example">
<pre class="example">ffmpeg -i fl -i fr -i fc -i sl -i sr -i lfe -filter_complex
'join=inputs=6:channel_layout=5.1:map=0.0-FL|1.0-FR|2.0-FC|3.0-SL|4.0-SR|5.0-LFE'
out
</pre></div>
<a name="ladspa"></a>
<h2 class="section"><a href="ffmpeg-filters.html#toc-ladspa">6.35 ladspa</a></h2>
<p>Load a LADSPA (Linux Audio Developer’s Simple Plugin API) plugin.
</p>
<p>To enable compilation of this filter you need to configure FFmpeg with
<code>--enable-ladspa</code>.
</p>
<dl compact="compact">
<dt>‘<samp>file, f</samp>’</dt>
<dd><p>Specifies the name of LADSPA plugin library to load. If the environment
variable <code>LADSPA_PATH</code> is defined, the LADSPA plugin is searched in
each one of the directories specified by the colon separated list in
<code>LADSPA_PATH</code>, otherwise in the standard LADSPA paths, which are in
this order: ‘<tt>HOME/.ladspa/lib/</tt>’, ‘<tt>/usr/local/lib/ladspa/</tt>’,
‘<tt>/usr/lib/ladspa/</tt>’.
</p>
</dd>
<dt>‘<samp>plugin, p</samp>’</dt>
<dd><p>Specifies the plugin within the library. Some libraries contain only
one plugin, but others contain many of them. If this is not set filter
will list all available plugins within the specified library.
</p>
</dd>
<dt>‘<samp>controls, c</samp>’</dt>
<dd><p>Set the ’|’ separated list of controls which are zero or more floating point
values that determine the behavior of the loaded plugin (for example delay,
threshold or gain).
Controls need to be defined using the following syntax:
c0=<var>value0</var>|c1=<var>value1</var>|c2=<var>value2</var>|..., where
<var>valuei</var> is the value set on the <var>i</var>-th control.
If ‘<samp>controls</samp>’ is set to <code>help</code>, all available controls and
their valid ranges are printed.
</p>
</dd>
<dt>‘<samp>sample_rate, s</samp>’</dt>
<dd><p>Specify the sample rate, default to 44100. Only used if plugin have
zero inputs.
</p>
</dd>
<dt>‘<samp>nb_samples, n</samp>’</dt>
<dd><p>Set the number of samples per channel per each output frame, default
is 1024. Only used if plugin have zero inputs.
</p>
</dd>
<dt>‘<samp>duration, d</samp>’</dt>
<dd><p>Set the minimum duration of the sourced audio. See
<a href="ffmpeg-utils.html#time-duration-syntax">(ffmpeg-utils)the Time duration section in the ffmpeg-utils(1) manual</a>
for the accepted syntax.
Note that the resulting duration may be greater than the specified duration,
as the generated audio is always cut at the end of a complete frame.
If not specified, or the expressed duration is negative, the audio is
supposed to be generated forever.
Only used if plugin have zero inputs.
</p>
</dd>
</dl>
<a name="Examples-35"></a>
<h3 class="subsection"><a href="ffmpeg-filters.html#toc-Examples-35">6.35.1 Examples</a></h3>
<ul>
<li>
List all available plugins within amp (LADSPA example plugin) library:
<div class="example">
<pre class="example">ladspa=file=amp
</pre></div>
</li><li>
List all available controls and their valid ranges for <code>vcf_notch</code>
plugin from <code>VCF</code> library:
<div class="example">
<pre class="example">ladspa=f=vcf:p=vcf_notch:c=help
</pre></div>
</li><li>
Simulate low quality audio equipment using <code>Computer Music Toolkit</code> (CMT)
plugin library:
<div class="example">
<pre class="example">ladspa=file=cmt:plugin=lofi:controls=c0=22|c1=12|c2=12
</pre></div>
</li><li>
Add reverberation to the audio using TAP-plugins
(Tom’s Audio Processing plugins):
<div class="example">
<pre class="example">ladspa=file=tap_reverb:tap_reverb
</pre></div>
</li><li>
Generate white noise, with 0.2 amplitude:
<div class="example">
<pre class="example">ladspa=file=cmt:noise_source_white:c=c0=.2
</pre></div>
</li><li>
Generate 20 bpm clicks using plugin <code>C* Click - Metronome</code> from the
<code>C* Audio Plugin Suite</code> (CAPS) library:
<div class="example">
<pre class="example">ladspa=file=caps:Click:c=c1=20'
</pre></div>
</li><li>
Apply <code>C* Eq10X2 - Stereo 10-band equaliser</code> effect:
<div class="example">
<pre class="example">ladspa=caps:Eq10X2:c=c0=-48|c9=-24|c3=12|c4=2
</pre></div>
</li></ul>
<a name="Commands"></a>
<h3 class="subsection"><a href="ffmpeg-filters.html#toc-Commands">6.35.2 Commands</a></h3>
<p>This filter supports the following commands:
</p><dl compact="compact">
<dt>‘<samp>cN</samp>’</dt>
<dd><p>Modify the <var>N</var>-th control value.
</p>
<p>If the specified value is not valid, it is ignored and prior one is kept.
</p></dd>
</dl>
<a name="lowpass"></a>
<h2 class="section"><a href="ffmpeg-filters.html#toc-lowpass">6.36 lowpass</a></h2>
<p>Apply a low-pass filter with 3dB point frequency.
The filter can be either single-pole or double-pole (the default).
The filter roll off at 6dB per pole per octave (20dB per pole per decade).
</p>
<p>The filter accepts the following options:
</p>
<dl compact="compact">
<dt>‘<samp>frequency, f</samp>’</dt>
<dd><p>Set frequency in Hz. Default is 500.
</p>
</dd>
<dt>‘<samp>poles, p</samp>’</dt>
<dd><p>Set number of poles. Default is 2.
</p>
</dd>
<dt>‘<samp>width_type</samp>’</dt>
<dd><p>Set method to specify band-width of filter.
</p><dl compact="compact">
<dt>‘<samp>h</samp>’</dt>
<dd><p>Hz
</p></dd>
<dt>‘<samp>q</samp>’</dt>
<dd><p>Q-Factor
</p></dd>
<dt>‘<samp>o</samp>’</dt>
<dd><p>octave
</p></dd>
<dt>‘<samp>s</samp>’</dt>
<dd><p>slope
</p></dd>
</dl>
</dd>
<dt>‘<samp>width, w</samp>’</dt>
<dd><p>Specify the band-width of a filter in width_type units.
Applies only to double-pole filter.
The default is 0.707q and gives a Butterworth response.
</p></dd>
</dl>
<a name="pan"></a>
<h2 class="section"><a href="ffmpeg-filters.html#toc-pan">6.37 pan</a></h2>
<p>Mix channels with specific gain levels. The filter accepts the output
channel layout followed by a set of channels definitions.
</p>
<p>This filter is also designed to remap efficiently the channels of an audio
stream.
</p>
<p>The filter accepts parameters of the form:
"<var>l</var>:<var>outdef</var>:<var>outdef</var>:..."
</p>
<dl compact="compact">
<dt>‘<samp>l</samp>’</dt>
<dd><p>output channel layout or number of channels
</p>
</dd>
<dt>‘<samp>outdef</samp>’</dt>
<dd><p>output channel specification, of the form:
"<var>out_name</var>=[<var>gain</var>*]<var>in_name</var>[+[<var>gain</var>*]<var>in_name</var>...]"
</p>
</dd>
<dt>‘<samp>out_name</samp>’</dt>
<dd><p>output channel to define, either a channel name (FL, FR, etc.) or a channel
number (c0, c1, etc.)
</p>
</dd>
<dt>‘<samp>gain</samp>’</dt>
<dd><p>multiplicative coefficient for the channel, 1 leaving the volume unchanged
</p>
</dd>
<dt>‘<samp>in_name</samp>’</dt>
<dd><p>input channel to use, see out_name for details; it is not possible to mix
named and numbered input channels
</p></dd>
</dl>
<p>If the ‘=’ in a channel specification is replaced by ‘<’, then the gains for
that specification will be renormalized so that the total is 1, thus
avoiding clipping noise.
</p>
<a name="Mixing-examples"></a>
<h3 class="subsection"><a href="ffmpeg-filters.html#toc-Mixing-examples">6.37.1 Mixing examples</a></h3>
<p>For example, if you want to down-mix from stereo to mono, but with a bigger
factor for the left channel:
</p><div class="example">
<pre class="example">pan=1:c0=0.9*c0+0.1*c1
</pre></div>
<p>A customized down-mix to stereo that works automatically for 3-, 4-, 5- and
7-channels surround:
</p><div class="example">
<pre class="example">pan=stereo: FL < FL + 0.5*FC + 0.6*BL + 0.6*SL : FR < FR + 0.5*FC + 0.6*BR + 0.6*SR
</pre></div>
<p>Note that <code>ffmpeg</code> integrates a default down-mix (and up-mix) system
that should be preferred (see "-ac" option) unless you have very specific
needs.
</p>
<a name="Remapping-examples"></a>
<h3 class="subsection"><a href="ffmpeg-filters.html#toc-Remapping-examples">6.37.2 Remapping examples</a></h3>
<p>The channel remapping will be effective if, and only if:
</p>
<ul>
<li> gain coefficients are zeroes or ones,
</li><li> only one input per channel output,
</li></ul>
<p>If all these conditions are satisfied, the filter will notify the user ("Pure
channel mapping detected"), and use an optimized and lossless method to do the
remapping.
</p>
<p>For example, if you have a 5.1 source and want a stereo audio stream by
dropping the extra channels:
</p><div class="example">
<pre class="example">pan="stereo: c0=FL : c1=FR"
</pre></div>
<p>Given the same source, you can also switch front left and front right channels
and keep the input channel layout:
</p><div class="example">
<pre class="example">pan="5.1: c0=c1 : c1=c0 : c2=c2 : c3=c3 : c4=c4 : c5=c5"
</pre></div>
<p>If the input is a stereo audio stream, you can mute the front left channel (and
still keep the stereo channel layout) with:
</p><div class="example">
<pre class="example">pan="stereo:c1=c1"
</pre></div>
<p>Still with a stereo audio stream input, you can copy the right channel in both
front left and right:
</p><div class="example">
<pre class="example">pan="stereo: c0=FR : c1=FR"
</pre></div>
<a name="replaygain"></a>
<h2 class="section"><a href="ffmpeg-filters.html#toc-replaygain">6.38 replaygain</a></h2>
<p>ReplayGain scanner filter. This filter takes an audio stream as an input and
outputs it unchanged.
At end of filtering it displays <code>track_gain</code> and <code>track_peak</code>.
</p>
<a name="resample"></a>
<h2 class="section"><a href="ffmpeg-filters.html#toc-resample">6.39 resample</a></h2>
<p>Convert the audio sample format, sample rate and channel layout. It is
not meant to be used directly.
</p>
<a name="silencedetect"></a>
<h2 class="section"><a href="ffmpeg-filters.html#toc-silencedetect">6.40 silencedetect</a></h2>
<p>Detect silence in an audio stream.
</p>
<p>This filter logs a message when it detects that the input audio volume is less
or equal to a noise tolerance value for a duration greater or equal to the
minimum detected noise duration.
</p>
<p>The printed times and duration are expressed in seconds.
</p>
<p>The filter accepts the following options:
</p>
<dl compact="compact">
<dt>‘<samp>duration, d</samp>’</dt>
<dd><p>Set silence duration until notification (default is 2 seconds).
</p>
</dd>
<dt>‘<samp>noise, n</samp>’</dt>
<dd><p>Set noise tolerance. Can be specified in dB (in case "dB" is appended to the
specified value) or amplitude ratio. Default is -60dB, or 0.001.
</p></dd>
</dl>
<a name="Examples-39"></a>
<h3 class="subsection"><a href="ffmpeg-filters.html#toc-Examples-39">6.40.1 Examples</a></h3>
<ul>
<li>
Detect 5 seconds of silence with -50dB noise tolerance:
<div class="example">
<pre class="example">silencedetect=n=-50dB:d=5
</pre></div>
</li><li>
Complete example with <code>ffmpeg</code> to detect silence with 0.0001 noise
tolerance in ‘<tt>silence.mp3</tt>’:
<div class="example">
<pre class="example">ffmpeg -i silence.mp3 -af silencedetect=noise=0.0001 -f null -
</pre></div>
</li></ul>
<a name="silenceremove"></a>
<h2 class="section"><a href="ffmpeg-filters.html#toc-silenceremove">6.41 silenceremove</a></h2>
<p>Remove silence from the beginning, middle or end of the audio.
</p>
<p>The filter accepts the following options:
</p>
<dl compact="compact">
<dt>‘<samp>start_periods</samp>’</dt>
<dd><p>This value is used to indicate if audio should be trimmed at beginning of
the audio. A value of zero indicates no silence should be trimmed from the
beginning. When specifying a non-zero value, it trims audio up until it
finds non-silence. Normally, when trimming silence from beginning of audio
the <var>start_periods</var> will be <code>1</code> but it can be increased to higher
values to trim all audio up to specific count of non-silence periods.
Default value is <code>0</code>.
</p>
</dd>
<dt>‘<samp>start_duration</samp>’</dt>
<dd><p>Specify the amount of time that non-silence must be detected before it stops
trimming audio. By increasing the duration, bursts of noises can be treated
as silence and trimmed off. Default value is <code>0</code>.
</p>
</dd>
<dt>‘<samp>start_threshold</samp>’</dt>
<dd><p>This indicates what sample value should be treated as silence. For digital
audio, a value of <code>0</code> may be fine but for audio recorded from analog,
you may wish to increase the value to account for background noise.
Can be specified in dB (in case "dB" is appended to the specified value)
or amplitude ratio. Default value is <code>0</code>.
</p>
</dd>
<dt>‘<samp>stop_periods</samp>’</dt>
<dd><p>Set the count for trimming silence from the end of audio.
To remove silence from the middle of a file, specify a <var>stop_periods</var>
that is negative. This value is then threated as a positive value and is
used to indicate the effect should restart processing as specified by
<var>start_periods</var>, making it suitable for removing periods of silence
in the middle of the audio.
Default value is <code>0</code>.
</p>
</dd>
<dt>‘<samp>stop_duration</samp>’</dt>
<dd><p>Specify a duration of silence that must exist before audio is not copied any
more. By specifying a higher duration, silence that is wanted can be left in
the audio.
Default value is <code>0</code>.
</p>
</dd>
<dt>‘<samp>stop_threshold</samp>’</dt>
<dd><p>This is the same as ‘<samp>start_threshold</samp>’ but for trimming silence from
the end of audio.
Can be specified in dB (in case "dB" is appended to the specified value)
or amplitude ratio. Default value is <code>0</code>.
</p>
</dd>
<dt>‘<samp>leave_silence</samp>’</dt>
<dd><p>This indicate that <var>stop_duration</var> length of audio should be left intact
at the beginning of each period of silence.
For example, if you want to remove long pauses between words but do not want
to remove the pauses completely. Default value is <code>0</code>.
</p>
</dd>
</dl>
<a name="Examples-58"></a>
<h3 class="subsection"><a href="ffmpeg-filters.html#toc-Examples-58">6.41.1 Examples</a></h3>
<ul>
<li>
The following example shows how this filter can be used to start a recording
that does not contain the delay at the start which usually occurs between
pressing the record button and the start of the performance:
<div class="example">
<pre class="example">silenceremove=1:5:0.02
</pre></div>
</li></ul>
<a name="treble"></a>
<h2 class="section"><a href="ffmpeg-filters.html#toc-treble">6.42 treble</a></h2>
<p>Boost or cut treble (upper) frequencies of the audio using a two-pole
shelving filter with a response similar to that of a standard
hi-fi’s tone-controls. This is also known as shelving equalisation (EQ).
</p>
<p>The filter accepts the following options:
</p>
<dl compact="compact">
<dt>‘<samp>gain, g</samp>’</dt>
<dd><p>Give the gain at whichever is the lower of ~22 kHz and the
Nyquist frequency. Its useful range is about -20 (for a large cut)
to +20 (for a large boost). Beware of clipping when using a positive gain.
</p>
</dd>
<dt>‘<samp>frequency, f</samp>’</dt>
<dd><p>Set the filter’s central frequency and so can be used
to extend or reduce the frequency range to be boosted or cut.
The default value is <code>3000</code> Hz.
</p>
</dd>
<dt>‘<samp>width_type</samp>’</dt>
<dd><p>Set method to specify band-width of filter.
</p><dl compact="compact">
<dt>‘<samp>h</samp>’</dt>
<dd><p>Hz
</p></dd>
<dt>‘<samp>q</samp>’</dt>
<dd><p>Q-Factor
</p></dd>
<dt>‘<samp>o</samp>’</dt>
<dd><p>octave
</p></dd>
<dt>‘<samp>s</samp>’</dt>
<dd><p>slope
</p></dd>
</dl>
</dd>
<dt>‘<samp>width, w</samp>’</dt>
<dd><p>Determine how steep is the filter’s shelf transition.
</p></dd>
</dl>
<a name="volume"></a>
<h2 class="section"><a href="ffmpeg-filters.html#toc-volume">6.43 volume</a></h2>
<p>Adjust the input audio volume.
</p>
<p>It accepts the following parameters:
</p><dl compact="compact">
<dt>‘<samp>volume</samp>’</dt>
<dd><p>Set audio volume expression.
</p>
<p>Output values are clipped to the maximum value.
</p>
<p>The output audio volume is given by the relation:
</p><div class="example">
<pre class="example"><var>output_volume</var> = <var>volume</var> * <var>input_volume</var>
</pre></div>
<p>The default value for <var>volume</var> is "1.0".
</p>
</dd>
<dt>‘<samp>precision</samp>’</dt>
<dd><p>This parameter represents the mathematical precision.
</p>
<p>It determines which input sample formats will be allowed, which affects the
precision of the volume scaling.
</p>
<dl compact="compact">
<dt>‘<samp>fixed</samp>’</dt>
<dd><p>8-bit fixed-point; this limits input sample format to U8, S16, and S32.
</p></dd>
<dt>‘<samp>float</samp>’</dt>
<dd><p>32-bit floating-point; this limits input sample format to FLT. (default)
</p></dd>
<dt>‘<samp>double</samp>’</dt>
<dd><p>64-bit floating-point; this limits input sample format to DBL.
</p></dd>
</dl>
</dd>
<dt>‘<samp>replaygain</samp>’</dt>
<dd><p>Choose the behaviour on encountering ReplayGain side data in input frames.
</p>
<dl compact="compact">
<dt>‘<samp>drop</samp>’</dt>
<dd><p>Remove ReplayGain side data, ignoring its contents (the default).
</p>
</dd>
<dt>‘<samp>ignore</samp>’</dt>
<dd><p>Ignore ReplayGain side data, but leave it in the frame.
</p>
</dd>
<dt>‘<samp>track</samp>’</dt>
<dd><p>Prefer the track gain, if present.
</p>
</dd>
<dt>‘<samp>album</samp>’</dt>
<dd><p>Prefer the album gain, if present.
</p></dd>
</dl>
</dd>
<dt>‘<samp>replaygain_preamp</samp>’</dt>
<dd><p>Pre-amplification gain in dB to apply to the selected replaygain gain.
</p>
<p>Default value for <var>replaygain_preamp</var> is 0.0.
</p>
</dd>
<dt>‘<samp>eval</samp>’</dt>
<dd><p>Set when the volume expression is evaluated.
</p>
<p>It accepts the following values:
</p><dl compact="compact">
<dt>‘<samp>once</samp>’</dt>
<dd><p>only evaluate expression once during the filter initialization, or
when the ‘<samp>volume</samp>’ command is sent
</p>
</dd>
<dt>‘<samp>frame</samp>’</dt>
<dd><p>evaluate expression for each incoming frame
</p></dd>
</dl>
<p>Default value is ‘<samp>once</samp>’.
</p></dd>
</dl>
<p>The volume expression can contain the following parameters.
</p>
<dl compact="compact">
<dt>‘<samp>n</samp>’</dt>
<dd><p>frame number (starting at zero)
</p></dd>
<dt>‘<samp>nb_channels</samp>’</dt>
<dd><p>number of channels
</p></dd>
<dt>‘<samp>nb_consumed_samples</samp>’</dt>
<dd><p>number of samples consumed by the filter
</p></dd>
<dt>‘<samp>nb_samples</samp>’</dt>
<dd><p>number of samples in the current frame
</p></dd>
<dt>‘<samp>pos</samp>’</dt>
<dd><p>original frame position in the file
</p></dd>
<dt>‘<samp>pts</samp>’</dt>
<dd><p>frame PTS
</p></dd>
<dt>‘<samp>sample_rate</samp>’</dt>
<dd><p>sample rate
</p></dd>
<dt>‘<samp>startpts</samp>’</dt>
<dd><p>PTS at start of stream
</p></dd>
<dt>‘<samp>startt</samp>’</dt>
<dd><p>time at start of stream
</p></dd>
<dt>‘<samp>t</samp>’</dt>
<dd><p>frame time
</p></dd>
<dt>‘<samp>tb</samp>’</dt>
<dd><p>timestamp timebase
</p></dd>
<dt>‘<samp>volume</samp>’</dt>
<dd><p>last set volume value
</p></dd>
</dl>
<p>Note that when ‘<samp>eval</samp>’ is set to ‘<samp>once</samp>’ only the
<var>sample_rate</var> and <var>tb</var> variables are available, all other
variables will evaluate to NAN.
</p>
<a name="Commands-1"></a>
<h3 class="subsection"><a href="ffmpeg-filters.html#toc-Commands-1">6.43.1 Commands</a></h3>
<p>This filter supports the following commands:
</p><dl compact="compact">
<dt>‘<samp>volume</samp>’</dt>
<dd><p>Modify the volume expression.
The command accepts the same syntax of the corresponding option.
</p>
<p>If the specified expression is not valid, it is kept at its current
value.
</p></dd>
<dt>‘<samp>replaygain_noclip</samp>’</dt>
<dd><p>Prevent clipping by limiting the gain applied.
</p>
<p>Default value for <var>replaygain_noclip</var> is 1.
</p>
</dd>
</dl>
<a name="Examples-68"></a>
<h3 class="subsection"><a href="ffmpeg-filters.html#toc-Examples-68">6.43.2 Examples</a></h3>
<ul>
<li>
Halve the input audio volume:
<div class="example">
<pre class="example">volume=volume=0.5
volume=volume=1/2
volume=volume=-6.0206dB
</pre></div>
<p>In all the above example the named key for ‘<samp>volume</samp>’ can be
omitted, for example like in:
</p><div class="example">
<pre class="example">volume=0.5
</pre></div>
</li><li>
Increase input audio power by 6 decibels using fixed-point precision:
<div class="example">
<pre class="example">volume=volume=6dB:precision=fixed
</pre></div>
</li><li>
Fade volume after time 10 with an annihilation period of 5 seconds:
<div class="example">
<pre class="example">volume='if(lt(t,10),1,max(1-(t-10)/5,0))':eval=frame
</pre></div>
</li></ul>
<a name="volumedetect"></a>
<h2 class="section"><a href="ffmpeg-filters.html#toc-volumedetect">6.44 volumedetect</a></h2>
<p>Detect the volume of the input video.
</p>
<p>The filter has no parameters. The input is not modified. Statistics about
the volume will be printed in the log when the input stream end is reached.
</p>
<p>In particular it will show the mean volume (root mean square), maximum
volume (on a per-sample basis), and the beginning of a histogram of the
registered volume values (from the maximum value to a cumulated 1/1000 of
the samples).
</p>
<p>All volumes are in decibels relative to the maximum PCM value.
</p>
<a name="Examples-22"></a>
<h3 class="subsection"><a href="ffmpeg-filters.html#toc-Examples-22">6.44.1 Examples</a></h3>
<p>Here is an excerpt of the output:
</p><div class="example">
<pre class="example">[Parsed_volumedetect_0 0xa23120] mean_volume: -27 dB
[Parsed_volumedetect_0 0xa23120] max_volume: -4 dB
[Parsed_volumedetect_0 0xa23120] histogram_4db: 6
[Parsed_volumedetect_0 0xa23120] histogram_5db: 62
[Parsed_volumedetect_0 0xa23120] histogram_6db: 286
[Parsed_volumedetect_0 0xa23120] histogram_7db: 1042
[Parsed_volumedetect_0 0xa23120] histogram_8db: 2551
[Parsed_volumedetect_0 0xa23120] histogram_9db: 4609
[Parsed_volumedetect_0 0xa23120] histogram_10db: 8409
</pre></div>
<p>It means that:
</p><ul>
<li>
The mean square energy is approximately -27 dB, or 10^-2.7.
</li><li>
The largest sample is at -4 dB, or more precisely between -4 dB and -5 dB.
</li><li>
There are 6 samples at -4 dB, 62 at -5 dB, 286 at -6 dB, etc.
</li></ul>
<p>In other words, raising the volume by +4 dB does not cause any clipping,
raising it by +5 dB causes clipping for 6 samples, etc.
</p>
<a name="Audio-Sources"></a>
<h1 class="chapter"><a href="ffmpeg-filters.html#toc-Audio-Sources">7 Audio Sources</a></h1>
<p>Below is a description of the currently available audio sources.
</p>
<a name="abuffer"></a>
<h2 class="section"><a href="ffmpeg-filters.html#toc-abuffer">7.1 abuffer</a></h2>
<p>Buffer audio frames, and make them available to the filter chain.
</p>
<p>This source is mainly intended for a programmatic use, in particular
through the interface defined in ‘<tt>libavfilter/asrc_abuffer.h</tt>’.
</p>
<p>It accepts the following parameters:
</p><dl compact="compact">
<dt>‘<samp>time_base</samp>’</dt>
<dd><p>The timebase which will be used for timestamps of submitted frames. It must be
either a floating-point number or in <var>numerator</var>/<var>denominator</var> form.
</p>
</dd>
<dt>‘<samp>sample_rate</samp>’</dt>
<dd><p>The sample rate of the incoming audio buffers.
</p>
</dd>
<dt>‘<samp>sample_fmt</samp>’</dt>
<dd><p>The sample format of the incoming audio buffers.
Either a sample format name or its corresponging integer representation from
the enum AVSampleFormat in ‘<tt>libavutil/samplefmt.h</tt>’
</p>
</dd>
<dt>‘<samp>channel_layout</samp>’</dt>
<dd><p>The channel layout of the incoming audio buffers.
Either a channel layout name from channel_layout_map in
‘<tt>libavutil/channel_layout.c</tt>’ or its corresponding integer representation
from the AV_CH_LAYOUT_* macros in ‘<tt>libavutil/channel_layout.h</tt>’
</p>
</dd>
<dt>‘<samp>channels</samp>’</dt>
<dd><p>The number of channels of the incoming audio buffers.
If both <var>channels</var> and <var>channel_layout</var> are specified, then they
must be consistent.
</p>
</dd>
</dl>
<a name="Examples-60"></a>
<h3 class="subsection"><a href="ffmpeg-filters.html#toc-Examples-60">7.1.1 Examples</a></h3>
<div class="example">
<pre class="example">abuffer=sample_rate=44100:sample_fmt=s16p:channel_layout=stereo
</pre></div>
<p>will instruct the source to accept planar 16bit signed stereo at 44100Hz.
Since the sample format with name "s16p" corresponds to the number
6 and the "stereo" channel layout corresponds to the value 0x3, this is
equivalent to:
</p><div class="example">
<pre class="example">abuffer=sample_rate=44100:sample_fmt=6:channel_layout=0x3
</pre></div>
<a name="aevalsrc"></a>
<h2 class="section"><a href="ffmpeg-filters.html#toc-aevalsrc">7.2 aevalsrc</a></h2>
<p>Generate an audio signal specified by an expression.
</p>
<p>This source accepts in input one or more expressions (one for each
channel), which are evaluated and used to generate a corresponding
audio signal.
</p>
<p>This source accepts the following options:
</p>
<dl compact="compact">
<dt>‘<samp>exprs</samp>’</dt>
<dd><p>Set the ’|’-separated expressions list for each separate channel. In case the
‘<samp>channel_layout</samp>’ option is not specified, the selected channel layout
depends on the number of provided expressions. Otherwise the last
specified expression is applied to the remaining output channels.
</p>
</dd>
<dt>‘<samp>channel_layout, c</samp>’</dt>
<dd><p>Set the channel layout. The number of channels in the specified layout
must be equal to the number of specified expressions.
</p>
</dd>
<dt>‘<samp>duration, d</samp>’</dt>
<dd><p>Set the minimum duration of the sourced audio. See
<a href="ffmpeg-utils.html#time-duration-syntax">(ffmpeg-utils)the Time duration section in the ffmpeg-utils(1) manual</a>
for the accepted syntax.
Note that the resulting duration may be greater than the specified
duration, as the generated audio is always cut at the end of a
complete frame.
</p>
<p>If not specified, or the expressed duration is negative, the audio is
supposed to be generated forever.
</p>
</dd>
<dt>‘<samp>nb_samples, n</samp>’</dt>
<dd><p>Set the number of samples per channel per each output frame,
default to 1024.
</p>
</dd>
<dt>‘<samp>sample_rate, s</samp>’</dt>
<dd><p>Specify the sample rate, default to 44100.
</p></dd>
</dl>
<p>Each expression in <var>exprs</var> can contain the following constants:
</p>
<dl compact="compact">
<dt>‘<samp>n</samp>’</dt>
<dd><p>number of the evaluated sample, starting from 0
</p>
</dd>
<dt>‘<samp>t</samp>’</dt>
<dd><p>time of the evaluated sample expressed in seconds, starting from 0
</p>
</dd>
<dt>‘<samp>s</samp>’</dt>
<dd><p>sample rate
</p>
</dd>
</dl>
<a name="Examples-11"></a>
<h3 class="subsection"><a href="ffmpeg-filters.html#toc-Examples-11">7.2.1 Examples</a></h3>
<ul>
<li>
Generate silence:
<div class="example">
<pre class="example">aevalsrc=0
</pre></div>
</li><li>
Generate a sin signal with frequency of 440 Hz, set sample rate to
8000 Hz:
<div class="example">
<pre class="example">aevalsrc="sin(440*2*PI*t):s=8000"
</pre></div>
</li><li>
Generate a two channels signal, specify the channel layout (Front
Center + Back Center) explicitly:
<div class="example">
<pre class="example">aevalsrc="sin(420*2*PI*t)|cos(430*2*PI*t):c=FC|BC"
</pre></div>
</li><li>
Generate white noise:
<div class="example">
<pre class="example">aevalsrc="-2+random(0)"
</pre></div>
</li><li>
Generate an amplitude modulated signal:
<div class="example">
<pre class="example">aevalsrc="sin(10*2*PI*t)*sin(880*2*PI*t)"
</pre></div>
</li><li>
Generate 2.5 Hz binaural beats on a 360 Hz carrier:
<div class="example">
<pre class="example">aevalsrc="0.1*sin(2*PI*(360-2.5/2)*t) | 0.1*sin(2*PI*(360+2.5/2)*t)"
</pre></div>
</li></ul>
<a name="anullsrc"></a>
<h2 class="section"><a href="ffmpeg-filters.html#toc-anullsrc">7.3 anullsrc</a></h2>
<p>The null audio source, return unprocessed audio frames. It is mainly useful
as a template and to be employed in analysis / debugging tools, or as
the source for filters which ignore the input data (for example the sox
synth filter).
</p>
<p>This source accepts the following options:
</p>
<dl compact="compact">
<dt>‘<samp>channel_layout, cl</samp>’</dt>
<dd>
<p>Specifies the channel layout, and can be either an integer or a string
representing a channel layout. The default value of <var>channel_layout</var>
is "stereo".
</p>
<p>Check the channel_layout_map definition in
‘<tt>libavutil/channel_layout.c</tt>’ for the mapping between strings and
channel layout values.
</p>
</dd>
<dt>‘<samp>sample_rate, r</samp>’</dt>
<dd><p>Specifies the sample rate, and defaults to 44100.
</p>
</dd>
<dt>‘<samp>nb_samples, n</samp>’</dt>
<dd><p>Set the number of samples per requested frames.
</p>
</dd>
</dl>
<a name="Examples-32"></a>
<h3 class="subsection"><a href="ffmpeg-filters.html#toc-Examples-32">7.3.1 Examples</a></h3>
<ul>
<li>
Set the sample rate to 48000 Hz and the channel layout to AV_CH_LAYOUT_MONO.
<div class="example">
<pre class="example">anullsrc=r=48000:cl=4
</pre></div>
</li><li>
Do the same operation with a more obvious syntax:
<div class="example">
<pre class="example">anullsrc=r=48000:cl=mono
</pre></div>
</li></ul>
<p>All the parameters need to be explicitly defined.
</p>
<a name="flite"></a>
<h2 class="section"><a href="ffmpeg-filters.html#toc-flite">7.4 flite</a></h2>
<p>Synthesize a voice utterance using the libflite library.
</p>
<p>To enable compilation of this filter you need to configure FFmpeg with
<code>--enable-libflite</code>.
</p>
<p>Note that the flite library is not thread-safe.
</p>
<p>The filter accepts the following options:
</p>
<dl compact="compact">
<dt>‘<samp>list_voices</samp>’</dt>
<dd><p>If set to 1, list the names of the available voices and exit
immediately. Default value is 0.
</p>
</dd>
<dt>‘<samp>nb_samples, n</samp>’</dt>
<dd><p>Set the maximum number of samples per frame. Default value is 512.
</p>
</dd>
<dt>‘<samp>textfile</samp>’</dt>
<dd><p>Set the filename containing the text to speak.
</p>
</dd>
<dt>‘<samp>text</samp>’</dt>
<dd><p>Set the text to speak.
</p>
</dd>
<dt>‘<samp>voice, v</samp>’</dt>
<dd><p>Set the voice to use for the speech synthesis. Default value is
<code>kal</code>. See also the <var>list_voices</var> option.
</p></dd>
</dl>
<a name="Examples-72"></a>
<h3 class="subsection"><a href="ffmpeg-filters.html#toc-Examples-72">7.4.1 Examples</a></h3>
<ul>
<li>
Read from file ‘<tt>speech.txt</tt>’, and synthetize the text using the
standard flite voice:
<div class="example">
<pre class="example">flite=textfile=speech.txt
</pre></div>
</li><li>
Read the specified text selecting the <code>slt</code> voice:
<div class="example">
<pre class="example">flite=text='So fare thee well, poor devil of a Sub-Sub, whose commentator I am':voice=slt
</pre></div>
</li><li>
Input text to ffmpeg:
<div class="example">
<pre class="example">ffmpeg -f lavfi -i flite=text='So fare thee well, poor devil of a Sub-Sub, whose commentator I am':voice=slt
</pre></div>
</li><li>
Make ‘<tt>ffplay</tt>’ speak the specified text, using <code>flite</code> and
the <code>lavfi</code> device:
<div class="example">
<pre class="example">ffplay -f lavfi flite=text='No more be grieved for which that thou hast done.'
</pre></div>
</li></ul>
<p>For more information about libflite, check:
<a href="http://www.speech.cs.cmu.edu/flite/">http://www.speech.cs.cmu.edu/flite/</a>
</p>
<a name="sine"></a>
<h2 class="section"><a href="ffmpeg-filters.html#toc-sine">7.5 sine</a></h2>
<p>Generate an audio signal made of a sine wave with amplitude 1/8.
</p>
<p>The audio signal is bit-exact.
</p>
<p>The filter accepts the following options:
</p>
<dl compact="compact">
<dt>‘<samp>frequency, f</samp>’</dt>
<dd><p>Set the carrier frequency. Default is 440 Hz.
</p>
</dd>
<dt>‘<samp>beep_factor, b</samp>’</dt>
<dd><p>Enable a periodic beep every second with frequency <var>beep_factor</var> times
the carrier frequency. Default is 0, meaning the beep is disabled.
</p>
</dd>
<dt>‘<samp>sample_rate, r</samp>’</dt>
<dd><p>Specify the sample rate, default is 44100.
</p>
</dd>
<dt>‘<samp>duration, d</samp>’</dt>
<dd><p>Specify the duration of the generated audio stream.
</p>
</dd>
<dt>‘<samp>samples_per_frame</samp>’</dt>
<dd><p>Set the number of samples per output frame, default is 1024.
</p></dd>
</dl>
<a name="Examples-8"></a>
<h3 class="subsection"><a href="ffmpeg-filters.html#toc-Examples-8">7.5.1 Examples</a></h3>
<ul>
<li>
Generate a simple 440 Hz sine wave:
<div class="example">
<pre class="example">sine
</pre></div>
</li><li>
Generate a 220 Hz sine wave with a 880 Hz beep each second, for 5 seconds:
<div class="example">
<pre class="example">sine=220:4:d=5
sine=f=220:b=4:d=5
sine=frequency=220:beep_factor=4:duration=5
</pre></div>
</li></ul>
<a name="Audio-Sinks"></a>
<h1 class="chapter"><a href="ffmpeg-filters.html#toc-Audio-Sinks">8 Audio Sinks</a></h1>
<p>Below is a description of the currently available audio sinks.
</p>
<a name="abuffersink"></a>
<h2 class="section"><a href="ffmpeg-filters.html#toc-abuffersink">8.1 abuffersink</a></h2>
<p>Buffer audio frames, and make them available to the end of filter chain.
</p>
<p>This sink is mainly intended for programmatic use, in particular
through the interface defined in ‘<tt>libavfilter/buffersink.h</tt>’
or the options system.
</p>
<p>It accepts a pointer to an AVABufferSinkContext structure, which
defines the incoming buffers’ formats, to be passed as the opaque
parameter to <code>avfilter_init_filter</code> for initialization.
</p><a name="anullsink"></a>
<h2 class="section"><a href="ffmpeg-filters.html#toc-anullsink">8.2 anullsink</a></h2>
<p>Null audio sink; do absolutely nothing with the input audio. It is
mainly useful as a template and for use in analysis / debugging
tools.
</p>
<a name="Video-Filters"></a>
<h1 class="chapter"><a href="ffmpeg-filters.html#toc-Video-Filters">9 Video Filters</a></h1>
<p>When you configure your FFmpeg build, you can disable any of the
existing filters using <code>--disable-filters</code>.
The configure output will show the video filters included in your
build.
</p>
<p>Below is a description of the currently available video filters.
</p>
<a name="alphaextract"></a>
<h2 class="section"><a href="ffmpeg-filters.html#toc-alphaextract">9.1 alphaextract</a></h2>
<p>Extract the alpha component from the input as a grayscale video. This
is especially useful with the <var>alphamerge</var> filter.
</p>
<a name="alphamerge"></a>
<h2 class="section"><a href="ffmpeg-filters.html#toc-alphamerge">9.2 alphamerge</a></h2>
<p>Add or replace the alpha component of the primary input with the
grayscale value of a second input. This is intended for use with
<var>alphaextract</var> to allow the transmission or storage of frame
sequences that have alpha in a format that doesn’t support an alpha
channel.
</p>
<p>For example, to reconstruct full frames from a normal YUV-encoded video
and a separate video created with <var>alphaextract</var>, you might use:
</p><div class="example">
<pre class="example">movie=in_alpha.mkv [alpha]; [in][alpha] alphamerge [out]
</pre></div>
<p>Since this filter is designed for reconstruction, it operates on frame
sequences without considering timestamps, and terminates when either
input reaches end of stream. This will cause problems if your encoding
pipeline drops frames. If you’re trying to apply an image as an
overlay to a video stream, consider the <var>overlay</var> filter instead.
</p>
<a name="ass"></a>
<h2 class="section"><a href="ffmpeg-filters.html#toc-ass">9.3 ass</a></h2>
<p>Same as the <a href="#subtitles">subtitles</a> filter, except that it doesn’t require libavcodec
and libavformat to work. On the other hand, it is limited to ASS (Advanced
Substation Alpha) subtitles files.
</p>
<a name="bbox"></a>
<h2 class="section"><a href="ffmpeg-filters.html#toc-bbox">9.4 bbox</a></h2>
<p>Compute the bounding box for the non-black pixels in the input frame
luminance plane.
</p>
<p>This filter computes the bounding box containing all the pixels with a
luminance value greater than the minimum allowed value.
The parameters describing the bounding box are printed on the filter
log.
</p>
<p>The filter accepts the following option:
</p>
<dl compact="compact">
<dt>‘<samp>min_val</samp>’</dt>
<dd><p>Set the minimal luminance value. Default is <code>16</code>.
</p></dd>
</dl>
<a name="blackdetect"></a>
<h2 class="section"><a href="ffmpeg-filters.html#toc-blackdetect">9.5 blackdetect</a></h2>
<p>Detect video intervals that are (almost) completely black. Can be
useful to detect chapter transitions, commercials, or invalid
recordings. Output lines contains the time for the start, end and
duration of the detected black interval expressed in seconds.
</p>
<p>In order to display the output lines, you need to set the loglevel at
least to the AV_LOG_INFO value.
</p>
<p>The filter accepts the following options:
</p>
<dl compact="compact">
<dt>‘<samp>black_min_duration, d</samp>’</dt>
<dd><p>Set the minimum detected black duration expressed in seconds. It must
be a non-negative floating point number.
</p>
<p>Default value is 2.0.
</p>
</dd>
<dt>‘<samp>picture_black_ratio_th, pic_th</samp>’</dt>
<dd><p>Set the threshold for considering a picture "black".
Express the minimum value for the ratio:
</p><div class="example">
<pre class="example"><var>nb_black_pixels</var> / <var>nb_pixels</var>
</pre></div>
<p>for which a picture is considered black.
Default value is 0.98.
</p>
</dd>
<dt>‘<samp>pixel_black_th, pix_th</samp>’</dt>
<dd><p>Set the threshold for considering a pixel "black".
</p>
<p>The threshold expresses the maximum pixel luminance value for which a
pixel is considered "black". The provided value is scaled according to
the following equation:
</p><div class="example">
<pre class="example"><var>absolute_threshold</var> = <var>luminance_minimum_value</var> + <var>pixel_black_th</var> * <var>luminance_range_size</var>
</pre></div>
<p><var>luminance_range_size</var> and <var>luminance_minimum_value</var> depend on
the input video format, the range is [0-255] for YUV full-range
formats and [16-235] for YUV non full-range formats.
</p>
<p>Default value is 0.10.
</p></dd>
</dl>
<p>The following example sets the maximum pixel threshold to the minimum
value, and detects only black intervals of 2 or more seconds:
</p><div class="example">
<pre class="example">blackdetect=d=2:pix_th=0.00
</pre></div>
<a name="blackframe"></a>
<h2 class="section"><a href="ffmpeg-filters.html#toc-blackframe">9.6 blackframe</a></h2>
<p>Detect frames that are (almost) completely black. Can be useful to
detect chapter transitions or commercials. Output lines consist of
the frame number of the detected frame, the percentage of blackness,
the position in the file if known or -1 and the timestamp in seconds.
</p>
<p>In order to display the output lines, you need to set the loglevel at
least to the AV_LOG_INFO value.
</p>
<p>It accepts the following parameters:
</p>
<dl compact="compact">
<dt>‘<samp>amount</samp>’</dt>
<dd><p>The percentage of the pixels that have to be below the threshold; it defaults to
<code>98</code>.
</p>
</dd>
<dt>‘<samp>threshold, thresh</samp>’</dt>
<dd><p>The threshold below which a pixel value is considered black; it defaults to
<code>32</code>.
</p>
</dd>
</dl>
<a name="blend"></a>
<h2 class="section"><a href="ffmpeg-filters.html#toc-blend">9.7 blend</a></h2>
<p>Blend two video frames into each other.
</p>
<p>It takes two input streams and outputs one stream, the first input is the
"top" layer and second input is "bottom" layer.
Output terminates when shortest input terminates.
</p>
<p>A description of the accepted options follows.
</p>
<dl compact="compact">
<dt>‘<samp>c0_mode</samp>’</dt>
<dt>‘<samp>c1_mode</samp>’</dt>
<dt>‘<samp>c2_mode</samp>’</dt>
<dt>‘<samp>c3_mode</samp>’</dt>
<dt>‘<samp>all_mode</samp>’</dt>
<dd><p>Set blend mode for specific pixel component or all pixel components in case
of <var>all_mode</var>. Default value is <code>normal</code>.
</p>
<p>Available values for component modes are:
</p><dl compact="compact">
<dt>‘<samp>addition</samp>’</dt>
<dt>‘<samp>and</samp>’</dt>
<dt>‘<samp>average</samp>’</dt>
<dt>‘<samp>burn</samp>’</dt>
<dt>‘<samp>darken</samp>’</dt>
<dt>‘<samp>difference</samp>’</dt>
<dt>‘<samp>divide</samp>’</dt>
<dt>‘<samp>dodge</samp>’</dt>
<dt>‘<samp>exclusion</samp>’</dt>
<dt>‘<samp>hardlight</samp>’</dt>
<dt>‘<samp>lighten</samp>’</dt>
<dt>‘<samp>multiply</samp>’</dt>
<dt>‘<samp>negation</samp>’</dt>
<dt>‘<samp>normal</samp>’</dt>
<dt>‘<samp>or</samp>’</dt>
<dt>‘<samp>overlay</samp>’</dt>
<dt>‘<samp>phoenix</samp>’</dt>
<dt>‘<samp>pinlight</samp>’</dt>
<dt>‘<samp>reflect</samp>’</dt>
<dt>‘<samp>screen</samp>’</dt>
<dt>‘<samp>softlight</samp>’</dt>
<dt>‘<samp>subtract</samp>’</dt>
<dt>‘<samp>vividlight</samp>’</dt>
<dt>‘<samp>xor</samp>’</dt>
</dl>
</dd>
<dt>‘<samp>c0_opacity</samp>’</dt>
<dt>‘<samp>c1_opacity</samp>’</dt>
<dt>‘<samp>c2_opacity</samp>’</dt>
<dt>‘<samp>c3_opacity</samp>’</dt>
<dt>‘<samp>all_opacity</samp>’</dt>
<dd><p>Set blend opacity for specific pixel component or all pixel components in case
of <var>all_opacity</var>. Only used in combination with pixel component blend modes.
</p>
</dd>
<dt>‘<samp>c0_expr</samp>’</dt>
<dt>‘<samp>c1_expr</samp>’</dt>
<dt>‘<samp>c2_expr</samp>’</dt>
<dt>‘<samp>c3_expr</samp>’</dt>
<dt>‘<samp>all_expr</samp>’</dt>
<dd><p>Set blend expression for specific pixel component or all pixel components in case
of <var>all_expr</var>. Note that related mode options will be ignored if those are set.
</p>
<p>The expressions can use the following variables:
</p>
<dl compact="compact">
<dt>‘<samp>N</samp>’</dt>
<dd><p>The sequential number of the filtered frame, starting from <code>0</code>.
</p>
</dd>
<dt>‘<samp>X</samp>’</dt>
<dt>‘<samp>Y</samp>’</dt>
<dd><p>the coordinates of the current sample
</p>
</dd>
<dt>‘<samp>W</samp>’</dt>
<dt>‘<samp>H</samp>’</dt>
<dd><p>the width and height of currently filtered plane
</p>
</dd>
<dt>‘<samp>SW</samp>’</dt>
<dt>‘<samp>SH</samp>’</dt>
<dd><p>Width and height scale depending on the currently filtered plane. It is the
ratio between the corresponding luma plane number of pixels and the current
plane ones. E.g. for YUV4:2:0 the values are <code>1,1</code> for the luma plane, and
<code>0.5,0.5</code> for chroma planes.
</p>
</dd>
<dt>‘<samp>T</samp>’</dt>
<dd><p>Time of the current frame, expressed in seconds.
</p>
</dd>
<dt>‘<samp>TOP, A</samp>’</dt>
<dd><p>Value of pixel component at current location for first video frame (top layer).
</p>
</dd>
<dt>‘<samp>BOTTOM, B</samp>’</dt>
<dd><p>Value of pixel component at current location for second video frame (bottom layer).
</p></dd>
</dl>
</dd>
<dt>‘<samp>shortest</samp>’</dt>
<dd><p>Force termination when the shortest input terminates. Default is <code>0</code>.
</p></dd>
<dt>‘<samp>repeatlast</samp>’</dt>
<dd><p>Continue applying the last bottom frame after the end of the stream. A value of
<code>0</code> disable the filter after the last frame of the bottom layer is reached.
Default is <code>1</code>.
</p></dd>
</dl>
<a name="Examples-78"></a>
<h3 class="subsection"><a href="ffmpeg-filters.html#toc-Examples-78">9.7.1 Examples</a></h3>
<ul>
<li>
Apply transition from bottom layer to top layer in first 10 seconds:
<div class="example">
<pre class="example">blend=all_expr='A*(if(gte(T,10),1,T/10))+B*(1-(if(gte(T,10),1,T/10)))'
</pre></div>
</li><li>
Apply 1x1 checkerboard effect:
<div class="example">
<pre class="example">blend=all_expr='if(eq(mod(X,2),mod(Y,2)),A,B)'
</pre></div>
</li><li>
Apply uncover left effect:
<div class="example">
<pre class="example">blend=all_expr='if(gte(N*SW+X,W),A,B)'
</pre></div>
</li><li>
Apply uncover down effect:
<div class="example">
<pre class="example">blend=all_expr='if(gte(Y-N*SH,0),A,B)'
</pre></div>
</li><li>
Apply uncover up-left effect:
<div class="example">
<pre class="example">blend=all_expr='if(gte(T*SH*40+Y,H)*gte((T*40*SW+X)*W/H,W),A,B)'
</pre></div>
</li></ul>
<a name="boxblur"></a>
<h2 class="section"><a href="ffmpeg-filters.html#toc-boxblur">9.8 boxblur</a></h2>
<p>Apply a boxblur algorithm to the input video.
</p>
<p>It accepts the following parameters:
</p>
<dl compact="compact">
<dt>‘<samp>luma_radius, lr</samp>’</dt>
<dt>‘<samp>luma_power, lp</samp>’</dt>
<dt>‘<samp>chroma_radius, cr</samp>’</dt>
<dt>‘<samp>chroma_power, cp</samp>’</dt>
<dt>‘<samp>alpha_radius, ar</samp>’</dt>
<dt>‘<samp>alpha_power, ap</samp>’</dt>
</dl>
<p>A description of the accepted options follows.
