1 @chapter Filtering Introduction
2 @c man begin FILTERING INTRODUCTION
4 Filtering in FFmpeg is enabled through the libavfilter library.
6 In libavfilter, a filter can have multiple inputs and multiple
8 To illustrate the sorts of things that are possible, we consider the
13 input --> split ---------------------> overlay --> output
16 +-----> crop --> vflip -------+
19 This filtergraph splits the input stream in two streams, then sends one
20 stream through the crop filter and the vflip filter, before merging it
21 back with the other stream by overlaying it on top. You can use the
22 following command to achieve this:
25 ffmpeg -i INPUT -vf "split [main][tmp]; [tmp] crop=iw:ih/2:0:0, vflip [flip]; [main][flip] overlay=0:H/2" OUTPUT
28 The result will be that the top half of the video is mirrored
29 onto the bottom half of the output video.
31 Filters in the same linear chain are separated by commas, and distinct
32 linear chains of filters are separated by semicolons. In our example,
33 @var{crop,vflip} are in one linear chain, @var{split} and
34 @var{overlay} are separately in another. The points where the linear
35 chains join are labelled by names enclosed in square brackets. In the
36 example, the split filter generates two outputs that are associated to
37 the labels @var{[main]} and @var{[tmp]}.
39 The stream sent to the second output of @var{split}, labelled as
40 @var{[tmp]}, is processed through the @var{crop} filter, which crops
41 away the lower half part of the video, and then vertically flipped. The
42 @var{overlay} filter takes in input the first unchanged output of the
43 split filter (which was labelled as @var{[main]}), and overlay on its
44 lower half the output generated by the @var{crop,vflip} filterchain.
46 Some filters take in input a list of parameters: they are specified
47 after the filter name and an equal sign, and are separated from each other
50 There exist so-called @var{source filters} that do not have an
51 audio/video input, and @var{sink filters} that will not have audio/video
54 @c man end FILTERING INTRODUCTION
57 @c man begin GRAPH2DOT
59 The @file{graph2dot} program included in the FFmpeg @file{tools}
60 directory can be used to parse a filtergraph description and issue a
61 corresponding textual representation in the dot language.
68 to see how to use @file{graph2dot}.
70 You can then pass the dot description to the @file{dot} program (from
71 the graphviz suite of programs) and obtain a graphical representation
74 For example the sequence of commands:
76 echo @var{GRAPH_DESCRIPTION} | \
77 tools/graph2dot -o graph.tmp && \
78 dot -Tpng graph.tmp -o graph.png && \
82 can be used to create and display an image representing the graph
83 described by the @var{GRAPH_DESCRIPTION} string. Note that this string must be
84 a complete self-contained graph, with its inputs and outputs explicitly defined.
85 For example if your command line is of the form:
87 ffmpeg -i infile -vf scale=640:360 outfile
89 your @var{GRAPH_DESCRIPTION} string will need to be of the form:
91 nullsrc,scale=640:360,nullsink
93 you may also need to set the @var{nullsrc} parameters and add a @var{format}
94 filter in order to simulate a specific input file.
98 @chapter Filtergraph description
99 @c man begin FILTERGRAPH DESCRIPTION
101 A filtergraph is a directed graph of connected filters. It can contain
102 cycles, and there can be multiple links between a pair of
103 filters. Each link has one input pad on one side connecting it to one
104 filter from which it takes its input, and one output pad on the other
105 side connecting it to one filter accepting its output.
107 Each filter in a filtergraph is an instance of a filter class
108 registered in the application, which defines the features and the
109 number of input and output pads of the filter.
111 A filter with no input pads is called a "source", and a filter with no
112 output pads is called a "sink".
114 @anchor{Filtergraph syntax}
115 @section Filtergraph syntax
117 A filtergraph has a textual representation, which is recognized by the
118 @option{-filter}/@option{-vf}/@option{-af} and
119 @option{-filter_complex} options in @command{ffmpeg} and
120 @option{-vf}/@option{-af} in @command{ffplay}, and by the
121 @code{avfilter_graph_parse_ptr()} function defined in
122 @file{libavfilter/avfilter.h}.
124 A filterchain consists of a sequence of connected filters, each one
125 connected to the previous one in the sequence. A filterchain is
126 represented by a list of ","-separated filter descriptions.
128 A filtergraph consists of a sequence of filterchains. A sequence of
129 filterchains is represented by a list of ";"-separated filterchain
132 A filter is represented by a string of the form:
133 [@var{in_link_1}]...[@var{in_link_N}]@var{filter_name}=@var{arguments}[@var{out_link_1}]...[@var{out_link_M}]
135 @var{filter_name} is the name of the filter class of which the
136 described filter is an instance of, and has to be the name of one of
137 the filter classes registered in the program.
138 The name of the filter class is optionally followed by a string
141 @var{arguments} is a string which contains the parameters used to
142 initialize the filter instance. It may have one of two forms:
146 A ':'-separated list of @var{key=value} pairs.
149 A ':'-separated list of @var{value}. In this case, the keys are assumed to be
150 the option names in the order they are declared. E.g. the @code{fade} filter
151 declares three options in this order -- @option{type}, @option{start_frame} and
152 @option{nb_frames}. Then the parameter list @var{in:0:30} means that the value
153 @var{in} is assigned to the option @option{type}, @var{0} to
154 @option{start_frame} and @var{30} to @option{nb_frames}.
157 A ':'-separated list of mixed direct @var{value} and long @var{key=value}
158 pairs. The direct @var{value} must precede the @var{key=value} pairs, and
159 follow the same constraints order of the previous point. The following
160 @var{key=value} pairs can be set in any preferred order.
164 If the option value itself is a list of items (e.g. the @code{format} filter
165 takes a list of pixel formats), the items in the list are usually separated by
168 The list of arguments can be quoted using the character @samp{'} as initial
169 and ending mark, and the character @samp{\} for escaping the characters
170 within the quoted text; otherwise the argument string is considered
171 terminated when the next special character (belonging to the set
172 @samp{[]=;,}) is encountered.
174 The name and arguments of the filter are optionally preceded and
175 followed by a list of link labels.
176 A link label allows one to name a link and associate it to a filter output
177 or input pad. The preceding labels @var{in_link_1}
178 ... @var{in_link_N}, are associated to the filter input pads,
179 the following labels @var{out_link_1} ... @var{out_link_M}, are
180 associated to the output pads.
182 When two link labels with the same name are found in the
183 filtergraph, a link between the corresponding input and output pad is
186 If an output pad is not labelled, it is linked by default to the first
187 unlabelled input pad of the next filter in the filterchain.
188 For example in the filterchain
190 nullsrc, split[L1], [L2]overlay, nullsink
192 the split filter instance has two output pads, and the overlay filter
193 instance two input pads. The first output pad of split is labelled
194 "L1", the first input pad of overlay is labelled "L2", and the second
195 output pad of split is linked to the second input pad of overlay,
196 which are both unlabelled.
198 In a filter description, if the input label of the first filter is not
199 specified, "in" is assumed; if the output label of the last filter is not
200 specified, "out" is assumed.
202 In a complete filterchain all the unlabelled filter input and output
203 pads must be connected. A filtergraph is considered valid if all the
204 filter input and output pads of all the filterchains are connected.
206 Libavfilter will automatically insert @ref{scale} filters where format
207 conversion is required. It is possible to specify swscale flags
208 for those automatically inserted scalers by prepending
209 @code{sws_flags=@var{flags};}
210 to the filtergraph description.