</p>
<dl compact="compact">
<dt>‘<samp>luma_radius, lr</samp>’</dt>
<dt>‘<samp>chroma_radius, cr</samp>’</dt>
<dt>‘<samp>alpha_radius, ar</samp>’</dt>
<dd><p>Set an expression for the box radius in pixels used for blurring the
corresponding input plane.
</p>
<p>The radius value must be a non-negative number, and must not be
greater than the value of the expression <code>min(w,h)/2</code> for the
luma and alpha planes, and of <code>min(cw,ch)/2</code> for the chroma
planes.
</p>
<p>Default value for ‘<samp>luma_radius</samp>’ is "2". If not specified,
‘<samp>chroma_radius</samp>’ and ‘<samp>alpha_radius</samp>’ default to the
corresponding value set for ‘<samp>luma_radius</samp>’.
</p>
<p>The expressions can contain the following constants:
</p><dl compact="compact">
<dt>‘<samp>w</samp>’</dt>
<dt>‘<samp>h</samp>’</dt>
<dd><p>The input width and height in pixels.
</p>
</dd>
<dt>‘<samp>cw</samp>’</dt>
<dt>‘<samp>ch</samp>’</dt>
<dd><p>The input chroma image width and height in pixels.
</p>
</dd>
<dt>‘<samp>hsub</samp>’</dt>
<dt>‘<samp>vsub</samp>’</dt>
<dd><p>The horizontal and vertical chroma subsample values. For example, for the
pixel format "yuv422p", <var>hsub</var> is 2 and <var>vsub</var> is 1.
</p></dd>
</dl>
</dd>
<dt>‘<samp>luma_power, lp</samp>’</dt>
<dt>‘<samp>chroma_power, cp</samp>’</dt>
<dt>‘<samp>alpha_power, ap</samp>’</dt>
<dd><p>Specify how many times the boxblur filter is applied to the
corresponding plane.
</p>
<p>Default value for ‘<samp>luma_power</samp>’ is 2. If not specified,
‘<samp>chroma_power</samp>’ and ‘<samp>alpha_power</samp>’ default to the
corresponding value set for ‘<samp>luma_power</samp>’.
</p>
<p>A value of 0 will disable the effect.
</p></dd>
</dl>
<a name="Examples-81"></a>
<h3 class="subsection"><a href="ffmpeg-filters.html#toc-Examples-81">9.8.1 Examples</a></h3>
<ul>
<li>
Apply a boxblur filter with the luma, chroma, and alpha radii
set to 2:
<div class="example">
<pre class="example">boxblur=luma_radius=2:luma_power=1
boxblur=2:1
</pre></div>
</li><li>
Set the luma radius to 2, and alpha and chroma radius to 0:
<div class="example">
<pre class="example">boxblur=2:1:cr=0:ar=0
</pre></div>
</li><li>
Set the luma and chroma radii to a fraction of the video dimension:
<div class="example">
<pre class="example">boxblur=luma_radius=min(h\,w)/10:luma_power=1:chroma_radius=min(cw\,ch)/10:chroma_power=1
</pre></div>
</li></ul>
<a name="codecview"></a>
<h2 class="section"><a href="ffmpeg-filters.html#toc-codecview">9.9 codecview</a></h2>
<p>Visualize information exported by some codecs.
</p>
<p>Some codecs can export information through frames using side-data or other
means. For example, some MPEG based codecs export motion vectors through the
<var>export_mvs</var> flag in the codec ‘<samp>flags2</samp>’ option.
</p>
<p>The filter accepts the following option:
</p>
<dl compact="compact">
<dt>‘<samp>mv</samp>’</dt>
<dd><p>Set motion vectors to visualize.
</p>
<p>Available flags for <var>mv</var> are:
</p>
<dl compact="compact">
<dt>‘<samp>pf</samp>’</dt>
<dd><p>forward predicted MVs of P-frames
</p></dd>
<dt>‘<samp>bf</samp>’</dt>
<dd><p>forward predicted MVs of B-frames
</p></dd>
<dt>‘<samp>bb</samp>’</dt>
<dd><p>backward predicted MVs of B-frames
</p></dd>
</dl>
</dd>
</dl>
<a name="Examples-6"></a>
<h3 class="subsection"><a href="ffmpeg-filters.html#toc-Examples-6">9.9.1 Examples</a></h3>
<ul>
<li>
Visualizes multi-directionals MVs from P and B-Frames using <code>ffplay</code>:
<div class="example">
<pre class="example">ffplay -flags2 +export_mvs input.mpg -vf codecview=mv=pf+bf+bb
</pre></div>
</li></ul>
<a name="colorbalance"></a>
<h2 class="section"><a href="ffmpeg-filters.html#toc-colorbalance">9.10 colorbalance</a></h2>
<p>Modify intensity of primary colors (red, green and blue) of input frames.
</p>
<p>The filter allows an input frame to be adjusted in the shadows, midtones or highlights
regions for the red-cyan, green-magenta or blue-yellow balance.
</p>
<p>A positive adjustment value shifts the balance towards the primary color, a negative
value towards the complementary color.
</p>
<p>The filter accepts the following options:
</p>
<dl compact="compact">
<dt>‘<samp>rs</samp>’</dt>
<dt>‘<samp>gs</samp>’</dt>
<dt>‘<samp>bs</samp>’</dt>
<dd><p>Adjust red, green and blue shadows (darkest pixels).
</p>
</dd>
<dt>‘<samp>rm</samp>’</dt>
<dt>‘<samp>gm</samp>’</dt>
<dt>‘<samp>bm</samp>’</dt>
<dd><p>Adjust red, green and blue midtones (medium pixels).
</p>
</dd>
<dt>‘<samp>rh</samp>’</dt>
<dt>‘<samp>gh</samp>’</dt>
<dt>‘<samp>bh</samp>’</dt>
<dd><p>Adjust red, green and blue highlights (brightest pixels).
</p>
<p>Allowed ranges for options are <code>[-1.0, 1.0]</code>. Defaults are <code>0</code>.
</p></dd>
</dl>
<a name="Examples-77"></a>
<h3 class="subsection"><a href="ffmpeg-filters.html#toc-Examples-77">9.10.1 Examples</a></h3>
<ul>
<li>
Add red color cast to shadows:
<div class="example">
<pre class="example">colorbalance=rs=.3
</pre></div>
</li></ul>
<a name="colorchannelmixer"></a>
<h2 class="section"><a href="ffmpeg-filters.html#toc-colorchannelmixer">9.11 colorchannelmixer</a></h2>
<p>Adjust video input frames by re-mixing color channels.
</p>
<p>This filter modifies a color channel by adding the values associated to
the other channels of the same pixels. For example if the value to
modify is red, the output value will be:
</p><div class="example">
<pre class="example"><var>red</var>=<var>red</var>*<var>rr</var> + <var>blue</var>*<var>rb</var> + <var>green</var>*<var>rg</var> + <var>alpha</var>*<var>ra</var>
</pre></div>
<p>The filter accepts the following options:
</p>
<dl compact="compact">
<dt>‘<samp>rr</samp>’</dt>
<dt>‘<samp>rg</samp>’</dt>
<dt>‘<samp>rb</samp>’</dt>
<dt>‘<samp>ra</samp>’</dt>
<dd><p>Adjust contribution of input red, green, blue and alpha channels for output red channel.
Default is <code>1</code> for <var>rr</var>, and <code>0</code> for <var>rg</var>, <var>rb</var> and <var>ra</var>.
</p>
</dd>
<dt>‘<samp>gr</samp>’</dt>
<dt>‘<samp>gg</samp>’</dt>
<dt>‘<samp>gb</samp>’</dt>
<dt>‘<samp>ga</samp>’</dt>
<dd><p>Adjust contribution of input red, green, blue and alpha channels for output green channel.
Default is <code>1</code> for <var>gg</var>, and <code>0</code> for <var>gr</var>, <var>gb</var> and <var>ga</var>.
</p>
</dd>
<dt>‘<samp>br</samp>’</dt>
<dt>‘<samp>bg</samp>’</dt>
<dt>‘<samp>bb</samp>’</dt>
<dt>‘<samp>ba</samp>’</dt>
<dd><p>Adjust contribution of input red, green, blue and alpha channels for output blue channel.
Default is <code>1</code> for <var>bb</var>, and <code>0</code> for <var>br</var>, <var>bg</var> and <var>ba</var>.
</p>
</dd>
<dt>‘<samp>ar</samp>’</dt>
<dt>‘<samp>ag</samp>’</dt>
<dt>‘<samp>ab</samp>’</dt>
<dt>‘<samp>aa</samp>’</dt>
<dd><p>Adjust contribution of input red, green, blue and alpha channels for output alpha channel.
Default is <code>1</code> for <var>aa</var>, and <code>0</code> for <var>ar</var>, <var>ag</var> and <var>ab</var>.
</p>
<p>Allowed ranges for options are <code>[-2.0, 2.0]</code>.
</p></dd>
</dl>
<a name="Examples-56"></a>
<h3 class="subsection"><a href="ffmpeg-filters.html#toc-Examples-56">9.11.1 Examples</a></h3>
<ul>
<li>
Convert source to grayscale:
<div class="example">
<pre class="example">colorchannelmixer=.3:.4:.3:0:.3:.4:.3:0:.3:.4:.3
</pre></div>
</li><li>
Simulate sepia tones:
<div class="example">
<pre class="example">colorchannelmixer=.393:.769:.189:0:.349:.686:.168:0:.272:.534:.131
</pre></div>
</li></ul>
<a name="colormatrix"></a>
<h2 class="section"><a href="ffmpeg-filters.html#toc-colormatrix">9.12 colormatrix</a></h2>
<p>Convert color matrix.
</p>
<p>The filter accepts the following options:
</p>
<dl compact="compact">
<dt>‘<samp>src</samp>’</dt>
<dt>‘<samp>dst</samp>’</dt>
<dd><p>Specify the source and destination color matrix. Both values must be
specified.
</p>
<p>The accepted values are:
</p><dl compact="compact">
<dt>‘<samp>bt709</samp>’</dt>
<dd><p>BT.709
</p>
</dd>
<dt>‘<samp>bt601</samp>’</dt>
<dd><p>BT.601
</p>
</dd>
<dt>‘<samp>smpte240m</samp>’</dt>
<dd><p>SMPTE-240M
</p>
</dd>
<dt>‘<samp>fcc</samp>’</dt>
<dd><p>FCC
</p></dd>
</dl>
</dd>
</dl>
<p>For example to convert from BT.601 to SMPTE-240M, use the command:
</p><div class="example">
<pre class="example">colormatrix=bt601:smpte240m
</pre></div>
<a name="copy"></a>
<h2 class="section"><a href="ffmpeg-filters.html#toc-copy">9.13 copy</a></h2>
<p>Copy the input source unchanged to the output. This is mainly useful for
testing purposes.
</p>
<a name="crop"></a>
<h2 class="section"><a href="ffmpeg-filters.html#toc-crop">9.14 crop</a></h2>
<p>Crop the input video to given dimensions.
</p>
<p>It accepts the following parameters:
</p>
<dl compact="compact">
<dt>‘<samp>w, out_w</samp>’</dt>
<dd><p>The width of the output video. It defaults to <code>iw</code>.
This expression is evaluated only once during the filter
configuration.
</p>
</dd>
<dt>‘<samp>h, out_h</samp>’</dt>
<dd><p>The height of the output video. It defaults to <code>ih</code>.
This expression is evaluated only once during the filter
configuration.
</p>
</dd>
<dt>‘<samp>x</samp>’</dt>
<dd><p>The horizontal position, in the input video, of the left edge of the output
video. It defaults to <code>(in_w-out_w)/2</code>.
This expression is evaluated per-frame.
</p>
</dd>
<dt>‘<samp>y</samp>’</dt>
<dd><p>The vertical position, in the input video, of the top edge of the output video.
It defaults to <code>(in_h-out_h)/2</code>.
This expression is evaluated per-frame.
</p>
</dd>
<dt>‘<samp>keep_aspect</samp>’</dt>
<dd><p>If set to 1 will force the output display aspect ratio
to be the same of the input, by changing the output sample aspect
ratio. It defaults to 0.
</p></dd>
</dl>
<p>The <var>out_w</var>, <var>out_h</var>, <var>x</var>, <var>y</var> parameters are
expressions containing the following constants:
</p>
<dl compact="compact">
<dt>‘<samp>x</samp>’</dt>
<dt>‘<samp>y</samp>’</dt>
<dd><p>The computed values for <var>x</var> and <var>y</var>. They are evaluated for
each new frame.
</p>
</dd>
<dt>‘<samp>in_w</samp>’</dt>
<dt>‘<samp>in_h</samp>’</dt>
<dd><p>The input width and height.
</p>
</dd>
<dt>‘<samp>iw</samp>’</dt>
<dt>‘<samp>ih</samp>’</dt>
<dd><p>These are the same as <var>in_w</var> and <var>in_h</var>.
</p>
</dd>
<dt>‘<samp>out_w</samp>’</dt>
<dt>‘<samp>out_h</samp>’</dt>
<dd><p>The output (cropped) width and height.
</p>
</dd>
<dt>‘<samp>ow</samp>’</dt>
<dt>‘<samp>oh</samp>’</dt>
<dd><p>These are the same as <var>out_w</var> and <var>out_h</var>.
</p>
</dd>
<dt>‘<samp>a</samp>’</dt>
<dd><p>same as <var>iw</var> / <var>ih</var>
</p>
</dd>
<dt>‘<samp>sar</samp>’</dt>
<dd><p>input sample aspect ratio
</p>
</dd>
<dt>‘<samp>dar</samp>’</dt>
<dd><p>input display aspect ratio, it is the same as (<var>iw</var> / <var>ih</var>) * <var>sar</var>
</p>
</dd>
<dt>‘<samp>hsub</samp>’</dt>
<dt>‘<samp>vsub</samp>’</dt>
<dd><p>horizontal and vertical chroma subsample values. For example for the
pixel format "yuv422p" <var>hsub</var> is 2 and <var>vsub</var> is 1.
</p>
</dd>
<dt>‘<samp>n</samp>’</dt>
<dd><p>The number of the input frame, starting from 0.
</p>
</dd>
<dt>‘<samp>pos</samp>’</dt>
<dd><p>the position in the file of the input frame, NAN if unknown
</p>
</dd>
<dt>‘<samp>t</samp>’</dt>
<dd><p>The timestamp expressed in seconds. It’s NAN if the input timestamp is unknown.
</p>
</dd>
</dl>
<p>The expression for <var>out_w</var> may depend on the value of <var>out_h</var>,
and the expression for <var>out_h</var> may depend on <var>out_w</var>, but they
cannot depend on <var>x</var> and <var>y</var>, as <var>x</var> and <var>y</var> are
evaluated after <var>out_w</var> and <var>out_h</var>.
</p>
<p>The <var>x</var> and <var>y</var> parameters specify the expressions for the
position of the top-left corner of the output (non-cropped) area. They
are evaluated for each frame. If the evaluated value is not valid, it
is approximated to the nearest valid value.
</p>
<p>The expression for <var>x</var> may depend on <var>y</var>, and the expression
for <var>y</var> may depend on <var>x</var>.
</p>
<a name="Examples-10"></a>
<h3 class="subsection"><a href="ffmpeg-filters.html#toc-Examples-10">9.14.1 Examples</a></h3>
<ul>
<li>
Crop area with size 100x100 at position (12,34).
<div class="example">
<pre class="example">crop=100:100:12:34
</pre></div>
<p>Using named options, the example above becomes:
</p><div class="example">
<pre class="example">crop=w=100:h=100:x=12:y=34
</pre></div>
</li><li>
Crop the central input area with size 100x100:
<div class="example">
<pre class="example">crop=100:100
</pre></div>
</li><li>
Crop the central input area with size 2/3 of the input video:
<div class="example">
<pre class="example">crop=2/3*in_w:2/3*in_h
</pre></div>
</li><li>
Crop the input video central square:
<div class="example">
<pre class="example">crop=out_w=in_h
crop=in_h
</pre></div>
</li><li>
Delimit the rectangle with the top-left corner placed at position
100:100 and the right-bottom corner corresponding to the right-bottom
corner of the input image.
<div class="example">
<pre class="example">crop=in_w-100:in_h-100:100:100
</pre></div>
</li><li>
Crop 10 pixels from the left and right borders, and 20 pixels from
the top and bottom borders
<div class="example">
<pre class="example">crop=in_w-2*10:in_h-2*20
</pre></div>
</li><li>
Keep only the bottom right quarter of the input image:
<div class="example">
<pre class="example">crop=in_w/2:in_h/2:in_w/2:in_h/2
</pre></div>
</li><li>
Crop height for getting Greek harmony:
<div class="example">
<pre class="example">crop=in_w:1/PHI*in_w
</pre></div>
</li><li>
Appply trembling effect:
<div class="example">
<pre class="example">crop=in_w/2:in_h/2:(in_w-out_w)/2+((in_w-out_w)/2)*sin(n/10):(in_h-out_h)/2 +((in_h-out_h)/2)*sin(n/7)
</pre></div>
</li><li>
Apply erratic camera effect depending on timestamp:
<div class="example">
<pre class="example">crop=in_w/2:in_h/2:(in_w-out_w)/2+((in_w-out_w)/2)*sin(t*10):(in_h-out_h)/2 +((in_h-out_h)/2)*sin(t*13)"
</pre></div>
</li><li>
Set x depending on the value of y:
<div class="example">
<pre class="example">crop=in_w/2:in_h/2:y:10+10*sin(n/10)
</pre></div>
</li></ul>
<a name="cropdetect"></a>
<h2 class="section"><a href="ffmpeg-filters.html#toc-cropdetect">9.15 cropdetect</a></h2>
<p>Auto-detect the crop size.
</p>
<p>It calculates the necessary cropping parameters and prints the
recommended parameters via the logging system. The detected dimensions
correspond to the non-black area of the input video.
</p>
<p>It accepts the following parameters:
</p>
<dl compact="compact">
<dt>‘<samp>limit</samp>’</dt>
<dd><p>Set higher black value threshold, which can be optionally specified
from nothing (0) to everything (255). An intensity value greater
to the set value is considered non-black. It defaults to 24.
</p>
</dd>
<dt>‘<samp>round</samp>’</dt>
<dd><p>The value which the width/height should be divisible by. It defaults to
16. The offset is automatically adjusted to center the video. Use 2 to
get only even dimensions (needed for 4:2:2 video). 16 is best when
encoding to most video codecs.
</p>
</dd>
<dt>‘<samp>reset_count, reset</samp>’</dt>
<dd><p>Set the counter that determines after how many frames cropdetect will
reset the previously detected largest video area and start over to
detect the current optimal crop area. Default value is 0.
</p>
<p>This can be useful when channel logos distort the video area. 0
indicates ’never reset’, and returns the largest area encountered during
playback.
</p></dd>
</dl>
<p><a name="curves"></a>
</p><a name="curves-1"></a>
<h2 class="section"><a href="ffmpeg-filters.html#toc-curves-1">9.16 curves</a></h2>
<p>Apply color adjustments using curves.
</p>
<p>This filter is similar to the Adobe Photoshop and GIMP curves tools. Each
component (red, green and blue) has its values defined by <var>N</var> key points
tied from each other using a smooth curve. The x-axis represents the pixel
values from the input frame, and the y-axis the new pixel values to be set for
the output frame.
</p>
<p>By default, a component curve is defined by the two points <var>(0;0)</var> and
<var>(1;1)</var>. This creates a straight line where each original pixel value is
"adjusted" to its own value, which means no change to the image.
</p>
<p>The filter allows you to redefine these two points and add some more. A new
curve (using a natural cubic spline interpolation) will be define to pass
smoothly through all these new coordinates. The new defined points needs to be
strictly increasing over the x-axis, and their <var>x</var> and <var>y</var> values must
be in the <var>[0;1]</var> interval. If the computed curves happened to go outside
the vector spaces, the values will be clipped accordingly.
</p>
<p>If there is no key point defined in <code>x=0</code>, the filter will automatically
insert a <var>(0;0)</var> point. In the same way, if there is no key point defined
in <code>x=1</code>, the filter will automatically insert a <var>(1;1)</var> point.
</p>
<p>The filter accepts the following options:
</p>
<dl compact="compact">
<dt>‘<samp>preset</samp>’</dt>
<dd><p>Select one of the available color presets. This option can be used in addition
to the ‘<samp>r</samp>’, ‘<samp>g</samp>’, ‘<samp>b</samp>’ parameters; in this case, the later
options takes priority on the preset values.
Available presets are:
</p><dl compact="compact">
<dt>‘<samp>none</samp>’</dt>
<dt>‘<samp>color_negative</samp>’</dt>
<dt>‘<samp>cross_process</samp>’</dt>
<dt>‘<samp>darker</samp>’</dt>
<dt>‘<samp>increase_contrast</samp>’</dt>
<dt>‘<samp>lighter</samp>’</dt>
<dt>‘<samp>linear_contrast</samp>’</dt>
<dt>‘<samp>medium_contrast</samp>’</dt>
<dt>‘<samp>negative</samp>’</dt>
<dt>‘<samp>strong_contrast</samp>’</dt>
<dt>‘<samp>vintage</samp>’</dt>
</dl>
<p>Default is <code>none</code>.
</p></dd>
<dt>‘<samp>master, m</samp>’</dt>
<dd><p>Set the master key points. These points will define a second pass mapping. It
is sometimes called a "luminance" or "value" mapping. It can be used with
‘<samp>r</samp>’, ‘<samp>g</samp>’, ‘<samp>b</samp>’ or ‘<samp>all</samp>’ since it acts like a
post-processing LUT.
</p></dd>
<dt>‘<samp>red, r</samp>’</dt>
<dd><p>Set the key points for the red component.
</p></dd>
<dt>‘<samp>green, g</samp>’</dt>
<dd><p>Set the key points for the green component.
</p></dd>
<dt>‘<samp>blue, b</samp>’</dt>
<dd><p>Set the key points for the blue component.
</p></dd>
<dt>‘<samp>all</samp>’</dt>
<dd><p>Set the key points for all components (not including master).
Can be used in addition to the other key points component
options. In this case, the unset component(s) will fallback on this
‘<samp>all</samp>’ setting.
</p></dd>
<dt>‘<samp>psfile</samp>’</dt>
<dd><p>Specify a Photoshop curves file (<code>.asv</code>) to import the settings from.
</p></dd>
</dl>
<p>To avoid some filtergraph syntax conflicts, each key points list need to be
defined using the following syntax: <code>x0/y0 x1/y1 x2/y2 ...</code>.
</p>
<a name="Examples-20"></a>
<h3 class="subsection"><a href="ffmpeg-filters.html#toc-Examples-20">9.16.1 Examples</a></h3>
<ul>
<li>
Increase slightly the middle level of blue:
<div class="example">
<pre class="example">curves=blue='0.5/0.58'
</pre></div>
</li><li>
Vintage effect:
<div class="example">
<pre class="example">curves=r='0/0.11 .42/.51 1/0.95':g='0.50/0.48':b='0/0.22 .49/.44 1/0.8'
</pre></div>
<p>Here we obtain the following coordinates for each components:
</p><dl compact="compact">
<dt><var>red</var></dt>
<dd><p><code>(0;0.11) (0.42;0.51) (1;0.95)</code>
</p></dd>
<dt><var>green</var></dt>
<dd><p><code>(0;0) (0.50;0.48) (1;1)</code>
</p></dd>
<dt><var>blue</var></dt>
<dd><p><code>(0;0.22) (0.49;0.44) (1;0.80)</code>
</p></dd>
</dl>
</li><li>
The previous example can also be achieved with the associated built-in preset:
<div class="example">
<pre class="example">curves=preset=vintage
</pre></div>
</li><li>
Or simply:
<div class="example">
<pre class="example">curves=vintage
</pre></div>
</li><li>
Use a Photoshop preset and redefine the points of the green component:
<div class="example">
<pre class="example">curves=psfile='MyCurvesPresets/purple.asv':green='0.45/0.53'
</pre></div>
</li></ul>
<a name="dctdnoiz"></a>
<h2 class="section"><a href="ffmpeg-filters.html#toc-dctdnoiz">9.17 dctdnoiz</a></h2>
<p>Denoise frames using 2D DCT (frequency domain filtering).
</p>
<p>This filter is not designed for real time.
</p>
<p>The filter accepts the following options:
</p>
<dl compact="compact">
<dt>‘<samp>sigma, s</samp>’</dt>
<dd><p>Set the noise sigma constant.
</p>
<p>This <var>sigma</var> defines a hard threshold of <code>3 * sigma</code>; every DCT
coefficient (absolute value) below this threshold with be dropped.
</p>
<p>If you need a more advanced filtering, see ‘<samp>expr</samp>’.
</p>
<p>Default is <code>0</code>.
</p>
</dd>
<dt>‘<samp>overlap</samp>’</dt>
<dd><p>Set number overlapping pixels for each block. Since the filter can be slow, you
may want to reduce this value, at the cost of a less effective filter and the
risk of various artefacts.
</p>
<p>If the overlapping value doesn’t permit processing the whole input width or
height, a warning will be displayed and according borders won’t be denoised.
</p>
<p>Default value is <var>blocksize</var>-1, which is the best possible setting.
</p>
</dd>
<dt>‘<samp>expr, e</samp>’</dt>
<dd><p>Set the coefficient factor expression.
</p>
<p>For each coefficient of a DCT block, this expression will be evaluated as a
multiplier value for the coefficient.
</p>
<p>If this is option is set, the ‘<samp>sigma</samp>’ option will be ignored.
</p>
<p>The absolute value of the coefficient can be accessed through the <var>c</var>
variable.
</p>
</dd>
<dt>‘<samp>n</samp>’</dt>
<dd><p>Set the <var>blocksize</var> using the number of bits. <code>1<<<var>n</var></code> defines the
<var>blocksize</var>, which is the width and height of the processed blocks.
</p>
<p>The default value is <var>3</var> (8x8) and can be raised to <var>4</var> for a
<var>blocksize</var> of 16x16. Note that changing this setting has huge consequences
on the speed processing. Also, a larger block size does not necessarily means a
better de-noising.
</p></dd>
</dl>
<a name="Examples-59"></a>
<h3 class="subsection"><a href="ffmpeg-filters.html#toc-Examples-59">9.17.1 Examples</a></h3>
<p>Apply a denoise with a ‘<samp>sigma</samp>’ of <code>4.5</code>:
</p><div class="example">
<pre class="example">dctdnoiz=4.5
</pre></div>
<p>The same operation can be achieved using the expression system:
</p><div class="example">
<pre class="example">dctdnoiz=e='gte(c, 4.5*3)'
</pre></div>
<p>Violent denoise using a block size of <code>16x16</code>:
</p><div class="example">
<pre class="example">dctdnoiz=15:n=4
</pre></div>
<p><a name="decimate"></a>
</p><a name="decimate-1"></a>
<h2 class="section"><a href="ffmpeg-filters.html#toc-decimate-1">9.18 decimate</a></h2>
<p>Drop duplicated frames at regular intervals.
</p>
<p>The filter accepts the following options:
</p>
<dl compact="compact">
<dt>‘<samp>cycle</samp>’</dt>
<dd><p>Set the number of frames from which one will be dropped. Setting this to
<var>N</var> means one frame in every batch of <var>N</var> frames will be dropped.
Default is <code>5</code>.
</p>
</dd>
<dt>‘<samp>dupthresh</samp>’</dt>
<dd><p>Set the threshold for duplicate detection. If the difference metric for a frame
is less than or equal to this value, then it is declared as duplicate. Default
is <code>1.1</code>
</p>
</dd>
<dt>‘<samp>scthresh</samp>’</dt>
<dd><p>Set scene change threshold. Default is <code>15</code>.
</p>
</dd>
<dt>‘<samp>blockx</samp>’</dt>
<dt>‘<samp>blocky</samp>’</dt>
<dd><p>Set the size of the x and y-axis blocks used during metric calculations.
Larger blocks give better noise suppression, but also give worse detection of
small movements. Must be a power of two. Default is <code>32</code>.
</p>
</dd>
<dt>‘<samp>ppsrc</samp>’</dt>
<dd><p>Mark main input as a pre-processed input and activate clean source input
stream. This allows the input to be pre-processed with various filters to help
the metrics calculation while keeping the frame selection lossless. When set to
<code>1</code>, the first stream is for the pre-processed input, and the second
stream is the clean source from where the kept frames are chosen. Default is
<code>0</code>.
</p>
</dd>
<dt>‘<samp>chroma</samp>’</dt>
<dd><p>Set whether or not chroma is considered in the metric calculations. Default is
<code>1</code>.
</p></dd>
</dl>
<a name="dejudder"></a>
<h2 class="section"><a href="ffmpeg-filters.html#toc-dejudder">9.19 dejudder</a></h2>
<p>Remove judder produced by partially interlaced telecined content.
</p>
<p>Judder can be introduced, for instance, by <a href="#pullup">pullup</a> filter. If the original
source was partially telecined content then the output of <code>pullup,dejudder</code>
will have a variable frame rate. May change the recorded frame rate of the
container. Aside from that change, this filter will not affect constant frame
rate video.
</p>
<p>The option available in this filter is:
</p><dl compact="compact">
<dt>‘<samp>cycle</samp>’</dt>
<dd><p>Specify the length of the window over which the judder repeats.
</p>
<p>Accepts any integer greater than 1. Useful values are:
</p><dl compact="compact">
<dt>‘<samp>4</samp>’</dt>
<dd><p>If the original was telecined from 24 to 30 fps (Film to NTSC).
</p>
</dd>
<dt>‘<samp>5</samp>’</dt>
<dd><p>If the original was telecined from 25 to 30 fps (PAL to NTSC).
</p>
</dd>
<dt>‘<samp>20</samp>’</dt>
<dd><p>If a mixture of the two.
</p></dd>
</dl>
<p>The default is ‘<samp>4</samp>’.
</p></dd>
</dl>
<a name="delogo"></a>
<h2 class="section"><a href="ffmpeg-filters.html#toc-delogo">9.20 delogo</a></h2>
<p>Suppress a TV station logo by a simple interpolation of the surrounding
pixels. Just set a rectangle covering the logo and watch it disappear
(and sometimes something even uglier appear - your mileage may vary).
</p>
<p>It accepts the following parameters:
</p><dl compact="compact">
<dt>‘<samp>x</samp>’</dt>
<dt>‘<samp>y</samp>’</dt>
<dd><p>Specify the top left corner coordinates of the logo. They must be
specified.
</p>
</dd>
<dt>‘<samp>w</samp>’</dt>
<dt>‘<samp>h</samp>’</dt>
<dd><p>Specify the width and height of the logo to clear. They must be
specified.
</p>
</dd>
<dt>‘<samp>band, t</samp>’</dt>
<dd><p>Specify the thickness of the fuzzy edge of the rectangle (added to
<var>w</var> and <var>h</var>). The default value is 4.
</p>
</dd>
<dt>‘<samp>show</samp>’</dt>
<dd><p>When set to 1, a green rectangle is drawn on the screen to simplify
finding the right <var>x</var>, <var>y</var>, <var>w</var>, and <var>h</var> parameters.
The default value is 0.
</p>
<p>The rectangle is drawn on the outermost pixels which will be (partly)
replaced with interpolated values. The values of the next pixels
immediately outside this rectangle in each direction will be used to
compute the interpolated pixel values inside the rectangle.
</p>
</dd>
</dl>
<a name="Examples-75"></a>
<h3 class="subsection"><a href="ffmpeg-filters.html#toc-Examples-75">9.20.1 Examples</a></h3>
<ul>
<li>
Set a rectangle covering the area with top left corner coordinates 0,0
and size 100x77, and a band of size 10:
<div class="example">
<pre class="example">delogo=x=0:y=0:w=100:h=77:band=10
</pre></div>
</li></ul>
<a name="deshake"></a>
<h2 class="section"><a href="ffmpeg-filters.html#toc-deshake">9.21 deshake</a></h2>
<p>Attempt to fix small changes in horizontal and/or vertical shift. This
filter helps remove camera shake from hand-holding a camera, bumping a
tripod, moving on a vehicle, etc.
</p>
<p>The filter accepts the following options:
</p>
<dl compact="compact">
<dt>‘<samp>x</samp>’</dt>
<dt>‘<samp>y</samp>’</dt>
<dt>‘<samp>w</samp>’</dt>
<dt>‘<samp>h</samp>’</dt>
<dd><p>Specify a rectangular area where to limit the search for motion
vectors.
If desired the search for motion vectors can be limited to a
rectangular area of the frame defined by its top left corner, width
and height. These parameters have the same meaning as the drawbox
filter which can be used to visualise the position of the bounding
box.
</p>
<p>This is useful when simultaneous movement of subjects within the frame
might be confused for camera motion by the motion vector search.
</p>
<p>If any or all of <var>x</var>, <var>y</var>, <var>w</var> and <var>h</var> are set to -1
then the full frame is used. This allows later options to be set
without specifying the bounding box for the motion vector search.
</p>
<p>Default - search the whole frame.
</p>
</dd>
<dt>‘<samp>rx</samp>’</dt>
<dt>‘<samp>ry</samp>’</dt>
<dd><p>Specify the maximum extent of movement in x and y directions in the
range 0-64 pixels. Default 16.
</p>
</dd>
<dt>‘<samp>edge</samp>’</dt>
<dd><p>Specify how to generate pixels to fill blanks at the edge of the
frame. Available values are:
</p><dl compact="compact">
<dt>‘<samp>blank, 0</samp>’</dt>
<dd><p>Fill zeroes at blank locations
</p></dd>
<dt>‘<samp>original, 1</samp>’</dt>
<dd><p>Original image at blank locations
</p></dd>
<dt>‘<samp>clamp, 2</samp>’</dt>
<dd><p>Extruded edge value at blank locations
</p></dd>
<dt>‘<samp>mirror, 3</samp>’</dt>
<dd><p>Mirrored edge at blank locations
</p></dd>
</dl>
<p>Default value is ‘<samp>mirror</samp>’.
</p>
</dd>
<dt>‘<samp>blocksize</samp>’</dt>
<dd><p>Specify the blocksize to use for motion search. Range 4-128 pixels,
default 8.
</p>
</dd>
<dt>‘<samp>contrast</samp>’</dt>
<dd><p>Specify the contrast threshold for blocks. Only blocks with more than
the specified contrast (difference between darkest and lightest
pixels) will be considered. Range 1-255, default 125.
</p>
</dd>
<dt>‘<samp>search</samp>’</dt>
<dd><p>Specify the search strategy. Available values are:
</p><dl compact="compact">
<dt>‘<samp>exhaustive, 0</samp>’</dt>
<dd><p>Set exhaustive search
</p></dd>
<dt>‘<samp>less, 1</samp>’</dt>
<dd><p>Set less exhaustive search.
</p></dd>
</dl>
<p>Default value is ‘<samp>exhaustive</samp>’.
</p>
</dd>
<dt>‘<samp>filename</samp>’</dt>
<dd><p>If set then a detailed log of the motion search is written to the
specified file.
</p>
</dd>
<dt>‘<samp>opencl</samp>’</dt>
<dd><p>If set to 1, specify using OpenCL capabilities, only available if
FFmpeg was configured with <code>--enable-opencl</code>. Default value is 0.
</p>
</dd>
</dl>
<a name="drawbox"></a>
<h2 class="section"><a href="ffmpeg-filters.html#toc-drawbox">9.22 drawbox</a></h2>
<p>Draw a colored box on the input image.
</p>
<p>It accepts the following parameters:
</p>
<dl compact="compact">
<dt>‘<samp>x</samp>’</dt>
<dt>‘<samp>y</samp>’</dt>
<dd><p>The expressions which specify the top left corner coordinates of the box. It defaults to 0.
</p>
</dd>
<dt>‘<samp>width, w</samp>’</dt>
<dt>‘<samp>height, h</samp>’</dt>
<dd><p>The expressions which specify the width and height of the box; if 0 they are interpreted as
the input width and height. It defaults to 0.
</p>
</dd>
<dt>‘<samp>color, c</samp>’</dt>
<dd><p>Specify the color of the box to write. For the general syntax of this option,
check the "Color" section in the ffmpeg-utils manual. If the special
value <code>invert</code> is used, the box edge color is the same as the
video with inverted luma.
</p>
</dd>
<dt>‘<samp>thickness, t</samp>’</dt>
<dd><p>The expression which sets the thickness of the box edge. Default value is <code>3</code>.
</p>
<p>See below for the list of accepted constants.
</p></dd>
</dl>
<p>The parameters for <var>x</var>, <var>y</var>, <var>w</var> and <var>h</var> and <var>t</var> are expressions containing the
following constants:
</p>
<dl compact="compact">
<dt>‘<samp>dar</samp>’</dt>
<dd><p>The input display aspect ratio, it is the same as (<var>w</var> / <var>h</var>) * <var>sar</var>.
</p>
</dd>
<dt>‘<samp>hsub</samp>’</dt>
<dt>‘<samp>vsub</samp>’</dt>
<dd><p>horizontal and vertical chroma subsample values. For example for the
pixel format "yuv422p" <var>hsub</var> is 2 and <var>vsub</var> is 1.
</p>
</dd>
<dt>‘<samp>in_h, ih</samp>’</dt>
<dt>‘<samp>in_w, iw</samp>’</dt>
<dd><p>The input width and height.
</p>
</dd>
<dt>‘<samp>sar</samp>’</dt>
<dd><p>The input sample aspect ratio.
</p>
</dd>
<dt>‘<samp>x</samp>’</dt>
<dt>‘<samp>y</samp>’</dt>
<dd><p>The x and y offset coordinates where the box is drawn.
</p>
</dd>
<dt>‘<samp>w</samp>’</dt>
<dt>‘<samp>h</samp>’</dt>
<dd><p>The width and height of the drawn box.
</p>
</dd>
<dt>‘<samp>t</samp>’</dt>
<dd><p>The thickness of the drawn box.
</p>
<p>These constants allow the <var>x</var>, <var>y</var>, <var>w</var>, <var>h</var> and <var>t</var> expressions to refer to
each other, so you may for example specify <code>y=x/dar</code> or <code>h=w/dar</code>.
</p>
</dd>
</dl>
<a name="Examples-16"></a>
<h3 class="subsection"><a href="ffmpeg-filters.html#toc-Examples-16">9.22.1 Examples</a></h3>
<ul>
<li>
Draw a black box around the edge of the input image:
<div class="example">
<pre class="example">drawbox
</pre></div>
</li><li>
Draw a box with color red and an opacity of 50%:
<div class="example">
<pre class="example">drawbox=10:20:200:60:red@0.5
</pre></div>
<p>The previous example can be specified as:
</p><div class="example">
<pre class="example">drawbox=x=10:y=20:w=200:h=60:color=red@0.5
</pre></div>
</li><li>
Fill the box with pink color:
<div class="example">
<pre class="example">drawbox=x=10:y=10:w=100:h=100:color=pink@0.5:t=max
</pre></div>
</li><li>
Draw a 2-pixel red 2.40:1 mask:
<div class="example">
<pre class="example">drawbox=x=-t:y=0.5*(ih-iw/2.4)-t:w=iw+t*2:h=iw/2.4+t*2:t=2:c=red
</pre></div>
</li></ul>
<a name="drawgrid"></a>
<h2 class="section"><a href="ffmpeg-filters.html#toc-drawgrid">9.23 drawgrid</a></h2>
<p>Draw a grid on the input image.
</p>
<p>It accepts the following parameters:
</p>
<dl compact="compact">
<dt>‘<samp>x</samp>’</dt>
<dt>‘<samp>y</samp>’</dt>
<dd><p>The expressions which specify the coordinates of some point of grid intersection (meant to configure offset). Both default to 0.
</p>
</dd>
<dt>‘<samp>width, w</samp>’</dt>
<dt>‘<samp>height, h</samp>’</dt>
<dd><p>The expressions which specify the width and height of the grid cell, if 0 they are interpreted as the
input width and height, respectively, minus <code>thickness</code>, so image gets
framed. Default to 0.
</p>
</dd>
<dt>‘<samp>color, c</samp>’</dt>
<dd><p>Specify the color of the grid. For the general syntax of this option,
check the "Color" section in the ffmpeg-utils manual. If the special
value <code>invert</code> is used, the grid color is the same as the
video with inverted luma.
</p>
</dd>
<dt>‘<samp>thickness, t</samp>’</dt>
<dd><p>The expression which sets the thickness of the grid line. Default value is <code>1</code>.
</p>
<p>See below for the list of accepted constants.
</p></dd>
</dl>
<p>The parameters for <var>x</var>, <var>y</var>, <var>w</var> and <var>h</var> and <var>t</var> are expressions containing the
following constants:
</p>
<dl compact="compact">
<dt>‘<samp>dar</samp>’</dt>
<dd><p>The input display aspect ratio, it is the same as (<var>w</var> / <var>h</var>) * <var>sar</var>.
</p>
</dd>
<dt>‘<samp>hsub</samp>’</dt>
<dt>‘<samp>vsub</samp>’</dt>
<dd><p>horizontal and vertical chroma subsample values. For example for the
pixel format "yuv422p" <var>hsub</var> is 2 and <var>vsub</var> is 1.
</p>
</dd>
<dt>‘<samp>in_h, ih</samp>’</dt>
<dt>‘<samp>in_w, iw</samp>’</dt>
<dd><p>The input grid cell width and height.
</p>
</dd>
<dt>‘<samp>sar</samp>’</dt>
<dd><p>The input sample aspect ratio.
</p>
</dd>
<dt>‘<samp>x</samp>’</dt>
<dt>‘<samp>y</samp>’</dt>
<dd><p>The x and y coordinates of some point of grid intersection (meant to configure offset).
</p>
</dd>
<dt>‘<samp>w</samp>’</dt>
<dt>‘<samp>h</samp>’</dt>
<dd><p>The width and height of the drawn cell.
</p>
</dd>
<dt>‘<samp>t</samp>’</dt>
<dd><p>The thickness of the drawn cell.
</p>
<p>These constants allow the <var>x</var>, <var>y</var>, <var>w</var>, <var>h</var> and <var>t</var> expressions to refer to
each other, so you may for example specify <code>y=x/dar</code> or <code>h=w/dar</code>.
</p>
</dd>
</dl>
<a name="Examples-29"></a>
<h3 class="subsection"><a href="ffmpeg-filters.html#toc-Examples-29">9.23.1 Examples</a></h3>
<ul>
<li>
Draw a grid with cell 100x100 pixels, thickness 2 pixels, with color red and an opacity of 50%:
<div class="example">
<pre class="example">drawgrid=width=100:height=100:thickness=2:color=red@0.5
</pre></div>
</li><li>
Draw a white 3x3 grid with an opacity of 50%:
<div class="example">
<pre class="example">drawgrid=w=iw/3:h=ih/3:t=2:c=white@0.5
</pre></div>
</li></ul>
<p><a name="drawtext"></a>
</p><a name="drawtext-1"></a>
<h2 class="section"><a href="ffmpeg-filters.html#toc-drawtext-1">9.24 drawtext</a></h2>
<p>Draw a text string or text from a specified file on top of a video, using the
libfreetype library.
</p>
<p>To enable compilation of this filter, you need to configure FFmpeg with
<code>--enable-libfreetype</code>.
To enable default font fallback and the <var>font</var> option you need to
configure FFmpeg with <code>--enable-libfontconfig</code>.
To enable the <var>text_shaping</var> option, you need to configure FFmpeg with
<code>--enable-libfribidi</code>.
</p>
<a name="Syntax"></a>
<h3 class="subsection"><a href="ffmpeg-filters.html#toc-Syntax">9.24.1 Syntax</a></h3>
<p>It accepts the following parameters:
</p>
<dl compact="compact">
<dt>‘<samp>box</samp>’</dt>
<dd><p>Used to draw a box around text using the background color.
The value must be either 1 (enable) or 0 (disable).
The default value of <var>box</var> is 0.
</p>
</dd>
<dt>‘<samp>boxcolor</samp>’</dt>
<dd><p>The color to be used for drawing box around text. For the syntax of this
option, check the "Color" section in the ffmpeg-utils manual.
</p>
<p>The default value of <var>boxcolor</var> is "white".
</p>
</dd>
<dt>‘<samp>borderw</samp>’</dt>
<dd><p>Set the width of the border to be drawn around the text using <var>bordercolor</var>.
The default value of <var>borderw</var> is 0.
</p>
</dd>
<dt>‘<samp>bordercolor</samp>’</dt>
<dd><p>Set the color to be used for drawing border around text. For the syntax of this
option, check the "Color" section in the ffmpeg-utils manual.
</p>
<p>The default value of <var>bordercolor</var> is "black".
</p>
</dd>
<dt>‘<samp>expansion</samp>’</dt>
<dd><p>Select how the <var>text</var> is expanded. Can be either <code>none</code>,
<code>strftime</code> (deprecated) or
<code>normal</code> (default). See the <a href="#drawtext_005fexpansion">Text expansion</a> section
below for details.
</p>
</dd>
<dt>‘<samp>fix_bounds</samp>’</dt>
<dd><p>If true, check and fix text coords to avoid clipping.
</p>
</dd>
<dt>‘<samp>fontcolor</samp>’</dt>
<dd><p>The color to be used for drawing fonts. For the syntax of this option, check
the "Color" section in the ffmpeg-utils manual.
</p>
<p>The default value of <var>fontcolor</var> is "black".
</p>
</dd>
<dt>‘<samp>fontcolor_expr</samp>’</dt>
<dd><p>String which is expanded the same way as <var>text</var> to obtain dynamic
<var>fontcolor</var> value. By default this option has empty value and is not
processed. When this option is set, it overrides <var>fontcolor</var> option.
</p>
</dd>
<dt>‘<samp>font</samp>’</dt>
<dd><p>The font family to be used for drawing text. By default Sans.
</p>
</dd>
<dt>‘<samp>fontfile</samp>’</dt>
<dd><p>The font file to be used for drawing text. The path must be included.
This parameter is mandatory if the fontconfig support is disabled.
</p>
</dd>
<dt>‘<samp>fontsize</samp>’</dt>
<dd><p>The font size to be used for drawing text.
The default value of <var>fontsize</var> is 16.
</p>
</dd>
<dt>‘<samp>text_shaping</samp>’</dt>
<dd><p>If set to 1, attempt to shape the text (for example, reverse the order of
right-to-left text and join Arabic characters) before drawing it.
Otherwise, just draw the text exactly as given.
By default 1 (if supported).
</p>
</dd>
<dt>‘<samp>ft_load_flags</samp>’</dt>
<dd><p>The flags to be used for loading the fonts.
</p>
<p>The flags map the corresponding flags supported by libfreetype, and are
a combination of the following values:
</p><dl compact="compact">
<dt><var>default</var></dt>
<dt><var>no_scale</var></dt>
<dt><var>no_hinting</var></dt>
<dt><var>render</var></dt>
<dt><var>no_bitmap</var></dt>
<dt><var>vertical_layout</var></dt>
<dt><var>force_autohint</var></dt>
<dt><var>crop_bitmap</var></dt>
<dt><var>pedantic</var></dt>
<dt><var>ignore_global_advance_width</var></dt>
<dt><var>no_recurse</var></dt>
<dt><var>ignore_transform</var></dt>
<dt><var>monochrome</var></dt>
<dt><var>linear_design</var></dt>
<dt><var>no_autohint</var></dt>
</dl>
<p>Default value is "default".
</p>
<p>For more information consult the documentation for the FT_LOAD_*
libfreetype flags.
</p>
</dd>
<dt>‘<samp>shadowcolor</samp>’</dt>
<dd><p>The color to be used for drawing a shadow behind the drawn text. For the
syntax of this option, check the "Color" section in the ffmpeg-utils manual.
</p>
<p>The default value of <var>shadowcolor</var> is "black".
</p>
</dd>
<dt>‘<samp>shadowx</samp>’</dt>
<dt>‘<samp>shadowy</samp>’</dt>
<dd><p>The x and y offsets for the text shadow position with respect to the
position of the text. They can be either positive or negative
values. The default value for both is "0".
</p>
</dd>
<dt>‘<samp>start_number</samp>’</dt>
<dd><p>The starting frame number for the n/frame_num variable. The default value
is "0".
</p>
</dd>
<dt>‘<samp>tabsize</samp>’</dt>
<dd><p>The size in number of spaces to use for rendering the tab.
Default value is 4.
</p>
</dd>
<dt>‘<samp>timecode</samp>’</dt>
<dd><p>Set the initial timecode representation in "hh:mm:ss[:;.]ff"
format. It can be used with or without text parameter. <var>timecode_rate</var>
option must be specified.
</p>
</dd>
<dt>‘<samp>timecode_rate, rate, r</samp>’</dt>
<dd><p>Set the timecode frame rate (timecode only).
</p>
</dd>
<dt>‘<samp>text</samp>’</dt>
<dd><p>The text string to be drawn. The text must be a sequence of UTF-8
encoded characters.
This parameter is mandatory if no file is specified with the parameter
<var>textfile</var>.
</p>
</dd>
<dt>‘<samp>textfile</samp>’</dt>
<dd><p>A text file containing text to be drawn. The text must be a sequence
of UTF-8 encoded characters.
</p>
<p>This parameter is mandatory if no text string is specified with the
parameter <var>text</var>.
</p>
<p>If both <var>text</var> and <var>textfile</var> are specified, an error is thrown.
</p>
</dd>
<dt>‘<samp>reload</samp>’</dt>
<dd><p>If set to 1, the <var>textfile</var> will be reloaded before each frame.
Be sure to update it atomically, or it may be read partially, or even fail.
</p>
</dd>
<dt>‘<samp>x</samp>’</dt>
<dt>‘<samp>y</samp>’</dt>
<dd><p>The expressions which specify the offsets where text will be drawn
within the video frame. They are relative to the top/left border of the
output image.
</p>
<p>The default value of <var>x</var> and <var>y</var> is "0".
</p>
<p>See below for the list of accepted constants and functions.