212 Here is a BNF description of the filtergraph syntax:
214 @var{NAME} ::= sequence of alphanumeric characters and '_'
215 @var{LINKLABEL} ::= "[" @var{NAME} "]"
216 @var{LINKLABELS} ::= @var{LINKLABEL} [@var{LINKLABELS}]
217 @var{FILTER_ARGUMENTS} ::= sequence of chars (possibly quoted)
218 @var{FILTER} ::= [@var{LINKLABELS}] @var{NAME} ["=" @var{FILTER_ARGUMENTS}] [@var{LINKLABELS}]
219 @var{FILTERCHAIN} ::= @var{FILTER} [,@var{FILTERCHAIN}]
220 @var{FILTERGRAPH} ::= [sws_flags=@var{flags};] @var{FILTERCHAIN} [;@var{FILTERGRAPH}]
223 @section Notes on filtergraph escaping
225 Filtergraph description composition entails several levels of
226 escaping. See @ref{quoting_and_escaping,,the "Quoting and escaping"
227 section in the ffmpeg-utils(1) manual,ffmpeg-utils} for more
228 information about the employed escaping procedure.
230 A first level escaping affects the content of each filter option
231 value, which may contain the special character @code{:} used to
232 separate values, or one of the escaping characters @code{\'}.
234 A second level escaping affects the whole filter description, which
235 may contain the escaping characters @code{\'} or the special
236 characters @code{[],;} used by the filtergraph description.
238 Finally, when you specify a filtergraph on a shell commandline, you
239 need to perform a third level escaping for the shell special
240 characters contained within it.
242 For example, consider the following string to be embedded in
243 the @ref{drawtext} filter description @option{text} value:
245 this is a 'string': may contain one, or more, special characters
248 This string contains the @code{'} special escaping character, and the
249 @code{:} special character, so it needs to be escaped in this way:
251 text=this is a \'string\'\: may contain one, or more, special characters
254 A second level of escaping is required when embedding the filter
255 description in a filtergraph description, in order to escape all the
256 filtergraph special characters. Thus the example above becomes:
258 drawtext=text=this is a \\\'string\\\'\\: may contain one\, or more\, special characters
260 (note that in addition to the @code{\'} escaping special characters,
261 also @code{,} needs to be escaped).
263 Finally an additional level of escaping is needed when writing the
264 filtergraph description in a shell command, which depends on the
265 escaping rules of the adopted shell. For example, assuming that
266 @code{\} is special and needs to be escaped with another @code{\}, the
267 previous string will finally result in:
269 -vf "drawtext=text=this is a \\\\\\'string\\\\\\'\\\\: may contain one\\, or more\\, special characters"
272 @chapter Timeline editing
274 Some filters support a generic @option{enable} option. For the filters
275 supporting timeline editing, this option can be set to an expression which is
276 evaluated before sending a frame to the filter. If the evaluation is non-zero,
277 the filter will be enabled, otherwise the frame will be sent unchanged to the
278 next filter in the filtergraph.
280 The expression accepts the following values:
283 timestamp expressed in seconds, NAN if the input timestamp is unknown
286 sequential number of the input frame, starting from 0
289 the position in the file of the input frame, NAN if unknown
293 width and height of the input frame if video
296 Additionally, these filters support an @option{enable} command that can be used
297 to re-define the expression.
299 Like any other filtering option, the @option{enable} option follows the same
302 For example, to enable a blur filter (@ref{smartblur}) from 10 seconds to 3
303 minutes, and a @ref{curves} filter starting at 3 seconds:
305 smartblur = enable='between(t,10,3*60)',
306 curves = enable='gte(t,3)' : preset=cross_process
309 @c man end FILTERGRAPH DESCRIPTION
311 @chapter Audio Filters
312 @c man begin AUDIO FILTERS
314 When you configure your FFmpeg build, you can disable any of the
315 existing filters using @code{--disable-filters}.
316 The configure output will show the audio filters included in your
319 Below is a description of the currently available audio filters.
323 A compressor is mainly used to reduce the dynamic range of a signal.
324 Especially modern music is mostly compressed at a high ratio to
325 improve the overall loudness. It's done to get the highest attention
326 of a listener, "fatten" the sound and bring more "power" to the track.
327 If a signal is compressed too much it may sound dull or "dead"
328 afterwards or it may start to "pump" (which could be a powerful effect
329 but can also destroy a track completely).
330 The right compression is the key to reach a professional sound and is
331 the high art of mixing and mastering. Because of its complex settings
332 it may take a long time to get the right feeling for this kind of effect.
334 Compression is done by detecting the volume above a chosen level
335 @code{threshold} and dividing it by the factor set with @code{ratio}.
336 So if you set the threshold to -12dB and your signal reaches -6dB a ratio
337 of 2:1 will result in a signal at -9dB. Because an exact manipulation of
338 the signal would cause distortion of the waveform the reduction can be
339 levelled over the time. This is done by setting "Attack" and "Release".
340 @code{attack} determines how long the signal has to rise above the threshold
341 before any reduction will occur and @code{release} sets the time the signal
342 has to fall below the threshold to reduce the reduction again. Shorter signals
343 than the chosen attack time will be left untouched.
344 The overall reduction of the signal can be made up afterwards with the
345 @code{makeup} setting. So compressing the peaks of a signal about 6dB and
346 raising the makeup to this level results in a signal twice as loud than the
347 source. To gain a softer entry in the compression the @code{knee} flattens the
348 hard edge at the threshold in the range of the chosen decibels.
350 The filter accepts the following options:
354 Set input gain. Default is 1. Range is between 0.015625 and 64.
357 If a signal of second stream rises above this level it will affect the gain
358 reduction of the first stream.
359 By default it is 0.125. Range is between 0.00097563 and 1.
362 Set a ratio by which the signal is reduced. 1:2 means that if the level
363 rose 4dB above the threshold, it will be only 2dB above after the reduction.
364 Default is 2. Range is between 1 and 20.
367 Amount of milliseconds the signal has to rise above the threshold before gain
368 reduction starts. Default is 20. Range is between 0.01 and 2000.
371 Amount of milliseconds the signal has to fall below the threshold before
372 reduction is decreased again. Default is 250. Range is between 0.01 and 9000.
375 Set the amount by how much signal will be amplified after processing.
376 Default is 2. Range is from 1 and 64.
379 Curve the sharp knee around the threshold to enter gain reduction more softly.
380 Default is 2.82843. Range is between 1 and 8.
383 Choose if the @code{average} level between all channels of input stream
384 or the louder(@code{maximum}) channel of input stream affects the
385 reduction. Default is @code{average}.
388 Should the exact signal be taken in case of @code{peak} or an RMS one in case
389 of @code{rms}. Default is @code{rms} which is mostly smoother.
392 How much to use compressed signal in output. Default is 1.
393 Range is between 0 and 1.
398 Apply cross fade from one input audio stream to another input audio stream.
399 The cross fade is applied for specified duration near the end of first stream.
401 The filter accepts the following options:
405 Specify the number of samples for which the cross fade effect has to last.
406 At the end of the cross fade effect the first input audio will be completely
407 silent. Default is 44100.
410 Specify the duration of the cross fade effect. See
411 @ref{time duration syntax,,the Time duration section in the ffmpeg-utils(1) manual,ffmpeg-utils}
412 for the accepted syntax.
413 By default the duration is determined by @var{nb_samples}.
414 If set this option is used instead of @var{nb_samples}.
417 Should first stream end overlap with second stream start. Default is enabled.
420 Set curve for cross fade transition for first stream.
423 Set curve for cross fade transition for second stream.
425 For description of available curve types see @ref{afade} filter description.
432 Cross fade from one input to another:
434 ffmpeg -i first.flac -i second.flac -filter_complex acrossfade=d=10:c1=exp:c2=exp output.flac
438 Cross fade from one input to another but without overlapping:
440 ffmpeg -i first.flac -i second.flac -filter_complex acrossfade=d=10:o=0:c1=exp:c2=exp output.flac
446 Delay one or more audio channels.
448 Samples in delayed channel are filled with silence.
450 The filter accepts the following option:
454 Set list of delays in milliseconds for each channel separated by '|'.
455 At least one delay greater than 0 should be provided.
456 Unused delays will be silently ignored. If number of given delays is
457 smaller than number of channels all remaining channels will not be delayed.