</p></dd>
</dl>
<p>The parameters for <var>x</var> and <var>y</var> are expressions containing the
following constants and functions:
</p>
<dl compact="compact">
<dt>‘<samp>dar</samp>’</dt>
<dd><p>input display aspect ratio, it is the same as (<var>w</var> / <var>h</var>) * <var>sar</var>
</p>
</dd>
<dt>‘<samp>hsub</samp>’</dt>
<dt>‘<samp>vsub</samp>’</dt>
<dd><p>horizontal and vertical chroma subsample values. For example for the
pixel format "yuv422p" <var>hsub</var> is 2 and <var>vsub</var> is 1.
</p>
</dd>
<dt>‘<samp>line_h, lh</samp>’</dt>
<dd><p>the height of each text line
</p>
</dd>
<dt>‘<samp>main_h, h, H</samp>’</dt>
<dd><p>the input height
</p>
</dd>
<dt>‘<samp>main_w, w, W</samp>’</dt>
<dd><p>the input width
</p>
</dd>
<dt>‘<samp>max_glyph_a, ascent</samp>’</dt>
<dd><p>the maximum distance from the baseline to the highest/upper grid
coordinate used to place a glyph outline point, for all the rendered
glyphs.
It is a positive value, due to the grid’s orientation with the Y axis
upwards.
</p>
</dd>
<dt>‘<samp>max_glyph_d, descent</samp>’</dt>
<dd><p>the maximum distance from the baseline to the lowest grid coordinate
used to place a glyph outline point, for all the rendered glyphs.
This is a negative value, due to the grid’s orientation, with the Y axis
upwards.
</p>
</dd>
<dt>‘<samp>max_glyph_h</samp>’</dt>
<dd><p>maximum glyph height, that is the maximum height for all the glyphs
contained in the rendered text, it is equivalent to <var>ascent</var> -
<var>descent</var>.
</p>
</dd>
<dt>‘<samp>max_glyph_w</samp>’</dt>
<dd><p>maximum glyph width, that is the maximum width for all the glyphs
contained in the rendered text
</p>
</dd>
<dt>‘<samp>n</samp>’</dt>
<dd><p>the number of input frame, starting from 0
</p>
</dd>
<dt>‘<samp>rand(min, max)</samp>’</dt>
<dd><p>return a random number included between <var>min</var> and <var>max</var>
</p>
</dd>
<dt>‘<samp>sar</samp>’</dt>
<dd><p>The input sample aspect ratio.
</p>
</dd>
<dt>‘<samp>t</samp>’</dt>
<dd><p>timestamp expressed in seconds, NAN if the input timestamp is unknown
</p>
</dd>
<dt>‘<samp>text_h, th</samp>’</dt>
<dd><p>the height of the rendered text
</p>
</dd>
<dt>‘<samp>text_w, tw</samp>’</dt>
<dd><p>the width of the rendered text
</p>
</dd>
<dt>‘<samp>x</samp>’</dt>
<dt>‘<samp>y</samp>’</dt>
<dd><p>the x and y offset coordinates where the text is drawn.
</p>
<p>These parameters allow the <var>x</var> and <var>y</var> expressions to refer
each other, so you can for example specify <code>y=x/dar</code>.
</p></dd>
</dl>
<p><a name="drawtext_005fexpansion"></a>
</p><a name="Text-expansion"></a>
<h3 class="subsection"><a href="ffmpeg-filters.html#toc-Text-expansion">9.24.2 Text expansion</a></h3>
<p>If ‘<samp>expansion</samp>’ is set to <code>strftime</code>,
the filter recognizes strftime() sequences in the provided text and
expands them accordingly. Check the documentation of strftime(). This
feature is deprecated.
</p>
<p>If ‘<samp>expansion</samp>’ is set to <code>none</code>, the text is printed verbatim.
</p>
<p>If ‘<samp>expansion</samp>’ is set to <code>normal</code> (which is the default),
the following expansion mechanism is used.
</p>
<p>The backslash character ’\’, followed by any character, always expands to
the second character.
</p>
<p>Sequence of the form <code>%{...}</code> are expanded. The text between the
braces is a function name, possibly followed by arguments separated by ’:’.
If the arguments contain special characters or delimiters (’:’ or ’}’),
they should be escaped.
</p>
<p>Note that they probably must also be escaped as the value for the
‘<samp>text</samp>’ option in the filter argument string and as the filter
argument in the filtergraph description, and possibly also for the shell,
that makes up to four levels of escaping; using a text file avoids these
problems.
</p>
<p>The following functions are available:
</p>
<dl compact="compact">
<dt><code>expr, e</code></dt>
<dd><p>The expression evaluation result.
</p>
<p>It must take one argument specifying the expression to be evaluated,
which accepts the same constants and functions as the <var>x</var> and
<var>y</var> values. Note that not all constants should be used, for
example the text size is not known when evaluating the expression, so
the constants <var>text_w</var> and <var>text_h</var> will have an undefined
value.
</p>
</dd>
<dt><code>expr_int_format, eif</code></dt>
<dd><p>Evaluate the expression’s value and output as formatted integer.
</p>
<p>The first argument is the expression to be evaluated, just as for the <var>expr</var> function.
The second argument specifies the output format. Allowed values are ’x’, ’X’, ’d’ and
’u’. They are treated exactly as in the printf function.
The third parameter is optional and sets the number of positions taken by the output.
It can be used to add padding with zeros from the left.
</p>
</dd>
<dt><code>gmtime</code></dt>
<dd><p>The time at which the filter is running, expressed in UTC.
It can accept an argument: a strftime() format string.
</p>
</dd>
<dt><code>localtime</code></dt>
<dd><p>The time at which the filter is running, expressed in the local time zone.
It can accept an argument: a strftime() format string.
</p>
</dd>
<dt><code>metadata</code></dt>
<dd><p>Frame metadata. It must take one argument specifying metadata key.
</p>
</dd>
<dt><code>n, frame_num</code></dt>
<dd><p>The frame number, starting from 0.
</p>
</dd>
<dt><code>pict_type</code></dt>
<dd><p>A 1 character description of the current picture type.
</p>
</dd>
<dt><code>pts</code></dt>
<dd><p>The timestamp of the current frame.
It can take up to two arguments.
</p>
<p>The first argument is the format of the timestamp; it defaults to <code>flt</code>
for seconds as a decimal number with microsecond accuracy; <code>hms</code> stands
for a formatted <var>[-]HH:MM:SS.mmm</var> timestamp with millisecond accuracy.
</p>
<p>The second argument is an offset added to the timestamp.
</p>
</dd>
</dl>
<a name="Examples-46"></a>
<h3 class="subsection"><a href="ffmpeg-filters.html#toc-Examples-46">9.24.3 Examples</a></h3>
<ul>
<li>
Draw "Test Text" with font FreeSerif, using the default values for the
optional parameters.
<div class="example">
<pre class="example">drawtext="fontfile=/usr/share/fonts/truetype/freefont/FreeSerif.ttf: text='Test Text'"
</pre></div>
</li><li>
Draw ’Test Text’ with font FreeSerif of size 24 at position x=100
and y=50 (counting from the top-left corner of the screen), text is
yellow with a red box around it. Both the text and the box have an
opacity of 20%.
<div class="example">
<pre class="example">drawtext="fontfile=/usr/share/fonts/truetype/freefont/FreeSerif.ttf: text='Test Text':\
x=100: y=50: fontsize=24: fontcolor=yellow@0.2: box=1: boxcolor=red@0.2"
</pre></div>
<p>Note that the double quotes are not necessary if spaces are not used
within the parameter list.
</p>
</li><li>
Show the text at the center of the video frame:
<div class="example">
<pre class="example">drawtext="fontsize=30:fontfile=FreeSerif.ttf:text='hello world':x=(w-text_w)/2:y=(h-text_h)/2"
</pre></div>
</li><li>
Show a text line sliding from right to left in the last row of the video
frame. The file ‘<tt>LONG_LINE</tt>’ is assumed to contain a single line
with no newlines.
<div class="example">
<pre class="example">drawtext="fontsize=15:fontfile=FreeSerif.ttf:text=LONG_LINE:y=h-line_h:x=-50*t"
</pre></div>
</li><li>
Show the content of file ‘<tt>CREDITS</tt>’ off the bottom of the frame and scroll up.
<div class="example">
<pre class="example">drawtext="fontsize=20:fontfile=FreeSerif.ttf:textfile=CREDITS:y=h-20*t"
</pre></div>
</li><li>
Draw a single green letter "g", at the center of the input video.
The glyph baseline is placed at half screen height.
<div class="example">
<pre class="example">drawtext="fontsize=60:fontfile=FreeSerif.ttf:fontcolor=green:text=g:x=(w-max_glyph_w)/2:y=h/2-ascent"
</pre></div>
</li><li>
Show text for 1 second every 3 seconds:
<div class="example">
<pre class="example">drawtext="fontfile=FreeSerif.ttf:fontcolor=white:x=100:y=x/dar:enable=lt(mod(t\,3)\,1):text='blink'"
</pre></div>
</li><li>
Use fontconfig to set the font. Note that the colons need to be escaped.
<div class="example">
<pre class="example">drawtext='fontfile=Linux Libertine O-40\:style=Semibold:text=FFmpeg'
</pre></div>
</li><li>
Print the date of a real-time encoding (see strftime(3)):
<div class="example">
<pre class="example">drawtext='fontfile=FreeSans.ttf:text=%{localtime:%a %b %d %Y}'
</pre></div>
</li><li>
Show text fading in and out (appearing/disappearing):
<div class="example">
<pre class="example">#!/bin/sh
DS=1.0 # display start
DE=10.0 # display end
FID=1.5 # fade in duration
FOD=5 # fade out duration
ffplay -f lavfi "color,drawtext=text=TEST:fontsize=50:fontfile=FreeSerif.ttf:fontcolor_expr=ff0000%{eif\\\\: clip(255*(1*between(t\\, $DS + $FID\\, $DE - $FOD) + ((t - $DS)/$FID)*between(t\\, $DS\\, $DS + $FID) + (-(t - $DE)/$FOD)*between(t\\, $DE - $FOD\\, $DE) )\\, 0\\, 255) \\\\: x\\\\: 2 }"
</pre></div>
</li></ul>
<p>For more information about libfreetype, check:
<a href="http://www.freetype.org/">http://www.freetype.org/</a>.
</p>
<p>For more information about fontconfig, check:
<a href="http://freedesktop.org/software/fontconfig/fontconfig-user.html">http://freedesktop.org/software/fontconfig/fontconfig-user.html</a>.
</p>
<p>For more information about libfribidi, check:
<a href="http://fribidi.org/">http://fribidi.org/</a>.
</p>
<a name="edgedetect"></a>
<h2 class="section"><a href="ffmpeg-filters.html#toc-edgedetect">9.25 edgedetect</a></h2>
<p>Detect and draw edges. The filter uses the Canny Edge Detection algorithm.
</p>
<p>The filter accepts the following options:
</p>
<dl compact="compact">
<dt>‘<samp>low</samp>’</dt>
<dt>‘<samp>high</samp>’</dt>
<dd><p>Set low and high threshold values used by the Canny thresholding
algorithm.
</p>
<p>The high threshold selects the "strong" edge pixels, which are then
connected through 8-connectivity with the "weak" edge pixels selected
by the low threshold.
</p>
<p><var>low</var> and <var>high</var> threshold values must be chosen in the range
[0,1], and <var>low</var> should be lesser or equal to <var>high</var>.
</p>
<p>Default value for <var>low</var> is <code>20/255</code>, and default value for <var>high</var>
is <code>50/255</code>.
</p>
</dd>
<dt>‘<samp>mode</samp>’</dt>
<dd><p>Define the drawing mode.
</p>
<dl compact="compact">
<dt>‘<samp>wires</samp>’</dt>
<dd><p>Draw white/gray wires on black background.
</p>
</dd>
<dt>‘<samp>colormix</samp>’</dt>
<dd><p>Mix the colors to create a paint/cartoon effect.
</p></dd>
</dl>
<p>Default value is <var>wires</var>.
</p></dd>
</dl>
<a name="Examples-52"></a>
<h3 class="subsection"><a href="ffmpeg-filters.html#toc-Examples-52">9.25.1 Examples</a></h3>
<ul>
<li>
Standard edge detection with custom values for the hysteresis thresholding:
<div class="example">
<pre class="example">edgedetect=low=0.1:high=0.4
</pre></div>
</li><li>
Painting effect without thresholding:
<div class="example">
<pre class="example">edgedetect=mode=colormix:high=0
</pre></div>
</li></ul>
<a name="extractplanes"></a>
<h2 class="section"><a href="ffmpeg-filters.html#toc-extractplanes">9.26 extractplanes</a></h2>
<p>Extract color channel components from input video stream into
separate grayscale video streams.
</p>
<p>The filter accepts the following option:
</p>
<dl compact="compact">
<dt>‘<samp>planes</samp>’</dt>
<dd><p>Set plane(s) to extract.
</p>
<p>Available values for planes are:
</p><dl compact="compact">
<dt>‘<samp>y</samp>’</dt>
<dt>‘<samp>u</samp>’</dt>
<dt>‘<samp>v</samp>’</dt>
<dt>‘<samp>a</samp>’</dt>
<dt>‘<samp>r</samp>’</dt>
<dt>‘<samp>g</samp>’</dt>
<dt>‘<samp>b</samp>’</dt>
</dl>
<p>Choosing planes not available in the input will result in an error.
That means you cannot select <code>r</code>, <code>g</code>, <code>b</code> planes
with <code>y</code>, <code>u</code>, <code>v</code> planes at same time.
</p></dd>
</dl>
<a name="Examples-37"></a>
<h3 class="subsection"><a href="ffmpeg-filters.html#toc-Examples-37">9.26.1 Examples</a></h3>
<ul>
<li>
Extract luma, u and v color channel component from input video frame
into 3 grayscale outputs:
<div class="example">
<pre class="example">ffmpeg -i video.avi -filter_complex 'extractplanes=y+u+v[y][u][v]' -map '[y]' y.avi -map '[u]' u.avi -map '[v]' v.avi
</pre></div>
</li></ul>
<a name="elbg"></a>
<h2 class="section"><a href="ffmpeg-filters.html#toc-elbg">9.27 elbg</a></h2>
<p>Apply a posterize effect using the ELBG (Enhanced LBG) algorithm.
</p>
<p>For each input image, the filter will compute the optimal mapping from
the input to the output given the codebook length, that is the number
of distinct output colors.
</p>
<p>This filter accepts the following options.
</p>
<dl compact="compact">
<dt>‘<samp>codebook_length, l</samp>’</dt>
<dd><p>Set codebook length. The value must be a positive integer, and
represents the number of distinct output colors. Default value is 256.
</p>
</dd>
<dt>‘<samp>nb_steps, n</samp>’</dt>
<dd><p>Set the maximum number of iterations to apply for computing the optimal
mapping. The higher the value the better the result and the higher the
computation time. Default value is 1.
</p>
</dd>
<dt>‘<samp>seed, s</samp>’</dt>
<dd><p>Set a random seed, must be an integer included between 0 and
UINT32_MAX. If not specified, or if explicitly set to -1, the filter
will try to use a good random seed on a best effort basis.
</p></dd>
</dl>
<a name="fade"></a>
<h2 class="section"><a href="ffmpeg-filters.html#toc-fade">9.28 fade</a></h2>
<p>Apply a fade-in/out effect to the input video.
</p>
<p>It accepts the following parameters:
</p>
<dl compact="compact">
<dt>‘<samp>type, t</samp>’</dt>
<dd><p>The effect type can be either "in" for a fade-in, or "out" for a fade-out
effect.
Default is <code>in</code>.
</p>
</dd>
<dt>‘<samp>start_frame, s</samp>’</dt>
<dd><p>Specify the number of the frame to start applying the fade
effect at. Default is 0.
</p>
</dd>
<dt>‘<samp>nb_frames, n</samp>’</dt>
<dd><p>The number of frames that the fade effect lasts. At the end of the
fade-in effect, the output video will have the same intensity as the input video.
At the end of the fade-out transition, the output video will be filled with the
selected ‘<samp>color</samp>’.
Default is 25.
</p>
</dd>
<dt>‘<samp>alpha</samp>’</dt>
<dd><p>If set to 1, fade only alpha channel, if one exists on the input.
Default value is 0.
</p>
</dd>
<dt>‘<samp>start_time, st</samp>’</dt>
<dd><p>Specify the timestamp (in seconds) of the frame to start to apply the fade
effect. If both start_frame and start_time are specified, the fade will start at
whichever comes last. Default is 0.
</p>
</dd>
<dt>‘<samp>duration, d</samp>’</dt>
<dd><p>The number of seconds for which the fade effect has to last. At the end of the
fade-in effect the output video will have the same intensity as the input video,
at the end of the fade-out transition the output video will be filled with the
selected ‘<samp>color</samp>’.
If both duration and nb_frames are specified, duration is used. Default is 0.
</p>
</dd>
<dt>‘<samp>color, c</samp>’</dt>
<dd><p>Specify the color of the fade. Default is "black".
</p></dd>
</dl>
<a name="Examples-69"></a>
<h3 class="subsection"><a href="ffmpeg-filters.html#toc-Examples-69">9.28.1 Examples</a></h3>
<ul>
<li>
Fade in the first 30 frames of video:
<div class="example">
<pre class="example">fade=in:0:30
</pre></div>
<p>The command above is equivalent to:
</p><div class="example">
<pre class="example">fade=t=in:s=0:n=30
</pre></div>
</li><li>
Fade out the last 45 frames of a 200-frame video:
<div class="example">
<pre class="example">fade=out:155:45
fade=type=out:start_frame=155:nb_frames=45
</pre></div>
</li><li>
Fade in the first 25 frames and fade out the last 25 frames of a 1000-frame video:
<div class="example">
<pre class="example">fade=in:0:25, fade=out:975:25
</pre></div>
</li><li>
Make the first 5 frames yellow, then fade in from frame 5-24:
<div class="example">
<pre class="example">fade=in:5:20:color=yellow
</pre></div>
</li><li>
Fade in alpha over first 25 frames of video:
<div class="example">
<pre class="example">fade=in:0:25:alpha=1
</pre></div>
</li><li>
Make the first 5.5 seconds black, then fade in for 0.5 seconds:
<div class="example">
<pre class="example">fade=t=in:st=5.5:d=0.5
</pre></div>
</li></ul>
<a name="field"></a>
<h2 class="section"><a href="ffmpeg-filters.html#toc-field">9.29 field</a></h2>
<p>Extract a single field from an interlaced image using stride
arithmetic to avoid wasting CPU time. The output frames are marked as
non-interlaced.
</p>
<p>The filter accepts the following options:
</p>
<dl compact="compact">
<dt>‘<samp>type</samp>’</dt>
<dd><p>Specify whether to extract the top (if the value is <code>0</code> or
<code>top</code>) or the bottom field (if the value is <code>1</code> or
<code>bottom</code>).
</p></dd>
</dl>
<a name="fieldmatch"></a>
<h2 class="section"><a href="ffmpeg-filters.html#toc-fieldmatch">9.30 fieldmatch</a></h2>
<p>Field matching filter for inverse telecine. It is meant to reconstruct the
progressive frames from a telecined stream. The filter does not drop duplicated
frames, so to achieve a complete inverse telecine <code>fieldmatch</code> needs to be
followed by a decimation filter such as <a href="#decimate">decimate</a> in the filtergraph.
</p>
<p>The separation of the field matching and the decimation is notably motivated by
the possibility of inserting a de-interlacing filter fallback between the two.
If the source has mixed telecined and real interlaced content,
<code>fieldmatch</code> will not be able to match fields for the interlaced parts.
But these remaining combed frames will be marked as interlaced, and thus can be
de-interlaced by a later filter such as <a href="#yadif">yadif</a> before decimation.
</p>
<p>In addition to the various configuration options, <code>fieldmatch</code> can take an
optional second stream, activated through the ‘<samp>ppsrc</samp>’ option. If
enabled, the frames reconstruction will be based on the fields and frames from
this second stream. This allows the first input to be pre-processed in order to
help the various algorithms of the filter, while keeping the output lossless
(assuming the fields are matched properly). Typically, a field-aware denoiser,
or brightness/contrast adjustments can help.
</p>
<p>Note that this filter uses the same algorithms as TIVTC/TFM (AviSynth project)
and VIVTC/VFM (VapourSynth project). The later is a light clone of TFM from
which <code>fieldmatch</code> is based on. While the semantic and usage are very
close, some behaviour and options names can differ.
</p>
<p>The filter accepts the following options:
</p>
<dl compact="compact">
<dt>‘<samp>order</samp>’</dt>
<dd><p>Specify the assumed field order of the input stream. Available values are:
</p>
<dl compact="compact">
<dt>‘<samp>auto</samp>’</dt>
<dd><p>Auto detect parity (use FFmpeg’s internal parity value).
</p></dd>
<dt>‘<samp>bff</samp>’</dt>
<dd><p>Assume bottom field first.
</p></dd>
<dt>‘<samp>tff</samp>’</dt>
<dd><p>Assume top field first.
</p></dd>
</dl>
<p>Note that it is sometimes recommended not to trust the parity announced by the
stream.
</p>
<p>Default value is <var>auto</var>.
</p>
</dd>
<dt>‘<samp>mode</samp>’</dt>
<dd><p>Set the matching mode or strategy to use. ‘<samp>pc</samp>’ mode is the safest in the
sense that it won’t risk creating jerkiness due to duplicate frames when
possible, but if there are bad edits or blended fields it will end up
outputting combed frames when a good match might actually exist. On the other
hand, ‘<samp>pcn_ub</samp>’ mode is the most risky in terms of creating jerkiness,
but will almost always find a good frame if there is one. The other values are
all somewhere in between ‘<samp>pc</samp>’ and ‘<samp>pcn_ub</samp>’ in terms of risking
jerkiness and creating duplicate frames versus finding good matches in sections
with bad edits, orphaned fields, blended fields, etc.
</p>
<p>More details about p/c/n/u/b are available in <a href="#p_002fc_002fn_002fu_002fb-meaning">p/c/n/u/b meaning</a> section.
</p>
<p>Available values are:
</p>
<dl compact="compact">
<dt>‘<samp>pc</samp>’</dt>
<dd><p>2-way matching (p/c)
</p></dd>
<dt>‘<samp>pc_n</samp>’</dt>
<dd><p>2-way matching, and trying 3rd match if still combed (p/c + n)
</p></dd>
<dt>‘<samp>pc_u</samp>’</dt>
<dd><p>2-way matching, and trying 3rd match (same order) if still combed (p/c + u)
</p></dd>
<dt>‘<samp>pc_n_ub</samp>’</dt>
<dd><p>2-way matching, trying 3rd match if still combed, and trying 4th/5th matches if
still combed (p/c + n + u/b)
</p></dd>
<dt>‘<samp>pcn</samp>’</dt>
<dd><p>3-way matching (p/c/n)
</p></dd>
<dt>‘<samp>pcn_ub</samp>’</dt>
<dd><p>3-way matching, and trying 4th/5th matches if all 3 of the original matches are
detected as combed (p/c/n + u/b)
</p></dd>
</dl>
<p>The parenthesis at the end indicate the matches that would be used for that
mode assuming ‘<samp>order</samp>’=<var>tff</var> (and ‘<samp>field</samp>’ on <var>auto</var> or
<var>top</var>).
</p>
<p>In terms of speed ‘<samp>pc</samp>’ mode is by far the fastest and ‘<samp>pcn_ub</samp>’ is
the slowest.
</p>
<p>Default value is <var>pc_n</var>.
</p>
</dd>
<dt>‘<samp>ppsrc</samp>’</dt>
<dd><p>Mark the main input stream as a pre-processed input, and enable the secondary
input stream as the clean source to pick the fields from. See the filter
introduction for more details. It is similar to the ‘<samp>clip2</samp>’ feature from
VFM/TFM.
</p>
<p>Default value is <code>0</code> (disabled).
</p>
</dd>
<dt>‘<samp>field</samp>’</dt>
<dd><p>Set the field to match from. It is recommended to set this to the same value as
‘<samp>order</samp>’ unless you experience matching failures with that setting. In
certain circumstances changing the field that is used to match from can have a
large impact on matching performance. Available values are:
</p>
<dl compact="compact">
<dt>‘<samp>auto</samp>’</dt>
<dd><p>Automatic (same value as ‘<samp>order</samp>’).
</p></dd>
<dt>‘<samp>bottom</samp>’</dt>
<dd><p>Match from the bottom field.
</p></dd>
<dt>‘<samp>top</samp>’</dt>
<dd><p>Match from the top field.
</p></dd>
</dl>
<p>Default value is <var>auto</var>.
</p>
</dd>
<dt>‘<samp>mchroma</samp>’</dt>
<dd><p>Set whether or not chroma is included during the match comparisons. In most
cases it is recommended to leave this enabled. You should set this to <code>0</code>
only if your clip has bad chroma problems such as heavy rainbowing or other
artifacts. Setting this to <code>0</code> could also be used to speed things up at
the cost of some accuracy.
</p>
<p>Default value is <code>1</code>.
</p>
</dd>
<dt>‘<samp>y0</samp>’</dt>
<dt>‘<samp>y1</samp>’</dt>
<dd><p>These define an exclusion band which excludes the lines between ‘<samp>y0</samp>’ and
‘<samp>y1</samp>’ from being included in the field matching decision. An exclusion
band can be used to ignore subtitles, a logo, or other things that may
interfere with the matching. ‘<samp>y0</samp>’ sets the starting scan line and
‘<samp>y1</samp>’ sets the ending line; all lines in between ‘<samp>y0</samp>’ and
‘<samp>y1</samp>’ (including ‘<samp>y0</samp>’ and ‘<samp>y1</samp>’) will be ignored. Setting
‘<samp>y0</samp>’ and ‘<samp>y1</samp>’ to the same value will disable the feature.
‘<samp>y0</samp>’ and ‘<samp>y1</samp>’ defaults to <code>0</code>.
</p>
</dd>
<dt>‘<samp>scthresh</samp>’</dt>
<dd><p>Set the scene change detection threshold as a percentage of maximum change on
the luma plane. Good values are in the <code>[8.0, 14.0]</code> range. Scene change
detection is only relevant in case ‘<samp>combmatch</samp>’=<var>sc</var>. The range for
‘<samp>scthresh</samp>’ is <code>[0.0, 100.0]</code>.
</p>
<p>Default value is <code>12.0</code>.
</p>
</dd>
<dt>‘<samp>combmatch</samp>’</dt>
<dd><p>When ‘<samp>combatch</samp>’ is not <var>none</var>, <code>fieldmatch</code> will take into
account the combed scores of matches when deciding what match to use as the
final match. Available values are:
</p>
<dl compact="compact">
<dt>‘<samp>none</samp>’</dt>
<dd><p>No final matching based on combed scores.
</p></dd>
<dt>‘<samp>sc</samp>’</dt>
<dd><p>Combed scores are only used when a scene change is detected.
</p></dd>
<dt>‘<samp>full</samp>’</dt>
<dd><p>Use combed scores all the time.
</p></dd>
</dl>
<p>Default is <var>sc</var>.
</p>
</dd>
<dt>‘<samp>combdbg</samp>’</dt>
<dd><p>Force <code>fieldmatch</code> to calculate the combed metrics for certain matches and
print them. This setting is known as ‘<samp>micout</samp>’ in TFM/VFM vocabulary.
Available values are:
</p>
<dl compact="compact">
<dt>‘<samp>none</samp>’</dt>
<dd><p>No forced calculation.
</p></dd>
<dt>‘<samp>pcn</samp>’</dt>
<dd><p>Force p/c/n calculations.
</p></dd>
<dt>‘<samp>pcnub</samp>’</dt>
<dd><p>Force p/c/n/u/b calculations.
</p></dd>
</dl>
<p>Default value is <var>none</var>.
</p>
</dd>
<dt>‘<samp>cthresh</samp>’</dt>
<dd><p>This is the area combing threshold used for combed frame detection. This
essentially controls how "strong" or "visible" combing must be to be detected.
Larger values mean combing must be more visible and smaller values mean combing
can be less visible or strong and still be detected. Valid settings are from
<code>-1</code> (every pixel will be detected as combed) to <code>255</code> (no pixel will
be detected as combed). This is basically a pixel difference value. A good
range is <code>[8, 12]</code>.
</p>
<p>Default value is <code>9</code>.
</p>
</dd>
<dt>‘<samp>chroma</samp>’</dt>
<dd><p>Sets whether or not chroma is considered in the combed frame decision. Only
disable this if your source has chroma problems (rainbowing, etc.) that are
causing problems for the combed frame detection with chroma enabled. Actually,
using ‘<samp>chroma</samp>’=<var>0</var> is usually more reliable, except for the case
where there is chroma only combing in the source.
</p>
<p>Default value is <code>0</code>.
</p>
</dd>
<dt>‘<samp>blockx</samp>’</dt>
<dt>‘<samp>blocky</samp>’</dt>
<dd><p>Respectively set the x-axis and y-axis size of the window used during combed
frame detection. This has to do with the size of the area in which
‘<samp>combpel</samp>’ pixels are required to be detected as combed for a frame to be
declared combed. See the ‘<samp>combpel</samp>’ parameter description for more info.
Possible values are any number that is a power of 2 starting at 4 and going up
to 512.
</p>
<p>Default value is <code>16</code>.
</p>
</dd>
<dt>‘<samp>combpel</samp>’</dt>
<dd><p>The number of combed pixels inside any of the ‘<samp>blocky</samp>’ by
‘<samp>blockx</samp>’ size blocks on the frame for the frame to be detected as
combed. While ‘<samp>cthresh</samp>’ controls how "visible" the combing must be, this
setting controls "how much" combing there must be in any localized area (a
window defined by the ‘<samp>blockx</samp>’ and ‘<samp>blocky</samp>’ settings) on the
frame. Minimum value is <code>0</code> and maximum is <code>blocky x blockx</code> (at
which point no frames will ever be detected as combed). This setting is known
as ‘<samp>MI</samp>’ in TFM/VFM vocabulary.
</p>
<p>Default value is <code>80</code>.
</p></dd>
</dl>
<p><a name="p_002fc_002fn_002fu_002fb-meaning"></a>
</p><a name="p_002fc_002fn_002fu_002fb-meaning-1"></a>
<h3 class="subsection"><a href="ffmpeg-filters.html#toc-p_002fc_002fn_002fu_002fb-meaning-1">9.30.1 p/c/n/u/b meaning</a></h3>
<a name="p_002fc_002fn"></a>
<h4 class="subsubsection"><a href="ffmpeg-filters.html#toc-p_002fc_002fn">9.30.1.1 p/c/n</a></h4>
<p>We assume the following telecined stream:
</p>
<div class="example">
<pre class="example">Top fields: 1 2 2 3 4
Bottom fields: 1 2 3 4 4
</pre></div>
<p>The numbers correspond to the progressive frame the fields relate to. Here, the
first two frames are progressive, the 3rd and 4th are combed, and so on.
</p>
<p>When <code>fieldmatch</code> is configured to run a matching from bottom
(‘<samp>field</samp>’=<var>bottom</var>) this is how this input stream get transformed:
</p>
<div class="example">
<pre class="example">Input stream:
T 1 2 2 3 4
B 1 2 3 4 4 <-- matching reference
Matches: c c n n c
Output stream:
T 1 2 3 4 4
B 1 2 3 4 4
</pre></div>
<p>As a result of the field matching, we can see that some frames get duplicated.
To perform a complete inverse telecine, you need to rely on a decimation filter
after this operation. See for instance the <a href="#decimate">decimate</a> filter.
</p>
<p>The same operation now matching from top fields (‘<samp>field</samp>’=<var>top</var>)
looks like this:
</p>
<div class="example">
<pre class="example">Input stream:
T 1 2 2 3 4 <-- matching reference
B 1 2 3 4 4
Matches: c c p p c
Output stream:
T 1 2 2 3 4
B 1 2 2 3 4
</pre></div>
<p>In these examples, we can see what <var>p</var>, <var>c</var> and <var>n</var> mean;
basically, they refer to the frame and field of the opposite parity:
</p>
<ul>
<li> <var>p</var> matches the field of the opposite parity in the previous frame
</li><li> <var>c</var> matches the field of the opposite parity in the current frame
</li><li> <var>n</var> matches the field of the opposite parity in the next frame
</li></ul>
<a name="u_002fb"></a>
<h4 class="subsubsection"><a href="ffmpeg-filters.html#toc-u_002fb">9.30.1.2 u/b</a></h4>
<p>The <var>u</var> and <var>b</var> matching are a bit special in the sense that they match
from the opposite parity flag. In the following examples, we assume that we are
currently matching the 2nd frame (Top:2, bottom:2). According to the match, a
’x’ is placed above and below each matched fields.
</p>
<p>With bottom matching (‘<samp>field</samp>’=<var>bottom</var>):
</p><div class="example">
<pre class="example">Match: c p n b u
x x x x x
Top 1 2 2 1 2 2 1 2 2 1 2 2 1 2 2
Bottom 1 2 3 1 2 3 1 2 3 1 2 3 1 2 3
x x x x x
Output frames:
2 1 2 2 2
2 2 2 1 3
</pre></div>
<p>With top matching (‘<samp>field</samp>’=<var>top</var>):
</p><div class="example">
<pre class="example">Match: c p n b u
x x x x x
Top 1 2 2 1 2 2 1 2 2 1 2 2 1 2 2
Bottom 1 2 3 1 2 3 1 2 3 1 2 3 1 2 3
x x x x x
Output frames:
2 2 2 1 2
2 1 3 2 2
</pre></div>
<a name="Examples-62"></a>
<h3 class="subsection"><a href="ffmpeg-filters.html#toc-Examples-62">9.30.2 Examples</a></h3>
<p>Simple IVTC of a top field first telecined stream:
</p><div class="example">
<pre class="example">fieldmatch=order=tff:combmatch=none, decimate
</pre></div>
<p>Advanced IVTC, with fallback on <a href="#yadif">yadif</a> for still combed frames:
</p><div class="example">
<pre class="example">fieldmatch=order=tff:combmatch=full, yadif=deint=interlaced, decimate
</pre></div>
<a name="fieldorder"></a>
<h2 class="section"><a href="ffmpeg-filters.html#toc-fieldorder">9.31 fieldorder</a></h2>
<p>Transform the field order of the input video.
</p>
<p>It accepts the following parameters:
</p>
<dl compact="compact">
<dt>‘<samp>order</samp>’</dt>
<dd><p>The output field order. Valid values are <var>tff</var> for top field first or <var>bff</var>
for bottom field first.
</p></dd>
</dl>
<p>The default value is ‘<samp>tff</samp>’.
</p>
<p>The transformation is done by shifting the picture content up or down
by one line, and filling the remaining line with appropriate picture content.
This method is consistent with most broadcast field order converters.
</p>
<p>If the input video is not flagged as being interlaced, or it is already
flagged as being of the required output field order, then this filter does
not alter the incoming video.
</p>
<p>It is very useful when converting to or from PAL DV material,
which is bottom field first.
</p>
<p>For example:
</p><div class="example">
<pre class="example">ffmpeg -i in.vob -vf "fieldorder=bff" out.dv
</pre></div>
<a name="fifo"></a>
<h2 class="section"><a href="ffmpeg-filters.html#toc-fifo">9.32 fifo</a></h2>
<p>Buffer input images and send them when they are requested.
</p>
<p>It is mainly useful when auto-inserted by the libavfilter
framework.
</p>
<p>It does not take parameters.
</p>
<p><a name="format"></a>
</p><a name="format-1"></a>
<h2 class="section"><a href="ffmpeg-filters.html#toc-format-1">9.33 format</a></h2>
<p>Convert the input video to one of the specified pixel formats.
Libavfilter will try to pick one that is suitable as input to
the next filter.
</p>
<p>It accepts the following parameters:
</p><dl compact="compact">
<dt>‘<samp>pix_fmts</samp>’</dt>
<dd><p>A ’|’-separated list of pixel format names, such as
"pix_fmts=yuv420p|monow|rgb24".
</p>
</dd>
</dl>
<a name="Examples-41"></a>
<h3 class="subsection"><a href="ffmpeg-filters.html#toc-Examples-41">9.33.1 Examples</a></h3>
<ul>
<li>
Convert the input video to the <var>yuv420p</var> format
<div class="example">
<pre class="example">format=pix_fmts=yuv420p
</pre></div>
<p>Convert the input video to any of the formats in the list
</p><div class="example">
<pre class="example">format=pix_fmts=yuv420p|yuv444p|yuv410p
</pre></div>
</li></ul>
<p><a name="fps"></a>
</p><a name="fps-1"></a>
<h2 class="section"><a href="ffmpeg-filters.html#toc-fps-1">9.34 fps</a></h2>
<p>Convert the video to specified constant frame rate by duplicating or dropping
frames as necessary.
</p>
<p>It accepts the following parameters:
</p><dl compact="compact">
<dt>‘<samp>fps</samp>’</dt>
<dd><p>The desired output frame rate. The default is <code>25</code>.
</p>
</dd>
<dt>‘<samp>round</samp>’</dt>
<dd><p>Rounding method.
</p>
<p>Possible values are:
</p><dl compact="compact">
<dt>‘<samp>zero</samp>’</dt>
<dd><p>zero round towards 0
</p></dd>
<dt>‘<samp>inf</samp>’</dt>
<dd><p>round away from 0
</p></dd>
<dt>‘<samp>down</samp>’</dt>
<dd><p>round towards -infinity
</p></dd>
<dt>‘<samp>up</samp>’</dt>
<dd><p>round towards +infinity
</p></dd>
<dt>‘<samp>near</samp>’</dt>
<dd><p>round to nearest
</p></dd>
</dl>
<p>The default is <code>near</code>.
</p>
</dd>
<dt>‘<samp>start_time</samp>’</dt>
<dd><p>Assume the first PTS should be the given value, in seconds. This allows for
padding/trimming at the start of stream. By default, no assumption is made
about the first frame’s expected PTS, so no padding or trimming is done.
For example, this could be set to 0 to pad the beginning with duplicates of
the first frame if a video stream starts after the audio stream or to trim any
frames with a negative PTS.
</p>
</dd>
</dl>
<p>Alternatively, the options can be specified as a flat string:
<var>fps</var>[:<var>round</var>].
</p>
<p>See also the <a href="#setpts">setpts</a> filter.
</p>
<a name="Examples-34"></a>
<h3 class="subsection"><a href="ffmpeg-filters.html#toc-Examples-34">9.34.1 Examples</a></h3>
<ul>
<li>
A typical usage in order to set the fps to 25:
<div class="example">
<pre class="example">fps=fps=25
</pre></div>
</li><li>
Sets the fps to 24, using abbreviation and rounding method to round to nearest:
<div class="example">
<pre class="example">fps=fps=film:round=near
</pre></div>
</li></ul>
<a name="framepack"></a>
<h2 class="section"><a href="ffmpeg-filters.html#toc-framepack">9.35 framepack</a></h2>
<p>Pack two different video streams into a stereoscopic video, setting proper
metadata on supported codecs. The two views should have the same size and
framerate and processing will stop when the shorter video ends. Please note
that you may conveniently adjust view properties with the <a href="#scale">scale</a> and
<a href="#fps">fps</a> filters.
</p>
<p>It accepts the following parameters:
</p><dl compact="compact">
<dt>‘<samp>format</samp>’</dt>
<dd><p>The desired packing format. Supported values are:
</p>
<dl compact="compact">
<dt>‘<samp>sbs</samp>’</dt>
<dd><p>The views are next to each other (default).
</p>
</dd>
<dt>‘<samp>tab</samp>’</dt>
<dd><p>The views are on top of each other.
</p>
</dd>
<dt>‘<samp>lines</samp>’</dt>
<dd><p>The views are packed by line.
</p>
</dd>
<dt>‘<samp>columns</samp>’</dt>
<dd><p>The views are packed by column.
</p>
</dd>
<dt>‘<samp>frameseq</samp>’</dt>
<dd><p>The views are temporally interleaved.
</p>
</dd>
</dl>
</dd>
</dl>
<p>Some examples:
</p>
<div class="example">
<pre class="example"># Convert left and right views into a frame-sequential video
ffmpeg -i LEFT -i RIGHT -filter_complex framepack=frameseq OUTPUT
# Convert views into a side-by-side video with the same output resolution as the input
ffmpeg -i LEFT -i RIGHT -filter_complex [0:v]scale=w=iw/2[left],[1:v]scale=w=iw/2[right],[left][right]framepack=sbs OUTPUT
</pre></div>
<a name="framestep"></a>
<h2 class="section"><a href="ffmpeg-filters.html#toc-framestep">9.36 framestep</a></h2>
<p>Select one frame every N-th frame.
</p>
<p>This filter accepts the following option:
</p><dl compact="compact">
<dt>‘<samp>step</samp>’</dt>
<dd><p>Select frame after every <code>step</code> frames.
Allowed values are positive integers higher than 0. Default value is <code>1</code>.
</p></dd>
</dl>
<p><a name="frei0r"></a>
</p><a name="frei0r-1"></a>
<h2 class="section"><a href="ffmpeg-filters.html#toc-frei0r-1">9.37 frei0r</a></h2>
<p>Apply a frei0r effect to the input video.
</p>
<p>To enable the compilation of this filter, you need to install the frei0r
header and configure FFmpeg with <code>--enable-frei0r</code>.
</p>
<p>It accepts the following parameters:
</p>
<dl compact="compact">
<dt>‘<samp>filter_name</samp>’</dt>
<dd><p>The name of the frei0r effect to load. If the environment variable
<code>FREI0R_PATH</code> is defined, the frei0r effect is searched for in each of the
directories specified by the colon-separated list in <code>FREIOR_PATH</code>.
Otherwise, the standard frei0r paths are searched, in this order:
‘<tt>HOME/.frei0r-1/lib/</tt>’, ‘<tt>/usr/local/lib/frei0r-1/</tt>’,
‘<tt>/usr/lib/frei0r-1/</tt>’.
</p>
</dd>
<dt>‘<samp>filter_params</samp>’</dt>
<dd><p>A ’|’-separated list of parameters to pass to the frei0r effect.
</p>
</dd>
</dl>
<p>A frei0r effect parameter can be a boolean (its value is either
"y" or "n"), a double, a color (specified as
<var>R</var>/<var>G</var>/<var>B</var>, where <var>R</var>, <var>G</var>, and <var>B</var> are floating point
numbers between 0.0 and 1.0, inclusive) or by a color description specified in the "Color"
section in the ffmpeg-utils manual), a position (specified as <var>X</var>/<var>Y</var>, where
<var>X</var> and <var>Y</var> are floating point numbers) and/or a string.
</p>
<p>The number and types of parameters depend on the loaded effect. If an
effect parameter is not specified, the default value is set.
</p>
<a name="Examples-36"></a>
<h3 class="subsection"><a href="ffmpeg-filters.html#toc-Examples-36">9.37.1 Examples</a></h3>
<ul>
<li>
Apply the distort0r effect, setting the first two double parameters:
<div class="example">
<pre class="example">frei0r=filter_name=distort0r:filter_params=0.5|0.01
</pre></div>
</li><li>
Apply the colordistance effect, taking a color as the first parameter:
<div class="example">
<pre class="example">frei0r=colordistance:0.2/0.3/0.4
frei0r=colordistance:violet
frei0r=colordistance:0x112233
</pre></div>
</li><li>
Apply the perspective effect, specifying the top left and top right image
positions:
<div class="example">
<pre class="example">frei0r=perspective:0.2/0.2|0.8/0.2
</pre></div>
</li></ul>
<p>For more information, see
<a href="http://frei0r.dyne.org">http://frei0r.dyne.org</a>
</p>
<a name="geq"></a>
<h2 class="section"><a href="ffmpeg-filters.html#toc-geq">9.38 geq</a></h2>
<p>The filter accepts the following options:
</p>
<dl compact="compact">
<dt>‘<samp>lum_expr, lum</samp>’</dt>
<dd><p>Set the luminance expression.
</p></dd>
<dt>‘<samp>cb_expr, cb</samp>’</dt>
<dd><p>Set the chrominance blue expression.
</p></dd>
<dt>‘<samp>cr_expr, cr</samp>’</dt>
<dd><p>Set the chrominance red expression.
</p></dd>
<dt>‘<samp>alpha_expr, a</samp>’</dt>
<dd><p>Set the alpha expression.
</p></dd>
<dt>‘<samp>red_expr, r</samp>’</dt>
<dd><p>Set the red expression.
</p></dd>
<dt>‘<samp>green_expr, g</samp>’</dt>
<dd><p>Set the green expression.
</p></dd>
<dt>‘<samp>blue_expr, b</samp>’</dt>
<dd><p>Set the blue expression.
</p></dd>
</dl>
<p>The colorspace is selected according to the specified options. If one
of the ‘<samp>lum_expr</samp>’, ‘<samp>cb_expr</samp>’, or ‘<samp>cr_expr</samp>’
options is specified, the filter will automatically select a YCbCr
colorspace. If one of the ‘<samp>red_expr</samp>’, ‘<samp>green_expr</samp>’, or
‘<samp>blue_expr</samp>’ options is specified, it will select an RGB
colorspace.
</p>
<p>If one of the chrominance expression is not defined, it falls back on the other
one. If no alpha expression is specified it will evaluate to opaque value.
If none of chrominance expressions are specified, they will evaluate
to the luminance expression.
</p>
<p>The expressions can use the following variables and functions:
</p>
<dl compact="compact">
<dt>‘<samp>N</samp>’</dt>
<dd><p>The sequential number of the filtered frame, starting from <code>0</code>.
</p>
</dd>
<dt>‘<samp>X</samp>’</dt>
<dt>‘<samp>Y</samp>’</dt>
<dd><p>The coordinates of the current sample.
</p>
</dd>
<dt>‘<samp>W</samp>’</dt>
<dt>‘<samp>H</samp>’</dt>
<dd><p>The width and height of the image.
</p>
</dd>
<dt>‘<samp>SW</samp>’</dt>
<dt>‘<samp>SH</samp>’</dt>
<dd><p>Width and height scale depending on the currently filtered plane. It is the
ratio between the corresponding luma plane number of pixels and the current
plane ones. E.g. for YUV4:2:0 the values are <code>1,1</code> for the luma plane, and
<code>0.5,0.5</code> for chroma planes.
</p>
</dd>
<dt>‘<samp>T</samp>’</dt>
<dd><p>Time of the current frame, expressed in seconds.
</p>
</dd>
<dt>‘<samp>p(x, y)</samp>’</dt>
<dd><p>Return the value of the pixel at location (<var>x</var>,<var>y</var>) of the current
plane.
</p>
</dd>
<dt>‘<samp>lum(x, y)</samp>’</dt>
<dd><p>Return the value of the pixel at location (<var>x</var>,<var>y</var>) of the luminance
plane.
</p>
</dd>
<dt>‘<samp>cb(x, y)</samp>’</dt>
<dd><p>Return the value of the pixel at location (<var>x</var>,<var>y</var>) of the
blue-difference chroma plane. Return 0 if there is no such plane.
</p>
</dd>
<dt>‘<samp>cr(x, y)</samp>’</dt>
<dd><p>Return the value of the pixel at location (<var>x</var>,<var>y</var>) of the
red-difference chroma plane. Return 0 if there is no such plane.
</p>
</dd>
<dt>‘<samp>r(x, y)</samp>’</dt>
<dt>‘<samp>g(x, y)</samp>’</dt>
<dt>‘<samp>b(x, y)</samp>’</dt>
<dd><p>Return the value of the pixel at location (<var>x</var>,<var>y</var>) of the
red/green/blue component. Return 0 if there is no such component.
</p>
</dd>
<dt>‘<samp>alpha(x, y)</samp>’</dt>
<dd><p>Return the value of the pixel at location (<var>x</var>,<var>y</var>) of the alpha
plane. Return 0 if there is no such plane.
</p></dd>
</dl>
<p>For functions, if <var>x</var> and <var>y</var> are outside the area, the value will be
automatically clipped to the closer edge.
</p>
<a name="Examples-19"></a>
<h3 class="subsection"><a href="ffmpeg-filters.html#toc-Examples-19">9.38.1 Examples</a></h3>
<ul>
<li>
Flip the image horizontally:
<div class="example">
<pre class="example">geq=p(W-X\,Y)
</pre></div>
</li><li>
Generate a bidimensional sine wave, with angle <code>PI/3</code> and a
wavelength of 100 pixels:
<div class="example">
<pre class="example">geq=128 + 100*sin(2*(PI/100)*(cos(PI/3)*(X-50*T) + sin(PI/3)*Y)):128:128
</pre></div>
</li><li>
Generate a fancy enigmatic moving light:
<div class="example">
<pre class="example">nullsrc=s=256x256,geq=random(1)/hypot(X-cos(N*0.07)*W/2-W/2\,Y-sin(N*0.09)*H/2-H/2)^2*1000000*sin(N*0.02):128:128
</pre></div>
</li><li>
Generate a quick emboss effect:
<div class="example">
<pre class="example">format=gray,geq=lum_expr='(p(X,Y)+(256-p(X-4,Y-4)))/2'
</pre></div>
</li><li>
Modify RGB components depending on pixel position:
<div class="example">
<pre class="example">geq=r='X/W*r(X,Y)':g='(1-X/W)*g(X,Y)':b='(H-Y)/H*b(X,Y)'
</pre></div>
</li></ul>
<a name="gradfun"></a>
<h2 class="section"><a href="ffmpeg-filters.html#toc-gradfun">9.39 gradfun</a></h2>
<p>Fix the banding artifacts that are sometimes introduced into nearly flat
regions by truncation to 8bit color depth.
Interpolate the gradients that should go where the bands are, and
dither them.
</p>
<p>It is designed for playback only. Do not use it prior to
lossy compression, because compression tends to lose the dither and
bring back the bands.
</p>
<p>It accepts the following parameters:
</p>
<dl compact="compact">
<dt>‘<samp>strength</samp>’</dt>
<dd><p>The maximum amount by which the filter will change any one pixel. This is also
the threshold for detecting nearly flat regions. Acceptable values range from
.51 to 64; the default value is 1.2. Out-of-range values will be clipped to the
valid range.
</p>
</dd>
<dt>‘<samp>radius</samp>’</dt>
<dd><p>The neighborhood to fit the gradient to. A larger radius makes for smoother
gradients, but also prevents the filter from modifying the pixels near detailed
regions. Acceptable values are 8-32; the default value is 16. Out-of-range
values will be clipped to the valid range.
</p>
</dd>
</dl>
<p>Alternatively, the options can be specified as a flat string:
<var>strength</var>[:<var>radius</var>]
</p>
<a name="Examples-44"></a>
<h3 class="subsection"><a href="ffmpeg-filters.html#toc-Examples-44">9.39.1 Examples</a></h3>
<ul>
<li>
Apply the filter with a <code>3.5</code> strength and radius of <code>8</code>:
<div class="example">
<pre class="example">gradfun=3.5:8
</pre></div>
</li><li>
Specify radius, omitting the strength (which will fall-back to the default
value):
<div class="example">
<pre class="example">gradfun=radius=8
</pre></div>
</li></ul>
<p><a name="haldclut"></a>
</p><a name="haldclut-1"></a>
<h2 class="section"><a href="ffmpeg-filters.html#toc-haldclut-1">9.40 haldclut</a></h2>
<p>Apply a Hald CLUT to a video stream.
</p>
<p>First input is the video stream to process, and second one is the Hald CLUT.
The Hald CLUT input can be a simple picture or a complete video stream.
</p>
<p>The filter accepts the following options:
</p>
<dl compact="compact">
<dt>‘<samp>shortest</samp>’</dt>
<dd><p>Force termination when the shortest input terminates. Default is <code>0</code>.
</p></dd>
<dt>‘<samp>repeatlast</samp>’</dt>
<dd><p>Continue applying the last CLUT after the end of the stream. A value of
<code>0</code> disable the filter after the last frame of the CLUT is reached.
Default is <code>1</code>.
</p></dd>
</dl>
<p><code>haldclut</code> also has the same interpolation options as <a href="#lut3d">lut3d</a> (both
filters share the same internals).
</p>
<p>More information about the Hald CLUT can be found on Eskil Steenberg’s website
(Hald CLUT author) at <a href="http://www.quelsolaar.com/technology/clut.html">http://www.quelsolaar.com/technology/clut.html</a>.
</p>
<a name="Workflow-examples"></a>
<h3 class="subsection"><a href="ffmpeg-filters.html#toc-Workflow-examples">9.40.1 Workflow examples</a></h3>
<a name="Hald-CLUT-video-stream"></a>
<h4 class="subsubsection"><a href="ffmpeg-filters.html#toc-Hald-CLUT-video-stream">9.40.1.1 Hald CLUT video stream</a></h4>
<p>Generate an identity Hald CLUT stream altered with various effects:
</p><div class="example">
<pre class="example">ffmpeg -f lavfi -i <a href="#haldclutsrc">haldclutsrc</a>=8 -vf "hue=H=2*PI*t:s=sin(2*PI*t)+1, curves=cross_process" -t 10 -c:v ffv1 clut.nut
</pre></div>
<p>Note: make sure you use a lossless codec.
</p>
<p>Then use it with <code>haldclut</code> to apply it on some random stream:
</p><div class="example">
<pre class="example">ffmpeg -f lavfi -i mandelbrot -i clut.nut -filter_complex '[0][1] haldclut' -t 20 mandelclut.mkv
</pre></div>
<p>The Hald CLUT will be applied to the 10 first seconds (duration of
‘<tt>clut.nut</tt>’), then the latest picture of that CLUT stream will be applied
to the remaining frames of the <code>mandelbrot</code> stream.
</p>
<a name="Hald-CLUT-with-preview"></a>
<h4 class="subsubsection"><a href="ffmpeg-filters.html#toc-Hald-CLUT-with-preview">9.40.1.2 Hald CLUT with preview</a></h4>
<p>A Hald CLUT is supposed to be a squared image of <code>Level*Level*Level</code> by
<code>Level*Level*Level</code> pixels. For a given Hald CLUT, FFmpeg will select the
biggest possible square starting at the top left of the picture. The remaining
padding pixels (bottom or right) will be ignored. This area can be used to add
a preview of the Hald CLUT.
</p>
<p>Typically, the following generated Hald CLUT will be supported by the
<code>haldclut</code> filter:
</p>
<div class="example">
<pre class="example">ffmpeg -f lavfi -i <a href="#haldclutsrc">haldclutsrc</a>=8 -vf "
pad=iw+320 [padded_clut];
smptebars=s=320x256, split [a][b];
[padded_clut][a] overlay=W-320:h, curves=color_negative [main];
[main][b] overlay=W-320" -frames:v 1 clut.png
</pre></div>
<p>It contains the original and a preview of the effect of the CLUT: SMPTE color
bars are displayed on the right-top, and below the same color bars processed by
the color changes.
</p>
<p>Then, the effect of this Hald CLUT can be visualized with:
</p><div class="example">
<pre class="example">ffplay input.mkv -vf "movie=clut.png, [in] haldclut"
</pre></div>
<a name="hflip"></a>
<h2 class="section"><a href="ffmpeg-filters.html#toc-hflip">9.41 hflip</a></h2>
<p>Flip the input video horizontally.
</p>
<p>For example, to horizontally flip the input video with <code>ffmpeg</code>:
</p><div class="example">
<pre class="example">ffmpeg -i in.avi -vf "hflip" out.avi
</pre></div>
<a name="histeq"></a>
<h2 class="section"><a href="ffmpeg-filters.html#toc-histeq">9.42 histeq</a></h2>
<p>This filter applies a global color histogram equalization on a
per-frame basis.
</p>
<p>It can be used to correct video that has a compressed range of pixel
intensities. The filter redistributes the pixel intensities to
equalize their distribution across the intensity range. It may be
viewed as an "automatically adjusting contrast filter". This filter is
useful only for correcting degraded or poorly captured source
video.
</p>
<p>The filter accepts the following options:
</p>
<dl compact="compact">
<dt>‘<samp>strength</samp>’</dt>
<dd><p>Determine the amount of equalization to be applied. As the strength
is reduced, the distribution of pixel intensities more-and-more
approaches that of the input frame. The value must be a float number
in the range [0,1] and defaults to 0.200.
</p>
</dd>
<dt>‘<samp>intensity</samp>’</dt>
<dd><p>Set the maximum intensity that can generated and scale the output
values appropriately. The strength should be set as desired and then
the intensity can be limited if needed to avoid washing-out. The value
must be a float number in the range [0,1] and defaults to 0.210.
</p>
</dd>
<dt>‘<samp>antibanding</samp>’</dt>
<dd><p>Set the antibanding level. If enabled the filter will randomly vary
the luminance of output pixels by a small amount to avoid banding of
the histogram. Possible values are <code>none</code>, <code>weak</code> or
<code>strong</code>. It defaults to <code>none</code>.
</p></dd>
</dl>
<a name="histogram"></a>
<h2 class="section"><a href="ffmpeg-filters.html#toc-histogram">9.43 histogram</a></h2>
<p>Compute and draw a color distribution histogram for the input video.
</p>
<p>The computed histogram is a representation of the color component
distribution in an image.
</p>
<p>The filter accepts the following options:
</p>
<dl compact="compact">
<dt>‘<samp>mode</samp>’</dt>
<dd><p>Set histogram mode.
</p>
<p>It accepts the following values:
</p><dl compact="compact">
<dt>‘<samp>levels</samp>’</dt>
<dd><p>Standard histogram that displays the color components distribution in an
image. Displays color graph for each color component. Shows distribution of
the Y, U, V, A or R, G, B components, depending on input format, in the
current frame. Below each graph a color component scale meter is shown.
</p>
</dd>
<dt>‘<samp>color</samp>’</dt>
<dd><p>Displays chroma values (U/V color placement) in a two dimensional
graph (which is called a vectorscope). The brighter a pixel in the
vectorscope, the more pixels of the input frame correspond to that pixel
(i.e., more pixels have this chroma value). The V component is displayed on
the horizontal (X) axis, with the leftmost side being V = 0 and the rightmost
side being V = 255. The U component is displayed on the vertical (Y) axis,
with the top representing U = 0 and the bottom representing U = 255.
</p>
<p>The position of a white pixel in the graph corresponds to the chroma value of
a pixel of the input clip. The graph can therefore be used to read the hue
(color flavor) and the saturation (the dominance of the hue in the color). As
the hue of a color changes, it moves around the square. At the center of the
square the saturation is zero, which means that the corresponding pixel has no
color. If the amount of a specific color is increased (while leaving the other
colors unchanged) the saturation increases, and the indicator moves towards
the edge of the square.
</p>
</dd>
<dt>‘<samp>color2</samp>’</dt>
<dd><p>Chroma values in vectorscope, similar as <code>color</code> but actual chroma values
are displayed.
</p>
</dd>
<dt>‘<samp>waveform</samp>’</dt>
<dd><p>Per row/column color component graph. In row mode, the graph on the left side
represents color component value 0 and the right side represents value = 255.
In column mode, the top side represents color component value = 0 and bottom
side represents value = 255.
</p></dd>
</dl>
<p>Default value is <code>levels</code>.
</p>
</dd>
<dt>‘<samp>level_height</samp>’</dt>
<dd><p>Set height of level in <code>levels</code>. Default value is <code>200</code>.
Allowed range is [50, 2048].
</p>
</dd>
<dt>‘<samp>scale_height</samp>’</dt>
<dd><p>Set height of color scale in <code>levels</code>. Default value is <code>12</code>.
Allowed range is [0, 40].
</p>
</dd>
<dt>‘<samp>step</samp>’</dt>
<dd><p>Set step for <code>waveform</code> mode. Smaller values are useful to find out how
many values of the same luminance are distributed across input rows/columns.
Default value is <code>10</code>. Allowed range is [1, 255].
</p>
</dd>
<dt>‘<samp>waveform_mode</samp>’</dt>
<dd><p>Set mode for <code>waveform</code>. Can be either <code>row</code>, or <code>column</code>.
Default is <code>row</code>.
</p>
</dd>
<dt>‘<samp>waveform_mirror</samp>’</dt>
<dd><p>Set mirroring mode for <code>waveform</code>. <code>0</code> means unmirrored, <code>1</code>
means mirrored. In mirrored mode, higher values will be represented on the left
side for <code>row</code> mode and at the top for <code>column</code> mode. Default is
<code>0</code> (unmirrored).
</p>
</dd>
<dt>‘<samp>display_mode</samp>’</dt>
<dd><p>Set display mode for <code>waveform</code> and <code>levels</code>.
It accepts the following values:
</p><dl compact="compact">
<dt>‘<samp>parade</samp>’</dt>
<dd><p>Display separate graph for the color components side by side in
<code>row</code> waveform mode or one below the other in <code>column</code> waveform mode
for <code>waveform</code> histogram mode. For <code>levels</code> histogram mode,
per color component graphs are placed below each other.
</p>
<p>Using this display mode in <code>waveform</code> histogram mode makes it easy to
spot color casts in the highlights and shadows of an image, by comparing the
contours of the top and the bottom graphs of each waveform. Since whites,
grays, and blacks are characterized by exactly equal amounts of red, green,
and blue, neutral areas of the picture should display three waveforms of
roughly equal width/height. If not, the correction is easy to perform by
making level adjustments the three waveforms.
</p>
</dd>
<dt>‘<samp>overlay</samp>’</dt>
<dd><p>Presents information identical to that in the <code>parade</code>, except
that the graphs representing color components are superimposed directly
over one another.
</p>
<p>This display mode in <code>waveform</code> histogram mode makes it easier to spot
relative differences or similarities in overlapping areas of the color
components that are supposed to be identical, such as neutral whites, grays,
or blacks.
</p></dd>
</dl>
<p>Default is <code>parade</code>.
</p>
</dd>
<dt>‘<samp>levels_mode</samp>’</dt>
<dd><p>Set mode for <code>levels</code>. Can be either <code>linear</code>, or <code>logarithmic</code>.
Default is <code>linear</code>.
</p></dd>
</dl>
<a name="Examples-55"></a>
<h3 class="subsection"><a href="ffmpeg-filters.html#toc-Examples-55">9.43.1 Examples</a></h3>
<ul>
<li>
Calculate and draw histogram:
<div class="example">
<pre class="example">ffplay -i input -vf histogram
</pre></div>
</li></ul>
<p><a name="hqdn3d"></a>
</p><a name="hqdn3d-1"></a>
<h2 class="section"><a href="ffmpeg-filters.html#toc-hqdn3d-1">9.44 hqdn3d</a></h2>
<p>This is a high precision/quality 3d denoise filter. It aims to reduce
image noise, producing smooth images and making still images really
still. It should enhance compressibility.
</p>
<p>It accepts the following optional parameters:
</p>
<dl compact="compact">
<dt>‘<samp>luma_spatial</samp>’</dt>
<dd><p>A non-negative floating point number which specifies spatial luma strength.
It defaults to 4.0.
</p>
</dd>
<dt>‘<samp>chroma_spatial</samp>’</dt>
<dd><p>A non-negative floating point number which specifies spatial chroma strength.
It defaults to 3.0*<var>luma_spatial</var>/4.0.
</p>
</dd>
<dt>‘<samp>luma_tmp</samp>’</dt>
<dd><p>A floating point number which specifies luma temporal strength. It defaults to
6.0*<var>luma_spatial</var>/4.0.
</p>
</dd>
<dt>‘<samp>chroma_tmp</samp>’</dt>
<dd><p>A floating point number which specifies chroma temporal strength. It defaults to
<var>luma_tmp</var>*<var>chroma_spatial</var>/<var>luma_spatial</var>.
</p></dd>
</dl>
<a name="hqx"></a>
<h2 class="section"><a href="ffmpeg-filters.html#toc-hqx">9.45 hqx</a></h2>
<p>Apply a high-quality magnification filter designed for pixel art. This filter
was originally created by Maxim Stepin.
</p>
<p>It accepts the following option:
</p>
<dl compact="compact">
<dt>‘<samp>n</samp>’</dt>
<dd><p>Set the scaling dimension: <code>2</code> for <code>hq2x</code>, <code>3</code> for
<code>hq3x</code> and <code>4</code> for <code>hq4x</code>.
Default is <code>3</code>.
</p></dd>
</dl>
<a name="hue"></a>
<h2 class="section"><a href="ffmpeg-filters.html#toc-hue">9.46 hue</a></h2>
<p>Modify the hue and/or the saturation of the input.
</p>
<p>It accepts the following parameters:
</p>
<dl compact="compact">
<dt>‘<samp>h</samp>’</dt>
<dd><p>Specify the hue angle as a number of degrees. It accepts an expression,
and defaults to "0".
</p>
</dd>
<dt>‘<samp>s</samp>’</dt>
<dd><p>Specify the saturation in the [-10,10] range. It accepts an expression and
defaults to "1".
</p>
</dd>
<dt>‘<samp>H</samp>’</dt>
<dd><p>Specify the hue angle as a number of radians. It accepts an
expression, and defaults to "0".
</p>
</dd>
<dt>‘<samp>b</samp>’</dt>
<dd><p>Specify the brightness in the [-10,10] range. It accepts an expression and
defaults to "0".
</p></dd>
</dl>
<p>‘<samp>h</samp>’ and ‘<samp>H</samp>’ are mutually exclusive, and can’t be
specified at the same time.
</p>
<p>The ‘<samp>b</samp>’, ‘<samp>h</samp>’, ‘<samp>H</samp>’ and ‘<samp>s</samp>’ option values are
expressions containing the following constants:
</p>
<dl compact="compact">
<dt>‘<samp>n</samp>’</dt>
<dd><p>frame count of the input frame starting from 0
</p>
</dd>
<dt>‘<samp>pts</samp>’</dt>
<dd><p>presentation timestamp of the input frame expressed in time base units
</p>
</dd>
<dt>‘<samp>r</samp>’</dt>
<dd><p>frame rate of the input video, NAN if the input frame rate is unknown
</p>
</dd>
<dt>‘<samp>t</samp>’</dt>
<dd><p>timestamp expressed in seconds, NAN if the input timestamp is unknown
</p>
</dd>
<dt>‘<samp>tb</samp>’</dt>
<dd><p>time base of the input video
</p></dd>
</dl>
<a name="Examples-26"></a>
<h3 class="subsection"><a href="ffmpeg-filters.html#toc-Examples-26">9.46.1 Examples</a></h3>
<ul>
<li>
Set the hue to 90 degrees and the saturation to 1.0:
<div class="example">
<pre class="example">hue=h=90:s=1
</pre></div>
</li><li>
Same command but expressing the hue in radians:
<div class="example">
<pre class="example">hue=H=PI/2:s=1
</pre></div>
</li><li>
Rotate hue and make the saturation swing between 0
and 2 over a period of 1 second:
<div class="example">
<pre class="example">hue="H=2*PI*t: s=sin(2*PI*t)+1"
</pre></div>
</li><li>
Apply a 3 seconds saturation fade-in effect starting at 0:
<div class="example">
<pre class="example">hue="s=min(t/3\,1)"
</pre></div>
<p>The general fade-in expression can be written as:
</p><div class="example">
<pre class="example">hue="s=min(0\, max((t-START)/DURATION\, 1))"
</pre></div>
</li><li>
Apply a 3 seconds saturation fade-out effect starting at 5 seconds:
<div class="example">
<pre class="example">hue="s=max(0\, min(1\, (8-t)/3))"
</pre></div>
<p>The general fade-out expression can be written as:
</p><div class="example">
<pre class="example">hue="s=max(0\, min(1\, (START+DURATION-t)/DURATION))"
</pre></div>
</li></ul>
<a name="Commands-4"></a>
<h3 class="subsection"><a href="ffmpeg-filters.html#toc-Commands-4">9.46.2 Commands</a></h3>
<p>This filter supports the following commands:
</p><dl compact="compact">
<dt>‘<samp>b</samp>’</dt>
<dt>‘<samp>s</samp>’</dt>
<dt>‘<samp>h</samp>’</dt>
<dt>‘<samp>H</samp>’</dt>
<dd><p>Modify the hue and/or the saturation and/or brightness of the input video.
The command accepts the same syntax of the corresponding option.
</p>
<p>If the specified expression is not valid, it is kept at its current
value.
</p></dd>
</dl>
<a name="idet"></a>
<h2 class="section"><a href="ffmpeg-filters.html#toc-idet">9.47 idet</a></h2>
<p>Detect video interlacing type.
</p>
<p>This filter tries to detect if the input is interlaced or progressive,
top or bottom field first.
</p>
<p>The filter accepts the following options:
</p>
<dl compact="compact">
<dt>‘<samp>intl_thres</samp>’</dt>
<dd><p>Set interlacing threshold.
</p></dd>
<dt>‘<samp>prog_thres</samp>’</dt>
<dd><p>Set progressive threshold.
</p></dd>
</dl>
<a name="il"></a>
<h2 class="section"><a href="ffmpeg-filters.html#toc-il">9.48 il</a></h2>
<p>Deinterleave or interleave fields.
</p>
<p>This filter allows one to process interlaced images fields without
deinterlacing them. Deinterleaving splits the input frame into 2
fields (so called half pictures). Odd lines are moved to the top
half of the output image, even lines to the bottom half.
You can process (filter) them independently and then re-interleave them.
</p>
<p>The filter accepts the following options:
</p>
<dl compact="compact">
<dt>‘<samp>luma_mode, l</samp>’</dt>
<dt>‘<samp>chroma_mode, c</samp>’</dt>
<dt>‘<samp>alpha_mode, a</samp>’</dt>
<dd><p>Available values for <var>luma_mode</var>, <var>chroma_mode</var> and
<var>alpha_mode</var> are:
</p>
<dl compact="compact">
<dt>‘<samp>none</samp>’</dt>
<dd><p>Do nothing.
</p>
</dd>
<dt>‘<samp>deinterleave, d</samp>’</dt>
<dd><p>Deinterleave fields, placing one above the other.
</p>
</dd>
<dt>‘<samp>interleave, i</samp>’</dt>
<dd><p>Interleave fields. Reverse the effect of deinterleaving.
</p></dd>
</dl>
<p>Default value is <code>none</code>.
</p>
</dd>
<dt>‘<samp>luma_swap, ls</samp>’</dt>
<dt>‘<samp>chroma_swap, cs</samp>’</dt>
<dt>‘<samp>alpha_swap, as</samp>’</dt>
<dd><p>Swap luma/chroma/alpha fields. Exchange even & odd lines. Default value is <code>0</code>.
</p></dd>
</dl>
<a name="interlace"></a>
<h2 class="section"><a href="ffmpeg-filters.html#toc-interlace">9.49 interlace</a></h2>
<p>Simple interlacing filter from progressive contents. This interleaves upper (or
lower) lines from odd frames with lower (or upper) lines from even frames,
halving the frame rate and preserving image height.
</p>
<div class="example">
<pre class="example"> Original Original New Frame
Frame 'j' Frame 'j+1' (tff)
========== =========== ==================
Line 0 --------------------> Frame 'j' Line 0
Line 1 Line 1 ----> Frame 'j+1' Line 1
Line 2 ---------------------> Frame 'j' Line 2
Line 3 Line 3 ----> Frame 'j+1' Line 3
... ... ...
New Frame + 1 will be generated by Frame 'j+2' and Frame 'j+3' and so on
</pre></div>
<p>It accepts the following optional parameters:
</p>
<dl compact="compact">
<dt>‘<samp>scan</samp>’</dt>
<dd><p>This determines whether the interlaced frame is taken from the even
(tff - default) or odd (bff) lines of the progressive frame.
</p>
</dd>
<dt>‘<samp>lowpass</samp>’</dt>
<dd><p>Enable (default) or disable the vertical lowpass filter to avoid twitter
interlacing and reduce moire patterns.
</p></dd>
</dl>
<a name="kerndeint"></a>
<h2 class="section"><a href="ffmpeg-filters.html#toc-kerndeint">9.50 kerndeint</a></h2>
<p>Deinterlace input video by applying Donald Graft’s adaptive kernel
deinterling. Work on interlaced parts of a video to produce
progressive frames.
</p>
<p>The description of the accepted parameters follows.
</p>
<dl compact="compact">
<dt>‘<samp>thresh</samp>’</dt>
<dd><p>Set the threshold which affects the filter’s tolerance when
determining if a pixel line must be processed. It must be an integer
in the range [0,255] and defaults to 10. A value of 0 will result in
applying the process on every pixels.
</p>
</dd>
<dt>‘<samp>map</samp>’</dt>
<dd><p>Paint pixels exceeding the threshold value to white if set to 1.
Default is 0.
</p>
</dd>
<dt>‘<samp>order</samp>’</dt>
<dd><p>Set the fields order. Swap fields if set to 1, leave fields alone if
0. Default is 0.
</p>
</dd>
<dt>‘<samp>sharp</samp>’</dt>
<dd><p>Enable additional sharpening if set to 1. Default is 0.
</p>
</dd>
<dt>‘<samp>twoway</samp>’</dt>
<dd><p>Enable twoway sharpening if set to 1. Default is 0.
</p></dd>
</dl>
<a name="Examples-74"></a>
<h3 class="subsection"><a href="ffmpeg-filters.html#toc-Examples-74">9.50.1 Examples</a></h3>
<ul>
<li>
Apply default values:
<div class="example">
<pre class="example">kerndeint=thresh=10:map=0:order=0:sharp=0:twoway=0
</pre></div>
</li><li>
Enable additional sharpening:
<div class="example">
<pre class="example">kerndeint=sharp=1
</pre></div>
</li><li>
Paint processed pixels in white:
<div class="example">
<pre class="example">kerndeint=map=1
</pre></div>
</li></ul>
<a name="lenscorrection"></a>
<h2 class="section"><a href="ffmpeg-filters.html#toc-lenscorrection">9.51 lenscorrection</a></h2>
<p>Correct radial lens distortion
</p>
<p>This filter can be used to correct for radial distortion as can result from the use
of wide angle lenses, and thereby re-rectify the image. To find the right parameters
one can use tools available for example as part of opencv or simply trial-and-error.
To use opencv use the calibration sample (under samples/cpp) from the opencv sources
and extract the k1 and k2 coefficients from the resulting matrix.
</p>
<p>Note that effectively the same filter is available in the open-source tools Krita and
Digikam from the KDE project.
</p>
<p>In contrast to the <a href="#vignette">vignette</a> filter, which can also be used to compensate lens errors,
this filter corrects the distortion of the image, whereas <a href="#vignette">vignette</a> corrects the
brightness distribution, so you may want to use both filters together in certain
cases, though you will have to take care of ordering, i.e. whether vignetting should
be applied before or after lens correction.
</p>
<a name="Options-1"></a>
<h3 class="subsection"><a href="ffmpeg-filters.html#toc-Options-1">9.51.1 Options</a></h3>
<p>The filter accepts the following options:
</p>
<dl compact="compact">
<dt>‘<samp>cx</samp>’</dt>
<dd><p>Relative x-coordinate of the focal point of the image, and thereby the center of the
distrortion. This value has a range [0,1] and is expressed as fractions of the image
width.
</p></dd>
<dt>‘<samp>cy</samp>’</dt>
<dd><p>Relative y-coordinate of the focal point of the image, and thereby the center of the
distrortion. This value has a range [0,1] and is expressed as fractions of the image
height.
</p></dd>
<dt>‘<samp>k1</samp>’</dt>
<dd><p>Coefficient of the quadratic correction term. 0.5 means no correction.
</p></dd>
<dt>‘<samp>k2</samp>’</dt>
<dd><p>Coefficient of the double quadratic correction term. 0.5 means no correction.
</p></dd>
</dl>
<p>The formula that generates the correction is:
</p>
<p><var>r_src</var> = <var>r_tgt</var> * (1 + <var>k1</var> * (<var>r_tgt</var> / <var>r_0</var>)^2 + <var>k2</var> * (<var>r_tgt</var> / <var>r_0</var>)^4)
</p>
<p>where <var>r_0</var> is halve of the image diagonal and <var>r_src</var> and <var>r_tgt</var> are the
distances from the focal point in the source and target images, respectively.
</p>
<p><a name="lut3d"></a>
</p><a name="lut3d-1"></a>
<h2 class="section"><a href="ffmpeg-filters.html#toc-lut3d-1">9.52 lut3d</a></h2>
<p>Apply a 3D LUT to an input video.
</p>
<p>The filter accepts the following options:
</p>
<dl compact="compact">
<dt>‘<samp>file</samp>’</dt>
<dd><p>Set the 3D LUT file name.
</p>
<p>Currently supported formats:
</p><dl compact="compact">
<dt>‘<samp>3dl</samp>’</dt>
<dd><p>AfterEffects
</p></dd>
<dt>‘<samp>cube</samp>’</dt>
<dd><p>Iridas
</p></dd>
<dt>‘<samp>dat</samp>’</dt>
<dd><p>DaVinci
</p></dd>
<dt>‘<samp>m3d</samp>’</dt>
<dd><p>Pandora
</p></dd>
</dl>
</dd>
<dt>‘<samp>interp</samp>’</dt>
<dd><p>Select interpolation mode.
</p>
<p>Available values are:
</p>
<dl compact="compact">
<dt>‘<samp>nearest</samp>’</dt>
<dd><p>Use values from the nearest defined point.
</p></dd>
<dt>‘<samp>trilinear</samp>’</dt>
<dd><p>Interpolate values using the 8 points defining a cube.
</p></dd>
<dt>‘<samp>tetrahedral</samp>’</dt>
<dd><p>Interpolate values using a tetrahedron.
</p></dd>
</dl>
</dd>
</dl>
<a name="lut_002c-lutrgb_002c-lutyuv"></a>
<h2 class="section"><a href="ffmpeg-filters.html#toc-lut_002c-lutrgb_002c-lutyuv">9.53 lut, lutrgb, lutyuv</a></h2>
<p>Compute a look-up table for binding each pixel component input value
to an output value, and apply it to the input video.
</p>
<p><var>lutyuv</var> applies a lookup table to a YUV input video, <var>lutrgb</var>
to an RGB input video.
</p>
<p>These filters accept the following parameters:
</p><dl compact="compact">
<dt>‘<samp>c0</samp>’</dt>
<dd><p>set first pixel component expression
</p></dd>
<dt>‘<samp>c1</samp>’</dt>
<dd><p>set second pixel component expression
</p></dd>
<dt>‘<samp>c2</samp>’</dt>
<dd><p>set third pixel component expression
</p></dd>
<dt>‘<samp>c3</samp>’</dt>
<dd><p>set fourth pixel component expression, corresponds to the alpha component
</p>
</dd>
<dt>‘<samp>r</samp>’</dt>
<dd><p>set red component expression
</p></dd>
<dt>‘<samp>g</samp>’</dt>
<dd><p>set green component expression
</p></dd>
<dt>‘<samp>b</samp>’</dt>
<dd><p>set blue component expression
</p></dd>
<dt>‘<samp>a</samp>’</dt>
<dd><p>alpha component expression
</p>
</dd>
<dt>‘<samp>y</samp>’</dt>
<dd><p>set Y/luminance component expression
</p></dd>
<dt>‘<samp>u</samp>’</dt>
<dd><p>set U/Cb component expression
</p></dd>
<dt>‘<samp>v</samp>’</dt>
<dd><p>set V/Cr component expression
</p></dd>
</dl>
<p>Each of them specifies the expression to use for computing the lookup table for
the corresponding pixel component values.
</p>
<p>The exact component associated to each of the <var>c*</var> options depends on the
format in input.
</p>
<p>The <var>lut</var> filter requires either YUV or RGB pixel formats in input,
<var>lutrgb</var> requires RGB pixel formats in input, and <var>lutyuv</var> requires YUV.
</p>
<p>The expressions can contain the following constants and functions:
</p>
<dl compact="compact">
<dt>‘<samp>w</samp>’</dt>
<dt>‘<samp>h</samp>’</dt>
<dd><p>The input width and height.
</p>
</dd>
<dt>‘<samp>val</samp>’</dt>
<dd><p>The input value for the pixel component.
</p>
</dd>
<dt>‘<samp>clipval</samp>’</dt>
<dd><p>The input value, clipped to the <var>minval</var>-<var>maxval</var> range.
</p>
</dd>
<dt>‘<samp>maxval</samp>’</dt>
<dd><p>The maximum value for the pixel component.
</p>
</dd>
<dt>‘<samp>minval</samp>’</dt>
<dd><p>The minimum value for the pixel component.
</p>
</dd>
<dt>‘<samp>negval</samp>’</dt>
<dd><p>The negated value for the pixel component value, clipped to the
<var>minval</var>-<var>maxval</var> range; it corresponds to the expression
"maxval-clipval+minval".
</p>
</dd>
<dt>‘<samp>clip(val)</samp>’</dt>
<dd><p>The computed value in <var>val</var>, clipped to the
<var>minval</var>-<var>maxval</var> range.
</p>
</dd>
<dt>‘<samp>gammaval(gamma)</samp>’</dt>
<dd><p>The computed gamma correction value of the pixel component value,
clipped to the <var>minval</var>-<var>maxval</var> range. It corresponds to the
expression
"pow((clipval-minval)/(maxval-minval)\,<var>gamma</var>)*(maxval-minval)+minval"
</p>
</dd>
</dl>
<p>All expressions default to "val".
</p>
<a name="Examples-53"></a>
<h3 class="subsection"><a href="ffmpeg-filters.html#toc-Examples-53">9.53.1 Examples</a></h3>
<ul>
<li>
Negate input video:
<div class="example">
<pre class="example">lutrgb="r=maxval+minval-val:g=maxval+minval-val:b=maxval+minval-val"
lutyuv="y=maxval+minval-val:u=maxval+minval-val:v=maxval+minval-val"
</pre></div>
<p>The above is the same as:
</p><div class="example">
<pre class="example">lutrgb="r=negval:g=negval:b=negval"
lutyuv="y=negval:u=negval:v=negval"
</pre></div>
</li><li>
Negate luminance:
<div class="example">
<pre class="example">lutyuv=y=negval
</pre></div>
</li><li>
Remove chroma components, turning the video into a graytone image:
<div class="example">
<pre class="example">lutyuv="u=128:v=128"
</pre></div>
</li><li>
Apply a luma burning effect:
<div class="example">
<pre class="example">lutyuv="y=2*val"
</pre></div>
</li><li>
Remove green and blue components:
<div class="example">
<pre class="example">lutrgb="g=0:b=0"
</pre></div>
</li><li>
Set a constant alpha channel value on input:
<div class="example">
<pre class="example">format=rgba,lutrgb=a="maxval-minval/2"
</pre></div>
</li><li>
Correct luminance gamma by a factor of 0.5:
<div class="example">
<pre class="example">lutyuv=y=gammaval(0.5)
</pre></div>
</li><li>
Discard least significant bits of luma:
<div class="example">
<pre class="example">lutyuv=y='bitand(val, 128+64+32)'
</pre></div>
</li></ul>
<a name="mergeplanes"></a>
<h2 class="section"><a href="ffmpeg-filters.html#toc-mergeplanes">9.54 mergeplanes</a></h2>
<p>Merge color channel components from several video streams.
</p>
<p>The filter accepts up to 4 input streams, and merge selected input
planes to the output video.
</p>
<p>This filter accepts the following options:
</p><dl compact="compact">
<dt>‘<samp>mapping</samp>’</dt>
<dd><p>Set input to output plane mapping. Default is <code>0</code>.
</p>
<p>The mappings is specified as a bitmap. It should be specified as a
hexadecimal number in the form 0xAa[Bb[Cc[Dd]]]. ’Aa’ describes the
mapping for the first plane of the output stream. ’A’ sets the number of
the input stream to use (from 0 to 3), and ’a’ the plane number of the
corresponding input to use (from 0 to 3). The rest of the mappings is
similar, ’Bb’ describes the mapping for the output stream second
plane, ’Cc’ describes the mapping for the output stream third plane and
’Dd’ describes the mapping for the output stream fourth plane.
</p>
</dd>
<dt>‘<samp>format</samp>’</dt>
<dd><p>Set output pixel format. Default is <code>yuva444p</code>.
</p></dd>
</dl>
<a name="Examples-50"></a>
<h3 class="subsection"><a href="ffmpeg-filters.html#toc-Examples-50">9.54.1 Examples</a></h3>
<ul>
<li>
Merge three gray video streams of same width and height into single video stream:
<div class="example">
<pre class="example">[a0][a1][a2]mergeplanes=0x001020:yuv444p
</pre></div>
</li><li>
Merge 1st yuv444p stream and 2nd gray video stream into yuva444p video stream:
<div class="example">
<pre class="example">[a0][a1]mergeplanes=0x00010210:yuva444p
</pre></div>
</li><li>
Swap Y and A plane in yuva444p stream:
<div class="example">
<pre class="example">format=yuva444p,mergeplanes=0x03010200:yuva444p
</pre></div>
</li><li>
Swap U and V plane in yuv420p stream:
<div class="example">
<pre class="example">format=yuv420p,mergeplanes=0x000201:yuv420p
</pre></div>
</li><li>
Cast a rgb24 clip to yuv444p:
<div class="example">
<pre class="example">format=rgb24,mergeplanes=0x000102:yuv444p
</pre></div>
</li></ul>
<a name="mcdeint"></a>
<h2 class="section"><a href="ffmpeg-filters.html#toc-mcdeint">9.55 mcdeint</a></h2>
<p>Apply motion-compensation deinterlacing.
</p>
<p>It needs one field per frame as input and must thus be used together
with yadif=1/3 or equivalent.
</p>
<p>This filter accepts the following options:
</p><dl compact="compact">
<dt>‘<samp>mode</samp>’</dt>
<dd><p>Set the deinterlacing mode.
</p>
<p>It accepts one of the following values:
</p><dl compact="compact">
<dt>‘<samp>fast</samp>’</dt>
<dt>‘<samp>medium</samp>’</dt>
<dt>‘<samp>slow</samp>’</dt>
<dd><p>use iterative motion estimation
</p></dd>
<dt>‘<samp>extra_slow</samp>’</dt>
<dd><p>like ‘<samp>slow</samp>’, but use multiple reference frames.
</p></dd>
</dl>
<p>Default value is ‘<samp>fast</samp>’.
</p>
</dd>
<dt>‘<samp>parity</samp>’</dt>
<dd><p>Set the picture field parity assumed for the input video. It must be
one of the following values:
</p>
<dl compact="compact">
<dt>‘<samp>0, tff</samp>’</dt>
<dd><p>assume top field first
</p></dd>
<dt>‘<samp>1, bff</samp>’</dt>
<dd><p>assume bottom field first
</p></dd>
</dl>
<p>Default value is ‘<samp>bff</samp>’.
</p>
</dd>
<dt>‘<samp>qp</samp>’</dt>
<dd><p>Set per-block quantization parameter (QP) used by the internal
encoder.
</p>
<p>Higher values should result in a smoother motion vector field but less
optimal individual vectors. Default value is 1.
</p></dd>
</dl>
<a name="mp"></a>
<h2 class="section"><a href="ffmpeg-filters.html#toc-mp">9.56 mp</a></h2>
<p>Apply an MPlayer filter to the input video.
</p>
<p>This filter provides a wrapper around some of the filters of
MPlayer/MEncoder.
</p>
<p>This wrapper is considered experimental. Some of the wrapped filters
may not work properly and we may drop support for them, as they will
be implemented natively into FFmpeg. Thus you should avoid
depending on them when writing portable scripts.
</p>
<p>The filter accepts the parameters:
<var>filter_name</var>[:=]<var>filter_params</var>
</p>
<p><var>filter_name</var> is the name of a supported MPlayer filter,
<var>filter_params</var> is a string containing the parameters accepted by
the named filter.
</p>
<p>The list of the currently supported filters follows:
</p><dl compact="compact">
<dt><var>eq2</var></dt>
<dt><var>eq</var></dt>
<dt><var>fspp</var></dt>
<dt><var>ilpack</var></dt>
<dt><var>pp7</var></dt>
<dt><var>softpulldown</var></dt>
<dt><var>uspp</var></dt>
</dl>
<p>The parameter syntax and behavior for the listed filters are the same
of the corresponding MPlayer filters. For detailed instructions check
the "VIDEO FILTERS" section in the MPlayer manual.
</p>
<a name="Examples-17"></a>
<h3 class="subsection"><a href="ffmpeg-filters.html#toc-Examples-17">9.56.1 Examples</a></h3>
<ul>
<li>
Adjust gamma, brightness, contrast:
<div class="example">
<pre class="example">mp=eq2=1.0:2:0.5
</pre></div>
</li></ul>
<p>See also mplayer(1), <a href="http://www.mplayerhq.hu/">http://www.mplayerhq.hu/</a>.
</p>
<a name="mpdecimate"></a>
<h2 class="section"><a href="ffmpeg-filters.html#toc-mpdecimate">9.57 mpdecimate</a></h2>
<p>Drop frames that do not differ greatly from the previous frame in
order to reduce frame rate.
</p>
<p>The main use of this filter is for very-low-bitrate encoding
(e.g. streaming over dialup modem), but it could in theory be used for
fixing movies that were inverse-telecined incorrectly.
</p>
<p>A description of the accepted options follows.
</p>
<dl compact="compact">
<dt>‘<samp>max</samp>’</dt>
<dd><p>Set the maximum number of consecutive frames which can be dropped (if
positive), or the minimum interval between dropped frames (if
negative). If the value is 0, the frame is dropped unregarding the
number of previous sequentially dropped frames.
</p>
<p>Default value is 0.
</p>
</dd>
<dt>‘<samp>hi</samp>’</dt>
<dt>‘<samp>lo</samp>’</dt>
<dt>‘<samp>frac</samp>’</dt>
<dd><p>Set the dropping threshold values.
</p>
<p>Values for ‘<samp>hi</samp>’ and ‘<samp>lo</samp>’ are for 8x8 pixel blocks and
represent actual pixel value differences, so a threshold of 64
corresponds to 1 unit of difference for each pixel, or the same spread
out differently over the block.
</p>
<p>A frame is a candidate for dropping if no 8x8 blocks differ by more
than a threshold of ‘<samp>hi</samp>’, and if no more than ‘<samp>frac</samp>’ blocks (1
meaning the whole image) differ by more than a threshold of ‘<samp>lo</samp>’.
</p>
<p>Default value for ‘<samp>hi</samp>’ is 64*12, default value for ‘<samp>lo</samp>’ is
64*5, and default value for ‘<samp>frac</samp>’ is 0.33.
</p></dd>
</dl>
<a name="negate"></a>
<h2 class="section"><a href="ffmpeg-filters.html#toc-negate">9.58 negate</a></h2>
<p>Negate input video.
</p>
<p>It accepts an integer in input; if non-zero it negates the
alpha component (if available). The default value in input is 0.
</p>
<a name="noformat"></a>
<h2 class="section"><a href="ffmpeg-filters.html#toc-noformat">9.59 noformat</a></h2>
<p>Force libavfilter not to use any of the specified pixel formats for the
input to the next filter.
</p>
<p>It accepts the following parameters:
</p><dl compact="compact">
<dt>‘<samp>pix_fmts</samp>’</dt>
<dd><p>A ’|’-separated list of pixel format names, such as
apix_fmts=yuv420p|monow|rgb24".
</p>
</dd>
</dl>
<a name="Examples-30"></a>
<h3 class="subsection"><a href="ffmpeg-filters.html#toc-Examples-30">9.59.1 Examples</a></h3>
<ul>
<li>
Force libavfilter to use a format different from <var>yuv420p</var> for the
input to the vflip filter:
<div class="example">
<pre class="example">noformat=pix_fmts=yuv420p,vflip
</pre></div>
</li><li>
Convert the input video to any of the formats not contained in the list:
<div class="example">
<pre class="example">noformat=yuv420p|yuv444p|yuv410p
</pre></div>
</li></ul>
<a name="noise"></a>
<h2 class="section"><a href="ffmpeg-filters.html#toc-noise">9.60 noise</a></h2>
<p>Add noise on video input frame.
</p>
<p>The filter accepts the following options:
</p>
<dl compact="compact">
<dt>‘<samp>all_seed</samp>’</dt>
<dt>‘<samp>c0_seed</samp>’</dt>
<dt>‘<samp>c1_seed</samp>’</dt>
<dt>‘<samp>c2_seed</samp>’</dt>
<dt>‘<samp>c3_seed</samp>’</dt>
<dd><p>Set noise seed for specific pixel component or all pixel components in case
of <var>all_seed</var>. Default value is <code>123457</code>.
</p>
</dd>
<dt>‘<samp>all_strength, alls</samp>’</dt>
<dt>‘<samp>c0_strength, c0s</samp>’</dt>
<dt>‘<samp>c1_strength, c1s</samp>’</dt>
<dt>‘<samp>c2_strength, c2s</samp>’</dt>
<dt>‘<samp>c3_strength, c3s</samp>’</dt>
<dd><p>Set noise strength for specific pixel component or all pixel components in case
<var>all_strength</var>. Default value is <code>0</code>. Allowed range is [0, 100].
</p>
</dd>
<dt>‘<samp>all_flags, allf</samp>’</dt>
<dt>‘<samp>c0_flags, c0f</samp>’</dt>
<dt>‘<samp>c1_flags, c1f</samp>’</dt>
<dt>‘<samp>c2_flags, c2f</samp>’</dt>
<dt>‘<samp>c3_flags, c3f</samp>’</dt>
<dd><p>Set pixel component flags or set flags for all components if <var>all_flags</var>.
Available values for component flags are:
</p><dl compact="compact">
<dt>‘<samp>a</samp>’</dt>
<dd><p>averaged temporal noise (smoother)
</p></dd>
<dt>‘<samp>p</samp>’</dt>
<dd><p>mix random noise with a (semi)regular pattern
</p></dd>
<dt>‘<samp>t</samp>’</dt>
<dd><p>temporal noise (noise pattern changes between frames)
</p></dd>
<dt>‘<samp>u</samp>’</dt>
<dd><p>uniform noise (gaussian otherwise)
</p></dd>
</dl>
</dd>
</dl>
<a name="Examples-33"></a>
<h3 class="subsection"><a href="ffmpeg-filters.html#toc-Examples-33">9.60.1 Examples</a></h3>
<p>Add temporal and uniform noise to input video:
</p><div class="example">
<pre class="example">noise=alls=20:allf=t+u
</pre></div>
<a name="null"></a>
<h2 class="section"><a href="ffmpeg-filters.html#toc-null">9.61 null</a></h2>
<p>Pass the video source unchanged to the output.
</p>
<a name="ocv"></a>
<h2 class="section"><a href="ffmpeg-filters.html#toc-ocv">9.62 ocv</a></h2>
<p>Apply a video transform using libopencv.
</p>
<p>To enable this filter, install the libopencv library and headers and
configure FFmpeg with <code>--enable-libopencv</code>.
</p>
<p>It accepts the following parameters:
</p>
<dl compact="compact">
<dt>‘<samp>filter_name</samp>’</dt>
<dd><p>The name of the libopencv filter to apply.
</p>
</dd>
<dt>‘<samp>filter_params</samp>’</dt>
<dd><p>The parameters to pass to the libopencv filter. If not specified, the default
values are assumed.
</p>
</dd>
</dl>
<p>Refer to the official libopencv documentation for more precise
information:
<a href="http://opencv.willowgarage.com/documentation/c/image_filtering.html">http://opencv.willowgarage.com/documentation/c/image_filtering.html</a>
</p>
<p>Several libopencv filters are supported; see the following subsections.
</p>
<p><a name="dilate"></a>
</p><a name="dilate-1"></a>
<h3 class="subsection"><a href="ffmpeg-filters.html#toc-dilate-1">9.62.1 dilate</a></h3>
<p>Dilate an image by using a specific structuring element.
It corresponds to the libopencv function <code>cvDilate</code>.
</p>
<p>It accepts the parameters: <var>struct_el</var>|<var>nb_iterations</var>.
</p>
<p><var>struct_el</var> represents a structuring element, and has the syntax:
<var>cols</var>x<var>rows</var>+<var>anchor_x</var>x<var>anchor_y</var>/<var>shape</var>
</p>
<p><var>cols</var> and <var>rows</var> represent the number of columns and rows of
the structuring element, <var>anchor_x</var> and <var>anchor_y</var> the anchor
point, and <var>shape</var> the shape for the structuring element. <var>shape</var>
must be "rect", "cross", "ellipse", or "custom".
</p>
<p>If the value for <var>shape</var> is "custom", it must be followed by a
string of the form "=<var>filename</var>". The file with name
<var>filename</var> is assumed to represent a binary image, with each
printable character corresponding to a bright pixel. When a custom
<var>shape</var> is used, <var>cols</var> and <var>rows</var> are ignored, the number
or columns and rows of the read file are assumed instead.
</p>
<p>The default value for <var>struct_el</var> is "3x3+0x0/rect".
</p>
<p><var>nb_iterations</var> specifies the number of times the transform is
applied to the image, and defaults to 1.
</p>
<p>Some examples:
</p><div class="example">
<pre class="example"># Use the default values
ocv=dilate
# Dilate using a structuring element with a 5x5 cross, iterating two times
ocv=filter_name=dilate:filter_params=5x5+2x2/cross|2
# Read the shape from the file diamond.shape, iterating two times.
# The file diamond.shape may contain a pattern of characters like this
# *
# ***
# *****
# ***
# *
# The specified columns and rows are ignored
# but the anchor point coordinates are not
ocv=dilate:0x0+2x2/custom=diamond.shape|2
</pre></div>
<a name="erode"></a>
<h3 class="subsection"><a href="ffmpeg-filters.html#toc-erode">9.62.2 erode</a></h3>
<p>Erode an image by using a specific structuring element.
It corresponds to the libopencv function <code>cvErode</code>.
</p>
<p>It accepts the parameters: <var>struct_el</var>:<var>nb_iterations</var>,
with the same syntax and semantics as the <a href="#dilate">dilate</a> filter.
</p>
<a name="smooth"></a>
<h3 class="subsection"><a href="ffmpeg-filters.html#toc-smooth">9.62.3 smooth</a></h3>
<p>Smooth the input video.
</p>
<p>The filter takes the following parameters:
<var>type</var>|<var>param1</var>|<var>param2</var>|<var>param3</var>|<var>param4</var>.
</p>
<p><var>type</var> is the type of smooth filter to apply, and must be one of
the following values: "blur", "blur_no_scale", "median", "gaussian",
or "bilateral". The default value is "gaussian".
</p>
<p>The meaning of <var>param1</var>, <var>param2</var>, <var>param3</var>, and <var>param4</var>
depend on the smooth type. <var>param1</var> and
<var>param2</var> accept integer positive values or 0. <var>param3</var> and
<var>param4</var> accept floating point values.
</p>
<p>The default value for <var>param1</var> is 3. The default value for the
other parameters is 0.
</p>
<p>These parameters correspond to the parameters assigned to the
libopencv function <code>cvSmooth</code>.
</p>
<p><a name="overlay"></a>
</p><a name="overlay-1"></a>
<h2 class="section"><a href="ffmpeg-filters.html#toc-overlay-1">9.63 overlay</a></h2>
<p>Overlay one video on top of another.
</p>
<p>It takes two inputs and has one output. The first input is the "main"
video on which the second input is overlayed.
</p>
<p>It accepts the following parameters:
</p>
<p>A description of the accepted options follows.
</p>
<dl compact="compact">
<dt>‘<samp>x</samp>’</dt>
<dt>‘<samp>y</samp>’</dt>
<dd><p>Set the expression for the x and y coordinates of the overlayed video
on the main video. Default value is "0" for both expressions. In case
the expression is invalid, it is set to a huge value (meaning that the
overlay will not be displayed within the output visible area).
</p>
</dd>
<dt>‘<samp>eof_action</samp>’</dt>
<dd><p>The action to take when EOF is encountered on the secondary input; it accepts
one of the following values:
</p>
<dl compact="compact">
<dt>‘<samp>repeat</samp>’</dt>
<dd><p>Repeat the last frame (the default).
</p></dd>
<dt>‘<samp>endall</samp>’</dt>
<dd><p>End both streams.
</p></dd>
<dt>‘<samp>pass</samp>’</dt>
<dd><p>Pass the main input through.
</p></dd>
</dl>
</dd>
<dt>‘<samp>eval</samp>’</dt>
<dd><p>Set when the expressions for ‘<samp>x</samp>’, and ‘<samp>y</samp>’ are evaluated.