464 Delay first channel by 1.5 seconds, the third channel by 0.5 seconds and leave
465 the second channel (and any other channels that may be present) unchanged.
473 Apply echoing to the input audio.
475 Echoes are reflected sound and can occur naturally amongst mountains
476 (and sometimes large buildings) when talking or shouting; digital echo
477 effects emulate this behaviour and are often used to help fill out the
478 sound of a single instrument or vocal. The time difference between the
479 original signal and the reflection is the @code{delay}, and the
480 loudness of the reflected signal is the @code{decay}.
481 Multiple echoes can have different delays and decays.
483 A description of the accepted parameters follows.
487 Set input gain of reflected signal. Default is @code{0.6}.
490 Set output gain of reflected signal. Default is @code{0.3}.
493 Set list of time intervals in milliseconds between original signal and reflections
494 separated by '|'. Allowed range for each @code{delay} is @code{(0 - 90000.0]}.
495 Default is @code{1000}.
498 Set list of loudnesses of reflected signals separated by '|'.
499 Allowed range for each @code{decay} is @code{(0 - 1.0]}.
500 Default is @code{0.5}.
507 Make it sound as if there are twice as many instruments as are actually playing:
509 aecho=0.8:0.88:60:0.4
513 If delay is very short, then it sound like a (metallic) robot playing music:
519 A longer delay will sound like an open air concert in the mountains:
521 aecho=0.8:0.9:1000:0.3
525 Same as above but with one more mountain:
527 aecho=0.8:0.9:1000|1800:0.3|0.25
532 Audio emphasis filter creates or restores material directly taken from LPs or
533 emphased CDs with different filter curves. E.g. to store music on vinyl the
534 signal has to be altered by a filter first to even out the disadvantages of
535 this recording medium.
536 Once the material is played back the inverse filter has to be applied to
537 restore the distortion of the frequency response.
539 The filter accepts the following options:
549 Set filter mode. For restoring material use @code{reproduction} mode, otherwise
550 use @code{production} mode. Default is @code{reproduction} mode.
553 Set filter type. Selects medium. Can be one of the following:
565 select Compact Disc (CD).
571 select 50µs (FM-KF).
573 select 75µs (FM-KF).
579 Modify an audio signal according to the specified expressions.
581 This filter accepts one or more expressions (one for each channel),
582 which are evaluated and used to modify a corresponding audio signal.
584 It accepts the following parameters:
588 Set the '|'-separated expressions list for each separate channel. If
589 the number of input channels is greater than the number of
590 expressions, the last specified expression is used for the remaining
593 @item channel_layout, c
594 Set output channel layout. If not specified, the channel layout is
595 specified by the number of expressions. If set to @samp{same}, it will
596 use by default the same input channel layout.
599 Each expression in @var{exprs} can contain the following constants and functions:
603 channel number of the current expression
606 number of the evaluated sample, starting from 0
612 time of the evaluated sample expressed in seconds
615 @item nb_out_channels
616 input and output number of channels
619 the value of input channel with number @var{CH}
622 Note: this filter is slow. For faster processing you should use a
631 aeval=val(ch)/2:c=same
635 Invert phase of the second channel:
644 Apply fade-in/out effect to input audio.
646 A description of the accepted parameters follows.
650 Specify the effect type, can be either @code{in} for fade-in, or
651 @code{out} for a fade-out effect. Default is @code{in}.
653 @item start_sample, ss
654 Specify the number of the start sample for starting to apply the fade
655 effect. Default is 0.