</p>
<p>It accepts the following values:
</p><dl compact="compact">
<dt>‘<samp>init</samp>’</dt>
<dd><p>only evaluate expressions once during the filter initialization or
when a command is processed
</p>
</dd>
<dt>‘<samp>frame</samp>’</dt>
<dd><p>evaluate expressions for each incoming frame
</p></dd>
</dl>
<p>Default value is ‘<samp>frame</samp>’.
</p>
</dd>
<dt>‘<samp>shortest</samp>’</dt>
<dd><p>If set to 1, force the output to terminate when the shortest input
terminates. Default value is 0.
</p>
</dd>
<dt>‘<samp>format</samp>’</dt>
<dd><p>Set the format for the output video.
</p>
<p>It accepts the following values:
</p><dl compact="compact">
<dt>‘<samp>yuv420</samp>’</dt>
<dd><p>force YUV420 output
</p>
</dd>
<dt>‘<samp>yuv422</samp>’</dt>
<dd><p>force YUV422 output
</p>
</dd>
<dt>‘<samp>yuv444</samp>’</dt>
<dd><p>force YUV444 output
</p>
</dd>
<dt>‘<samp>rgb</samp>’</dt>
<dd><p>force RGB output
</p></dd>
</dl>
<p>Default value is ‘<samp>yuv420</samp>’.
</p>
</dd>
<dt>‘<samp>rgb <em>(deprecated)</em></samp>’</dt>
<dd><p>If set to 1, force the filter to accept inputs in the RGB
color space. Default value is 0. This option is deprecated, use
‘<samp>format</samp>’ instead.
</p>
</dd>
<dt>‘<samp>repeatlast</samp>’</dt>
<dd><p>If set to 1, force the filter to draw the last overlay frame over the
main input until the end of the stream. A value of 0 disables this
behavior. Default value is 1.
</p></dd>
</dl>
<p>The ‘<samp>x</samp>’, and ‘<samp>y</samp>’ expressions can contain the following
parameters.
</p>
<dl compact="compact">
<dt>‘<samp>main_w, W</samp>’</dt>
<dt>‘<samp>main_h, H</samp>’</dt>
<dd><p>The main input width and height.
</p>
</dd>
<dt>‘<samp>overlay_w, w</samp>’</dt>
<dt>‘<samp>overlay_h, h</samp>’</dt>
<dd><p>The overlay input width and height.
</p>
</dd>
<dt>‘<samp>x</samp>’</dt>
<dt>‘<samp>y</samp>’</dt>
<dd><p>The computed values for <var>x</var> and <var>y</var>. They are evaluated for
each new frame.
</p>
</dd>
<dt>‘<samp>hsub</samp>’</dt>
<dt>‘<samp>vsub</samp>’</dt>
<dd><p>horizontal and vertical chroma subsample values of the output
format. For example for the pixel format "yuv422p" <var>hsub</var> is 2 and
<var>vsub</var> is 1.
</p>
</dd>
<dt>‘<samp>n</samp>’</dt>
<dd><p>the number of input frame, starting from 0
</p>
</dd>
<dt>‘<samp>pos</samp>’</dt>
<dd><p>the position in the file of the input frame, NAN if unknown
</p>
</dd>
<dt>‘<samp>t</samp>’</dt>
<dd><p>The timestamp, expressed in seconds. It’s NAN if the input timestamp is unknown.
</p>
</dd>
</dl>
<p>Note that the <var>n</var>, <var>pos</var>, <var>t</var> variables are available only
when evaluation is done <em>per frame</em>, and will evaluate to NAN
when ‘<samp>eval</samp>’ is set to ‘<samp>init</samp>’.
</p>
<p>Be aware that frames are taken from each input video in timestamp
order, hence, if their initial timestamps differ, it is a good idea
to pass the two inputs through a <var>setpts=PTS-STARTPTS</var> filter to
have them begin in the same zero timestamp, as the example for
the <var>movie</var> filter does.
</p>
<p>You can chain together more overlays but you should test the
efficiency of such approach.
</p>
<a name="Commands-5"></a>
<h3 class="subsection"><a href="ffmpeg-filters.html#toc-Commands-5">9.63.1 Commands</a></h3>
<p>This filter supports the following commands:
</p><dl compact="compact">
<dt>‘<samp>x</samp>’</dt>
<dt>‘<samp>y</samp>’</dt>
<dd><p>Modify the x and y of the overlay input.
The command accepts the same syntax of the corresponding option.
</p>
<p>If the specified expression is not valid, it is kept at its current
value.
</p></dd>
</dl>
<a name="Examples-76"></a>
<h3 class="subsection"><a href="ffmpeg-filters.html#toc-Examples-76">9.63.2 Examples</a></h3>
<ul>
<li>
Draw the overlay at 10 pixels from the bottom right corner of the main
video:
<div class="example">
<pre class="example">overlay=main_w-overlay_w-10:main_h-overlay_h-10
</pre></div>
<p>Using named options the example above becomes:
</p><div class="example">
<pre class="example">overlay=x=main_w-overlay_w-10:y=main_h-overlay_h-10
</pre></div>
</li><li>
Insert a transparent PNG logo in the bottom left corner of the input,
using the <code>ffmpeg</code> tool with the <code>-filter_complex</code> option:
<div class="example">
<pre class="example">ffmpeg -i input -i logo -filter_complex 'overlay=10:main_h-overlay_h-10' output
</pre></div>
</li><li>
Insert 2 different transparent PNG logos (second logo on bottom
right corner) using the <code>ffmpeg</code> tool:
<div class="example">
<pre class="example">ffmpeg -i input -i logo1 -i logo2 -filter_complex 'overlay=x=10:y=H-h-10,overlay=x=W-w-10:y=H-h-10' output
</pre></div>
</li><li>
Add a transparent color layer on top of the main video; <code>WxH</code>
must specify the size of the main input to the overlay filter:
<div class="example">
<pre class="example">color=color=red@.3:size=WxH [over]; [in][over] overlay [out]
</pre></div>
</li><li>
Play an original video and a filtered version (here with the deshake
filter) side by side using the <code>ffplay</code> tool:
<div class="example">
<pre class="example">ffplay input.avi -vf 'split[a][b]; [a]pad=iw*2:ih[src]; [b]deshake[filt]; [src][filt]overlay=w'
</pre></div>
<p>The above command is the same as:
</p><div class="example">
<pre class="example">ffplay input.avi -vf 'split[b], pad=iw*2[src], [b]deshake, [src]overlay=w'
</pre></div>
</li><li>
Make a sliding overlay appearing from the left to the right top part of the
screen starting since time 2:
<div class="example">
<pre class="example">overlay=x='if(gte(t,2), -w+(t-2)*20, NAN)':y=0
</pre></div>
</li><li>
Compose output by putting two input videos side to side:
<div class="example">
<pre class="example">ffmpeg -i left.avi -i right.avi -filter_complex "
nullsrc=size=200x100 [background];
[0:v] setpts=PTS-STARTPTS, scale=100x100 [left];
[1:v] setpts=PTS-STARTPTS, scale=100x100 [right];
[background][left] overlay=shortest=1 [background+left];
[background+left][right] overlay=shortest=1:x=100 [left+right]
"
</pre></div>
</li><li>
Mask 10-20 seconds of a video by applying the delogo filter to a section
<div class="example">
<pre class="example">ffmpeg -i test.avi -codec:v:0 wmv2 -ar 11025 -b:v 9000k
-vf '[in]split[split_main][split_delogo];[split_delogo]trim=start=360:end=371,delogo=0:0:640:480[delogoed];[split_main][delogoed]overlay=eof_action=pass[out]'
masked.avi
</pre></div>
</li><li>
Chain several overlays in cascade:
<div class="example">
<pre class="example">nullsrc=s=200x200 [bg];
testsrc=s=100x100, split=4 [in0][in1][in2][in3];
[in0] lutrgb=r=0, [bg] overlay=0:0 [mid0];
[in1] lutrgb=g=0, [mid0] overlay=100:0 [mid1];
[in2] lutrgb=b=0, [mid1] overlay=0:100 [mid2];
[in3] null, [mid2] overlay=100:100 [out0]
</pre></div>
</li></ul>
<a name="owdenoise"></a>
<h2 class="section"><a href="ffmpeg-filters.html#toc-owdenoise">9.64 owdenoise</a></h2>
<p>Apply Overcomplete Wavelet denoiser.
</p>
<p>The filter accepts the following options:
</p>
<dl compact="compact">
<dt>‘<samp>depth</samp>’</dt>
<dd><p>Set depth.
</p>
<p>Larger depth values will denoise lower frequency components more, but
slow down filtering.
</p>
<p>Must be an int in the range 8-16, default is <code>8</code>.
</p>
</dd>
<dt>‘<samp>luma_strength, ls</samp>’</dt>
<dd><p>Set luma strength.
</p>
<p>Must be a double value in the range 0-1000, default is <code>1.0</code>.
</p>
</dd>
<dt>‘<samp>chroma_strength, cs</samp>’</dt>
<dd><p>Set chroma strength.
</p>
<p>Must be a double value in the range 0-1000, default is <code>1.0</code>.
</p></dd>
</dl>
<a name="pad"></a>
<h2 class="section"><a href="ffmpeg-filters.html#toc-pad">9.65 pad</a></h2>
<p>Add paddings to the input image, and place the original input at the
provided <var>x</var>, <var>y</var> coordinates.
</p>
<p>It accepts the following parameters:
</p>
<dl compact="compact">
<dt>‘<samp>width, w</samp>’</dt>
<dt>‘<samp>height, h</samp>’</dt>
<dd><p>Specify an expression for the size of the output image with the
paddings added. If the value for <var>width</var> or <var>height</var> is 0, the
corresponding input size is used for the output.
</p>
<p>The <var>width</var> expression can reference the value set by the
<var>height</var> expression, and vice versa.
</p>
<p>The default value of <var>width</var> and <var>height</var> is 0.
</p>
</dd>
<dt>‘<samp>x</samp>’</dt>
<dt>‘<samp>y</samp>’</dt>
<dd><p>Specify the offsets to place the input image at within the padded area,
with respect to the top/left border of the output image.
</p>
<p>The <var>x</var> expression can reference the value set by the <var>y</var>
expression, and vice versa.
</p>
<p>The default value of <var>x</var> and <var>y</var> is 0.
</p>
</dd>
<dt>‘<samp>color</samp>’</dt>
<dd><p>Specify the color of the padded area. For the syntax of this option,
check the "Color" section in the ffmpeg-utils manual.
</p>
<p>The default value of <var>color</var> is "black".
</p></dd>
</dl>
<p>The value for the <var>width</var>, <var>height</var>, <var>x</var>, and <var>y</var>
options are expressions containing the following constants:
</p>
<dl compact="compact">
<dt>‘<samp>in_w</samp>’</dt>
<dt>‘<samp>in_h</samp>’</dt>
<dd><p>The input video width and height.
</p>
</dd>
<dt>‘<samp>iw</samp>’</dt>
<dt>‘<samp>ih</samp>’</dt>
<dd><p>These are the same as <var>in_w</var> and <var>in_h</var>.
</p>
</dd>
<dt>‘<samp>out_w</samp>’</dt>
<dt>‘<samp>out_h</samp>’</dt>
<dd><p>The output width and height (the size of the padded area), as
specified by the <var>width</var> and <var>height</var> expressions.
</p>
</dd>
<dt>‘<samp>ow</samp>’</dt>
<dt>‘<samp>oh</samp>’</dt>
<dd><p>These are the same as <var>out_w</var> and <var>out_h</var>.
</p>
</dd>
<dt>‘<samp>x</samp>’</dt>
<dt>‘<samp>y</samp>’</dt>
<dd><p>The x and y offsets as specified by the <var>x</var> and <var>y</var>
expressions, or NAN if not yet specified.
</p>
</dd>
<dt>‘<samp>a</samp>’</dt>
<dd><p>same as <var>iw</var> / <var>ih</var>
</p>
</dd>
<dt>‘<samp>sar</samp>’</dt>
<dd><p>input sample aspect ratio
</p>
</dd>
<dt>‘<samp>dar</samp>’</dt>
<dd><p>input display aspect ratio, it is the same as (<var>iw</var> / <var>ih</var>) * <var>sar</var>
</p>
</dd>
<dt>‘<samp>hsub</samp>’</dt>
<dt>‘<samp>vsub</samp>’</dt>
<dd><p>The horizontal and vertical chroma subsample values. For example for the
pixel format "yuv422p" <var>hsub</var> is 2 and <var>vsub</var> is 1.
</p></dd>
</dl>
<a name="Examples-12"></a>
<h3 class="subsection"><a href="ffmpeg-filters.html#toc-Examples-12">9.65.1 Examples</a></h3>
<ul>
<li>
Add paddings with the color "violet" to the input video. The output video
size is 640x480, and the top-left corner of the input video is placed at
column 0, row 40
<div class="example">
<pre class="example">pad=640:480:0:40:violet
</pre></div>
<p>The example above is equivalent to the following command:
</p><div class="example">
<pre class="example">pad=width=640:height=480:x=0:y=40:color=violet
</pre></div>
</li><li>
Pad the input to get an output with dimensions increased by 3/2,
and put the input video at the center of the padded area:
<div class="example">
<pre class="example">pad="3/2*iw:3/2*ih:(ow-iw)/2:(oh-ih)/2"
</pre></div>
</li><li>
Pad the input to get a squared output with size equal to the maximum
value between the input width and height, and put the input video at
the center of the padded area:
<div class="example">
<pre class="example">pad="max(iw\,ih):ow:(ow-iw)/2:(oh-ih)/2"
</pre></div>
</li><li>
Pad the input to get a final w/h ratio of 16:9:
<div class="example">
<pre class="example">pad="ih*16/9:ih:(ow-iw)/2:(oh-ih)/2"
</pre></div>
</li><li>
In case of anamorphic video, in order to set the output display aspect
correctly, it is necessary to use <var>sar</var> in the expression,
according to the relation:
<div class="example">
<pre class="example">(ih * X / ih) * sar = output_dar
X = output_dar / sar
</pre></div>
<p>Thus the previous example needs to be modified to:
</p><div class="example">
<pre class="example">pad="ih*16/9/sar:ih:(ow-iw)/2:(oh-ih)/2"
</pre></div>
</li><li>
Double the output size and put the input video in the bottom-right
corner of the output padded area:
<div class="example">
<pre class="example">pad="2*iw:2*ih:ow-iw:oh-ih"
</pre></div>
</li></ul>
<a name="perspective"></a>
<h2 class="section"><a href="ffmpeg-filters.html#toc-perspective">9.66 perspective</a></h2>
<p>Correct perspective of video not recorded perpendicular to the screen.
</p>
<p>A description of the accepted parameters follows.
</p>
<dl compact="compact">
<dt>‘<samp>x0</samp>’</dt>
<dt>‘<samp>y0</samp>’</dt>
<dt>‘<samp>x1</samp>’</dt>
<dt>‘<samp>y1</samp>’</dt>
<dt>‘<samp>x2</samp>’</dt>
<dt>‘<samp>y2</samp>’</dt>
<dt>‘<samp>x3</samp>’</dt>
<dt>‘<samp>y3</samp>’</dt>
<dd><p>Set coordinates expression for top left, top right, bottom left and bottom right corners.
Default values are <code>0:0:W:0:0:H:W:H</code> with which perspective will remain unchanged.
</p>
<p>The expressions can use the following variables:
</p>
<dl compact="compact">
<dt>‘<samp>W</samp>’</dt>
<dt>‘<samp>H</samp>’</dt>
<dd><p>the width and height of video frame.
</p></dd>
</dl>
</dd>
<dt>‘<samp>interpolation</samp>’</dt>
<dd><p>Set interpolation for perspective correction.
</p>
<p>It accepts the following values:
</p><dl compact="compact">
<dt>‘<samp>linear</samp>’</dt>
<dt>‘<samp>cubic</samp>’</dt>
</dl>
<p>Default value is ‘<samp>linear</samp>’.
</p></dd>
</dl>
<a name="phase"></a>
<h2 class="section"><a href="ffmpeg-filters.html#toc-phase">9.67 phase</a></h2>
<p>Delay interlaced video by one field time so that the field order changes.
</p>
<p>The intended use is to fix PAL movies that have been captured with the
opposite field order to the film-to-video transfer.
</p>
<p>A description of the accepted parameters follows.
</p>
<dl compact="compact">
<dt>‘<samp>mode</samp>’</dt>
<dd><p>Set phase mode.
</p>
<p>It accepts the following values:
</p><dl compact="compact">
<dt>‘<samp>t</samp>’</dt>
<dd><p>Capture field order top-first, transfer bottom-first.
Filter will delay the bottom field.
</p>
</dd>
<dt>‘<samp>b</samp>’</dt>
<dd><p>Capture field order bottom-first, transfer top-first.
Filter will delay the top field.
</p>
</dd>
<dt>‘<samp>p</samp>’</dt>
<dd><p>Capture and transfer with the same field order. This mode only exists
for the documentation of the other options to refer to, but if you
actually select it, the filter will faithfully do nothing.
</p>
</dd>
<dt>‘<samp>a</samp>’</dt>
<dd><p>Capture field order determined automatically by field flags, transfer
opposite.
Filter selects among ‘<samp>t</samp>’ and ‘<samp>b</samp>’ modes on a frame by frame
basis using field flags. If no field information is available,
then this works just like ‘<samp>u</samp>’.
</p>
</dd>
<dt>‘<samp>u</samp>’</dt>
<dd><p>Capture unknown or varying, transfer opposite.
Filter selects among ‘<samp>t</samp>’ and ‘<samp>b</samp>’ on a frame by frame basis by
analyzing the images and selecting the alternative that produces best
match between the fields.
</p>
</dd>
<dt>‘<samp>T</samp>’</dt>
<dd><p>Capture top-first, transfer unknown or varying.
Filter selects among ‘<samp>t</samp>’ and ‘<samp>p</samp>’ using image analysis.
</p>
</dd>
<dt>‘<samp>B</samp>’</dt>
<dd><p>Capture bottom-first, transfer unknown or varying.
Filter selects among ‘<samp>b</samp>’ and ‘<samp>p</samp>’ using image analysis.
</p>
</dd>
<dt>‘<samp>A</samp>’</dt>
<dd><p>Capture determined by field flags, transfer unknown or varying.
Filter selects among ‘<samp>t</samp>’, ‘<samp>b</samp>’ and ‘<samp>p</samp>’ using field flags and
image analysis. If no field information is available, then this works just
like ‘<samp>U</samp>’. This is the default mode.
</p>
</dd>
<dt>‘<samp>U</samp>’</dt>
<dd><p>Both capture and transfer unknown or varying.
Filter selects among ‘<samp>t</samp>’, ‘<samp>b</samp>’ and ‘<samp>p</samp>’ using image analysis only.
</p></dd>
</dl>
</dd>
</dl>
<a name="pixdesctest"></a>
<h2 class="section"><a href="ffmpeg-filters.html#toc-pixdesctest">9.68 pixdesctest</a></h2>
<p>Pixel format descriptor test filter, mainly useful for internal
testing. The output video should be equal to the input video.
</p>
<p>For example:
</p><div class="example">
<pre class="example">format=monow, pixdesctest
</pre></div>
<p>can be used to test the monowhite pixel format descriptor definition.
</p>
<a name="pp"></a>
<h2 class="section"><a href="ffmpeg-filters.html#toc-pp">9.69 pp</a></h2>
<p>Enable the specified chain of postprocessing subfilters using libpostproc. This
library should be automatically selected with a GPL build (<code>--enable-gpl</code>).
Subfilters must be separated by ’/’ and can be disabled by prepending a ’-’.
Each subfilter and some options have a short and a long name that can be used
interchangeably, i.e. dr/dering are the same.
</p>
<p>The filters accept the following options:
</p>
<dl compact="compact">
<dt>‘<samp>subfilters</samp>’</dt>
<dd><p>Set postprocessing subfilters string.
</p></dd>
</dl>
<p>All subfilters share common options to determine their scope:
</p>
<dl compact="compact">
<dt>‘<samp>a/autoq</samp>’</dt>
<dd><p>Honor the quality commands for this subfilter.
</p>
</dd>
<dt>‘<samp>c/chrom</samp>’</dt>
<dd><p>Do chrominance filtering, too (default).
</p>
</dd>
<dt>‘<samp>y/nochrom</samp>’</dt>
<dd><p>Do luminance filtering only (no chrominance).
</p>
</dd>
<dt>‘<samp>n/noluma</samp>’</dt>
<dd><p>Do chrominance filtering only (no luminance).
</p></dd>
</dl>
<p>These options can be appended after the subfilter name, separated by a ’|’.
</p>
<p>Available subfilters are:
</p>
<dl compact="compact">
<dt>‘<samp>hb/hdeblock[|difference[|flatness]]</samp>’</dt>
<dd><p>Horizontal deblocking filter
</p><dl compact="compact">
<dt>‘<samp>difference</samp>’</dt>
<dd><p>Difference factor where higher values mean more deblocking (default: <code>32</code>).
</p></dd>
<dt>‘<samp>flatness</samp>’</dt>
<dd><p>Flatness threshold where lower values mean more deblocking (default: <code>39</code>).
</p></dd>
</dl>
</dd>
<dt>‘<samp>vb/vdeblock[|difference[|flatness]]</samp>’</dt>
<dd><p>Vertical deblocking filter
</p><dl compact="compact">
<dt>‘<samp>difference</samp>’</dt>
<dd><p>Difference factor where higher values mean more deblocking (default: <code>32</code>).
</p></dd>
<dt>‘<samp>flatness</samp>’</dt>
<dd><p>Flatness threshold where lower values mean more deblocking (default: <code>39</code>).
</p></dd>
</dl>
</dd>
<dt>‘<samp>ha/hadeblock[|difference[|flatness]]</samp>’</dt>
<dd><p>Accurate horizontal deblocking filter
</p><dl compact="compact">
<dt>‘<samp>difference</samp>’</dt>
<dd><p>Difference factor where higher values mean more deblocking (default: <code>32</code>).
</p></dd>
<dt>‘<samp>flatness</samp>’</dt>
<dd><p>Flatness threshold where lower values mean more deblocking (default: <code>39</code>).
</p></dd>
</dl>
</dd>
<dt>‘<samp>va/vadeblock[|difference[|flatness]]</samp>’</dt>
<dd><p>Accurate vertical deblocking filter
</p><dl compact="compact">
<dt>‘<samp>difference</samp>’</dt>
<dd><p>Difference factor where higher values mean more deblocking (default: <code>32</code>).
</p></dd>
<dt>‘<samp>flatness</samp>’</dt>
<dd><p>Flatness threshold where lower values mean more deblocking (default: <code>39</code>).
</p></dd>
</dl>
</dd>
</dl>
<p>The horizontal and vertical deblocking filters share the difference and
flatness values so you cannot set different horizontal and vertical
thresholds.
</p>
<dl compact="compact">
<dt>‘<samp>h1/x1hdeblock</samp>’</dt>
<dd><p>Experimental horizontal deblocking filter
</p>
</dd>
<dt>‘<samp>v1/x1vdeblock</samp>’</dt>
<dd><p>Experimental vertical deblocking filter
</p>
</dd>
<dt>‘<samp>dr/dering</samp>’</dt>
<dd><p>Deringing filter
</p>
</dd>
<dt>‘<samp>tn/tmpnoise[|threshold1[|threshold2[|threshold3]]], temporal noise reducer</samp>’</dt>
<dd><dl compact="compact">
<dt>‘<samp>threshold1</samp>’</dt>
<dd><p>larger -> stronger filtering
</p></dd>
<dt>‘<samp>threshold2</samp>’</dt>
<dd><p>larger -> stronger filtering
</p></dd>
<dt>‘<samp>threshold3</samp>’</dt>
<dd><p>larger -> stronger filtering
</p></dd>
</dl>
</dd>
<dt>‘<samp>al/autolevels[:f/fullyrange], automatic brightness / contrast correction</samp>’</dt>
<dd><dl compact="compact">
<dt>‘<samp>f/fullyrange</samp>’</dt>
<dd><p>Stretch luminance to <code>0-255</code>.
</p></dd>
</dl>
</dd>
<dt>‘<samp>lb/linblenddeint</samp>’</dt>
<dd><p>Linear blend deinterlacing filter that deinterlaces the given block by
filtering all lines with a <code>(1 2 1)</code> filter.
</p>
</dd>
<dt>‘<samp>li/linipoldeint</samp>’</dt>
<dd><p>Linear interpolating deinterlacing filter that deinterlaces the given block by
linearly interpolating every second line.
</p>
</dd>
<dt>‘<samp>ci/cubicipoldeint</samp>’</dt>
<dd><p>Cubic interpolating deinterlacing filter deinterlaces the given block by
cubically interpolating every second line.
</p>
</dd>
<dt>‘<samp>md/mediandeint</samp>’</dt>
<dd><p>Median deinterlacing filter that deinterlaces the given block by applying a
median filter to every second line.
</p>
</dd>
<dt>‘<samp>fd/ffmpegdeint</samp>’</dt>
<dd><p>FFmpeg deinterlacing filter that deinterlaces the given block by filtering every
second line with a <code>(-1 4 2 4 -1)</code> filter.
</p>
</dd>
<dt>‘<samp>l5/lowpass5</samp>’</dt>
<dd><p>Vertically applied FIR lowpass deinterlacing filter that deinterlaces the given
block by filtering all lines with a <code>(-1 2 6 2 -1)</code> filter.
</p>
</dd>
<dt>‘<samp>fq/forceQuant[|quantizer]</samp>’</dt>
<dd><p>Overrides the quantizer table from the input with the constant quantizer you
specify.
</p><dl compact="compact">
<dt>‘<samp>quantizer</samp>’</dt>
<dd><p>Quantizer to use
</p></dd>
</dl>
</dd>
<dt>‘<samp>de/default</samp>’</dt>
<dd><p>Default pp filter combination (<code>hb|a,vb|a,dr|a</code>)
</p>
</dd>
<dt>‘<samp>fa/fast</samp>’</dt>
<dd><p>Fast pp filter combination (<code>h1|a,v1|a,dr|a</code>)
</p>
</dd>
<dt>‘<samp>ac</samp>’</dt>
<dd><p>High quality pp filter combination (<code>ha|a|128|7,va|a,dr|a</code>)
</p></dd>
</dl>
<a name="Examples-18"></a>
<h3 class="subsection"><a href="ffmpeg-filters.html#toc-Examples-18">9.69.1 Examples</a></h3>
<ul>
<li>
Apply horizontal and vertical deblocking, deringing and automatic
brightness/contrast:
<div class="example">
<pre class="example">pp=hb/vb/dr/al
</pre></div>
</li><li>
Apply default filters without brightness/contrast correction:
<div class="example">
<pre class="example">pp=de/-al
</pre></div>
</li><li>
Apply default filters and temporal denoiser:
<div class="example">
<pre class="example">pp=default/tmpnoise|1|2|3
</pre></div>
</li><li>
Apply deblocking on luminance only, and switch vertical deblocking on or off
automatically depending on available CPU time:
<div class="example">
<pre class="example">pp=hb|y/vb|a
</pre></div>
</li></ul>
<a name="psnr"></a>
<h2 class="section"><a href="ffmpeg-filters.html#toc-psnr">9.70 psnr</a></h2>
<p>Obtain the average, maximum and minimum PSNR (Peak Signal to Noise
Ratio) between two input videos.
</p>
<p>This filter takes in input two input videos, the first input is
considered the "main" source and is passed unchanged to the
output. The second input is used as a "reference" video for computing
the PSNR.
</p>
<p>Both video inputs must have the same resolution and pixel format for
this filter to work correctly. Also it assumes that both inputs
have the same number of frames, which are compared one by one.
</p>
<p>The obtained average PSNR is printed through the logging system.
</p>
<p>The filter stores the accumulated MSE (mean squared error) of each
frame, and at the end of the processing it is averaged across all frames
equally, and the following formula is applied to obtain the PSNR:
</p>
<div class="example">
<pre class="example">PSNR = 10*log10(MAX^2/MSE)
</pre></div>
<p>Where MAX is the average of the maximum values of each component of the
image.
</p>
<p>The description of the accepted parameters follows.
</p>
<dl compact="compact">
<dt>‘<samp>stats_file, f</samp>’</dt>
<dd><p>If specified the filter will use the named file to save the PSNR of
each individual frame.
</p></dd>
</dl>
<p>The file printed if <var>stats_file</var> is selected, contains a sequence of
key/value pairs of the form <var>key</var>:<var>value</var> for each compared
couple of frames.
</p>
<p>A description of each shown parameter follows:
</p>
<dl compact="compact">
<dt>‘<samp>n</samp>’</dt>
<dd><p>sequential number of the input frame, starting from 1
</p>
</dd>
<dt>‘<samp>mse_avg</samp>’</dt>
<dd><p>Mean Square Error pixel-by-pixel average difference of the compared
frames, averaged over all the image components.
</p>
</dd>
<dt>‘<samp>mse_y, mse_u, mse_v, mse_r, mse_g, mse_g, mse_a</samp>’</dt>
<dd><p>Mean Square Error pixel-by-pixel average difference of the compared
frames for the component specified by the suffix.
</p>
</dd>
<dt>‘<samp>psnr_y, psnr_u, psnr_v, psnr_r, psnr_g, psnr_b, psnr_a</samp>’</dt>
<dd><p>Peak Signal to Noise ratio of the compared frames for the component
specified by the suffix.
</p></dd>
</dl>
<p>For example:
</p><div class="example">
<pre class="example">movie=ref_movie.mpg, setpts=PTS-STARTPTS [main];
[main][ref] psnr="stats_file=stats.log" [out]
</pre></div>
<p>On this example the input file being processed is compared with the
reference file ‘<tt>ref_movie.mpg</tt>’. The PSNR of each individual frame
is stored in ‘<tt>stats.log</tt>’.
</p>
<p><a name="pullup"></a>
</p><a name="pullup-1"></a>
<h2 class="section"><a href="ffmpeg-filters.html#toc-pullup-1">9.71 pullup</a></h2>
<p>Pulldown reversal (inverse telecine) filter, capable of handling mixed
hard-telecine, 24000/1001 fps progressive, and 30000/1001 fps progressive
content.
</p>
<p>The pullup filter is designed to take advantage of future context in making
its decisions. This filter is stateless in the sense that it does not lock
onto a pattern to follow, but it instead looks forward to the following
fields in order to identify matches and rebuild progressive frames.
</p>
<p>To produce content with an even framerate, insert the fps filter after
pullup, use <code>fps=24000/1001</code> if the input frame rate is 29.97fps,
<code>fps=24</code> for 30fps and the (rare) telecined 25fps input.
</p>
<p>The filter accepts the following options:
</p>
<dl compact="compact">
<dt>‘<samp>jl</samp>’</dt>
<dt>‘<samp>jr</samp>’</dt>
<dt>‘<samp>jt</samp>’</dt>
<dt>‘<samp>jb</samp>’</dt>
<dd><p>These options set the amount of "junk" to ignore at the left, right, top, and
bottom of the image, respectively. Left and right are in units of 8 pixels,
while top and bottom are in units of 2 lines.
The default is 8 pixels on each side.
</p>
</dd>
<dt>‘<samp>sb</samp>’</dt>
<dd><p>Set the strict breaks. Setting this option to 1 will reduce the chances of
filter generating an occasional mismatched frame, but it may also cause an
excessive number of frames to be dropped during high motion sequences.
Conversely, setting it to -1 will make filter match fields more easily.
This may help processing of video where there is slight blurring between
the fields, but may also cause there to be interlaced frames in the output.
Default value is <code>0</code>.
</p>
</dd>
<dt>‘<samp>mp</samp>’</dt>
<dd><p>Set the metric plane to use. It accepts the following values:
</p><dl compact="compact">
<dt>‘<samp>l</samp>’</dt>
<dd><p>Use luma plane.
</p>
</dd>
<dt>‘<samp>u</samp>’</dt>
<dd><p>Use chroma blue plane.
</p>
</dd>
<dt>‘<samp>v</samp>’</dt>
<dd><p>Use chroma red plane.
</p></dd>
</dl>
<p>This option may be set to use chroma plane instead of the default luma plane
for doing filter’s computations. This may improve accuracy on very clean
source material, but more likely will decrease accuracy, especially if there
is chroma noise (rainbow effect) or any grayscale video.
The main purpose of setting ‘<samp>mp</samp>’ to a chroma plane is to reduce CPU
load and make pullup usable in realtime on slow machines.
</p></dd>
</dl>
<p>For best results (without duplicated frames in the output file) it is
necessary to change the output frame rate. For example, to inverse
telecine NTSC input:
</p><div class="example">
<pre class="example">ffmpeg -i input -vf pullup -r 24000/1001 ...
</pre></div>
<a name="removelogo"></a>
<h2 class="section"><a href="ffmpeg-filters.html#toc-removelogo">9.72 removelogo</a></h2>
<p>Suppress a TV station logo, using an image file to determine which
pixels comprise the logo. It works by filling in the pixels that
comprise the logo with neighboring pixels.
</p>
<p>The filter accepts the following options:
</p>
<dl compact="compact">
<dt>‘<samp>filename, f</samp>’</dt>
<dd><p>Set the filter bitmap file, which can be any image format supported by
libavformat. The width and height of the image file must match those of the
video stream being processed.
</p></dd>
</dl>
<p>Pixels in the provided bitmap image with a value of zero are not
considered part of the logo, non-zero pixels are considered part of
the logo. If you use white (255) for the logo and black (0) for the
rest, you will be safe. For making the filter bitmap, it is
recommended to take a screen capture of a black frame with the logo
visible, and then using a threshold filter followed by the erode
filter once or twice.
</p>
<p>If needed, little splotches can be fixed manually. Remember that if
logo pixels are not covered, the filter quality will be much
reduced. Marking too many pixels as part of the logo does not hurt as
much, but it will increase the amount of blurring needed to cover over
the image and will destroy more information than necessary, and extra
pixels will slow things down on a large logo.
</p>
<a name="rotate"></a>
<h2 class="section"><a href="ffmpeg-filters.html#toc-rotate">9.73 rotate</a></h2>
<p>Rotate video by an arbitrary angle expressed in radians.
</p>
<p>The filter accepts the following options:
</p>
<p>A description of the optional parameters follows.
</p><dl compact="compact">
<dt>‘<samp>angle, a</samp>’</dt>
<dd><p>Set an expression for the angle by which to rotate the input video
clockwise, expressed as a number of radians. A negative value will
result in a counter-clockwise rotation. By default it is set to "0".
</p>
<p>This expression is evaluated for each frame.
</p>
</dd>
<dt>‘<samp>out_w, ow</samp>’</dt>
<dd><p>Set the output width expression, default value is "iw".
This expression is evaluated just once during configuration.
</p>
</dd>
<dt>‘<samp>out_h, oh</samp>’</dt>
<dd><p>Set the output height expression, default value is "ih".
This expression is evaluated just once during configuration.
</p>
</dd>
<dt>‘<samp>bilinear</samp>’</dt>
<dd><p>Enable bilinear interpolation if set to 1, a value of 0 disables
it. Default value is 1.
</p>
</dd>
<dt>‘<samp>fillcolor, c</samp>’</dt>
<dd><p>Set the color used to fill the output area not covered by the rotated
image. For the generalsyntax of this option, check the "Color" section in the
ffmpeg-utils manual. If the special value "none" is selected then no
background is printed (useful for example if the background is never shown).
</p>
<p>Default value is "black".
</p></dd>
</dl>
<p>The expressions for the angle and the output size can contain the
following constants and functions:
</p>
<dl compact="compact">
<dt>‘<samp>n</samp>’</dt>
<dd><p>sequential number of the input frame, starting from 0. It is always NAN
before the first frame is filtered.
</p>
</dd>
<dt>‘<samp>t</samp>’</dt>
<dd><p>time in seconds of the input frame, it is set to 0 when the filter is
configured. It is always NAN before the first frame is filtered.
</p>
</dd>
<dt>‘<samp>hsub</samp>’</dt>
<dt>‘<samp>vsub</samp>’</dt>
<dd><p>horizontal and vertical chroma subsample values. For example for the
pixel format "yuv422p" <var>hsub</var> is 2 and <var>vsub</var> is 1.
</p>
</dd>
<dt>‘<samp>in_w, iw</samp>’</dt>
<dt>‘<samp>in_h, ih</samp>’</dt>
<dd><p>the input video width and height
</p>
</dd>
<dt>‘<samp>out_w, ow</samp>’</dt>
<dt>‘<samp>out_h, oh</samp>’</dt>
<dd><p>the output width and height, that is the size of the padded area as
specified by the <var>width</var> and <var>height</var> expressions
</p>
</dd>
<dt>‘<samp>rotw(a)</samp>’</dt>
<dt>‘<samp>roth(a)</samp>’</dt>
<dd><p>the minimal width/height required for completely containing the input
video rotated by <var>a</var> radians.
</p>
<p>These are only available when computing the ‘<samp>out_w</samp>’ and
‘<samp>out_h</samp>’ expressions.
</p></dd>
</dl>
<a name="Examples-73"></a>
<h3 class="subsection"><a href="ffmpeg-filters.html#toc-Examples-73">9.73.1 Examples</a></h3>
<ul>
<li>
Rotate the input by PI/6 radians clockwise:
<div class="example">
<pre class="example">rotate=PI/6
</pre></div>
</li><li>
Rotate the input by PI/6 radians counter-clockwise:
<div class="example">
<pre class="example">rotate=-PI/6
</pre></div>
</li><li>
Rotate the input by 45 degrees clockwise:
<div class="example">
<pre class="example">rotate=45*PI/180
</pre></div>
</li><li>
Apply a constant rotation with period T, starting from an angle of PI/3:
<div class="example">
<pre class="example">rotate=PI/3+2*PI*t/T
</pre></div>
</li><li>
Make the input video rotation oscillating with a period of T
seconds and an amplitude of A radians:
<div class="example">
<pre class="example">rotate=A*sin(2*PI/T*t)
</pre></div>
</li><li>
Rotate the video, output size is chosen so that the whole rotating
input video is always completely contained in the output:
<div class="example">
<pre class="example">rotate='2*PI*t:ow=hypot(iw,ih):oh=ow'
</pre></div>
</li><li>
Rotate the video, reduce the output size so that no background is ever
shown:
<div class="example">
<pre class="example">rotate=2*PI*t:ow='min(iw,ih)/sqrt(2)':oh=ow:c=none
</pre></div>
</li></ul>
<a name="Commands-3"></a>
<h3 class="subsection"><a href="ffmpeg-filters.html#toc-Commands-3">9.73.2 Commands</a></h3>
<p>The filter supports the following commands:
</p>
<dl compact="compact">
<dt>‘<samp>a, angle</samp>’</dt>
<dd><p>Set the angle expression.
The command accepts the same syntax of the corresponding option.
</p>
<p>If the specified expression is not valid, it is kept at its current
value.
</p></dd>
</dl>
<a name="sab"></a>
<h2 class="section"><a href="ffmpeg-filters.html#toc-sab">9.74 sab</a></h2>
<p>Apply Shape Adaptive Blur.
</p>
<p>The filter accepts the following options:
</p>
<dl compact="compact">
<dt>‘<samp>luma_radius, lr</samp>’</dt>
<dd><p>Set luma blur filter strength, must be a value in range 0.1-4.0, default
value is 1.0. A greater value will result in a more blurred image, and
in slower processing.
</p>
</dd>
<dt>‘<samp>luma_pre_filter_radius, lpfr</samp>’</dt>
<dd><p>Set luma pre-filter radius, must be a value in the 0.1-2.0 range, default
value is 1.0.
</p>
</dd>
<dt>‘<samp>luma_strength, ls</samp>’</dt>
<dd><p>Set luma maximum difference between pixels to still be considered, must
be a value in the 0.1-100.0 range, default value is 1.0.
</p>
</dd>
<dt>‘<samp>chroma_radius, cr</samp>’</dt>
<dd><p>Set chroma blur filter strength, must be a value in range 0.1-4.0. A
greater value will result in a more blurred image, and in slower
processing.
</p>
</dd>
<dt>‘<samp>chroma_pre_filter_radius, cpfr</samp>’</dt>
<dd><p>Set chroma pre-filter radius, must be a value in the 0.1-2.0 range.
</p>
</dd>
<dt>‘<samp>chroma_strength, cs</samp>’</dt>
<dd><p>Set chroma maximum difference between pixels to still be considered,
must be a value in the 0.1-100.0 range.
</p></dd>
</dl>
<p>Each chroma option value, if not explicitly specified, is set to the
corresponding luma option value.
</p>
<p><a name="scale"></a>
</p><a name="scale-1"></a>
<h2 class="section"><a href="ffmpeg-filters.html#toc-scale-1">9.75 scale</a></h2>
<p>Scale (resize) the input video, using the libswscale library.
</p>
<p>The scale filter forces the output display aspect ratio to be the same
of the input, by changing the output sample aspect ratio.
</p>
<p>If the input image format is different from the format requested by
the next filter, the scale filter will convert the input to the
requested format.
</p>
<a name="Options"></a>
<h3 class="subsection"><a href="ffmpeg-filters.html#toc-Options">9.75.1 Options</a></h3>
<p>The filter accepts the following options, or any of the options
supported by the libswscale scaler.
</p>
<p>See <a href="ffmpeg-scaler.html#scaler_005foptions">(ffmpeg-scaler)the ffmpeg-scaler manual</a> for
the complete list of scaler options.
</p>
<dl compact="compact">
<dt>‘<samp>width, w</samp>’</dt>
<dt>‘<samp>height, h</samp>’</dt>
<dd><p>Set the output video dimension expression. Default value is the input
dimension.
</p>
<p>If the value is 0, the input width is used for the output.
</p>
<p>If one of the values is -1, the scale filter will use a value that
maintains the aspect ratio of the input image, calculated from the
other specified dimension. If both of them are -1, the input size is
used
</p>
<p>If one of the values is -n with n > 1, the scale filter will also use a value
that maintains the aspect ratio of the input image, calculated from the other
specified dimension. After that it will, however, make sure that the calculated
dimension is divisible by n and adjust the value if necessary.
</p>
<p>See below for the list of accepted constants for use in the dimension
expression.
</p>
</dd>
<dt>‘<samp>interl</samp>’</dt>
<dd><p>Set the interlacing mode. It accepts the following values:
</p>
<dl compact="compact">
<dt>‘<samp>1</samp>’</dt>
<dd><p>Force interlaced aware scaling.
</p>
</dd>
<dt>‘<samp>0</samp>’</dt>
<dd><p>Do not apply interlaced scaling.
</p>
</dd>
<dt>‘<samp>-1</samp>’</dt>
<dd><p>Select interlaced aware scaling depending on whether the source frames
are flagged as interlaced or not.
</p></dd>
</dl>
<p>Default value is ‘<samp>0</samp>’.
</p>
</dd>
<dt>‘<samp>flags</samp>’</dt>
<dd><p>Set libswscale scaling flags. See
<a href="ffmpeg-scaler.html#sws_005fflags">(ffmpeg-scaler)the ffmpeg-scaler manual</a> for the
complete list of values. If not explicitly specified the filter applies
the default flags.
</p>
</dd>
<dt>‘<samp>size, s</samp>’</dt>
<dd><p>Set the video size. For the syntax of this option, check the "Video size"
section in the ffmpeg-utils manual.
</p>
</dd>
<dt>‘<samp>in_color_matrix</samp>’</dt>
<dt>‘<samp>out_color_matrix</samp>’</dt>
<dd><p>Set in/output YCbCr color space type.
</p>
<p>This allows the autodetected value to be overridden as well as allows forcing
a specific value used for the output and encoder.
</p>
<p>If not specified, the color space type depends on the pixel format.
</p>
<p>Possible values:
</p>
<dl compact="compact">
<dt>‘<samp>auto</samp>’</dt>
<dd><p>Choose automatically.
</p>
</dd>
<dt>‘<samp>bt709</samp>’</dt>
<dd><p>Format conforming to International Telecommunication Union (ITU)
Recommendation BT.709.
</p>
</dd>
<dt>‘<samp>fcc</samp>’</dt>
<dd><p>Set color space conforming to the United States Federal Communications
Commission (FCC) Code of Federal Regulations (CFR) Title 47 (2003) 73.682 (a).
</p>
</dd>
<dt>‘<samp>bt601</samp>’</dt>
<dd><p>Set color space conforming to:
</p>
<ul>
<li>
ITU Radiocommunication Sector (ITU-R) Recommendation BT.601
</li><li>
ITU-R Rec. BT.470-6 (1998) Systems B, B1, and G
</li><li>
Society of Motion Picture and Television Engineers (SMPTE) ST 170:2004
</li></ul>
</dd>
<dt>‘<samp>smpte240m</samp>’</dt>
<dd><p>Set color space conforming to SMPTE ST 240:1999.
</p></dd>
</dl>
</dd>
<dt>‘<samp>in_range</samp>’</dt>
<dt>‘<samp>out_range</samp>’</dt>
<dd><p>Set in/output YCbCr sample range.
</p>
<p>This allows the autodetected value to be overridden as well as allows forcing
a specific value used for the output and encoder. If not specified, the
range depends on the pixel format. Possible values:
</p>
<dl compact="compact">
<dt>‘<samp>auto</samp>’</dt>
<dd><p>Choose automatically.
</p>
</dd>
<dt>‘<samp>jpeg/full/pc</samp>’</dt>
<dd><p>Set full range (0-255 in case of 8-bit luma).
</p>
</dd>
<dt>‘<samp>mpeg/tv</samp>’</dt>
<dd><p>Set "MPEG" range (16-235 in case of 8-bit luma).
</p></dd>
</dl>
</dd>
<dt>‘<samp>force_original_aspect_ratio</samp>’</dt>
<dd><p>Enable decreasing or increasing output video width or height if necessary to
keep the original aspect ratio. Possible values:
</p>
<dl compact="compact">
<dt>‘<samp>disable</samp>’</dt>
<dd><p>Scale the video as specified and disable this feature.
</p>
</dd>
<dt>‘<samp>decrease</samp>’</dt>
<dd><p>The output video dimensions will automatically be decreased if needed.
</p>
</dd>
<dt>‘<samp>increase</samp>’</dt>
<dd><p>The output video dimensions will automatically be increased if needed.
</p>
</dd>
</dl>
<p>One useful instance of this option is that when you know a specific device’s
maximum allowed resolution, you can use this to limit the output video to
that, while retaining the aspect ratio. For example, device A allows
1280x720 playback, and your video is 1920x800. Using this option (set it to
decrease) and specifying 1280x720 to the command line makes the output
1280x533.
</p>
<p>Please note that this is a different thing than specifying -1 for ‘<samp>w</samp>’
or ‘<samp>h</samp>’, you still need to specify the output resolution for this option
to work.
</p>
</dd>
</dl>
<p>The values of the ‘<samp>w</samp>’ and ‘<samp>h</samp>’ options are expressions
containing the following constants:
</p>
<dl compact="compact">
<dt><var>in_w</var></dt>
<dt><var>in_h</var></dt>
<dd><p>The input width and height
</p>
</dd>
<dt><var>iw</var></dt>
<dt><var>ih</var></dt>
<dd><p>These are the same as <var>in_w</var> and <var>in_h</var>.
</p>
</dd>
<dt><var>out_w</var></dt>
<dt><var>out_h</var></dt>
<dd><p>The output (scaled) width and height
</p>
</dd>
<dt><var>ow</var></dt>
<dt><var>oh</var></dt>
<dd><p>These are the same as <var>out_w</var> and <var>out_h</var>
</p>
</dd>
<dt><var>a</var></dt>
<dd><p>The same as <var>iw</var> / <var>ih</var>
</p>
</dd>
<dt><var>sar</var></dt>
<dd><p>input sample aspect ratio
</p>
</dd>
<dt><var>dar</var></dt>
<dd><p>The input display aspect ratio. Calculated from <code>(iw / ih) * sar</code>.
</p>
</dd>
<dt><var>hsub</var></dt>
<dt><var>vsub</var></dt>
<dd><p>horizontal and vertical input chroma subsample values. For example for the
pixel format "yuv422p" <var>hsub</var> is 2 and <var>vsub</var> is 1.
</p>
</dd>
<dt><var>ohsub</var></dt>
<dt><var>ovsub</var></dt>
<dd><p>horizontal and vertical output chroma subsample values. For example for the
pixel format "yuv422p" <var>hsub</var> is 2 and <var>vsub</var> is 1.
</p></dd>
</dl>
<a name="Examples-67"></a>
<h3 class="subsection"><a href="ffmpeg-filters.html#toc-Examples-67">9.75.2 Examples</a></h3>
<ul>
<li>
Scale the input video to a size of 200x100
<div class="example">
<pre class="example">scale=w=200:h=100
</pre></div>
<p>This is equivalent to:
</p><div class="example">
<pre class="example">scale=200:100
</pre></div>
<p>or:
</p><div class="example">
<pre class="example">scale=200x100
</pre></div>
</li><li>
Specify a size abbreviation for the output size:
<div class="example">
<pre class="example">scale=qcif
</pre></div>
<p>which can also be written as:
</p><div class="example">
<pre class="example">scale=size=qcif
</pre></div>
</li><li>
Scale the input to 2x:
<div class="example">
<pre class="example">scale=w=2*iw:h=2*ih
</pre></div>
</li><li>
The above is the same as:
<div class="example">
<pre class="example">scale=2*in_w:2*in_h
</pre></div>
</li><li>
Scale the input to 2x with forced interlaced scaling:
<div class="example">
<pre class="example">scale=2*iw:2*ih:interl=1
</pre></div>
</li><li>
Scale the input to half size:
<div class="example">
<pre class="example">scale=w=iw/2:h=ih/2
</pre></div>
</li><li>
Increase the width, and set the height to the same size:
<div class="example">
<pre class="example">scale=3/2*iw:ow
</pre></div>
</li><li>
Seek Greek harmony:
<div class="example">
<pre class="example">scale=iw:1/PHI*iw
scale=ih*PHI:ih
</pre></div>
</li><li>
Increase the height, and set the width to 3/2 of the height:
<div class="example">
<pre class="example">scale=w=3/2*oh:h=3/5*ih
</pre></div>
</li><li>
Increase the size, making the size a multiple of the chroma
subsample values:
<div class="example">
<pre class="example">scale="trunc(3/2*iw/hsub)*hsub:trunc(3/2*ih/vsub)*vsub"
</pre></div>
</li><li>
Increase the width to a maximum of 500 pixels,
keeping the same aspect ratio as the input:
<div class="example">
<pre class="example">scale=w='min(500\, iw*3/2):h=-1'
</pre></div>
</li></ul>
<a name="separatefields"></a>
<h2 class="section"><a href="ffmpeg-filters.html#toc-separatefields">9.76 separatefields</a></h2>
<p>The <code>separatefields</code> takes a frame-based video input and splits
each frame into its components fields, producing a new half height clip
with twice the frame rate and twice the frame count.
</p>
<p>This filter use field-dominance information in frame to decide which
of each pair of fields to place first in the output.
If it gets it wrong use <a href="#setfield">setfield</a> filter before <code>separatefields</code> filter.
</p>
<a name="setdar_002c-setsar"></a>
<h2 class="section"><a href="ffmpeg-filters.html#toc-setdar_002c-setsar">9.77 setdar, setsar</a></h2>
<p>The <code>setdar</code> filter sets the Display Aspect Ratio for the filter
output video.
</p>
<p>This is done by changing the specified Sample (aka Pixel) Aspect
Ratio, according to the following equation:
</p><div class="example">
<pre class="example"><var>DAR</var> = <var>HORIZONTAL_RESOLUTION</var> / <var>VERTICAL_RESOLUTION</var> * <var>SAR</var>
</pre></div>
<p>Keep in mind that the <code>setdar</code> filter does not modify the pixel
dimensions of the video frame. Also, the display aspect ratio set by
this filter may be changed by later filters in the filterchain,
e.g. in case of scaling or if another "setdar" or a "setsar" filter is
applied.
</p>
<p>The <code>setsar</code> filter sets the Sample (aka Pixel) Aspect Ratio for
the filter output video.
</p>
<p>Note that as a consequence of the application of this filter, the
output display aspect ratio will change according to the equation
above.
</p>
<p>Keep in mind that the sample aspect ratio set by the <code>setsar</code>
filter may be changed by later filters in the filterchain, e.g. if
another "setsar" or a "setdar" filter is applied.
</p>
<p>It accepts the following parameters:
</p>
<dl compact="compact">
<dt>‘<samp>r, ratio, dar (<code>setdar</code> only), sar (<code>setsar</code> only)</samp>’</dt>
<dd><p>Set the aspect ratio used by the filter.
</p>
<p>The parameter can be a floating point number string, an expression, or
a string of the form <var>num</var>:<var>den</var>, where <var>num</var> and
<var>den</var> are the numerator and denominator of the aspect ratio. If
the parameter is not specified, it is assumed the value "0".
In case the form "<var>num</var>:<var>den</var>" is used, the <code>:</code> character
should be escaped.
</p>
</dd>
<dt>‘<samp>max</samp>’</dt>
<dd><p>Set the maximum integer value to use for expressing numerator and
denominator when reducing the expressed aspect ratio to a rational.
Default value is <code>100</code>.
</p>
</dd>
</dl>
<p>The parameter <var>sar</var> is an expression containing
the following constants:
</p>
<dl compact="compact">
<dt>‘<samp>E, PI, PHI</samp>’</dt>
<dd><p>These are approximated values for the mathematical constants e
(Euler’s number), pi (Greek pi), and phi (the golden ratio).
</p>
</dd>
<dt>‘<samp>w, h</samp>’</dt>
<dd><p>The input width and height.
</p>
</dd>
<dt>‘<samp>a</samp>’</dt>
<dd><p>These are the same as <var>w</var> / <var>h</var>.
</p>
</dd>
<dt>‘<samp>sar</samp>’</dt>
<dd><p>The input sample aspect ratio.
</p>
</dd>
<dt>‘<samp>dar</samp>’</dt>
<dd><p>The input display aspect ratio. It is the same as
(<var>w</var> / <var>h</var>) * <var>sar</var>.
</p>
</dd>
<dt>‘<samp>hsub, vsub</samp>’</dt>
<dd><p>Horizontal and vertical chroma subsample values. For example, for the
pixel format "yuv422p" <var>hsub</var> is 2 and <var>vsub</var> is 1.
</p></dd>
</dl>
<a name="Examples-65"></a>
<h3 class="subsection"><a href="ffmpeg-filters.html#toc-Examples-65">9.77.1 Examples</a></h3>
<ul>
<li>
To change the display aspect ratio to 16:9, specify one of the following:
<div class="example">
<pre class="example">setdar=dar=1.77777
setdar=dar=16/9
setdar=dar=1.77777
</pre></div>
</li><li>
To change the sample aspect ratio to 10:11, specify:
<div class="example">
<pre class="example">setsar=sar=10/11
</pre></div>
</li><li>
To set a display aspect ratio of 16:9, and specify a maximum integer value of
1000 in the aspect ratio reduction, use the command:
<div class="example">
<pre class="example">setdar=ratio=16/9:max=1000
</pre></div>
</li></ul>
<p><a name="setfield"></a>
</p><a name="setfield-1"></a>
<h2 class="section"><a href="ffmpeg-filters.html#toc-setfield-1">9.78 setfield</a></h2>
<p>Force field for the output video frame.
</p>
<p>The <code>setfield</code> filter marks the interlace type field for the
output frames. It does not change the input frame, but only sets the
corresponding property, which affects how the frame is treated by
following filters (e.g. <code>fieldorder</code> or <code>yadif</code>).
</p>
<p>The filter accepts the following options:
</p>
<dl compact="compact">
<dt>‘<samp>mode</samp>’</dt>
<dd><p>Available values are:
</p>
<dl compact="compact">
<dt>‘<samp>auto</samp>’</dt>
<dd><p>Keep the same field property.
</p>
</dd>
<dt>‘<samp>bff</samp>’</dt>
<dd><p>Mark the frame as bottom-field-first.
</p>
</dd>
<dt>‘<samp>tff</samp>’</dt>
<dd><p>Mark the frame as top-field-first.
</p>
</dd>
<dt>‘<samp>prog</samp>’</dt>
<dd><p>Mark the frame as progressive.
</p></dd>
</dl>
</dd>
</dl>
<a name="showinfo"></a>
<h2 class="section"><a href="ffmpeg-filters.html#toc-showinfo">9.79 showinfo</a></h2>
<p>Show a line containing various information for each input video frame.
The input video is not modified.
</p>
<p>The shown line contains a sequence of key/value pairs of the form
<var>key</var>:<var>value</var>.
</p>
<p>The following values are shown in the output:
</p>
<dl compact="compact">
<dt>‘<samp>n</samp>’</dt>
<dd><p>The (sequential) number of the input frame, starting from 0.
</p>
</dd>
<dt>‘<samp>pts</samp>’</dt>
<dd><p>The Presentation TimeStamp of the input frame, expressed as a number of
time base units. The time base unit depends on the filter input pad.
</p>
</dd>
<dt>‘<samp>pts_time</samp>’</dt>
<dd><p>The Presentation TimeStamp of the input frame, expressed as a number of
seconds.
</p>
</dd>
<dt>‘<samp>pos</samp>’</dt>
<dd><p>The position of the frame in the input stream, or -1 if this information is
unavailable and/or meaningless (for example in case of synthetic video).
</p>
</dd>
<dt>‘<samp>fmt</samp>’</dt>
<dd><p>The pixel format name.
</p>
</dd>
<dt>‘<samp>sar</samp>’</dt>
<dd><p>The sample aspect ratio of the input frame, expressed in the form
<var>num</var>/<var>den</var>.
</p>
</dd>
<dt>‘<samp>s</samp>’</dt>
<dd><p>The size of the input frame. For the syntax of this option, check the "Video size"
section in the ffmpeg-utils manual.
</p>
</dd>
<dt>‘<samp>i</samp>’</dt>
<dd><p>The type of interlaced mode ("P" for "progressive", "T" for top field first, "B"
for bottom field first).
</p>
</dd>
<dt>‘<samp>iskey</samp>’</dt>
<dd><p>This is 1 if the frame is a key frame, 0 otherwise.
</p>
</dd>
<dt>‘<samp>type</samp>’</dt>
<dd><p>The picture type of the input frame ("I" for an I-frame, "P" for a
P-frame, "B" for a B-frame, or "?" for an unknown type).
Also refer to the documentation of the <code>AVPictureType</code> enum and of
the <code>av_get_picture_type_char</code> function defined in
‘<tt>libavutil/avutil.h</tt>’.
</p>
</dd>
<dt>‘<samp>checksum</samp>’</dt>
<dd><p>The Adler-32 checksum (printed in hexadecimal) of all the planes of the input frame.
</p>
</dd>
<dt>‘<samp>plane_checksum</samp>’</dt>
<dd><p>The Adler-32 checksum (printed in hexadecimal) of each plane of the input frame,
expressed in the form "[<var>c0</var> <var>c1</var> <var>c2</var> <var>c3</var>]".
</p></dd>
</dl>
<a name="shuffleplanes"></a>
<h2 class="section"><a href="ffmpeg-filters.html#toc-shuffleplanes">9.80 shuffleplanes</a></h2>
<p>Reorder and/or duplicate video planes.
</p>
<p>It accepts the following parameters:
</p>
<dl compact="compact">
<dt>‘<samp>map0</samp>’</dt>
<dd><p>The index of the input plane to be used as the first output plane.
</p>
</dd>
<dt>‘<samp>map1</samp>’</dt>
<dd><p>The index of the input plane to be used as the second output plane.
</p>
</dd>
<dt>‘<samp>map2</samp>’</dt>
<dd><p>The index of the input plane to be used as the third output plane.
</p>
</dd>
<dt>‘<samp>map3</samp>’</dt>
<dd><p>The index of the input plane to be used as the fourth output plane.
</p>
</dd>
</dl>
<p>The first plane has the index 0. The default is to keep the input unchanged.
</p>
<p>Swap the second and third planes of the input:
</p><div class="example">
<pre class="example">ffmpeg -i INPUT -vf shuffleplanes=0:2:1:3 OUTPUT
</pre></div>
<a name="signalstats"></a>
<h2 class="section"><a href="ffmpeg-filters.html#toc-signalstats">9.81 signalstats</a></h2>
<p>Evaluate various visual metrics that assist in determining issues associated
with the digitization of analog video media.
</p>
<p>By default the filter will log these metadata values:
</p>
<dl compact="compact">
<dt>‘<samp>YMIN</samp>’</dt>
<dd><p>Display the minimal Y value contained within the input frame. Expressed in
range of [0-255].
</p>
</dd>
<dt>‘<samp>YLOW</samp>’</dt>
<dd><p>Display the Y value at the 10% percentile within the input frame. Expressed in
range of [0-255].
</p>
</dd>
<dt>‘<samp>YAVG</samp>’</dt>
<dd><p>Display the average Y value within the input frame. Expressed in range of
[0-255].
</p>
</dd>
<dt>‘<samp>YHIGH</samp>’</dt>
<dd><p>Display the Y value at the 90% percentile within the input frame. Expressed in
range of [0-255].
</p>
</dd>
<dt>‘<samp>YMAX</samp>’</dt>
<dd><p>Display the maximum Y value contained within the input frame. Expressed in
range of [0-255].
</p>
</dd>
<dt>‘<samp>UMIN</samp>’</dt>
<dd><p>Display the minimal U value contained within the input frame. Expressed in
range of [0-255].
</p>
</dd>
<dt>‘<samp>ULOW</samp>’</dt>
<dd><p>Display the U value at the 10% percentile within the input frame. Expressed in
range of [0-255].
</p>
</dd>
<dt>‘<samp>UAVG</samp>’</dt>
<dd><p>Display the average U value within the input frame. Expressed in range of
[0-255].
</p>
</dd>
<dt>‘<samp>UHIGH</samp>’</dt>
<dd><p>Display the U value at the 90% percentile within the input frame. Expressed in
range of [0-255].
</p>
</dd>
<dt>‘<samp>UMAX</samp>’</dt>
<dd><p>Display the maximum U value contained within the input frame. Expressed in
range of [0-255].
</p>
</dd>
<dt>‘<samp>VMIN</samp>’</dt>
<dd><p>Display the minimal V value contained within the input frame. Expressed in
range of [0-255].
</p>
</dd>
<dt>‘<samp>VLOW</samp>’</dt>
<dd><p>Display the V value at the 10% percentile within the input frame. Expressed in
range of [0-255].
</p>
</dd>
<dt>‘<samp>VAVG</samp>’</dt>
<dd><p>Display the average V value within the input frame. Expressed in range of
[0-255].
</p>
</dd>
<dt>‘<samp>VHIGH</samp>’</dt>
<dd><p>Display the V value at the 90% percentile within the input frame. Expressed in
range of [0-255].
</p>
</dd>
<dt>‘<samp>VMAX</samp>’</dt>
<dd><p>Display the maximum V value contained within the input frame. Expressed in
range of [0-255].
</p>
</dd>
<dt>‘<samp>SATMIN</samp>’</dt>
<dd><p>Display the minimal saturation value contained within the input frame.
Expressed in range of [0-~181.02].
</p>
</dd>
<dt>‘<samp>SATLOW</samp>’</dt>
<dd><p>Display the saturation value at the 10% percentile within the input frame.
Expressed in range of [0-~181.02].
</p>
</dd>
<dt>‘<samp>SATAVG</samp>’</dt>
<dd><p>Display the average saturation value within the input frame. Expressed in range
of [0-~181.02].
</p>
</dd>
<dt>‘<samp>SATHIGH</samp>’</dt>
<dd><p>Display the saturation value at the 90% percentile within the input frame.
Expressed in range of [0-~181.02].
</p>
</dd>
<dt>‘<samp>SATMAX</samp>’</dt>
<dd><p>Display the maximum saturation value contained within the input frame.
Expressed in range of [0-~181.02].
</p>
</dd>
<dt>‘<samp>HUEMED</samp>’</dt>
<dd><p>Display the median value for hue within the input frame. Expressed in range of
[0-360].
</p>
</dd>
<dt>‘<samp>HUEAVG</samp>’</dt>
<dd><p>Display the average value for hue within the input frame. Expressed in range of
[0-360].
</p>
</dd>
<dt>‘<samp>YDIF</samp>’</dt>
<dd><p>Display the average of sample value difference between all values of the Y
plane in the current frame and corresponding values of the previous input frame.
Expressed in range of [0-255].
</p>
</dd>
<dt>‘<samp>UDIF</samp>’</dt>
<dd><p>Display the average of sample value difference between all values of the U
plane in the current frame and corresponding values of the previous input frame.
Expressed in range of [0-255].
</p>
</dd>
<dt>‘<samp>VDIF</samp>’</dt>
<dd><p>Display the average of sample value difference between all values of the V
plane in the current frame and corresponding values of the previous input frame.
Expressed in range of [0-255].
</p></dd>
</dl>
<p>The filter accepts the following options:
</p>
<dl compact="compact">
<dt>‘<samp>stat</samp>’</dt>
<dt>‘<samp>out</samp>’</dt>
<dd>
<p>‘<samp>stat</samp>’ specify an additional form of image analysis.
‘<samp>out</samp>’ output video with the specified type of pixel highlighted.
</p>
<p>Both options accept the following values:
</p>
<dl compact="compact">
<dt>‘<samp>tout</samp>’</dt>
<dd><p>Identify <var>temporal outliers</var> pixels. A <var>temporal outlier</var> is a pixel
unlike the neighboring pixels of the same field. Examples of temporal outliers
include the results of video dropouts, head clogs, or tape tracking issues.
</p>
</dd>
<dt>‘<samp>vrep</samp>’</dt>
<dd><p>Identify <var>vertical line repetition</var>. Vertical line repetition includes
similar rows of pixels within a frame. In born-digital video vertical line
repetition is common, but this pattern is uncommon in video digitized from an
analog source. When it occurs in video that results from the digitization of an
analog source it can indicate concealment from a dropout compensator.
</p>
</dd>
<dt>‘<samp>brng</samp>’</dt>
<dd><p>Identify pixels that fall outside of legal broadcast range.
</p></dd>
</dl>
</dd>
<dt>‘<samp>color, c</samp>’</dt>
<dd><p>Set the highlight color for the ‘<samp>out</samp>’ option. The default color is
yellow.
</p></dd>
</dl>
<a name="Examples-21"></a>
<h3 class="subsection"><a href="ffmpeg-filters.html#toc-Examples-21">9.81.1 Examples</a></h3>
<ul>
<li>
Output data of various video metrics:
<div class="example">
<pre class="example">ffprobe -f lavfi movie=example.mov,signalstats="stat=tout+vrep+brng" -show_frames
</pre></div>
</li><li>
Output specific data about the minimum and maximum values of the Y plane per frame:
<div class="example">
<pre class="example">ffprobe -f lavfi movie=example.mov,signalstats -show_entries frame_tags=lavfi.signalstats.YMAX,lavfi.signalstats.YMIN
</pre></div>
</li><li>
Playback video while highlighting pixels that are outside of broadcast range in red.
<div class="example">
<pre class="example">ffplay example.mov -vf signalstats="out=brng:color=red"
</pre></div>
</li><li>
Playback video with signalstats metadata drawn over the frame.
<div class="example">
<pre class="example">ffplay example.mov -vf signalstats=stat=brng+vrep+tout,drawtext=fontfile=FreeSerif.ttf:textfile=signalstat_drawtext.txt
</pre></div>
<p>The contents of signalstat_drawtext.txt used in the command are:
</p><div class="example">
<pre class="example">time %{pts:hms}
Y (%{metadata:lavfi.signalstats.YMIN}-%{metadata:lavfi.signalstats.YMAX})
U (%{metadata:lavfi.signalstats.UMIN}-%{metadata:lavfi.signalstats.UMAX})
V (%{metadata:lavfi.signalstats.VMIN}-%{metadata:lavfi.signalstats.VMAX})
saturation maximum: %{metadata:lavfi.signalstats.SATMAX}
</pre></div>
</li></ul>
<p><a name="smartblur"></a>
</p><a name="smartblur-1"></a>
<h2 class="section"><a href="ffmpeg-filters.html#toc-smartblur-1">9.82 smartblur</a></h2>
<p>Blur the input video without impacting the outlines.
</p>
<p>It accepts the following options:
</p>
<dl compact="compact">
<dt>‘<samp>luma_radius, lr</samp>’</dt>
<dd><p>Set the luma radius. The option value must be a float number in
the range [0.1,5.0] that specifies the variance of the gaussian filter
used to blur the image (slower if larger). Default value is 1.0.
</p>
</dd>
<dt>‘<samp>luma_strength, ls</samp>’</dt>
<dd><p>Set the luma strength. The option value must be a float number
in the range [-1.0,1.0] that configures the blurring. A value included
in [0.0,1.0] will blur the image whereas a value included in
[-1.0,0.0] will sharpen the image. Default value is 1.0.
</p>
</dd>
<dt>‘<samp>luma_threshold, lt</samp>’</dt>
<dd><p>Set the luma threshold used as a coefficient to determine
whether a pixel should be blurred or not. The option value must be an
integer in the range [-30,30]. A value of 0 will filter all the image,
a value included in [0,30] will filter flat areas and a value included
in [-30,0] will filter edges. Default value is 0.
</p>
</dd>
<dt>‘<samp>chroma_radius, cr</samp>’</dt>
<dd><p>Set the chroma radius. The option value must be a float number in
the range [0.1,5.0] that specifies the variance of the gaussian filter
used to blur the image (slower if larger). Default value is 1.0.
</p>
</dd>
<dt>‘<samp>chroma_strength, cs</samp>’</dt>
<dd><p>Set the chroma strength. The option value must be a float number
in the range [-1.0,1.0] that configures the blurring. A value included
in [0.0,1.0] will blur the image whereas a value included in
[-1.0,0.0] will sharpen the image. Default value is 1.0.
</p>
</dd>
<dt>‘<samp>chroma_threshold, ct</samp>’</dt>
<dd><p>Set the chroma threshold used as a coefficient to determine
whether a pixel should be blurred or not. The option value must be an
integer in the range [-30,30]. A value of 0 will filter all the image,
a value included in [0,30] will filter flat areas and a value included
in [-30,0] will filter edges. Default value is 0.
</p></dd>
</dl>
<p>If a chroma option is not explicitly set, the corresponding luma value
is set.
</p>
<a name="stereo3d"></a>
<h2 class="section"><a href="ffmpeg-filters.html#toc-stereo3d">9.83 stereo3d</a></h2>
<p>Convert between different stereoscopic image formats.
</p>
<p>The filters accept the following options:
</p>
<dl compact="compact">
<dt>‘<samp>in</samp>’</dt>
<dd><p>Set stereoscopic image format of input.
</p>
<p>Available values for input image formats are:
</p><dl compact="compact">
<dt>‘<samp>sbsl</samp>’</dt>
<dd><p>side by side parallel (left eye left, right eye right)
</p>
</dd>
<dt>‘<samp>sbsr</samp>’</dt>
<dd><p>side by side crosseye (right eye left, left eye right)
</p>
</dd>
<dt>‘<samp>sbs2l</samp>’</dt>
<dd><p>side by side parallel with half width resolution
(left eye left, right eye right)
</p>
</dd>
<dt>‘<samp>sbs2r</samp>’</dt>
<dd><p>side by side crosseye with half width resolution
(right eye left, left eye right)
</p>
</dd>
<dt>‘<samp>abl</samp>’</dt>
<dd><p>above-below (left eye above, right eye below)
</p>
</dd>
<dt>‘<samp>abr</samp>’</dt>
<dd><p>above-below (right eye above, left eye below)
</p>
</dd>
<dt>‘<samp>ab2l</samp>’</dt>
<dd><p>above-below with half height resolution
(left eye above, right eye below)
</p>
</dd>
<dt>‘<samp>ab2r</samp>’</dt>
<dd><p>above-below with half height resolution
(right eye above, left eye below)
</p>
</dd>
<dt>‘<samp>al</samp>’</dt>
<dd><p>alternating frames (left eye first, right eye second)
</p>
</dd>
<dt>‘<samp>ar</samp>’</dt>
<dd><p>alternating frames (right eye first, left eye second)
</p>
<p>Default value is ‘<samp>sbsl</samp>’.
</p></dd>
</dl>
</dd>
<dt>‘<samp>out</samp>’</dt>
<dd><p>Set stereoscopic image format of output.
</p>
<p>Available values for output image formats are all the input formats as well as:
</p><dl compact="compact">
<dt>‘<samp>arbg</samp>’</dt>
<dd><p>anaglyph red/blue gray
(red filter on left eye, blue filter on right eye)
</p>
</dd>
<dt>‘<samp>argg</samp>’</dt>
<dd><p>anaglyph red/green gray
(red filter on left eye, green filter on right eye)
</p>
</dd>
<dt>‘<samp>arcg</samp>’</dt>
<dd><p>anaglyph red/cyan gray
(red filter on left eye, cyan filter on right eye)
</p>
</dd>
<dt>‘<samp>arch</samp>’</dt>
<dd><p>anaglyph red/cyan half colored
(red filter on left eye, cyan filter on right eye)
</p>
</dd>
<dt>‘<samp>arcc</samp>’</dt>
<dd><p>anaglyph red/cyan color
(red filter on left eye, cyan filter on right eye)
</p>
</dd>
<dt>‘<samp>arcd</samp>’</dt>
<dd><p>anaglyph red/cyan color optimized with the least squares projection of dubois
(red filter on left eye, cyan filter on right eye)
</p>
</dd>
<dt>‘<samp>agmg</samp>’</dt>
<dd><p>anaglyph green/magenta gray
(green filter on left eye, magenta filter on right eye)
</p>
</dd>
<dt>‘<samp>agmh</samp>’</dt>
<dd><p>anaglyph green/magenta half colored
(green filter on left eye, magenta filter on right eye)
</p>
</dd>
<dt>‘<samp>agmc</samp>’</dt>
<dd><p>anaglyph green/magenta colored
(green filter on left eye, magenta filter on right eye)
</p>
</dd>
<dt>‘<samp>agmd</samp>’</dt>
<dd><p>anaglyph green/magenta color optimized with the least squares projection of dubois
(green filter on left eye, magenta filter on right eye)
</p>
</dd>
<dt>‘<samp>aybg</samp>’</dt>
<dd><p>anaglyph yellow/blue gray
(yellow filter on left eye, blue filter on right eye)
</p>
</dd>
<dt>‘<samp>aybh</samp>’</dt>
<dd><p>anaglyph yellow/blue half colored
(yellow filter on left eye, blue filter on right eye)
</p>
</dd>
<dt>‘<samp>aybc</samp>’</dt>
<dd><p>anaglyph yellow/blue colored
(yellow filter on left eye, blue filter on right eye)
</p>
</dd>
<dt>‘<samp>aybd</samp>’</dt>
<dd><p>anaglyph yellow/blue color optimized with the least squares projection of dubois
(yellow filter on left eye, blue filter on right eye)
</p>
</dd>
<dt>‘<samp>irl</samp>’</dt>
<dd><p>interleaved rows (left eye has top row, right eye starts on next row)
</p>
</dd>
<dt>‘<samp>irr</samp>’</dt>
<dd><p>interleaved rows (right eye has top row, left eye starts on next row)
</p>
</dd>
<dt>‘<samp>ml</samp>’</dt>
<dd><p>mono output (left eye only)
</p>
</dd>
<dt>‘<samp>mr</samp>’</dt>
<dd><p>mono output (right eye only)
</p></dd>
</dl>
<p>Default value is ‘<samp>arcd</samp>’.
</p></dd>
</dl>
<a name="Examples-38"></a>
<h3 class="subsection"><a href="ffmpeg-filters.html#toc-Examples-38">9.83.1 Examples</a></h3>
<ul>
<li>
Convert input video from side by side parallel to anaglyph yellow/blue dubois:
<div class="example">
<pre class="example">stereo3d=sbsl:aybd
</pre></div>
</li><li>
Convert input video from above bellow (left eye above, right eye below) to side by side crosseye.
<div class="example">
<pre class="example">stereo3d=abl:sbsr
</pre></div>
</li></ul>
<a name="spp"></a>
<h2 class="section"><a href="ffmpeg-filters.html#toc-spp">9.84 spp</a></h2>
<p>Apply a simple postprocessing filter that compresses and decompresses the image
at several (or - in the case of ‘<samp>quality</samp>’ level <code>6</code> - all) shifts
and average the results.
</p>
<p>The filter accepts the following options:
</p>
<dl compact="compact">
<dt>‘<samp>quality</samp>’</dt>
<dd><p>Set quality. This option defines the number of levels for averaging. It accepts
an integer in the range 0-6. If set to <code>0</code>, the filter will have no
effect. A value of <code>6</code> means the higher quality. For each increment of
that value the speed drops by a factor of approximately 2. Default value is
<code>3</code>.
</p>
</dd>
<dt>‘<samp>qp</samp>’</dt>
<dd><p>Force a constant quantization parameter. If not set, the filter will use the QP
from the video stream (if available).
</p>
</dd>
<dt>‘<samp>mode</samp>’</dt>
<dd><p>Set thresholding mode. Available modes are:
</p>
<dl compact="compact">
<dt>‘<samp>hard</samp>’</dt>
<dd><p>Set hard thresholding (default).
</p></dd>
<dt>‘<samp>soft</samp>’</dt>
<dd><p>Set soft thresholding (better de-ringing effect, but likely blurrier).
</p></dd>
</dl>
</dd>
<dt>‘<samp>use_bframe_qp</samp>’</dt>
<dd><p>Enable the use of the QP from the B-Frames if set to <code>1</code>. Using this
option may cause flicker since the B-Frames have often larger QP. Default is
<code>0</code> (not enabled).
</p></dd>
</dl>
<p><a name="subtitles"></a>
</p><a name="subtitles-1"></a>
<h2 class="section"><a href="ffmpeg-filters.html#toc-subtitles-1">9.85 subtitles</a></h2>
<p>Draw subtitles on top of input video using the libass library.
</p>
<p>To enable compilation of this filter you need to configure FFmpeg with
<code>--enable-libass</code>. This filter also requires a build with libavcodec and
libavformat to convert the passed subtitles file to ASS (Advanced Substation
Alpha) subtitles format.
</p>
<p>The filter accepts the following options:
</p>
<dl compact="compact">
<dt>‘<samp>filename, f</samp>’</dt>
<dd><p>Set the filename of the subtitle file to read. It must be specified.
</p>
</dd>
<dt>‘<samp>original_size</samp>’</dt>
<dd><p>Specify the size of the original video, the video for which the ASS file
was composed. For the syntax of this option, check the "Video size" section in
the ffmpeg-utils manual. Due to a misdesign in ASS aspect ratio arithmetic,
this is necessary to correctly scale the fonts if the aspect ratio has been
changed.
</p>
</dd>
<dt>‘<samp>charenc</samp>’</dt>
<dd><p>Set subtitles input character encoding. <code>subtitles</code> filter only. Only
useful if not UTF-8.
</p>
</dd>
<dt>‘<samp>stream_index, si</samp>’</dt>
<dd><p>Set subtitles stream index. <code>subtitles</code> filter only.
</p></dd>
</dl>
<p>If the first key is not specified, it is assumed that the first value
specifies the ‘<samp>filename</samp>’.
</p>
<p>For example, to render the file ‘<tt>sub.srt</tt>’ on top of the input
video, use the command:
</p><div class="example">
<pre class="example">subtitles=sub.srt
</pre></div>
<p>which is equivalent to:
</p><div class="example">
<pre class="example">subtitles=filename=sub.srt
</pre></div>
<p>To render the default subtitles stream from file ‘<tt>video.mkv</tt>’, use:
</p><div class="example">
<pre class="example">subtitles=video.mkv
</pre></div>
<p>To render the second subtitles stream from that file, use:
</p><div class="example">
<pre class="example">subtitles=video.mkv:si=1
</pre></div>
<a name="super2xsai"></a>
<h2 class="section"><a href="ffmpeg-filters.html#toc-super2xsai">9.86 super2xsai</a></h2>
<p>Scale the input by 2x and smooth using the Super2xSaI (Scale and
Interpolate) pixel art scaling algorithm.
</p>
<p>Useful for enlarging pixel art images without reducing sharpness.
</p>
<a name="swapuv"></a>
<h2 class="section"><a href="ffmpeg-filters.html#toc-swapuv">9.87 swapuv</a></h2>
<p>Swap U & V plane.
</p>
<a name="telecine"></a>
<h2 class="section"><a href="ffmpeg-filters.html#toc-telecine">9.88 telecine</a></h2>
<p>Apply telecine process to the video.
</p>
<p>This filter accepts the following options:
</p>
<dl compact="compact">
<dt>‘<samp>first_field</samp>’</dt>
<dd><dl compact="compact">
<dt>‘<samp>top, t</samp>’</dt>
<dd><p>top field first
</p></dd>
<dt>‘<samp>bottom, b</samp>’</dt>
<dd><p>bottom field first
The default value is <code>top</code>.
</p></dd>
</dl>
</dd>
<dt>‘<samp>pattern</samp>’</dt>
<dd><p>A string of numbers representing the pulldown pattern you wish to apply.
The default value is <code>23</code>.
</p></dd>
</dl>
<div class="example">
<pre class="example">Some typical patterns:
NTSC output (30i):
27.5p: 32222
24p: 23 (classic)
24p: 2332 (preferred)
20p: 33
18p: 334
16p: 3444
PAL output (25i):
27.5p: 12222
24p: 222222222223 ("Euro pulldown")
16.67p: 33
16p: 33333334
</pre></div>
<a name="thumbnail"></a>
<h2 class="section"><a href="ffmpeg-filters.html#toc-thumbnail">9.89 thumbnail</a></h2>
<p>Select the most representative frame in a given sequence of consecutive frames.
</p>
<p>The filter accepts the following options:
</p>
<dl compact="compact">
<dt>‘<samp>n</samp>’</dt>
<dd><p>Set the frames batch size to analyze; in a set of <var>n</var> frames, the filter
will pick one of them, and then handle the next batch of <var>n</var> frames until
the end. Default is <code>100</code>.
</p></dd>
</dl>
<p>Since the filter keeps track of the whole frames sequence, a bigger <var>n</var>
value will result in a higher memory usage, so a high value is not recommended.
</p>
<a name="Examples-3"></a>
<h3 class="subsection"><a href="ffmpeg-filters.html#toc-Examples-3">9.89.1 Examples</a></h3>
<ul>
<li>
Extract one picture each 50 frames:
<div class="example">
<pre class="example">thumbnail=50
</pre></div>
</li><li>
Complete example of a thumbnail creation with <code>ffmpeg</code>:
<div class="example">
<pre class="example">ffmpeg -i in.avi -vf thumbnail,scale=300:200 -frames:v 1 out.png
</pre></div>
</li></ul>
<a name="tile"></a>
<h2 class="section"><a href="ffmpeg-filters.html#toc-tile">9.90 tile</a></h2>
<p>Tile several successive frames together.
</p>
<p>The filter accepts the following options:
</p>
<dl compact="compact">
<dt>‘<samp>layout</samp>’</dt>
<dd><p>Set the grid size (i.e. the number of lines and columns). For the syntax of
this option, check the "Video size" section in the ffmpeg-utils manual.
</p>
</dd>
<dt>‘<samp>nb_frames</samp>’</dt>
<dd><p>Set the maximum number of frames to render in the given area. It must be less
than or equal to <var>w</var>x<var>h</var>. The default value is <code>0</code>, meaning all
the area will be used.
</p>
</dd>
<dt>‘<samp>margin</samp>’</dt>
<dd><p>Set the outer border margin in pixels.
</p>
</dd>
<dt>‘<samp>padding</samp>’</dt>
<dd><p>Set the inner border thickness (i.e. the number of pixels between frames). For
more advanced padding options (such as having different values for the edges),
refer to the pad video filter.
</p>
</dd>
<dt>‘<samp>color</samp>’</dt>
<dd><p>Specify the color of the unused areaFor the syntax of this option, check the
"Color" section in the ffmpeg-utils manual. The default value of <var>color</var>
is "black".
</p></dd>
</dl>
<a name="Examples-24"></a>
<h3 class="subsection"><a href="ffmpeg-filters.html#toc-Examples-24">9.90.1 Examples</a></h3>
<ul>
<li>
Produce 8x8 PNG tiles of all keyframes (‘<samp>-skip_frame nokey</samp>’) in a movie:
<div class="example">
<pre class="example">ffmpeg -skip_frame nokey -i file.avi -vf 'scale=128:72,tile=8x8' -an -vsync 0 keyframes%03d.png
</pre></div>
<p>The ‘<samp>-vsync 0</samp>’ is necessary to prevent <code>ffmpeg</code> from
duplicating each output frame to accommodate the originally detected frame
rate.
</p>
</li><li>
Display <code>5</code> pictures in an area of <code>3x2</code> frames,
with <code>7</code> pixels between them, and <code>2</code> pixels of initial margin, using
mixed flat and named options:
<div class="example">
<pre class="example">tile=3x2:nb_frames=5:padding=7:margin=2
</pre></div>
</li></ul>
<a name="tinterlace"></a>
<h2 class="section"><a href="ffmpeg-filters.html#toc-tinterlace">9.91 tinterlace</a></h2>
<p>Perform various types of temporal field interlacing.
</p>
<p>Frames are counted starting from 1, so the first input frame is
considered odd.
</p>
<p>The filter accepts the following options:
</p>
<dl compact="compact">
<dt>‘<samp>mode</samp>’</dt>
<dd><p>Specify the mode of the interlacing. This option can also be specified
as a value alone. See below for a list of values for this option.
</p>
<p>Available values are:
</p>
<dl compact="compact">
<dt>‘<samp>merge, 0</samp>’</dt>
<dd><p>Move odd frames into the upper field, even into the lower field,
generating a double height frame at half frame rate.
</p>
</dd>
<dt>‘<samp>drop_odd, 1</samp>’</dt>
<dd><p>Only output even frames, odd frames are dropped, generating a frame with
unchanged height at half frame rate.
</p>
</dd>
<dt>‘<samp>drop_even, 2</samp>’</dt>
<dd><p>Only output odd frames, even frames are dropped, generating a frame with
unchanged height at half frame rate.
</p>
</dd>
<dt>‘<samp>pad, 3</samp>’</dt>
<dd><p>Expand each frame to full height, but pad alternate lines with black,
generating a frame with double height at the same input frame rate.
</p>
</dd>
<dt>‘<samp>interleave_top, 4</samp>’</dt>
<dd><p>Interleave the upper field from odd frames with the lower field from
even frames, generating a frame with unchanged height at half frame rate.
</p>
</dd>
<dt>‘<samp>interleave_bottom, 5</samp>’</dt>
<dd><p>Interleave the lower field from odd frames with the upper field from
even frames, generating a frame with unchanged height at half frame rate.
</p>
</dd>
<dt>‘<samp>interlacex2, 6</samp>’</dt>
<dd><p>Double frame rate with unchanged height. Frames are inserted each
containing the second temporal field from the previous input frame and
the first temporal field from the next input frame. This mode relies on
the top_field_first flag. Useful for interlaced video displays with no
field synchronisation.
</p></dd>
</dl>
<p>Numeric values are deprecated but are accepted for backward
compatibility reasons.
</p>
<p>Default mode is <code>merge</code>.
</p>
</dd>
<dt>‘<samp>flags</samp>’</dt>
<dd><p>Specify flags influencing the filter process.
</p>
<p>Available value for <var>flags</var> is:
</p>
<dl compact="compact">
<dt>‘<samp>low_pass_filter, vlfp</samp>’</dt>
<dd><p>Enable vertical low-pass filtering in the filter.
Vertical low-pass filtering is required when creating an interlaced
destination from a progressive source which contains high-frequency
vertical detail. Filtering will reduce interlace ’twitter’ and Moire
patterning.
</p>
<p>Vertical low-pass filtering can only be enabled for ‘<samp>mode</samp>’
<var>interleave_top</var> and <var>interleave_bottom</var>.
</p>
</dd>
</dl>
</dd>
</dl>
<a name="transpose"></a>
<h2 class="section"><a href="ffmpeg-filters.html#toc-transpose">9.92 transpose</a></h2>
<p>Transpose rows with columns in the input video and optionally flip it.
</p>
<p>It accepts the following parameters:
</p>
<dl compact="compact">
<dt>‘<samp>dir</samp>’</dt>
<dd><p>Specify the transposition direction.
</p>
<p>Can assume the following values:
</p><dl compact="compact">
<dt>‘<samp>0, 4, cclock_flip</samp>’</dt>
<dd><p>Rotate by 90 degrees counterclockwise and vertically flip (default), that is:
</p><div class="example">
<pre class="example">L.R L.l
. . -> . .
l.r R.r
</pre></div>
</dd>
<dt>‘<samp>1, 5, clock</samp>’</dt>
<dd><p>Rotate by 90 degrees clockwise, that is:
</p><div class="example">
<pre class="example">L.R l.L
. . -> . .
l.r r.R
</pre></div>
</dd>
<dt>‘<samp>2, 6, cclock</samp>’</dt>
<dd><p>Rotate by 90 degrees counterclockwise, that is:
</p><div class="example">
<pre class="example">L.R R.r
. . -> . .
l.r L.l
</pre></div>
</dd>
<dt>‘<samp>3, 7, clock_flip</samp>’</dt>
<dd><p>Rotate by 90 degrees clockwise and vertically flip, that is:
</p><div class="example">
<pre class="example">L.R r.R
. . -> . .
l.r l.L
</pre></div>
</dd>
</dl>
<p>For values between 4-7, the transposition is only done if the input
video geometry is portrait and not landscape. These values are
deprecated, the <code>passthrough</code> option should be used instead.
</p>
<p>Numerical values are deprecated, and should be dropped in favor of
symbolic constants.
</p>
</dd>
<dt>‘<samp>passthrough</samp>’</dt>
<dd><p>Do not apply the transposition if the input geometry matches the one
specified by the specified value. It accepts the following values:
</p><dl compact="compact">
<dt>‘<samp>none</samp>’</dt>
<dd><p>Always apply transposition.
</p></dd>
<dt>‘<samp>portrait</samp>’</dt>
<dd><p>Preserve portrait geometry (when <var>height</var> >= <var>width</var>).
</p></dd>
<dt>‘<samp>landscape</samp>’</dt>
<dd><p>Preserve landscape geometry (when <var>width</var> >= <var>height</var>).
</p></dd>
</dl>
<p>Default value is <code>none</code>.
</p></dd>
</dl>
<p>For example to rotate by 90 degrees clockwise and preserve portrait
layout:
</p><div class="example">
<pre class="example">transpose=dir=1:passthrough=portrait
</pre></div>
<p>The command above can also be specified as:
</p><div class="example">
<pre class="example">transpose=1:portrait
</pre></div>
<a name="trim"></a>
<h2 class="section"><a href="ffmpeg-filters.html#toc-trim">9.93 trim</a></h2>
<p>Trim the input so that the output contains one continuous subpart of the input.
</p>
<p>It accepts the following parameters:
</p><dl compact="compact">
<dt>‘<samp>start</samp>’</dt>
<dd><p>Specify the time of the start of the kept section, i.e. the frame with the
timestamp <var>start</var> will be the first frame in the output.
</p>
</dd>
<dt>‘<samp>end</samp>’</dt>
<dd><p>Specify the time of the first frame that will be dropped, i.e. the frame
immediately preceding the one with the timestamp <var>end</var> will be the last
frame in the output.
</p>
</dd>
<dt>‘<samp>start_pts</samp>’</dt>
<dd><p>This is the same as <var>start</var>, except this option sets the start timestamp
in timebase units instead of seconds.
</p>
</dd>
<dt>‘<samp>end_pts</samp>’</dt>
<dd><p>This is the same as <var>end</var>, except this option sets the end timestamp
in timebase units instead of seconds.
</p>
</dd>
<dt>‘<samp>duration</samp>’</dt>
<dd><p>The maximum duration of the output in seconds.
</p>
</dd>
<dt>‘<samp>start_frame</samp>’</dt>
<dd><p>The number of the first frame that should be passed to the output.
</p>
</dd>
<dt>‘<samp>end_frame</samp>’</dt>
<dd><p>The number of the first frame that should be dropped.
</p></dd>
</dl>
<p>‘<samp>start</samp>’, ‘<samp>end</samp>’, and ‘<samp>duration</samp>’ are expressed as time
duration specifications; see
<a href="ffmpeg-utils.html#time-duration-syntax">(ffmpeg-utils)the Time duration section in the ffmpeg-utils(1) manual</a>
for the accepted syntax.
</p>
<p>Note that the first two sets of the start/end options and the ‘<samp>duration</samp>’
option look at the frame timestamp, while the _frame variants simply count the
frames that pass through the filter. Also note that this filter does not modify
the timestamps. If you wish for the output timestamps to start at zero, insert a
setpts filter after the trim filter.
</p>
<p>If multiple start or end options are set, this filter tries to be greedy and
keep all the frames that match at least one of the specified constraints. To keep
only the part that matches all the constraints at once, chain multiple trim
filters.
</p>
<p>The defaults are such that all the input is kept. So it is possible to set e.g.
just the end values to keep everything before the specified time.
</p>
<p>Examples:
</p><ul>
<li>
Drop everything except the second minute of input:
<div class="example">
<pre class="example">ffmpeg -i INPUT -vf trim=60:120
</pre></div>
</li><li>
Keep only the first second:
<div class="example">
<pre class="example">ffmpeg -i INPUT -vf trim=duration=1
</pre></div>
</li></ul>
<a name="unsharp"></a>
<h2 class="section"><a href="ffmpeg-filters.html#toc-unsharp">9.94 unsharp</a></h2>
<p>Sharpen or blur the input video.
</p>
<p>It accepts the following parameters:
</p>
<dl compact="compact">
<dt>‘<samp>luma_msize_x, lx</samp>’</dt>
<dd><p>Set the luma matrix horizontal size. It must be an odd integer between
3 and 63. The default value is 5.
</p>
</dd>
<dt>‘<samp>luma_msize_y, ly</samp>’</dt>
<dd><p>Set the luma matrix vertical size. It must be an odd integer between 3
and 63. The default value is 5.
</p>
</dd>
<dt>‘<samp>luma_amount, la</samp>’</dt>
<dd><p>Set the luma effect strength. It must be a floating point number, reasonable
values lay between -1.5 and 1.5.
</p>
<p>Negative values will blur the input video, while positive values will
sharpen it, a value of zero will disable the effect.
</p>
<p>Default value is 1.0.
</p>
</dd>
<dt>‘<samp>chroma_msize_x, cx</samp>’</dt>
<dd><p>Set the chroma matrix horizontal size. It must be an odd integer
between 3 and 63. The default value is 5.
</p>
</dd>
<dt>‘<samp>chroma_msize_y, cy</samp>’</dt>
<dd><p>Set the chroma matrix vertical size. It must be an odd integer
between 3 and 63. The default value is 5.
</p>
</dd>
<dt>‘<samp>chroma_amount, ca</samp>’</dt>
<dd><p>Set the chroma effect strength. It must be a floating point number, reasonable
values lay between -1.5 and 1.5.
</p>
<p>Negative values will blur the input video, while positive values will
sharpen it, a value of zero will disable the effect.
</p>
<p>Default value is 0.0.
</p>
</dd>
<dt>‘<samp>opencl</samp>’</dt>
<dd><p>If set to 1, specify using OpenCL capabilities, only available if
FFmpeg was configured with <code>--enable-opencl</code>. Default value is 0.
</p>
</dd>
</dl>
<p>All parameters are optional and default to the equivalent of the
string ’5:5:1.0:5:5:0.0’.
</p>
<a name="Examples-23"></a>
<h3 class="subsection"><a href="ffmpeg-filters.html#toc-Examples-23">9.94.1 Examples</a></h3>
<ul>
<li>
Apply strong luma sharpen effect:
<div class="example">
<pre class="example">unsharp=luma_msize_x=7:luma_msize_y=7:luma_amount=2.5
</pre></div>
</li><li>
Apply a strong blur of both luma and chroma parameters:
<div class="example">
<pre class="example">unsharp=7:7:-2:7:7:-2
</pre></div>
</li></ul>
<p><a name="vidstabdetect"></a>
</p><a name="vidstabdetect-1"></a>
<h2 class="section"><a href="ffmpeg-filters.html#toc-vidstabdetect-1">9.95 vidstabdetect</a></h2>
<p>Analyze video stabilization/deshaking. Perform pass 1 of 2, see
<a href="#vidstabtransform">vidstabtransform</a> for pass 2.
</p>
<p>This filter generates a file with relative translation and rotation
transform information about subsequent frames, which is then used by
the <a href="#vidstabtransform">vidstabtransform</a> filter.
</p>
<p>To enable compilation of this filter you need to configure FFmpeg with
<code>--enable-libvidstab</code>.
</p>
<p>This filter accepts the following options:
</p>
<dl compact="compact">
<dt>‘<samp>result</samp>’</dt>
<dd><p>Set the path to the file used to write the transforms information.
Default value is ‘<tt>transforms.trf</tt>’.
</p>
</dd>
<dt>‘<samp>shakiness</samp>’</dt>
<dd><p>Set how shaky the video is and how quick the camera is. It accepts an
integer in the range 1-10, a value of 1 means little shakiness, a
value of 10 means strong shakiness. Default value is 5.
</p>
</dd>
<dt>‘<samp>accuracy</samp>’</dt>
<dd><p>Set the accuracy of the detection process. It must be a value in the
range 1-15. A value of 1 means low accuracy, a value of 15 means high
accuracy. Default value is 15.
</p>
</dd>
<dt>‘<samp>stepsize</samp>’</dt>
<dd><p>Set stepsize of the search process. The region around minimum is
scanned with 1 pixel resolution. Default value is 6.
</p>
</dd>
<dt>‘<samp>mincontrast</samp>’</dt>
<dd><p>Set minimum contrast. Below this value a local measurement field is
discarded. Must be a floating point value in the range 0-1. Default
value is 0.3.
</p>
</dd>
<dt>‘<samp>tripod</samp>’</dt>
<dd><p>Set reference frame number for tripod mode.
</p>
<p>If enabled, the motion of the frames is compared to a reference frame
in the filtered stream, identified by the specified number. The idea
is to compensate all movements in a more-or-less static scene and keep
the camera view absolutely still.
</p>
<p>If set to 0, it is disabled. The frames are counted starting from 1.
</p>
</dd>
<dt>‘<samp>show</samp>’</dt>
<dd><p>Show fields and transforms in the resulting frames. It accepts an
integer in the range 0-2. Default value is 0, which disables any
visualization.
</p></dd>
</dl>
<a name="Examples-25"></a>
<h3 class="subsection"><a href="ffmpeg-filters.html#toc-Examples-25">9.95.1 Examples</a></h3>
<ul>
<li>
Use default values:
<div class="example">
<pre class="example">vidstabdetect
</pre></div>
</li><li>
Analyze strongly shaky movie and put the results in file
‘<tt>mytransforms.trf</tt>’:
<div class="example">
<pre class="example">vidstabdetect=shakiness=10:accuracy=15:result="mytransforms.trf"
</pre></div>
</li><li>
Visualize the result of internal transformations in the resulting
video:
<div class="example">
<pre class="example">vidstabdetect=show=1
</pre></div>
</li><li>
Analyze a video with medium shakiness using <code>ffmpeg</code>:
<div class="example">
<pre class="example">ffmpeg -i input -vf vidstabdetect=shakiness=5:show=1 dummy.avi
</pre></div>
</li></ul>
<p><a name="vidstabtransform"></a>
</p><a name="vidstabtransform-1"></a>
<h2 class="section"><a href="ffmpeg-filters.html#toc-vidstabtransform-1">9.96 vidstabtransform</a></h2>
<p>Video stabilization/deshaking: pass 2 of 2,
see <a href="#vidstabdetect">vidstabdetect</a> for pass 1.
</p>
<p>Read a file with transform information for each frame and
apply/compensate them. Together with the <a href="#vidstabdetect">vidstabdetect</a>
filter this can be used to deshake videos. See also
<a href="http://public.hronopik.de/vid.stab">http://public.hronopik.de/vid.stab</a>. It is important to also use
the unsharp filter, see below.
</p>
<p>To enable compilation of this filter you need to configure FFmpeg with
<code>--enable-libvidstab</code>.