658 Specify the number of samples for which the fade effect has to last. At
659 the end of the fade-in effect the output audio will have the same
660 volume as the input audio, at the end of the fade-out transition
661 the output audio will be silence. Default is 44100.
664 Specify the start time of the fade effect. Default is 0.
665 The value must be specified as a time duration; see
666 @ref{time duration syntax,,the Time duration section in the ffmpeg-utils(1) manual,ffmpeg-utils}
667 for the accepted syntax.
668 If set this option is used instead of @var{start_sample}.
671 Specify the duration of the fade effect. See
672 @ref{time duration syntax,,the Time duration section in the ffmpeg-utils(1) manual,ffmpeg-utils}
673 for the accepted syntax.
674 At the end of the fade-in effect the output audio will have the same
675 volume as the input audio, at the end of the fade-out transition
676 the output audio will be silence.
677 By default the duration is determined by @var{nb_samples}.
678 If set this option is used instead of @var{nb_samples}.
681 Set curve for fade transition.
683 It accepts the following values:
686 select triangular, linear slope (default)
688 select quarter of sine wave
690 select half of sine wave
692 select exponential sine wave
696 select inverted parabola
710 select inverted quarter of sine wave
712 select inverted half of sine wave
714 select double-exponential seat
716 select double-exponential sigmoid
724 Fade in first 15 seconds of audio:
730 Fade out last 25 seconds of a 900 seconds audio:
732 afade=t=out:st=875:d=25
739 Set output format constraints for the input audio. The framework will
740 negotiate the most appropriate format to minimize conversions.
742 It accepts the following parameters:
746 A '|'-separated list of requested sample formats.
749 A '|'-separated list of requested sample rates.
751 @item channel_layouts
752 A '|'-separated list of requested channel layouts.
754 See @ref{channel layout syntax,,the Channel Layout section in the ffmpeg-utils(1) manual,ffmpeg-utils}
755 for the required syntax.
758 If a parameter is omitted, all values are allowed.
760 Force the output to either unsigned 8-bit or signed 16-bit stereo
762 aformat=sample_fmts=u8|s16:channel_layouts=stereo
767 A gate is mainly used to reduce lower parts of a signal. This kind of signal
768 processing reduces disturbing noise between useful signals.
770 Gating is done by detecting the volume below a chosen level @var{threshold}
771 and divide it by the factor set with @var{ratio}. The bottom of the noise
772 floor is set via @var{range}. Because an exact manipulation of the signal
773 would cause distortion of the waveform the reduction can be levelled over
774 time. This is done by setting @var{attack} and @var{release}.
776 @var{attack} determines how long the signal has to fall below the threshold
777 before any reduction will occur and @var{release} sets the time the signal
778 has to raise above the threshold to reduce the reduction again.
779 Shorter signals than the chosen attack time will be left untouched.
783 Set input level before filtering.
784 Default is 1. Allowed range is from 0.015625 to 64.
787 Set the level of gain reduction when the signal is below the threshold.
788 Default is 0.06125. Allowed range is from 0 to 1.
791 If a signal rises above this level the gain reduction is released.
792 Default is 0.125. Allowed range is from 0 to 1.
795 Set a ratio about which the signal is reduced.
796 Default is 2. Allowed range is from 1 to 9000.
799 Amount of milliseconds the signal has to rise above the threshold before gain
801 Default is 20 milliseconds. Allowed range is from 0.01 to 9000.
804 Amount of milliseconds the signal has to fall below the threshold before the
805 reduction is increased again. Default is 250 milliseconds.