</p>
<a name="Options-2"></a>
<h3 class="subsection"><a href="ffmpeg-filters.html#toc-Options-2">9.96.1 Options</a></h3>
<dl compact="compact">
<dt>‘<samp>input</samp>’</dt>
<dd><p>Set path to the file used to read the transforms. Default value is
‘<tt>transforms.trf</tt>’).
</p>
</dd>
<dt>‘<samp>smoothing</samp>’</dt>
<dd><p>Set the number of frames (value*2 + 1) used for lowpass filtering the
camera movements. Default value is 10.
</p>
<p>For example a number of 10 means that 21 frames are used (10 in the
past and 10 in the future) to smoothen the motion in the video. A
larger values leads to a smoother video, but limits the acceleration
of the camera (pan/tilt movements). 0 is a special case where a
static camera is simulated.
</p>
</dd>
<dt>‘<samp>optalgo</samp>’</dt>
<dd><p>Set the camera path optimization algorithm.
</p>
<p>Accepted values are:
</p><dl compact="compact">
<dt>‘<samp>gauss</samp>’</dt>
<dd><p>gaussian kernel low-pass filter on camera motion (default)
</p></dd>
<dt>‘<samp>avg</samp>’</dt>
<dd><p>averaging on transformations
</p></dd>
</dl>
</dd>
<dt>‘<samp>maxshift</samp>’</dt>
<dd><p>Set maximal number of pixels to translate frames. Default value is -1,
meaning no limit.
</p>
</dd>
<dt>‘<samp>maxangle</samp>’</dt>
<dd><p>Set maximal angle in radians (degree*PI/180) to rotate frames. Default
value is -1, meaning no limit.
</p>
</dd>
<dt>‘<samp>crop</samp>’</dt>
<dd><p>Specify how to deal with borders that may be visible due to movement
compensation.
</p>
<p>Available values are:
</p><dl compact="compact">
<dt>‘<samp>keep</samp>’</dt>
<dd><p>keep image information from previous frame (default)
</p></dd>
<dt>‘<samp>black</samp>’</dt>
<dd><p>fill the border black
</p></dd>
</dl>
</dd>
<dt>‘<samp>invert</samp>’</dt>
<dd><p>Invert transforms if set to 1. Default value is 0.
</p>
</dd>
<dt>‘<samp>relative</samp>’</dt>
<dd><p>Consider transforms as relative to previsou frame if set to 1,
absolute if set to 0. Default value is 0.
</p>
</dd>
<dt>‘<samp>zoom</samp>’</dt>
<dd><p>Set percentage to zoom. A positive value will result in a zoom-in
effect, a negative value in a zoom-out effect. Default value is 0 (no
zoom).
</p>
</dd>
<dt>‘<samp>optzoom</samp>’</dt>
<dd><p>Set optimal zooming to avoid borders.
</p>
<p>Accepted values are:
</p><dl compact="compact">
<dt>‘<samp>0</samp>’</dt>
<dd><p>disabled
</p></dd>
<dt>‘<samp>1</samp>’</dt>
<dd><p>optimal static zoom value is determined (only very strong movements
will lead to visible borders) (default)
</p></dd>
<dt>‘<samp>2</samp>’</dt>
<dd><p>optimal adaptive zoom value is determined (no borders will be
visible), see ‘<samp>zoomspeed</samp>’
</p></dd>
</dl>
<p>Note that the value given at zoom is added to the one calculated here.
</p>
</dd>
<dt>‘<samp>zoomspeed</samp>’</dt>
<dd><p>Set percent to zoom maximally each frame (enabled when
‘<samp>optzoom</samp>’ is set to 2). Range is from 0 to 5, default value is
0.25.
</p>
</dd>
<dt>‘<samp>interpol</samp>’</dt>
<dd><p>Specify type of interpolation.
</p>
<p>Available values are:
</p><dl compact="compact">
<dt>‘<samp>no</samp>’</dt>
<dd><p>no interpolation
</p></dd>
<dt>‘<samp>linear</samp>’</dt>
<dd><p>linear only horizontal
</p></dd>
<dt>‘<samp>bilinear</samp>’</dt>
<dd><p>linear in both directions (default)
</p></dd>
<dt>‘<samp>bicubic</samp>’</dt>
<dd><p>cubic in both directions (slow)
</p></dd>
</dl>
</dd>
<dt>‘<samp>tripod</samp>’</dt>
<dd><p>Enable virtual tripod mode if set to 1, which is equivalent to
<code>relative=0:smoothing=0</code>. Default value is 0.
</p>
<p>Use also <code>tripod</code> option of <a href="#vidstabdetect">vidstabdetect</a>.
</p>
</dd>
<dt>‘<samp>debug</samp>’</dt>
<dd><p>Increase log verbosity if set to 1. Also the detected global motions
are written to the temporary file ‘<tt>global_motions.trf</tt>’. Default
value is 0.
</p></dd>
</dl>
<a name="Examples-54"></a>
<h3 class="subsection"><a href="ffmpeg-filters.html#toc-Examples-54">9.96.2 Examples</a></h3>
<ul>
<li>
Use <code>ffmpeg</code> for a typical stabilization with default values:
<div class="example">
<pre class="example">ffmpeg -i inp.mpeg -vf vidstabtransform,unsharp=5:5:0.8:3:3:0.4 inp_stabilized.mpeg
</pre></div>
<p>Note the use of the unsharp filter which is always recommended.
</p>
</li><li>
Zoom in a bit more and load transform data from a given file:
<div class="example">
<pre class="example">vidstabtransform=zoom=5:input="mytransforms.trf"
</pre></div>
</li><li>
Smoothen the video even more:
<div class="example">
<pre class="example">vidstabtransform=smoothing=30
</pre></div>
</li></ul>
<a name="vflip"></a>
<h2 class="section"><a href="ffmpeg-filters.html#toc-vflip">9.97 vflip</a></h2>
<p>Flip the input video vertically.
</p>
<p>For example, to vertically flip a video with <code>ffmpeg</code>:
</p><div class="example">
<pre class="example">ffmpeg -i in.avi -vf "vflip" out.avi
</pre></div>
<p><a name="vignette"></a>
</p><a name="vignette-1"></a>
<h2 class="section"><a href="ffmpeg-filters.html#toc-vignette-1">9.98 vignette</a></h2>
<p>Make or reverse a natural vignetting effect.
</p>
<p>The filter accepts the following options:
</p>
<dl compact="compact">
<dt>‘<samp>angle, a</samp>’</dt>
<dd><p>Set lens angle expression as a number of radians.
</p>
<p>The value is clipped in the <code>[0,PI/2]</code> range.
</p>
<p>Default value: <code>"PI/5"</code>
</p>
</dd>
<dt>‘<samp>x0</samp>’</dt>
<dt>‘<samp>y0</samp>’</dt>
<dd><p>Set center coordinates expressions. Respectively <code>"w/2"</code> and <code>"h/2"</code>
by default.
</p>
</dd>
<dt>‘<samp>mode</samp>’</dt>
<dd><p>Set forward/backward mode.
</p>
<p>Available modes are:
</p><dl compact="compact">
<dt>‘<samp>forward</samp>’</dt>
<dd><p>The larger the distance from the central point, the darker the image becomes.
</p>
</dd>
<dt>‘<samp>backward</samp>’</dt>
<dd><p>The larger the distance from the central point, the brighter the image becomes.
This can be used to reverse a vignette effect, though there is no automatic
detection to extract the lens ‘<samp>angle</samp>’ and other settings (yet). It can
also be used to create a burning effect.
</p></dd>
</dl>
<p>Default value is ‘<samp>forward</samp>’.
</p>
</dd>
<dt>‘<samp>eval</samp>’</dt>
<dd><p>Set evaluation mode for the expressions (‘<samp>angle</samp>’, ‘<samp>x0</samp>’, ‘<samp>y0</samp>’).
</p>
<p>It accepts the following values:
</p><dl compact="compact">
<dt>‘<samp>init</samp>’</dt>
<dd><p>Evaluate expressions only once during the filter initialization.
</p>
</dd>
<dt>‘<samp>frame</samp>’</dt>
<dd><p>Evaluate expressions for each incoming frame. This is way slower than the
‘<samp>init</samp>’ mode since it requires all the scalers to be re-computed, but it
allows advanced dynamic expressions.
</p></dd>
</dl>
<p>Default value is ‘<samp>init</samp>’.
</p>
</dd>
<dt>‘<samp>dither</samp>’</dt>
<dd><p>Set dithering to reduce the circular banding effects. Default is <code>1</code>
(enabled).
</p>
</dd>
<dt>‘<samp>aspect</samp>’</dt>
<dd><p>Set vignette aspect. This setting allows one to adjust the shape of the vignette.
Setting this value to the SAR of the input will make a rectangular vignetting
following the dimensions of the video.
</p>
<p>Default is <code>1/1</code>.
</p></dd>
</dl>
<a name="Expressions"></a>
<h3 class="subsection"><a href="ffmpeg-filters.html#toc-Expressions">9.98.1 Expressions</a></h3>
<p>The ‘<samp>alpha</samp>’, ‘<samp>x0</samp>’ and ‘<samp>y0</samp>’ expressions can contain the
following parameters.
</p>
<dl compact="compact">
<dt>‘<samp>w</samp>’</dt>
<dt>‘<samp>h</samp>’</dt>
<dd><p>input width and height
</p>
</dd>
<dt>‘<samp>n</samp>’</dt>
<dd><p>the number of input frame, starting from 0
</p>
</dd>
<dt>‘<samp>pts</samp>’</dt>
<dd><p>the PTS (Presentation TimeStamp) time of the filtered video frame, expressed in
<var>TB</var> units, NAN if undefined
</p>
</dd>
<dt>‘<samp>r</samp>’</dt>
<dd><p>frame rate of the input video, NAN if the input frame rate is unknown
</p>
</dd>
<dt>‘<samp>t</samp>’</dt>
<dd><p>the PTS (Presentation TimeStamp) of the filtered video frame,
expressed in seconds, NAN if undefined
</p>
</dd>
<dt>‘<samp>tb</samp>’</dt>
<dd><p>time base of the input video
</p></dd>
</dl>
<a name="Examples-42"></a>
<h3 class="subsection"><a href="ffmpeg-filters.html#toc-Examples-42">9.98.2 Examples</a></h3>
<ul>
<li>
Apply simple strong vignetting effect:
<div class="example">
<pre class="example">vignette=PI/4
</pre></div>
</li><li>
Make a flickering vignetting:
<div class="example">
<pre class="example">vignette='PI/4+random(1)*PI/50':eval=frame
</pre></div>
</li></ul>
<a name="w3fdif"></a>
<h2 class="section"><a href="ffmpeg-filters.html#toc-w3fdif">9.99 w3fdif</a></h2>
<p>Deinterlace the input video ("w3fdif" stands for "Weston 3 Field
Deinterlacing Filter").
</p>
<p>Based on the process described by Martin Weston for BBC R&D, and
implemented based on the de-interlace algorithm written by Jim
Easterbrook for BBC R&D, the Weston 3 field deinterlacing filter
uses filter coefficients calculated by BBC R&D.
</p>
<p>There are two sets of filter coefficients, so called "simple":
and "complex". Which set of filter coefficients is used can
be set by passing an optional parameter:
</p>
<dl compact="compact">
<dt>‘<samp>filter</samp>’</dt>
<dd><p>Set the interlacing filter coefficients. Accepts one of the following values:
</p>
<dl compact="compact">
<dt>‘<samp>simple</samp>’</dt>
<dd><p>Simple filter coefficient set.
</p></dd>
<dt>‘<samp>complex</samp>’</dt>
<dd><p>More-complex filter coefficient set.
</p></dd>
</dl>
<p>Default value is ‘<samp>complex</samp>’.
</p>
</dd>
<dt>‘<samp>deint</samp>’</dt>
<dd><p>Specify which frames to deinterlace. Accept one of the following values:
</p>
<dl compact="compact">
<dt>‘<samp>all</samp>’</dt>
<dd><p>Deinterlace all frames,
</p></dd>
<dt>‘<samp>interlaced</samp>’</dt>
<dd><p>Only deinterlace frames marked as interlaced.
</p></dd>
</dl>
<p>Default value is ‘<samp>all</samp>’.
</p></dd>
</dl>
<p><a name="yadif"></a>
</p><a name="yadif-1"></a>
<h2 class="section"><a href="ffmpeg-filters.html#toc-yadif-1">9.100 yadif</a></h2>
<p>Deinterlace the input video ("yadif" means "yet another deinterlacing
filter").
</p>
<p>It accepts the following parameters:
</p>
<dl compact="compact">
<dt>‘<samp>mode</samp>’</dt>
<dd><p>The interlacing mode to adopt. It accepts one of the following values:
</p>
<dl compact="compact">
<dt>‘<samp>0, send_frame</samp>’</dt>
<dd><p>Output one frame for each frame.
</p></dd>
<dt>‘<samp>1, send_field</samp>’</dt>
<dd><p>Output one frame for each field.
</p></dd>
<dt>‘<samp>2, send_frame_nospatial</samp>’</dt>
<dd><p>Like <code>send_frame</code>, but it skips the spatial interlacing check.
</p></dd>
<dt>‘<samp>3, send_field_nospatial</samp>’</dt>
<dd><p>Like <code>send_field</code>, but it skips the spatial interlacing check.
</p></dd>
</dl>
<p>The default value is <code>send_frame</code>.
</p>
</dd>
<dt>‘<samp>parity</samp>’</dt>
<dd><p>The picture field parity assumed for the input interlaced video. It accepts one
of the following values:
</p>
<dl compact="compact">
<dt>‘<samp>0, tff</samp>’</dt>
<dd><p>Assume the top field is first.
</p></dd>
<dt>‘<samp>1, bff</samp>’</dt>
<dd><p>Assume the bottom field is first.
</p></dd>
<dt>‘<samp>-1, auto</samp>’</dt>
<dd><p>Enable automatic detection of field parity.
</p></dd>
</dl>
<p>The default value is <code>auto</code>.
If the interlacing is unknown or the decoder does not export this information,
top field first will be assumed.
</p>
</dd>
<dt>‘<samp>deint</samp>’</dt>
<dd><p>Specify which frames to deinterlace. Accept one of the following
values:
</p>
<dl compact="compact">
<dt>‘<samp>0, all</samp>’</dt>
<dd><p>Deinterlace all frames.
</p></dd>
<dt>‘<samp>1, interlaced</samp>’</dt>
<dd><p>Only deinterlace frames marked as interlaced.
</p></dd>
</dl>
<p>The default value is <code>all</code>.
</p></dd>
</dl>
<a name="zoompan"></a>
<h2 class="section"><a href="ffmpeg-filters.html#toc-zoompan">9.101 zoompan</a></h2>
<p>Apply Zoom & Pan effect.
</p>
<p>This filter accepts the following options:
</p>
<dl compact="compact">
<dt>‘<samp>zoom, z</samp>’</dt>
<dd><p>Set the zoom expression. Default is 1.
</p>
</dd>
<dt>‘<samp>x</samp>’</dt>
<dt>‘<samp>y</samp>’</dt>
<dd><p>Set the x and y expression. Default is 0.
</p>
</dd>
<dt>‘<samp>d</samp>’</dt>
<dd><p>Set the duration expression in number of frames.
This sets for how many number of frames effect will last for
single input image.
</p>
</dd>
<dt>‘<samp>s</samp>’</dt>
<dd><p>Set the output image size, default is ’hd720’.
</p></dd>
</dl>
<p>Each expression can contain the following constants:
</p>
<dl compact="compact">
<dt>‘<samp>in_w, iw</samp>’</dt>
<dd><p>Input width.
</p>
</dd>
<dt>‘<samp>in_h, ih</samp>’</dt>
<dd><p>Input height.
</p>
</dd>
<dt>‘<samp>out_w, ow</samp>’</dt>
<dd><p>Output width.
</p>
</dd>
<dt>‘<samp>out_h, oh</samp>’</dt>
<dd><p>Output height.
</p>
</dd>
<dt>‘<samp>in</samp>’</dt>
<dd><p>Input frame count.
</p>
</dd>
<dt>‘<samp>on</samp>’</dt>
<dd><p>Output frame count.
</p>
</dd>
<dt>‘<samp>x</samp>’</dt>
<dt>‘<samp>y</samp>’</dt>
<dd><p>Last calculated ’x’ and ’y’ position from ’x’ and ’y’ expression
for current input frame.
</p>
</dd>
<dt>‘<samp>px</samp>’</dt>
<dt>‘<samp>py</samp>’</dt>
<dd><p>’x’ and ’y’ of last output frame of previous input frame or 0 when there was
not yet such frame (first input frame).
</p>
</dd>
<dt>‘<samp>zoom</samp>’</dt>
<dd><p>Last calculated zoom from ’z’ expression for current input frame.
</p>
</dd>
<dt>‘<samp>pzoom</samp>’</dt>
<dd><p>Last calculated zoom of last output frame of previous input frame.
</p>
</dd>
<dt>‘<samp>duration</samp>’</dt>
<dd><p>Number of output frames for current input frame. Calculated from ’d’ expression
for each input frame.
</p>
</dd>
<dt>‘<samp>pduration</samp>’</dt>
<dd><p>number of output frames created for previous input frame
</p>
</dd>
<dt>‘<samp>a</samp>’</dt>
<dd><p>Rational number: input width / input height
</p>
</dd>
<dt>‘<samp>sar</samp>’</dt>
<dd><p>sample aspect ratio
</p>
</dd>
<dt>‘<samp>dar</samp>’</dt>
<dd><p>display aspect ratio
</p>
</dd>
</dl>
<a name="Examples-13"></a>
<h3 class="subsection"><a href="ffmpeg-filters.html#toc-Examples-13">9.101.1 Examples</a></h3>
<ul>
<li>
Zoom-in up to 1.5 and pan at same time to some spot near center of picture:
<div class="example">
<pre class="example">zoompan=z='min(zoom+0.0015,1.5)':d=700:x='if(gte(zoom,1.5),x,x+1/a)':y='if(gte(zoom,1.5),y,y+1)':s=640x360
</pre></div>
</li></ul>
<a name="Video-Sources"></a>
<h1 class="chapter"><a href="ffmpeg-filters.html#toc-Video-Sources">10 Video Sources</a></h1>
<p>Below is a description of the currently available video sources.
</p>
<a name="buffer"></a>
<h2 class="section"><a href="ffmpeg-filters.html#toc-buffer">10.1 buffer</a></h2>
<p>Buffer video frames, and make them available to the filter chain.
</p>
<p>This source is mainly intended for a programmatic use, in particular
through the interface defined in ‘<tt>libavfilter/vsrc_buffer.h</tt>’.
</p>
<p>It accepts the following parameters:
</p>
<dl compact="compact">
<dt>‘<samp>video_size</samp>’</dt>
<dd><p>Specify the size (width and height) of the buffered video frames. For the
syntax of this option, check the "Video size" section in the ffmpeg-utils
manual.
</p>
</dd>
<dt>‘<samp>width</samp>’</dt>
<dd><p>The input video width.
</p>
</dd>
<dt>‘<samp>height</samp>’</dt>
<dd><p>The input video height.
</p>
</dd>
<dt>‘<samp>pix_fmt</samp>’</dt>
<dd><p>A string representing the pixel format of the buffered video frames.
It may be a number corresponding to a pixel format, or a pixel format
name.
</p>
</dd>
<dt>‘<samp>time_base</samp>’</dt>
<dd><p>Specify the timebase assumed by the timestamps of the buffered frames.
</p>
</dd>
<dt>‘<samp>frame_rate</samp>’</dt>
<dd><p>Specify the frame rate expected for the video stream.
</p>
</dd>
<dt>‘<samp>pixel_aspect, sar</samp>’</dt>
<dd><p>The sample (pixel) aspect ratio of the input video.
</p>
</dd>
<dt>‘<samp>sws_param</samp>’</dt>
<dd><p>Specify the optional parameters to be used for the scale filter which
is automatically inserted when an input change is detected in the
input size or format.
</p></dd>
</dl>
<p>For example:
</p><div class="example">
<pre class="example">buffer=width=320:height=240:pix_fmt=yuv410p:time_base=1/24:sar=1
</pre></div>
<p>will instruct the source to accept video frames with size 320x240 and
with format "yuv410p", assuming 1/24 as the timestamps timebase and
square pixels (1:1 sample aspect ratio).
Since the pixel format with name "yuv410p" corresponds to the number 6
(check the enum AVPixelFormat definition in ‘<tt>libavutil/pixfmt.h</tt>’),
this example corresponds to:
</p><div class="example">
<pre class="example">buffer=size=320x240:pixfmt=6:time_base=1/24:pixel_aspect=1/1
</pre></div>
<p>Alternatively, the options can be specified as a flat string, but this
syntax is deprecated:
</p>
<p><var>width</var>:<var>height</var>:<var>pix_fmt</var>:<var>time_base.num</var>:<var>time_base.den</var>:<var>pixel_aspect.num</var>:<var>pixel_aspect.den</var>[:<var>sws_param</var>]
</p>
<a name="cellauto"></a>
<h2 class="section"><a href="ffmpeg-filters.html#toc-cellauto">10.2 cellauto</a></h2>
<p>Create a pattern generated by an elementary cellular automaton.
</p>
<p>The initial state of the cellular automaton can be defined through the
‘<samp>filename</samp>’, and ‘<samp>pattern</samp>’ options. If such options are
not specified an initial state is created randomly.
</p>
<p>At each new frame a new row in the video is filled with the result of
the cellular automaton next generation. The behavior when the whole
frame is filled is defined by the ‘<samp>scroll</samp>’ option.
</p>
<p>This source accepts the following options:
</p>
<dl compact="compact">
<dt>‘<samp>filename, f</samp>’</dt>
<dd><p>Read the initial cellular automaton state, i.e. the starting row, from
the specified file.
In the file, each non-whitespace character is considered an alive
cell, a newline will terminate the row, and further characters in the
file will be ignored.
</p>
</dd>
<dt>‘<samp>pattern, p</samp>’</dt>
<dd><p>Read the initial cellular automaton state, i.e. the starting row, from
the specified string.
</p>
<p>Each non-whitespace character in the string is considered an alive
cell, a newline will terminate the row, and further characters in the
string will be ignored.
</p>
</dd>
<dt>‘<samp>rate, r</samp>’</dt>
<dd><p>Set the video rate, that is the number of frames generated per second.
Default is 25.
</p>
</dd>
<dt>‘<samp>random_fill_ratio, ratio</samp>’</dt>
<dd><p>Set the random fill ratio for the initial cellular automaton row. It
is a floating point number value ranging from 0 to 1, defaults to
1/PHI.
</p>
<p>This option is ignored when a file or a pattern is specified.
</p>
</dd>
<dt>‘<samp>random_seed, seed</samp>’</dt>
<dd><p>Set the seed for filling randomly the initial row, must be an integer
included between 0 and UINT32_MAX. If not specified, or if explicitly
set to -1, the filter will try to use a good random seed on a best
effort basis.
</p>
</dd>
<dt>‘<samp>rule</samp>’</dt>
<dd><p>Set the cellular automaton rule, it is a number ranging from 0 to 255.
Default value is 110.
</p>
</dd>
<dt>‘<samp>size, s</samp>’</dt>
<dd><p>Set the size of the output video. For the syntax of this option, check
the "Video size" section in the ffmpeg-utils manual.
</p>
<p>If ‘<samp>filename</samp>’ or ‘<samp>pattern</samp>’ is specified, the size is set
by default to the width of the specified initial state row, and the
height is set to <var>width</var> * PHI.
</p>
<p>If ‘<samp>size</samp>’ is set, it must contain the width of the specified
pattern string, and the specified pattern will be centered in the
larger row.
</p>
<p>If a filename or a pattern string is not specified, the size value
defaults to "320x518" (used for a randomly generated initial state).
</p>
</dd>
<dt>‘<samp>scroll</samp>’</dt>
<dd><p>If set to 1, scroll the output upward when all the rows in the output
have been already filled. If set to 0, the new generated row will be
written over the top row just after the bottom row is filled.
Defaults to 1.
</p>
</dd>
<dt>‘<samp>start_full, full</samp>’</dt>
<dd><p>If set to 1, completely fill the output with generated rows before
outputting the first frame.
This is the default behavior, for disabling set the value to 0.
</p>
</dd>
<dt>‘<samp>stitch</samp>’</dt>
<dd><p>If set to 1, stitch the left and right row edges together.
This is the default behavior, for disabling set the value to 0.
</p></dd>
</dl>
<a name="Examples-43"></a>
<h3 class="subsection"><a href="ffmpeg-filters.html#toc-Examples-43">10.2.1 Examples</a></h3>
<ul>
<li>
Read the initial state from ‘<tt>pattern</tt>’, and specify an output of
size 200x400.
<div class="example">
<pre class="example">cellauto=f=pattern:s=200x400
</pre></div>
</li><li>
Generate a random initial row with a width of 200 cells, with a fill
ratio of 2/3:
<div class="example">
<pre class="example">cellauto=ratio=2/3:s=200x200
</pre></div>
</li><li>
Create a pattern generated by rule 18 starting by a single alive cell
centered on an initial row with width 100:
<div class="example">
<pre class="example">cellauto=p=@:s=100x400:full=0:rule=18
</pre></div>
</li><li>
Specify a more elaborated initial pattern:
<div class="example">
<pre class="example">cellauto=p='@@ @ @@':s=100x400:full=0:rule=18
</pre></div>
</li></ul>
<a name="mandelbrot"></a>
<h2 class="section"><a href="ffmpeg-filters.html#toc-mandelbrot">10.3 mandelbrot</a></h2>
<p>Generate a Mandelbrot set fractal, and progressively zoom towards the
point specified with <var>start_x</var> and <var>start_y</var>.
</p>
<p>This source accepts the following options:
</p>
<dl compact="compact">
<dt>‘<samp>end_pts</samp>’</dt>
<dd><p>Set the terminal pts value. Default value is 400.
</p>
</dd>
<dt>‘<samp>end_scale</samp>’</dt>
<dd><p>Set the terminal scale value.
Must be a floating point value. Default value is 0.3.
</p>
</dd>
<dt>‘<samp>inner</samp>’</dt>
<dd><p>Set the inner coloring mode, that is the algorithm used to draw the
Mandelbrot fractal internal region.
</p>
<p>It shall assume one of the following values:
</p><dl compact="compact">
<dt>‘<samp>black</samp>’</dt>
<dd><p>Set black mode.
</p></dd>
<dt>‘<samp>convergence</samp>’</dt>
<dd><p>Show time until convergence.
</p></dd>
<dt>‘<samp>mincol</samp>’</dt>
<dd><p>Set color based on point closest to the origin of the iterations.
</p></dd>
<dt>‘<samp>period</samp>’</dt>
<dd><p>Set period mode.
</p></dd>
</dl>
<p>Default value is <var>mincol</var>.
</p>
</dd>
<dt>‘<samp>bailout</samp>’</dt>
<dd><p>Set the bailout value. Default value is 10.0.
</p>
</dd>
<dt>‘<samp>maxiter</samp>’</dt>
<dd><p>Set the maximum of iterations performed by the rendering
algorithm. Default value is 7189.
</p>
</dd>
<dt>‘<samp>outer</samp>’</dt>
<dd><p>Set outer coloring mode.
It shall assume one of following values:
</p><dl compact="compact">
<dt>‘<samp>iteration_count</samp>’</dt>
<dd><p>Set iteration cound mode.
</p></dd>
<dt>‘<samp>normalized_iteration_count</samp>’</dt>
<dd><p>set normalized iteration count mode.
</p></dd>
</dl>
<p>Default value is <var>normalized_iteration_count</var>.
</p>
</dd>
<dt>‘<samp>rate, r</samp>’</dt>
<dd><p>Set frame rate, expressed as number of frames per second. Default
value is "25".
</p>
</dd>
<dt>‘<samp>size, s</samp>’</dt>
<dd><p>Set frame size. For the syntax of this option, check the "Video
size" section in the ffmpeg-utils manual. Default value is "640x480".
</p>
</dd>
<dt>‘<samp>start_scale</samp>’</dt>
<dd><p>Set the initial scale value. Default value is 3.0.
</p>
</dd>
<dt>‘<samp>start_x</samp>’</dt>
<dd><p>Set the initial x position. Must be a floating point value between
-100 and 100. Default value is -0.743643887037158704752191506114774.
</p>
</dd>
<dt>‘<samp>start_y</samp>’</dt>
<dd><p>Set the initial y position. Must be a floating point value between
-100 and 100. Default value is -0.131825904205311970493132056385139.
</p></dd>
</dl>
<a name="mptestsrc"></a>
<h2 class="section"><a href="ffmpeg-filters.html#toc-mptestsrc">10.4 mptestsrc</a></h2>
<p>Generate various test patterns, as generated by the MPlayer test filter.
</p>
<p>The size of the generated video is fixed, and is 256x256.
This source is useful in particular for testing encoding features.
</p>
<p>This source accepts the following options:
</p>
<dl compact="compact">
<dt>‘<samp>rate, r</samp>’</dt>
<dd><p>Specify the frame rate of the sourced video, as the number of frames
generated per second. It has to be a string in the format
<var>frame_rate_num</var>/<var>frame_rate_den</var>, an integer number, a floating point
number or a valid video frame rate abbreviation. The default value is
"25".
</p>
</dd>
<dt>‘<samp>duration, d</samp>’</dt>
<dd><p>Set the duration of the sourced video. See
<a href="ffmpeg-utils.html#time-duration-syntax">(ffmpeg-utils)the Time duration section in the ffmpeg-utils(1) manual</a>
for the accepted syntax.
</p>
<p>If not specified, or the expressed duration is negative, the video is
supposed to be generated forever.
</p>
</dd>
<dt>‘<samp>test, t</samp>’</dt>
<dd>
<p>Set the number or the name of the test to perform. Supported tests are:
</p><dl compact="compact">
<dt>‘<samp>dc_luma</samp>’</dt>
<dt>‘<samp>dc_chroma</samp>’</dt>
<dt>‘<samp>freq_luma</samp>’</dt>
<dt>‘<samp>freq_chroma</samp>’</dt>
<dt>‘<samp>amp_luma</samp>’</dt>
<dt>‘<samp>amp_chroma</samp>’</dt>
<dt>‘<samp>cbp</samp>’</dt>
<dt>‘<samp>mv</samp>’</dt>
<dt>‘<samp>ring1</samp>’</dt>
<dt>‘<samp>ring2</samp>’</dt>
<dt>‘<samp>all</samp>’</dt>
</dl>
<p>Default value is "all", which will cycle through the list of all tests.
</p></dd>
</dl>
<p>Some examples:
</p><div class="example">
<pre class="example">mptestsrc=t=dc_luma
</pre></div>
<p>will generate a "dc_luma" test pattern.
</p>
<a name="frei0r_005fsrc"></a>
<h2 class="section"><a href="ffmpeg-filters.html#toc-frei0r_005fsrc">10.5 frei0r_src</a></h2>
<p>Provide a frei0r source.
</p>
<p>To enable compilation of this filter you need to install the frei0r
header and configure FFmpeg with <code>--enable-frei0r</code>.
</p>
<p>This source accepts the following parameters:
</p>
<dl compact="compact">
<dt>‘<samp>size</samp>’</dt>
<dd><p>The size of the video to generate. For the syntax of this option, check the
"Video size" section in the ffmpeg-utils manual.
</p>
</dd>
<dt>‘<samp>framerate</samp>’</dt>
<dd><p>The framerate of the generated video. It may be a string of the form
<var>num</var>/<var>den</var> or a frame rate abbreviation.
</p>
</dd>
<dt>‘<samp>filter_name</samp>’</dt>
<dd><p>The name to the frei0r source to load. For more information regarding frei0r and
how to set the parameters, read the <a href="#frei0r">frei0r</a> section in the video filters
documentation.
</p>
</dd>
<dt>‘<samp>filter_params</samp>’</dt>
<dd><p>A ’|’-separated list of parameters to pass to the frei0r source.
</p>
</dd>
</dl>
<p>For example, to generate a frei0r partik0l source with size 200x200
and frame rate 10 which is overlayed on the overlay filter main input:
</p><div class="example">
<pre class="example">frei0r_src=size=200x200:framerate=10:filter_name=partik0l:filter_params=1234 [overlay]; [in][overlay] overlay
</pre></div>
<a name="life"></a>
<h2 class="section"><a href="ffmpeg-filters.html#toc-life">10.6 life</a></h2>
<p>Generate a life pattern.
</p>
<p>This source is based on a generalization of John Conway’s life game.
</p>
<p>The sourced input represents a life grid, each pixel represents a cell
which can be in one of two possible states, alive or dead. Every cell
interacts with its eight neighbours, which are the cells that are
horizontally, vertically, or diagonally adjacent.
</p>
<p>At each interaction the grid evolves according to the adopted rule,
which specifies the number of neighbor alive cells which will make a
cell stay alive or born. The ‘<samp>rule</samp>’ option allows one to specify
the rule to adopt.
</p>
<p>This source accepts the following options:
</p>
<dl compact="compact">
<dt>‘<samp>filename, f</samp>’</dt>
<dd><p>Set the file from which to read the initial grid state. In the file,
each non-whitespace character is considered an alive cell, and newline
is used to delimit the end of each row.
</p>
<p>If this option is not specified, the initial grid is generated
randomly.
</p>
</dd>
<dt>‘<samp>rate, r</samp>’</dt>
<dd><p>Set the video rate, that is the number of frames generated per second.
Default is 25.
</p>
</dd>
<dt>‘<samp>random_fill_ratio, ratio</samp>’</dt>
<dd><p>Set the random fill ratio for the initial random grid. It is a
floating point number value ranging from 0 to 1, defaults to 1/PHI.
It is ignored when a file is specified.
</p>
</dd>
<dt>‘<samp>random_seed, seed</samp>’</dt>
<dd><p>Set the seed for filling the initial random grid, must be an integer
included between 0 and UINT32_MAX. If not specified, or if explicitly
set to -1, the filter will try to use a good random seed on a best
effort basis.
</p>
</dd>
<dt>‘<samp>rule</samp>’</dt>
<dd><p>Set the life rule.
</p>
<p>A rule can be specified with a code of the kind "S<var>NS</var>/B<var>NB</var>",
where <var>NS</var> and <var>NB</var> are sequences of numbers in the range 0-8,
<var>NS</var> specifies the number of alive neighbor cells which make a
live cell stay alive, and <var>NB</var> the number of alive neighbor cells
which make a dead cell to become alive (i.e. to "born").
"s" and "b" can be used in place of "S" and "B", respectively.
</p>
<p>Alternatively a rule can be specified by an 18-bits integer. The 9
high order bits are used to encode the next cell state if it is alive
for each number of neighbor alive cells, the low order bits specify
the rule for "borning" new cells. Higher order bits encode for an
higher number of neighbor cells.
For example the number 6153 = <code>(12<<9)+9</code> specifies a stay alive
rule of 12 and a born rule of 9, which corresponds to "S23/B03".
</p>
<p>Default value is "S23/B3", which is the original Conway’s game of life
rule, and will keep a cell alive if it has 2 or 3 neighbor alive
cells, and will born a new cell if there are three alive cells around
a dead cell.
</p>
</dd>
<dt>‘<samp>size, s</samp>’</dt>
<dd><p>Set the size of the output video. For the syntax of this option, check the
"Video size" section in the ffmpeg-utils manual.
</p>
<p>If ‘<samp>filename</samp>’ is specified, the size is set by default to the
same size of the input file. If ‘<samp>size</samp>’ is set, it must contain
the size specified in the input file, and the initial grid defined in
that file is centered in the larger resulting area.
</p>
<p>If a filename is not specified, the size value defaults to "320x240"
(used for a randomly generated initial grid).
</p>
</dd>
<dt>‘<samp>stitch</samp>’</dt>
<dd><p>If set to 1, stitch the left and right grid edges together, and the
top and bottom edges also. Defaults to 1.
</p>
</dd>
<dt>‘<samp>mold</samp>’</dt>
<dd><p>Set cell mold speed. If set, a dead cell will go from ‘<samp>death_color</samp>’ to
‘<samp>mold_color</samp>’ with a step of ‘<samp>mold</samp>’. ‘<samp>mold</samp>’ can have a
value from 0 to 255.
</p>
</dd>
<dt>‘<samp>life_color</samp>’</dt>
<dd><p>Set the color of living (or new born) cells.
</p>
</dd>
<dt>‘<samp>death_color</samp>’</dt>
<dd><p>Set the color of dead cells. If ‘<samp>mold</samp>’ is set, this is the first color
used to represent a dead cell.
</p>
</dd>
<dt>‘<samp>mold_color</samp>’</dt>
<dd><p>Set mold color, for definitely dead and moldy cells.
</p>
<p>For the syntax of these 3 color options, check the "Color" section in the
ffmpeg-utils manual.
</p></dd>
</dl>
<a name="Examples-7"></a>
<h3 class="subsection"><a href="ffmpeg-filters.html#toc-Examples-7">10.6.1 Examples</a></h3>
<ul>
<li>
Read a grid from ‘<tt>pattern</tt>’, and center it on a grid of size
300x300 pixels:
<div class="example">
<pre class="example">life=f=pattern:s=300x300
</pre></div>
</li><li>
Generate a random grid of size 200x200, with a fill ratio of 2/3:
<div class="example">
<pre class="example">life=ratio=2/3:s=200x200
</pre></div>
</li><li>
Specify a custom rule for evolving a randomly generated grid:
<div class="example">
<pre class="example">life=rule=S14/B34
</pre></div>
</li><li>
Full example with slow death effect (mold) using <code>ffplay</code>:
<div class="example">
<pre class="example">ffplay -f lavfi life=s=300x200:mold=10:r=60:ratio=0.1:death_color=#C83232:life_color=#00ff00,scale=1200:800:flags=16
</pre></div>
</li></ul>
<p><a name="color"></a>
<a name="haldclutsrc"></a>
<a name="nullsrc"></a>
<a name="rgbtestsrc"></a>
<a name="smptebars"></a>
<a name="smptehdbars"></a>
<a name="testsrc"></a>
</p><a name="color_002c-haldclutsrc_002c-nullsrc_002c-rgbtestsrc_002c-smptebars_002c-smptehdbars_002c-testsrc"></a>
<h2 class="section"><a href="ffmpeg-filters.html#toc-color_002c-haldclutsrc_002c-nullsrc_002c-rgbtestsrc_002c-smptebars_002c-smptehdbars_002c-testsrc">10.7 color, haldclutsrc, nullsrc, rgbtestsrc, smptebars, smptehdbars, testsrc</a></h2>
<p>The <code>color</code> source provides an uniformly colored input.
</p>
<p>The <code>haldclutsrc</code> source provides an identity Hald CLUT. See also
<a href="#haldclut">haldclut</a> filter.
</p>
<p>The <code>nullsrc</code> source returns unprocessed video frames. It is
mainly useful to be employed in analysis / debugging tools, or as the
source for filters which ignore the input data.
</p>
<p>The <code>rgbtestsrc</code> source generates an RGB test pattern useful for
detecting RGB vs BGR issues. You should see a red, green and blue
stripe from top to bottom.
</p>
<p>The <code>smptebars</code> source generates a color bars pattern, based on
the SMPTE Engineering Guideline EG 1-1990.
</p>
<p>The <code>smptehdbars</code> source generates a color bars pattern, based on
the SMPTE RP 219-2002.
</p>
<p>The <code>testsrc</code> source generates a test video pattern, showing a
color pattern, a scrolling gradient and a timestamp. This is mainly
intended for testing purposes.
</p>
<p>The sources accept the following parameters:
</p>
<dl compact="compact">
<dt>‘<samp>color, c</samp>’</dt>
<dd><p>Specify the color of the source, only available in the <code>color</code>
source. For the syntax of this option, check the "Color" section in the
ffmpeg-utils manual.
</p>
</dd>
<dt>‘<samp>level</samp>’</dt>
<dd><p>Specify the level of the Hald CLUT, only available in the <code>haldclutsrc</code>
source. A level of <code>N</code> generates a picture of <code>N*N*N</code> by <code>N*N*N</code>
pixels to be used as identity matrix for 3D lookup tables. Each component is
coded on a <code>1/(N*N)</code> scale.
</p>
</dd>
<dt>‘<samp>size, s</samp>’</dt>
<dd><p>Specify the size of the sourced video. For the syntax of this option, check the
"Video size" section in the ffmpeg-utils manual. The default value is
"320x240".
</p>
<p>This option is not available with the <code>haldclutsrc</code> filter.
</p>
</dd>
<dt>‘<samp>rate, r</samp>’</dt>
<dd><p>Specify the frame rate of the sourced video, as the number of frames
generated per second. It has to be a string in the format
<var>frame_rate_num</var>/<var>frame_rate_den</var>, an integer number, a floating point
number or a valid video frame rate abbreviation. The default value is
"25".
</p>
</dd>
<dt>‘<samp>sar</samp>’</dt>
<dd><p>Set the sample aspect ratio of the sourced video.
</p>
</dd>
<dt>‘<samp>duration, d</samp>’</dt>
<dd><p>Set the duration of the sourced video. See
<a href="ffmpeg-utils.html#time-duration-syntax">(ffmpeg-utils)the Time duration section in the ffmpeg-utils(1) manual</a>
for the accepted syntax.
</p>
<p>If not specified, or the expressed duration is negative, the video is
supposed to be generated forever.
</p>
</dd>
<dt>‘<samp>decimals, n</samp>’</dt>
<dd><p>Set the number of decimals to show in the timestamp, only available in the
<code>testsrc</code> source.
</p>
<p>The displayed timestamp value will correspond to the original
timestamp value multiplied by the power of 10 of the specified
value. Default value is 0.
</p></dd>
</dl>
<p>For example the following:
</p><div class="example">
<pre class="example">testsrc=duration=5.3:size=qcif:rate=10
</pre></div>
<p>will generate a video with a duration of 5.3 seconds, with size
176x144 and a frame rate of 10 frames per second.
</p>
<p>The following graph description will generate a red source
with an opacity of 0.2, with size "qcif" and a frame rate of 10
frames per second.
</p><div class="example">
<pre class="example">color=c=red@0.2:s=qcif:r=10
</pre></div>
<p>If the input content is to be ignored, <code>nullsrc</code> can be used. The
following command generates noise in the luminance plane by employing
the <code>geq</code> filter:
</p><div class="example">
<pre class="example">nullsrc=s=256x256, geq=random(1)*255:128:128
</pre></div>
<a name="Commands-2"></a>
<h3 class="subsection"><a href="ffmpeg-filters.html#toc-Commands-2">10.7.1 Commands</a></h3>
<p>The <code>color</code> source supports the following commands:
</p>
<dl compact="compact">
<dt>‘<samp>c, color</samp>’</dt>
<dd><p>Set the color of the created image. Accepts the same syntax of the
corresponding ‘<samp>color</samp>’ option.
</p></dd>
</dl>
<a name="Video-Sinks"></a>
<h1 class="chapter"><a href="ffmpeg-filters.html#toc-Video-Sinks">11 Video Sinks</a></h1>
<p>Below is a description of the currently available video sinks.
</p>
<a name="buffersink"></a>
<h2 class="section"><a href="ffmpeg-filters.html#toc-buffersink">11.1 buffersink</a></h2>
<p>Buffer video frames, and make them available to the end of the filter
graph.
</p>
<p>This sink is mainly intended for programmatic use, in particular
through the interface defined in ‘<tt>libavfilter/buffersink.h</tt>’
or the options system.
</p>
<p>It accepts a pointer to an AVBufferSinkContext structure, which
defines the incoming buffers’ formats, to be passed as the opaque
parameter to <code>avfilter_init_filter</code> for initialization.
</p>
<a name="nullsink"></a>
<h2 class="section"><a href="ffmpeg-filters.html#toc-nullsink">11.2 nullsink</a></h2>
<p>Null video sink: do absolutely nothing with the input video. It is
mainly useful as a template and for use in analysis / debugging
tools.
</p>
<a name="Multimedia-Filters"></a>
<h1 class="chapter"><a href="ffmpeg-filters.html#toc-Multimedia-Filters">12 Multimedia Filters</a></h1>
<p>Below is a description of the currently available multimedia filters.
</p>
<a name="avectorscope"></a>
<h2 class="section"><a href="ffmpeg-filters.html#toc-avectorscope">12.1 avectorscope</a></h2>
<p>Convert input audio to a video output, representing the audio vector
scope.
</p>
<p>The filter is used to measure the difference between channels of stereo
audio stream. A monoaural signal, consisting of identical left and right
signal, results in straight vertical line. Any stereo separation is visible
as a deviation from this line, creating a Lissajous figure.
If the straight (or deviation from it) but horizontal line appears this
indicates that the left and right channels are out of phase.
</p>
<p>The filter accepts the following options:
</p>
<dl compact="compact">
<dt>‘<samp>mode, m</samp>’</dt>
<dd><p>Set the vectorscope mode.
</p>
<p>Available values are:
</p><dl compact="compact">
<dt>‘<samp>lissajous</samp>’</dt>
<dd><p>Lissajous rotated by 45 degrees.
</p>
</dd>
<dt>‘<samp>lissajous_xy</samp>’</dt>
<dd><p>Same as above but not rotated.
</p></dd>
</dl>
<p>Default value is ‘<samp>lissajous</samp>’.
</p>
</dd>
<dt>‘<samp>size, s</samp>’</dt>
<dd><p>Set the video size for the output. For the syntax of this option, check the "Video size"
section in the ffmpeg-utils manual. Default value is <code>400x400</code>.
</p>
</dd>
<dt>‘<samp>rate, r</samp>’</dt>
<dd><p>Set the output frame rate. Default value is <code>25</code>.
</p>
</dd>
<dt>‘<samp>rc</samp>’</dt>
<dt>‘<samp>gc</samp>’</dt>
<dt>‘<samp>bc</samp>’</dt>
<dd><p>Specify the red, green and blue contrast. Default values are <code>40</code>, <code>160</code> and <code>80</code>.
Allowed range is <code>[0, 255]</code>.
</p>
</dd>
<dt>‘<samp>rf</samp>’</dt>
<dt>‘<samp>gf</samp>’</dt>
<dt>‘<samp>bf</samp>’</dt>
<dd><p>Specify the red, green and blue fade. Default values are <code>15</code>, <code>10</code> and <code>5</code>.
Allowed range is <code>[0, 255]</code>.
</p>
</dd>
<dt>‘<samp>zoom</samp>’</dt>
<dd><p>Set the zoom factor. Default value is <code>1</code>. Allowed range is <code>[1, 10]</code>.
</p></dd>
</dl>
<a name="Examples-45"></a>
<h3 class="subsection"><a href="ffmpeg-filters.html#toc-Examples-45">12.1.1 Examples</a></h3>
<ul>
<li>
Complete example using <code>ffplay</code>:
<div class="example">
<pre class="example">ffplay -f lavfi 'amovie=input.mp3, asplit [a][out1];
[a] avectorscope=zoom=1.3:rc=2:gc=200:bc=10:rf=1:gf=8:bf=7 [out0]'
</pre></div>
</li></ul>
<a name="concat"></a>
<h2 class="section"><a href="ffmpeg-filters.html#toc-concat">12.2 concat</a></h2>
<p>Concatenate audio and video streams, joining them together one after the
other.
</p>
<p>The filter works on segments of synchronized video and audio streams. All
segments must have the same number of streams of each type, and that will
also be the number of streams at output.
</p>
<p>The filter accepts the following options:
</p>
<dl compact="compact">
<dt>‘<samp>n</samp>’</dt>
<dd><p>Set the number of segments. Default is 2.
</p>
</dd>
<dt>‘<samp>v</samp>’</dt>
<dd><p>Set the number of output video streams, that is also the number of video
streams in each segment. Default is 1.
</p>
</dd>
<dt>‘<samp>a</samp>’</dt>
<dd><p>Set the number of output audio streams, that is also the number of audio
streams in each segment. Default is 0.
</p>
</dd>
<dt>‘<samp>unsafe</samp>’</dt>
<dd><p>Activate unsafe mode: do not fail if segments have a different format.
</p>
</dd>
</dl>
<p>The filter has <var>v</var>+<var>a</var> outputs: first <var>v</var> video outputs, then
<var>a</var> audio outputs.
</p>
<p>There are <var>n</var>x(<var>v</var>+<var>a</var>) inputs: first the inputs for the first
segment, in the same order as the outputs, then the inputs for the second
segment, etc.
</p>
<p>Related streams do not always have exactly the same duration, for various
reasons including codec frame size or sloppy authoring. For that reason,
related synchronized streams (e.g. a video and its audio track) should be
concatenated at once. The concat filter will use the duration of the longest
stream in each segment (except the last one), and if necessary pad shorter
audio streams with silence.
</p>
<p>For this filter to work correctly, all segments must start at timestamp 0.
</p>
<p>All corresponding streams must have the same parameters in all segments; the
filtering system will automatically select a common pixel format for video
streams, and a common sample format, sample rate and channel layout for
audio streams, but other settings, such as resolution, must be converted
explicitly by the user.
</p>
<p>Different frame rates are acceptable but will result in variable frame rate
at output; be sure to configure the output file to handle it.
</p>
<a name="Examples-31"></a>
<h3 class="subsection"><a href="ffmpeg-filters.html#toc-Examples-31">12.2.1 Examples</a></h3>
<ul>
<li>
Concatenate an opening, an episode and an ending, all in bilingual version
(video in stream 0, audio in streams 1 and 2):
<div class="example">
<pre class="example">ffmpeg -i opening.mkv -i episode.mkv -i ending.mkv -filter_complex \
'[0:0] [0:1] [0:2] [1:0] [1:1] [1:2] [2:0] [2:1] [2:2]
concat=n=3:v=1:a=2 [v] [a1] [a2]' \
-map '[v]' -map '[a1]' -map '[a2]' output.mkv
</pre></div>
</li><li>
Concatenate two parts, handling audio and video separately, using the
(a)movie sources, and adjusting the resolution:
<div class="example">
<pre class="example">movie=part1.mp4, scale=512:288 [v1] ; amovie=part1.mp4 [a1] ;
movie=part2.mp4, scale=512:288 [v2] ; amovie=part2.mp4 [a2] ;
[v1] [v2] concat [outv] ; [a1] [a2] concat=v=0:a=1 [outa]
</pre></div>
<p>Note that a desync will happen at the stitch if the audio and video streams
do not have exactly the same duration in the first file.
</p>
</li></ul>
<a name="ebur128"></a>
<h2 class="section"><a href="ffmpeg-filters.html#toc-ebur128">12.3 ebur128</a></h2>
<p>EBU R128 scanner filter. This filter takes an audio stream as input and outputs
it unchanged. By default, it logs a message at a frequency of 10Hz with the
Momentary loudness (identified by <code>M</code>), Short-term loudness (<code>S</code>),
Integrated loudness (<code>I</code>) and Loudness Range (<code>LRA</code>).
</p>
<p>The filter also has a video output (see the <var>video</var> option) with a real
time graph to observe the loudness evolution. The graphic contains the logged
message mentioned above, so it is not printed anymore when this option is set,
unless the verbose logging is set. The main graphing area contains the
short-term loudness (3 seconds of analysis), and the gauge on the right is for
the momentary loudness (400 milliseconds).
</p>
<p>More information about the Loudness Recommendation EBU R128 on
<a href="http://tech.ebu.ch/loudness">http://tech.ebu.ch/loudness</a>.
</p>
<p>The filter accepts the following options:
</p>
<dl compact="compact">
<dt>‘<samp>video</samp>’</dt>
<dd><p>Activate the video output. The audio stream is passed unchanged whether this
option is set or no. The video stream will be the first output stream if
activated. Default is <code>0</code>.
</p>
</dd>
<dt>‘<samp>size</samp>’</dt>
<dd><p>Set the video size. This option is for video only. For the syntax of this
option, check the "Video size" section in the ffmpeg-utils manual. Default
and minimum resolution is <code>640x480</code>.
</p>
</dd>
<dt>‘<samp>meter</samp>’</dt>
<dd><p>Set the EBU scale meter. Default is <code>9</code>. Common values are <code>9</code> and
<code>18</code>, respectively for EBU scale meter +9 and EBU scale meter +18. Any
other integer value between this range is allowed.
</p>
</dd>
<dt>‘<samp>metadata</samp>’</dt>
<dd><p>Set metadata injection. If set to <code>1</code>, the audio input will be segmented
into 100ms output frames, each of them containing various loudness information
in metadata. All the metadata keys are prefixed with <code>lavfi.r128.</code>.
</p>
<p>Default is <code>0</code>.
</p>
</dd>
<dt>‘<samp>framelog</samp>’</dt>
<dd><p>Force the frame logging level.
</p>
<p>Available values are:
</p><dl compact="compact">
<dt>‘<samp>info</samp>’</dt>
<dd><p>information logging level
</p></dd>
<dt>‘<samp>verbose</samp>’</dt>
<dd><p>verbose logging level
</p></dd>
</dl>
<p>By default, the logging level is set to <var>info</var>. If the ‘<samp>video</samp>’ or
the ‘<samp>metadata</samp>’ options are set, it switches to <var>verbose</var>.
</p>
</dd>
<dt>‘<samp>peak</samp>’</dt>
<dd><p>Set peak mode(s).
</p>
<p>Available modes can be cumulated (the option is a <code>flag</code> type). Possible
values are:
</p><dl compact="compact">
<dt>‘<samp>none</samp>’</dt>
<dd><p>Disable any peak mode (default).
</p></dd>
<dt>‘<samp>sample</samp>’</dt>
<dd><p>Enable sample-peak mode.
</p>
<p>Simple peak mode looking for the higher sample value. It logs a message
for sample-peak (identified by <code>SPK</code>).
</p></dd>
<dt>‘<samp>true</samp>’</dt>
<dd><p>Enable true-peak mode.
</p>
<p>If enabled, the peak lookup is done on an over-sampled version of the input
stream for better peak accuracy. It logs a message for true-peak.
(identified by <code>TPK</code>) and true-peak per frame (identified by <code>FTPK</code>).
This mode requires a build with <code>libswresample</code>.
</p></dd>
</dl>
</dd>
</dl>
<a name="Examples-71"></a>
<h3 class="subsection"><a href="ffmpeg-filters.html#toc-Examples-71">12.3.1 Examples</a></h3>
<ul>
<li>
Real-time graph using <code>ffplay</code>, with a EBU scale meter +18:
<div class="example">
<pre class="example">ffplay -f lavfi -i "amovie=input.mp3,ebur128=video=1:meter=18 [out0][out1]"
</pre></div>
</li><li>
Run an analysis with <code>ffmpeg</code>:
<div class="example">
<pre class="example">ffmpeg -nostats -i input.mp3 -filter_complex ebur128 -f null -
</pre></div>
</li></ul>
<a name="interleave_002c-ainterleave"></a>
<h2 class="section"><a href="ffmpeg-filters.html#toc-interleave_002c-ainterleave">12.4 interleave, ainterleave</a></h2>
<p>Temporally interleave frames from several inputs.
</p>
<p><code>interleave</code> works with video inputs, <code>ainterleave</code> with audio.
</p>
<p>These filters read frames from several inputs and send the oldest
queued frame to the output.
</p>
<p>Input streams must have a well defined, monotonically increasing frame
timestamp values.
</p>
<p>In order to submit one frame to output, these filters need to enqueue
at least one frame for each input, so they cannot work in case one
input is not yet terminated and will not receive incoming frames.
</p>
<p>For example consider the case when one input is a <code>select</code> filter
which always drop input frames. The <code>interleave</code> filter will keep
reading from that input, but it will never be able to send new frames
to output until the input will send an end-of-stream signal.
</p>
<p>Also, depending on inputs synchronization, the filters will drop
frames in case one input receives more frames than the other ones, and
the queue is already filled.
</p>
<p>These filters accept the following options:
</p>
<dl compact="compact">
<dt>‘<samp>nb_inputs, n</samp>’</dt>
<dd><p>Set the number of different inputs, it is 2 by default.
</p></dd>
</dl>
<a name="Examples-49"></a>
<h3 class="subsection"><a href="ffmpeg-filters.html#toc-Examples-49">12.4.1 Examples</a></h3>
<ul>
<li>
Interleave frames belonging to different streams using <code>ffmpeg</code>:
<div class="example">
<pre class="example">ffmpeg -i bambi.avi -i pr0n.mkv -filter_complex "[0:v][1:v] interleave" out.avi
</pre></div>
</li><li>
Add flickering blur effect:
<div class="example">
<pre class="example">select='if(gt(random(0), 0.2), 1, 2)':n=2 [tmp], boxblur=2:2, [tmp] interleave
</pre></div>
</li></ul>
<a name="perms_002c-aperms"></a>
<h2 class="section"><a href="ffmpeg-filters.html#toc-perms_002c-aperms">12.5 perms, aperms</a></h2>
<p>Set read/write permissions for the output frames.
</p>
<p>These filters are mainly aimed at developers to test direct path in the
following filter in the filtergraph.
</p>
<p>The filters accept the following options:
</p>
<dl compact="compact">
<dt>‘<samp>mode</samp>’</dt>
<dd><p>Select the permissions mode.
</p>
<p>It accepts the following values:
</p><dl compact="compact">
<dt>‘<samp>none</samp>’</dt>
<dd><p>Do nothing. This is the default.
</p></dd>
<dt>‘<samp>ro</samp>’</dt>
<dd><p>Set all the output frames read-only.
</p></dd>
<dt>‘<samp>rw</samp>’</dt>
<dd><p>Set all the output frames directly writable.
</p></dd>
<dt>‘<samp>toggle</samp>’</dt>
<dd><p>Make the frame read-only if writable, and writable if read-only.
</p></dd>
<dt>‘<samp>random</samp>’</dt>
<dd><p>Set each output frame read-only or writable randomly.
</p></dd>
</dl>
</dd>
<dt>‘<samp>seed</samp>’</dt>
<dd><p>Set the seed for the <var>random</var> mode, must be an integer included between
<code>0</code> and <code>UINT32_MAX</code>. If not specified, or if explicitly set to
<code>-1</code>, the filter will try to use a good random seed on a best effort
basis.
</p></dd>
</dl>
<p>Note: in case of auto-inserted filter between the permission filter and the
following one, the permission might not be received as expected in that
following filter. Inserting a <a href="#format">format</a> or <a href="#aformat">aformat</a> filter before the
perms/aperms filter can avoid this problem.
</p>
<a name="select_002c-aselect"></a>
<h2 class="section"><a href="ffmpeg-filters.html#toc-select_002c-aselect">12.6 select, aselect</a></h2>
<p>Select frames to pass in output.
</p>
<p>This filter accepts the following options:
</p>
<dl compact="compact">
<dt>‘<samp>expr, e</samp>’</dt>
<dd><p>Set expression, which is evaluated for each input frame.
</p>
<p>If the expression is evaluated to zero, the frame is discarded.
</p>
<p>If the evaluation result is negative or NaN, the frame is sent to the
first output; otherwise it is sent to the output with index
<code>ceil(val)-1</code>, assuming that the input index starts from 0.
</p>
<p>For example a value of <code>1.2</code> corresponds to the output with index
<code>ceil(1.2)-1 = 2-1 = 1</code>, that is the second output.
</p>
</dd>
<dt>‘<samp>outputs, n</samp>’</dt>
<dd><p>Set the number of outputs. The output to which to send the selected
frame is based on the result of the evaluation. Default value is 1.
</p></dd>
</dl>
<p>The expression can contain the following constants:
</p>
<dl compact="compact">
<dt>‘<samp>n</samp>’</dt>
<dd><p>The (sequential) number of the filtered frame, starting from 0.
</p>
</dd>
<dt>‘<samp>selected_n</samp>’</dt>
<dd><p>The (sequential) number of the selected frame, starting from 0.
</p>
</dd>
<dt>‘<samp>prev_selected_n</samp>’</dt>
<dd><p>The sequential number of the last selected frame. It’s NAN if undefined.
</p>
</dd>
<dt>‘<samp>TB</samp>’</dt>
<dd><p>The timebase of the input timestamps.
</p>
</dd>
<dt>‘<samp>pts</samp>’</dt>
<dd><p>The PTS (Presentation TimeStamp) of the filtered video frame,
expressed in <var>TB</var> units. It’s NAN if undefined.
</p>
</dd>
<dt>‘<samp>t</samp>’</dt>
<dd><p>The PTS of the filtered video frame,
expressed in seconds. It’s NAN if undefined.
</p>
</dd>
<dt>‘<samp>prev_pts</samp>’</dt>
<dd><p>The PTS of the previously filtered video frame. It’s NAN if undefined.
</p>
</dd>
<dt>‘<samp>prev_selected_pts</samp>’</dt>
<dd><p>The PTS of the last previously filtered video frame. It’s NAN if undefined.
</p>
</dd>
<dt>‘<samp>prev_selected_t</samp>’</dt>
<dd><p>The PTS of the last previously selected video frame. It’s NAN if undefined.
</p>
</dd>
<dt>‘<samp>start_pts</samp>’</dt>
<dd><p>The PTS of the first video frame in the video. It’s NAN if undefined.
</p>
</dd>
<dt>‘<samp>start_t</samp>’</dt>
<dd><p>The time of the first video frame in the video. It’s NAN if undefined.
</p>
</dd>
<dt>‘<samp>pict_type <em>(video only)</em></samp>’</dt>
<dd><p>The type of the filtered frame. It can assume one of the following
values:
</p><dl compact="compact">
<dt>‘<samp>I</samp>’</dt>
<dt>‘<samp>P</samp>’</dt>
<dt>‘<samp>B</samp>’</dt>
<dt>‘<samp>S</samp>’</dt>
<dt>‘<samp>SI</samp>’</dt>
<dt>‘<samp>SP</samp>’</dt>
<dt>‘<samp>BI</samp>’</dt>
</dl>
</dd>
<dt>‘<samp>interlace_type <em>(video only)</em></samp>’</dt>
<dd><p>The frame interlace type. It can assume one of the following values:
</p><dl compact="compact">
<dt>‘<samp>PROGRESSIVE</samp>’</dt>
<dd><p>The frame is progressive (not interlaced).
</p></dd>
<dt>‘<samp>TOPFIRST</samp>’</dt>
<dd><p>The frame is top-field-first.
</p></dd>
<dt>‘<samp>BOTTOMFIRST</samp>’</dt>
<dd><p>The frame is bottom-field-first.
</p></dd>
</dl>
</dd>
<dt>‘<samp>consumed_sample_n <em>(audio only)</em></samp>’</dt>
<dd><p>the number of selected samples before the current frame
</p>
</dd>
<dt>‘<samp>samples_n <em>(audio only)</em></samp>’</dt>
<dd><p>the number of samples in the current frame
</p>
</dd>
<dt>‘<samp>sample_rate <em>(audio only)</em></samp>’</dt>
<dd><p>the input sample rate
</p>
</dd>
<dt>‘<samp>key</samp>’</dt>
<dd><p>This is 1 if the filtered frame is a key-frame, 0 otherwise.
</p>
</dd>
<dt>‘<samp>pos</samp>’</dt>
<dd><p>the position in the file of the filtered frame, -1 if the information
is not available (e.g. for synthetic video)
</p>
</dd>
<dt>‘<samp>scene <em>(video only)</em></samp>’</dt>
<dd><p>value between 0 and 1 to indicate a new scene; a low value reflects a low
probability for the current frame to introduce a new scene, while a higher
value means the current frame is more likely to be one (see the example below)
</p>
</dd>
</dl>
<p>The default value of the select expression is "1".
</p>
<a name="Examples-1"></a>
<h3 class="subsection"><a href="ffmpeg-filters.html#toc-Examples-1">12.6.1 Examples</a></h3>
<ul>
<li>
Select all frames in input:
<div class="example">
<pre class="example">select
</pre></div>
<p>The example above is the same as:
</p><div class="example">
<pre class="example">select=1
</pre></div>
</li><li>
Skip all frames:
<div class="example">
<pre class="example">select=0
</pre></div>
</li><li>
Select only I-frames:
<div class="example">
<pre class="example">select='eq(pict_type\,I)'
</pre></div>
</li><li>
Select one frame every 100:
<div class="example">
<pre class="example">select='not(mod(n\,100))'
</pre></div>
</li><li>
Select only frames contained in the 10-20 time interval:
<div class="example">
<pre class="example">select=between(t\,10\,20)
</pre></div>
</li><li>
Select only I frames contained in the 10-20 time interval:
<div class="example">
<pre class="example">select=between(t\,10\,20)*eq(pict_type\,I)
</pre></div>
</li><li>
Select frames with a minimum distance of 10 seconds:
<div class="example">
<pre class="example">select='isnan(prev_selected_t)+gte(t-prev_selected_t\,10)'
</pre></div>
</li><li>
Use aselect to select only audio frames with samples number > 100:
<div class="example">
<pre class="example">aselect='gt(samples_n\,100)'
</pre></div>
</li><li>
Create a mosaic of the first scenes:
<div class="example">
<pre class="example">ffmpeg -i video.avi -vf select='gt(scene\,0.4)',scale=160:120,tile -frames:v 1 preview.png
</pre></div>
<p>Comparing <var>scene</var> against a value between 0.3 and 0.5 is generally a sane
choice.
</p>
</li><li>
Send even and odd frames to separate outputs, and compose them:
<div class="example">
<pre class="example">select=n=2:e='mod(n, 2)+1' [odd][even]; [odd] pad=h=2*ih [tmp]; [tmp][even] overlay=y=h
</pre></div>
</li></ul>
<a name="sendcmd_002c-asendcmd"></a>
<h2 class="section"><a href="ffmpeg-filters.html#toc-sendcmd_002c-asendcmd">12.7 sendcmd, asendcmd</a></h2>
<p>Send commands to filters in the filtergraph.
</p>
<p>These filters read commands to be sent to other filters in the
filtergraph.
</p>
<p><code>sendcmd</code> must be inserted between two video filters,
<code>asendcmd</code> must be inserted between two audio filters, but apart
from that they act the same way.
</p>
<p>The specification of commands can be provided in the filter arguments
with the <var>commands</var> option, or in a file specified by the
<var>filename</var> option.
</p>
<p>These filters accept the following options:
</p><dl compact="compact">
<dt>‘<samp>commands, c</samp>’</dt>
<dd><p>Set the commands to be read and sent to the other filters.
</p></dd>
<dt>‘<samp>filename, f</samp>’</dt>
<dd><p>Set the filename of the commands to be read and sent to the other
filters.
</p></dd>
</dl>
<a name="Commands-syntax"></a>
<h3 class="subsection"><a href="ffmpeg-filters.html#toc-Commands-syntax">12.7.1 Commands syntax</a></h3>
<p>A commands description consists of a sequence of interval
specifications, comprising a list of commands to be executed when a
particular event related to that interval occurs. The occurring event
is typically the current frame time entering or leaving a given time
interval.
</p>
<p>An interval is specified by the following syntax:
</p><div class="example">
<pre class="example"><var>START</var>[-<var>END</var>] <var>COMMANDS</var>;
</pre></div>
<p>The time interval is specified by the <var>START</var> and <var>END</var> times.
<var>END</var> is optional and defaults to the maximum time.
</p>
<p>The current frame time is considered within the specified interval if
it is included in the interval [<var>START</var>, <var>END</var>), that is when
the time is greater or equal to <var>START</var> and is lesser than
<var>END</var>.
</p>
<p><var>COMMANDS</var> consists of a sequence of one or more command
specifications, separated by ",", relating to that interval. The
syntax of a command specification is given by:
</p><div class="example">
<pre class="example">[<var>FLAGS</var>] <var>TARGET</var> <var>COMMAND</var> <var>ARG</var>
</pre></div>
<p><var>FLAGS</var> is optional and specifies the type of events relating to
the time interval which enable sending the specified command, and must
be a non-null sequence of identifier flags separated by "+" or "|" and
enclosed between "[" and "]".
</p>
<p>The following flags are recognized:
</p><dl compact="compact">
<dt>‘<samp>enter</samp>’</dt>
<dd><p>The command is sent when the current frame timestamp enters the
specified interval. In other words, the command is sent when the
previous frame timestamp was not in the given interval, and the
current is.
</p>
</dd>
<dt>‘<samp>leave</samp>’</dt>
<dd><p>The command is sent when the current frame timestamp leaves the
specified interval. In other words, the command is sent when the
previous frame timestamp was in the given interval, and the
current is not.
</p></dd>
</dl>
<p>If <var>FLAGS</var> is not specified, a default value of <code>[enter]</code> is
assumed.
</p>
<p><var>TARGET</var> specifies the target of the command, usually the name of
the filter class or a specific filter instance name.
</p>
<p><var>COMMAND</var> specifies the name of the command for the target filter.
</p>
<p><var>ARG</var> is optional and specifies the optional list of argument for
the given <var>COMMAND</var>.
</p>
<p>Between one interval specification and another, whitespaces, or
sequences of characters starting with <code>#</code> until the end of line,
are ignored and can be used to annotate comments.
</p>
<p>A simplified BNF description of the commands specification syntax
follows:
</p><div class="example">
<pre class="example"><var>COMMAND_FLAG</var> ::= "enter" | "leave"
<var>COMMAND_FLAGS</var> ::= <var>COMMAND_FLAG</var> [(+|"|")<var>COMMAND_FLAG</var>]
<var>COMMAND</var> ::= ["[" <var>COMMAND_FLAGS</var> "]"] <var>TARGET</var> <var>COMMAND</var> [<var>ARG</var>]
<var>COMMANDS</var> ::= <var>COMMAND</var> [,<var>COMMANDS</var>]
<var>INTERVAL</var> ::= <var>START</var>[-<var>END</var>] <var>COMMANDS</var>
<var>INTERVALS</var> ::= <var>INTERVAL</var>[;<var>INTERVALS</var>]
</pre></div>
<a name="Examples-63"></a>
<h3 class="subsection"><a href="ffmpeg-filters.html#toc-Examples-63">12.7.2 Examples</a></h3>
<ul>
<li>
Specify audio tempo change at second 4:
<div class="example">
<pre class="example">asendcmd=c='4.0 atempo tempo 1.5',atempo
</pre></div>
</li><li>
Specify a list of drawtext and hue commands in a file.
<div class="example">
<pre class="example"># show text in the interval 5-10
5.0-10.0 [enter] drawtext reinit 'fontfile=FreeSerif.ttf:text=hello world',
[leave] drawtext reinit 'fontfile=FreeSerif.ttf:text=';
# desaturate the image in the interval 15-20
15.0-20.0 [enter] hue s 0,
[enter] drawtext reinit 'fontfile=FreeSerif.ttf:text=nocolor',
[leave] hue s 1,
[leave] drawtext reinit 'fontfile=FreeSerif.ttf:text=color';
# apply an exponential saturation fade-out effect, starting from time 25
25 [enter] hue s exp(25-t)
</pre></div>
<p>A filtergraph allowing to read and process the above command list
stored in a file ‘<tt>test.cmd</tt>’, can be specified with:
</p><div class="example">
<pre class="example">sendcmd=f=test.cmd,drawtext=fontfile=FreeSerif.ttf:text='',hue
</pre></div>
</li></ul>
<p><a name="setpts"></a>
</p><a name="setpts_002c-asetpts"></a>
<h2 class="section"><a href="ffmpeg-filters.html#toc-setpts_002c-asetpts">12.8 setpts, asetpts</a></h2>
<p>Change the PTS (presentation timestamp) of the input frames.
</p>
<p><code>setpts</code> works on video frames, <code>asetpts</code> on audio frames.
</p>
<p>This filter accepts the following options:
</p>
<dl compact="compact">
<dt>‘<samp>expr</samp>’</dt>
<dd><p>The expression which is evaluated for each frame to construct its timestamp.
</p>
</dd>
</dl>
<p>The expression is evaluated through the eval API and can contain the following
constants:
</p>
<dl compact="compact">
<dt>‘<samp>FRAME_RATE</samp>’</dt>
<dd><p>frame rate, only defined for constant frame-rate video
</p>
</dd>
<dt>‘<samp>PTS</samp>’</dt>
<dd><p>The presentation timestamp in input
</p>
</dd>
<dt>‘<samp>N</samp>’</dt>
<dd><p>The count of the input frame for video or the number of consumed samples,
not including the current frame for audio, starting from 0.
</p>
</dd>
<dt>‘<samp>NB_CONSUMED_SAMPLES</samp>’</dt>
<dd><p>The number of consumed samples, not including the current frame (only
audio)
</p>
</dd>
<dt>‘<samp>NB_SAMPLES, S</samp>’</dt>
<dd><p>The number of samples in the current frame (only audio)
</p>
</dd>
<dt>‘<samp>SAMPLE_RATE, SR</samp>’</dt>
<dd><p>The audio sample rate.
</p>
</dd>
<dt>‘<samp>STARTPTS</samp>’</dt>
<dd><p>The PTS of the first frame.
</p>
</dd>
<dt>‘<samp>STARTT</samp>’</dt>
<dd><p>the time in seconds of the first frame
</p>
</dd>
<dt>‘<samp>INTERLACED</samp>’</dt>
<dd><p>State whether the current frame is interlaced.
</p>
</dd>
<dt>‘<samp>T</samp>’</dt>
<dd><p>the time in seconds of the current frame
</p>
</dd>
<dt>‘<samp>POS</samp>’</dt>
<dd><p>original position in the file of the frame, or undefined if undefined
for the current frame
</p>
</dd>
<dt>‘<samp>PREV_INPTS</samp>’</dt>
<dd><p>The previous input PTS.
</p>
</dd>
<dt>‘<samp>PREV_INT</samp>’</dt>
<dd><p>previous input time in seconds
</p>
</dd>
<dt>‘<samp>PREV_OUTPTS</samp>’</dt>
<dd><p>The previous output PTS.
</p>
</dd>
<dt>‘<samp>PREV_OUTT</samp>’</dt>
<dd><p>previous output time in seconds
</p>
</dd>
<dt>‘<samp>RTCTIME</samp>’</dt>
<dd><p>The wallclock (RTC) time in microseconds.. This is deprecated, use time(0)
instead.
</p>
</dd>
<dt>‘<samp>RTCSTART</samp>’</dt>
<dd><p>The wallclock (RTC) time at the start of the movie in microseconds.
</p>
</dd>
<dt>‘<samp>TB</samp>’</dt>
<dd><p>The timebase of the input timestamps.
</p>
</dd>
</dl>
<a name="Examples-80"></a>
<h3 class="subsection"><a href="ffmpeg-filters.html#toc-Examples-80">12.8.1 Examples</a></h3>
<ul>
<li>
Start counting PTS from zero
<div class="example">
<pre class="example">setpts=PTS-STARTPTS
</pre></div>
</li><li>
Apply fast motion effect:
<div class="example">
<pre class="example">setpts=0.5*PTS
</pre></div>
</li><li>
Apply slow motion effect:
<div class="example">
<pre class="example">setpts=2.0*PTS
</pre></div>
</li><li>
Set fixed rate of 25 frames per second:
<div class="example">
<pre class="example">setpts=N/(25*TB)
</pre></div>
</li><li>
Set fixed rate 25 fps with some jitter:
<div class="example">
<pre class="example">setpts='1/(25*TB) * (N + 0.05 * sin(N*2*PI/25))'
</pre></div>
</li><li>
Apply an offset of 10 seconds to the input PTS:
<div class="example">
<pre class="example">setpts=PTS+10/TB
</pre></div>
</li><li>
Generate timestamps from a "live source" and rebase onto the current timebase:
<div class="example">
<pre class="example">setpts='(RTCTIME - RTCSTART) / (TB * 1000000)'
</pre></div>
</li><li>
Generate timestamps by counting samples:
<div class="example">
<pre class="example">asetpts=N/SR/TB
</pre></div>
</li></ul>
<a name="settb_002c-asettb"></a>
<h2 class="section"><a href="ffmpeg-filters.html#toc-settb_002c-asettb">12.9 settb, asettb</a></h2>
<p>Set the timebase to use for the output frames timestamps.
It is mainly useful for testing timebase configuration.
</p>
<p>It accepts the following parameters:
</p>
<dl compact="compact">
<dt>‘<samp>expr, tb</samp>’</dt>
<dd><p>The expression which is evaluated into the output timebase.
</p>
</dd>
</dl>
<p>The value for ‘<samp>tb</samp>’ is an arithmetic expression representing a
rational. The expression can contain the constants "AVTB" (the default
timebase), "intb" (the input timebase) and "sr" (the sample rate,
audio only). Default value is "intb".
</p>
<a name="Examples-66"></a>
<h3 class="subsection"><a href="ffmpeg-filters.html#toc-Examples-66">12.9.1 Examples</a></h3>
<ul>
<li>
Set the timebase to 1/25:
<div class="example">
<pre class="example">settb=expr=1/25
</pre></div>
</li><li>
Set the timebase to 1/10:
<div class="example">
<pre class="example">settb=expr=0.1
</pre></div>
</li><li>
Set the timebase to 1001/1000:
<div class="example">
<pre class="example">settb=1+0.001
</pre></div>
</li><li>
Set the timebase to 2*intb:
<div class="example">
<pre class="example">settb=2*intb
</pre></div>
</li><li>
Set the default timebase value:
<div class="example">
<pre class="example">settb=AVTB
</pre></div>
</li></ul>
<a name="showcqt"></a>
<h2 class="section"><a href="ffmpeg-filters.html#toc-showcqt">12.10 showcqt</a></h2>
<p>Convert input audio to a video output representing
frequency spectrum logarithmically (using constant Q transform with
Brown-Puckette algorithm), with musical tone scale, from E0 to D#10 (10 octaves).
</p>
<p>The filter accepts the following options:
</p>
<dl compact="compact">
<dt>‘<samp>volume</samp>’</dt>
<dd><p>Specify transform volume (multiplier) expression. The expression can contain
variables:
</p><dl compact="compact">
<dt>‘<samp>frequency, freq, f</samp>’</dt>
<dd><p>the frequency where transform is evaluated
</p></dd>
<dt>‘<samp>timeclamp, tc</samp>’</dt>
<dd><p>value of timeclamp option
</p></dd>
</dl>
<p>and functions:
</p><dl compact="compact">
<dt>‘<samp>a_weighting(f)</samp>’</dt>
<dd><p>A-weighting of equal loudness
</p></dd>
<dt>‘<samp>b_weighting(f)</samp>’</dt>
<dd><p>B-weighting of equal loudness
</p></dd>
<dt>‘<samp>c_weighting(f)</samp>’</dt>
<dd><p>C-weighting of equal loudness
</p></dd>
</dl>
<p>Default value is <code>16</code>.
</p>
</dd>
<dt>‘<samp>tlength</samp>’</dt>
<dd><p>Specify transform length expression. The expression can contain variables:
</p><dl compact="compact">
<dt>‘<samp>frequency, freq, f</samp>’</dt>
<dd><p>the frequency where transform is evaluated
</p></dd>
<dt>‘<samp>timeclamp, tc</samp>’</dt>
<dd><p>value of timeclamp option
</p></dd>
</dl>
<p>Default value is <code>384/f*tc/(384/f+tc)</code>.
</p>
</dd>
<dt>‘<samp>timeclamp</samp>’</dt>
<dd><p>Specify the transform timeclamp. At low frequency, there is trade-off between
accuracy in time domain and frequency domain. If timeclamp is lower,
event in time domain is represented more accurately (such as fast bass drum),
otherwise event in frequency domain is represented more accurately
(such as bass guitar). Acceptable value is [0.1, 1.0]. Default value is <code>0.17</code>.
</p>
</dd>
<dt>‘<samp>coeffclamp</samp>’</dt>
<dd><p>Specify the transform coeffclamp. If coeffclamp is lower, transform is
more accurate, otherwise transform is faster. Acceptable value is [0.1, 10.0].
Default value is <code>1.0</code>.
</p>
</dd>
<dt>‘<samp>gamma</samp>’</dt>
<dd><p>Specify gamma. Lower gamma makes the spectrum more contrast, higher gamma
makes the spectrum having more range. Acceptable value is [1.0, 7.0].
Default value is <code>3.0</code>.
</p>
</dd>
<dt>‘<samp>fontfile</samp>’</dt>
<dd><p>Specify font file for use with freetype. If not specified, use embedded font.
</p>
</dd>
<dt>‘<samp>fontcolor</samp>’</dt>
<dd><p>Specify font color expression. This is arithmetic expression that should return
integer value 0xRRGGBB. The expression can contain variables:
</p><dl compact="compact">
<dt>‘<samp>frequency, freq, f</samp>’</dt>
<dd><p>the frequency where transform is evaluated
</p></dd>
<dt>‘<samp>timeclamp, tc</samp>’</dt>
<dd><p>value of timeclamp option
</p></dd>
</dl>
<p>and functions:
</p><dl compact="compact">
<dt>‘<samp>midi(f)</samp>’</dt>
<dd><p>midi number of frequency f, some midi numbers: E0(16), C1(24), C2(36), A4(69)
</p></dd>
<dt>‘<samp>r(x), g(x), b(x)</samp>’</dt>
<dd><p>red, green, and blue value of intensity x
</p></dd>
</dl>
<p>Default value is <code>st(0, (midi(f)-59.5)/12);
st(1, if(between(ld(0),0,1), 0.5-0.5*cos(2*PI*ld(0)), 0));
r(1-ld(1)) + b(ld(1))</code>
</p>
</dd>
<dt>‘<samp>fullhd</samp>’</dt>
<dd><p>If set to 1 (the default), the video size is 1920x1080 (full HD),
if set to 0, the video size is 960x540. Use this option to make CPU usage lower.
</p>
</dd>
<dt>‘<samp>fps</samp>’</dt>
<dd><p>Specify video fps. Default value is <code>25</code>.
</p>
</dd>
<dt>‘<samp>count</samp>’</dt>
<dd><p>Specify number of transform per frame, so there are fps*count transforms
per second. Note that audio data rate must be divisible by fps*count.
Default value is <code>6</code>.
</p>
</dd>
</dl>
<a name="Examples-4"></a>
<h3 class="subsection"><a href="ffmpeg-filters.html#toc-Examples-4">12.10.1 Examples</a></h3>
<ul>
<li>
Playing audio while showing the spectrum:
<div class="example">
<pre class="example">ffplay -f lavfi 'amovie=a.mp3, asplit [a][out1]; [a] showcqt [out0]'
</pre></div>
</li><li>
Same as above, but with frame rate 30 fps:
<div class="example">
<pre class="example">ffplay -f lavfi 'amovie=a.mp3, asplit [a][out1]; [a] showcqt=fps=30:count=5 [out0]'
</pre></div>
</li><li>
Playing at 960x540 and lower CPU usage:
<div class="example">
<pre class="example">ffplay -f lavfi 'amovie=a.mp3, asplit [a][out1]; [a] showcqt=fullhd=0:count=3 [out0]'
</pre></div>
</li><li>
A1 and its harmonics: A1, A2, (near)E3, A3:
<div class="example">
<pre class="example">ffplay -f lavfi 'aevalsrc=0.1*sin(2*PI*55*t)+0.1*sin(4*PI*55*t)+0.1*sin(6*PI*55*t)+0.1*sin(8*PI*55*t),
asplit[a][out1]; [a] showcqt [out0]'
</pre></div>
</li><li>
Same as above, but with more accuracy in frequency domain (and slower):
<div class="example">
<pre class="example">ffplay -f lavfi 'aevalsrc=0.1*sin(2*PI*55*t)+0.1*sin(4*PI*55*t)+0.1*sin(6*PI*55*t)+0.1*sin(8*PI*55*t),
asplit[a][out1]; [a] showcqt=timeclamp=0.5 [out0]'
</pre></div>
</li><li>
B-weighting of equal loudness
<div class="example">
<pre class="example">volume=16*b_weighting(f)
</pre></div>
</li><li>
Lower Q factor
<div class="example">
<pre class="example">tlength=100/f*tc/(100/f+tc)
</pre></div>
</li><li>
Custom fontcolor, C-note is colored green, others are colored blue
<div class="example">
<pre class="example">fontcolor='if(mod(floor(midi(f)+0.5),12), 0x0000FF, g(1))'
</pre></div>
</li></ul>
<a name="showspectrum"></a>
<h2 class="section"><a href="ffmpeg-filters.html#toc-showspectrum">12.11 showspectrum</a></h2>
<p>Convert input audio to a video output, representing the audio frequency
spectrum.
</p>
<p>The filter accepts the following options:
</p>
<dl compact="compact">
<dt>‘<samp>size, s</samp>’</dt>
<dd><p>Specify the video size for the output. For the syntax of this option, check
the "Video size" section in the ffmpeg-utils manual. Default value is
<code>640x512</code>.
</p>
</dd>
<dt>‘<samp>slide</samp>’</dt>
<dd><p>Specify how the spectrum should slide along the window.
</p>
<p>It accepts the following values:
</p><dl compact="compact">
<dt>‘<samp>replace</samp>’</dt>
<dd><p>the samples start again on the left when they reach the right
</p></dd>
<dt>‘<samp>scroll</samp>’</dt>
<dd><p>the samples scroll from right to left
</p></dd>
<dt>‘<samp>fullframe</samp>’</dt>
<dd><p>frames are only produced when the samples reach the right
</p></dd>
</dl>
<p>Default value is <code>replace</code>.
</p>
</dd>
<dt>‘<samp>mode</samp>’</dt>
<dd><p>Specify display mode.
</p>
<p>It accepts the following values:
</p><dl compact="compact">
<dt>‘<samp>combined</samp>’</dt>
<dd><p>all channels are displayed in the same row
</p></dd>
<dt>‘<samp>separate</samp>’</dt>
<dd><p>all channels are displayed in separate rows
</p></dd>
</dl>
<p>Default value is ‘<samp>combined</samp>’.
</p>
</dd>
<dt>‘<samp>color</samp>’</dt>
<dd><p>Specify display color mode.
</p>
<p>It accepts the following values:
</p><dl compact="compact">
<dt>‘<samp>channel</samp>’</dt>
<dd><p>each channel is displayed in a separate color
</p></dd>
<dt>‘<samp>intensity</samp>’</dt>
<dd><p>each channel is is displayed using the same color scheme
</p></dd>
</dl>
<p>Default value is ‘<samp>channel</samp>’.
</p>
</dd>
<dt>‘<samp>scale</samp>’</dt>
<dd><p>Specify scale used for calculating intensity color values.
</p>
<p>It accepts the following values:
</p><dl compact="compact">
<dt>‘<samp>lin</samp>’</dt>
<dd><p>linear
</p></dd>
<dt>‘<samp>sqrt</samp>’</dt>
<dd><p>square root, default
</p></dd>
<dt>‘<samp>cbrt</samp>’</dt>
<dd><p>cubic root
</p></dd>
<dt>‘<samp>log</samp>’</dt>
<dd><p>logarithmic
</p></dd>
</dl>
<p>Default value is ‘<samp>sqrt</samp>’.
</p>
</dd>
<dt>‘<samp>saturation</samp>’</dt>
<dd><p>Set saturation modifier for displayed colors. Negative values provide
alternative color scheme. <code>0</code> is no saturation at all.
Saturation must be in [-10.0, 10.0] range.
Default value is <code>1</code>.
</p>
</dd>
<dt>‘<samp>win_func</samp>’</dt>
<dd><p>Set window function.
</p>
<p>It accepts the following values:
</p><dl compact="compact">
<dt>‘<samp>none</samp>’</dt>
<dd><p>No samples pre-processing (do not expect this to be faster)
</p></dd>
<dt>‘<samp>hann</samp>’</dt>
<dd><p>Hann window
</p></dd>
<dt>‘<samp>hamming</samp>’</dt>
<dd><p>Hamming window
</p></dd>
<dt>‘<samp>blackman</samp>’</dt>
<dd><p>Blackman window
</p></dd>
</dl>
<p>Default value is <code>hann</code>.
</p></dd>
</dl>
<p>The usage is very similar to the showwaves filter; see the examples in that
section.
</p>
<a name="Examples-57"></a>
<h3 class="subsection"><a href="ffmpeg-filters.html#toc-Examples-57">12.11.1 Examples</a></h3>
<ul>
<li>
Large window with logarithmic color scaling:
<div class="example">
<pre class="example">showspectrum=s=1280x480:scale=log
</pre></div>
</li><li>
Complete example for a colored and sliding spectrum per channel using <code>ffplay</code>:
<div class="example">
<pre class="example">ffplay -f lavfi 'amovie=input.mp3, asplit [a][out1];
[a] showspectrum=mode=separate:color=intensity:slide=1:scale=cbrt [out0]'
</pre></div>
</li></ul>
<a name="showwaves"></a>
<h2 class="section"><a href="ffmpeg-filters.html#toc-showwaves">12.12 showwaves</a></h2>
<p>Convert input audio to a video output, representing the samples waves.
</p>
<p>The filter accepts the following options:
</p>
<dl compact="compact">
<dt>‘<samp>size, s</samp>’</dt>
<dd><p>Specify the video size for the output. For the syntax of this option, check
the "Video size" section in the ffmpeg-utils manual. Default value
is "600x240".
</p>
</dd>
<dt>‘<samp>mode</samp>’</dt>
<dd><p>Set display mode.
</p>
<p>Available values are:
</p><dl compact="compact">
<dt>‘<samp>point</samp>’</dt>
<dd><p>Draw a point for each sample.
</p>
</dd>
<dt>‘<samp>line</samp>’</dt>
<dd><p>Draw a vertical line for each sample.
</p>
</dd>
<dt>‘<samp>p2p</samp>’</dt>
<dd><p>Draw a point for each sample and a line between them.
</p>
</dd>
<dt>‘<samp>cline</samp>’</dt>
<dd><p>Draw a centered vertical line for each sample.
</p></dd>
</dl>
<p>Default value is <code>point</code>.
</p>
</dd>
<dt>‘<samp>n</samp>’</dt>
<dd><p>Set the number of samples which are printed on the same column. A
larger value will decrease the frame rate. Must be a positive
integer. This option can be set only if the value for <var>rate</var>
is not explicitly specified.
</p>
</dd>
<dt>‘<samp>rate, r</samp>’</dt>
<dd><p>Set the (approximate) output frame rate. This is done by setting the
option <var>n</var>. Default value is "25".
</p>
</dd>
<dt>‘<samp>split_channels</samp>’</dt>
<dd><p>Set if channels should be drawn separately or overlap. Default value is 0.
</p>
</dd>
</dl>
<a name="Examples-51"></a>
<h3 class="subsection"><a href="ffmpeg-filters.html#toc-Examples-51">12.12.1 Examples</a></h3>
<ul>
<li>
Output the input file audio and the corresponding video representation
at the same time:
<div class="example">
<pre class="example">amovie=a.mp3,asplit[out0],showwaves[out1]
</pre></div>
</li><li>
Create a synthetic signal and show it with showwaves, forcing a
frame rate of 30 frames per second:
<div class="example">
<pre class="example">aevalsrc=sin(1*2*PI*t)*sin(880*2*PI*t):cos(2*PI*200*t),asplit[out0],showwaves=r=30[out1]
</pre></div>
</li></ul>
<a name="split_002c-asplit"></a>
<h2 class="section"><a href="ffmpeg-filters.html#toc-split_002c-asplit">12.13 split, asplit</a></h2>
<p>Split input into several identical outputs.
</p>
<p><code>asplit</code> works with audio input, <code>split</code> with video.
</p>
<p>The filter accepts a single parameter which specifies the number of outputs. If
unspecified, it defaults to 2.
</p>
<a name="Examples-9"></a>
<h3 class="subsection"><a href="ffmpeg-filters.html#toc-Examples-9">12.13.1 Examples</a></h3>
<ul>
<li>
Create two separate outputs from the same input:
<div class="example">
<pre class="example">[in] split [out0][out1]
</pre></div>
</li><li>
To create 3 or more outputs, you need to specify the number of
outputs, like in:
<div class="example">
<pre class="example">[in] asplit=3 [out0][out1][out2]
</pre></div>
</li><li>
Create two separate outputs from the same input, one cropped and
one padded:
<div class="example">
<pre class="example">[in] split [splitout1][splitout2];
[splitout1] crop=100:100:0:0 [cropout];
[splitout2] pad=200:200:100:100 [padout];
</pre></div>
</li><li>
Create 5 copies of the input audio with <code>ffmpeg</code>:
<div class="example">
<pre class="example">ffmpeg -i INPUT -filter_complex asplit=5 OUTPUT
</pre></div>
</li></ul>
<a name="zmq_002c-azmq"></a>
<h2 class="section"><a href="ffmpeg-filters.html#toc-zmq_002c-azmq">12.14 zmq, azmq</a></h2>
<p>Receive commands sent through a libzmq client, and forward them to
filters in the filtergraph.
</p>
<p><code>zmq</code> and <code>azmq</code> work as a pass-through filters. <code>zmq</code>
must be inserted between two video filters, <code>azmq</code> between two
audio filters.
</p>
<p>To enable these filters you need to install the libzmq library and
headers and configure FFmpeg with <code>--enable-libzmq</code>.
</p>
<p>For more information about libzmq see:
<a href="http://www.zeromq.org/">http://www.zeromq.org/</a>
</p>
<p>The <code>zmq</code> and <code>azmq</code> filters work as a libzmq server, which
receives messages sent through a network interface defined by the
‘<samp>bind_address</samp>’ option.
</p>
<p>The received message must be in the form:
</p><div class="example">
<pre class="example"><var>TARGET</var> <var>COMMAND</var> [<var>ARG</var>]
</pre></div>
<p><var>TARGET</var> specifies the target of the command, usually the name of
the filter class or a specific filter instance name.
</p>
<p><var>COMMAND</var> specifies the name of the command for the target filter.
</p>
<p><var>ARG</var> is optional and specifies the optional argument list for the
given <var>COMMAND</var>.
</p>
<p>Upon reception, the message is processed and the corresponding command
is injected into the filtergraph. Depending on the result, the filter
will send a reply to the client, adopting the format:
</p><div class="example">
<pre class="example"><var>ERROR_CODE</var> <var>ERROR_REASON</var>
<var>MESSAGE</var>
</pre></div>
<p><var>MESSAGE</var> is optional.
</p>
<a name="Examples-2"></a>
<h3 class="subsection"><a href="ffmpeg-filters.html#toc-Examples-2">12.14.1 Examples</a></h3>
<p>Look at ‘<tt>tools/zmqsend</tt>’ for an example of a zmq client which can
be used to send commands processed by these filters.
</p>
<p>Consider the following filtergraph generated by <code>ffplay</code>
</p><div class="example">
<pre class="example">ffplay -dumpgraph 1 -f lavfi "
color=s=100x100:c=red [l];
color=s=100x100:c=blue [r];
nullsrc=s=200x100, zmq [bg];
[bg][l] overlay [bg+l];
[bg+l][r] overlay=x=100 "
</pre></div>
<p>To change the color of the left side of the video, the following
command can be used:
</p><div class="example">
<pre class="example">echo Parsed_color_0 c yellow | tools/zmqsend
</pre></div>
<p>To change the right side:
</p><div class="example">
<pre class="example">echo Parsed_color_1 c pink | tools/zmqsend
</pre></div>
<a name="Multimedia-Sources"></a>
<h1 class="chapter"><a href="ffmpeg-filters.html#toc-Multimedia-Sources">13 Multimedia Sources</a></h1>
<p>Below is a description of the currently available multimedia sources.
</p>
<a name="amovie"></a>
<h2 class="section"><a href="ffmpeg-filters.html#toc-amovie">13.1 amovie</a></h2>
<p>This is the same as <a href="#movie">movie</a> source, except it selects an audio
stream by default.
</p>
<p><a name="movie"></a>
</p><a name="movie-1"></a>
<h2 class="section"><a href="ffmpeg-filters.html#toc-movie-1">13.2 movie</a></h2>
<p>Read audio and/or video stream(s) from a movie container.
</p>
<p>It accepts the following parameters:
</p>
<dl compact="compact">
<dt>‘<samp>filename</samp>’</dt>
<dd><p>The name of the resource to read (not necessarily a file; it can also be a
device or a stream accessed through some protocol).
</p>
</dd>
<dt>‘<samp>format_name, f</samp>’</dt>
<dd><p>Specifies the format assumed for the movie to read, and can be either
the name of a container or an input device. If not specified, the
format is guessed from <var>movie_name</var> or by probing.
</p>
</dd>
<dt>‘<samp>seek_point, sp</samp>’</dt>
<dd><p>Specifies the seek point in seconds. The frames will be output
starting from this seek point. The parameter is evaluated with
<code>av_strtod</code>, so the numerical value may be suffixed by an IS
postfix. The default value is "0".
</p>
</dd>
<dt>‘<samp>streams, s</samp>’</dt>
<dd><p>Specifies the streams to read. Several streams can be specified,
separated by "+". The source will then have as many outputs, in the
same order. The syntax is explained in the “Stream specifiers”
section in the ffmpeg manual. Two special names, "dv" and "da" specify
respectively the default (best suited) video and audio stream. Default
is "dv", or "da" if the filter is called as "amovie".
</p>
</dd>
<dt>‘<samp>stream_index, si</samp>’</dt>
<dd><p>Specifies the index of the video stream to read. If the value is -1,
the most suitable video stream will be automatically selected. The default
value is "-1". Deprecated. If the filter is called "amovie", it will select
audio instead of video.
</p>
</dd>
<dt>‘<samp>loop</samp>’</dt>
<dd><p>Specifies how many times to read the stream in sequence.
If the value is less than 1, the stream will be read again and again.
Default value is "1".
</p>
<p>Note that when the movie is looped the source timestamps are not
changed, so it will generate non monotonically increasing timestamps.
</p></dd>
</dl>
<p>It allows overlaying a second video on top of the main input of
a filtergraph, as shown in this graph:
</p><div class="example">
<pre class="example">input -----------> deltapts0 --> overlay --> output
^
|
movie --> scale--> deltapts1 -------+
</pre></div>
<a name="Examples-64"></a>
<h3 class="subsection"><a href="ffmpeg-filters.html#toc-Examples-64">13.2.1 Examples</a></h3>
<ul>
<li>
Skip 3.2 seconds from the start of the AVI file in.avi, and overlay it
on top of the input labelled "in":
<div class="example">
<pre class="example">movie=in.avi:seek_point=3.2, scale=180:-1, setpts=PTS-STARTPTS [over];
[in] setpts=PTS-STARTPTS [main];
[main][over] overlay=16:16 [out]
</pre></div>
</li><li>
Read from a video4linux2 device, and overlay it on top of the input
labelled "in":
<div class="example">
<pre class="example">movie=/dev/video0:f=video4linux2, scale=180:-1, setpts=PTS-STARTPTS [over];
[in] setpts=PTS-STARTPTS [main];
[main][over] overlay=16:16 [out]
</pre></div>
</li><li>
Read the first video stream and the audio stream with id 0x81 from
dvd.vob; the video is connected to the pad named "video" and the audio is
connected to the pad named "audio":
<div class="example">
<pre class="example">movie=dvd.vob:s=v:0+#0x81 [video] [audio]
</pre></div>
</li></ul>
<a name="See-Also"></a>
<h1 class="chapter"><a href="ffmpeg-filters.html#toc-See-Also">14 See Also</a></h1>
<p><a href="ffmpeg.html">ffmpeg</a>, <a href="ffplay.html">ffplay</a>, <a href="ffprobe.html">ffprobe</a>, <a href="ffserver.html">ffserver</a>,
<a href="libavfilter.html">libavfilter</a>
</p>
<a name="Authors"></a>
<h1 class="chapter"><a href="ffmpeg-filters.html#toc-Authors">15 Authors</a></h1>
<p>The FFmpeg developers.
</p>
<p>For details about the authorship, see the Git history of the project
(git://source.ffmpeg.org/ffmpeg), e.g. by typing the command
<code>git log</code> in the FFmpeg source directory, or browsing the
online repository at <a href="http://source.ffmpeg.org">http://source.ffmpeg.org</a>.
</p>
<p>Maintainers for the specific components are listed in the file
‘<tt>MAINTAINERS</tt>’ in the source code tree.
</p>
</div>
</body>
</html>
distrib
>
Mageia
>
5
>
i586
>
media
>
tainted-updates
>
by-pkgid
>
f31562021404636e2cdd7f56c175561c
>
files
>
15