806 Allowed range is from 0.01 to 9000.
809 Set amount of amplification of signal after processing.
810 Default is 1. Allowed range is from 1 to 64.
813 Curve the sharp knee around the threshold to enter gain reduction more softly.
814 Default is 2.828427125. Allowed range is from 1 to 8.
817 Choose if exact signal should be taken for detection or an RMS like one.
818 Default is rms. Can be peak or rms.
821 Choose if the average level between all channels or the louder channel affects
823 Default is average. Can be average or maximum.
828 The limiter prevents input signal from raising over a desired threshold.
829 This limiter uses lookahead technology to prevent your signal from distorting.
830 It means that there is a small delay after signal is processed. Keep in mind
831 that the delay it produces is the attack time you set.
833 The filter accepts the following options:
837 Set input gain. Default is 1.
840 Set output gain. Default is 1.
843 Don't let signals above this level pass the limiter. Default is 1.
846 The limiter will reach its attenuation level in this amount of time in
847 milliseconds. Default is 5 milliseconds.
850 Come back from limiting to attenuation 1.0 in this amount of milliseconds.
851 Default is 50 milliseconds.
854 When gain reduction is always needed ASC takes care of releasing to an
855 average reduction level rather than reaching a reduction of 0 in the release
859 Select how much the release time is affected by ASC, 0 means nearly no changes
860 in release time while 1 produces higher release times.
863 Auto level output signal. Default is enabled.
864 This normalizes audio back to 0dB if enabled.
867 Depending on picked setting it is recommended to upsample input 2x or 4x times
868 with @ref{aresample} before applying this filter.
872 Apply a two-pole all-pass filter with central frequency (in Hz)
873 @var{frequency}, and filter-width @var{width}.
874 An all-pass filter changes the audio's frequency to phase relationship
875 without changing its frequency to amplitude relationship.
877 The filter accepts the following options:
884 Set method to specify band-width of filter.
897 Specify the band-width of a filter in width_type units.
903 Merge two or more audio streams into a single multi-channel stream.
905 The filter accepts the following options:
910 Set the number of inputs. Default is 2.
914 If the channel layouts of the inputs are disjoint, and therefore compatible,
915 the channel layout of the output will be set accordingly and the channels
916 will be reordered as necessary. If the channel layouts of the inputs are not
917 disjoint, the output will have all the channels of the first input then all
918 the channels of the second input, in that order, and the channel layout of
919 the output will be the default value corresponding to the total number of
922 For example, if the first input is in 2.1 (FL+FR+LF) and the second input
923 is FC+BL+BR, then the output will be in 5.1, with the channels in the
924 following order: a1, a2, b1, a3, b2, b3 (a1 is the first channel of the
925 first input, b1 is the first channel of the second input).
927 On the other hand, if both input are in stereo, the output channels will be
928 in the default order: a1, a2, b1, b2, and the channel layout will be
929 arbitrarily set to 4.0, which may or may not be the expected value.
931 All inputs must have the same sample rate, and format.
933 If inputs do not have the same duration, the output will stop with the
940 Merge two mono files into a stereo stream:
942 amovie=left.wav [l] ; amovie=right.mp3 [r] ; [l] [r] amerge
946 Multiple merges assuming 1 video stream and 6 audio streams in @file{input.mkv}:
948 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
954 Mixes multiple audio inputs into a single output.
956 Note that this filter only supports float samples (the @var{amerge}
957 and @var{pan} audio filters support many formats). If the @var{amix}
958 input has integer samples then @ref{aresample} will be automatically
959 inserted to perform the conversion to float samples.
963 ffmpeg -i INPUT1 -i INPUT2 -i INPUT3 -filter_complex amix=inputs=3:duration=first:dropout_transition=3 OUTPUT
965 will mix 3 input audio streams to a single output with the same duration as the
966 first input and a dropout transition time of 3 seconds.
968 It accepts the following parameters:
972 The number of inputs. If unspecified, it defaults to 2.
975 How to determine the end-of-stream.
979 The duration of the longest input. (default)
982 The duration of the shortest input.
985 The duration of the first input.
989 @item dropout_transition
990 The transition time, in seconds, for volume renormalization when an input
991 stream ends. The default value is 2 seconds.
997 High-order parametric multiband equalizer for each channel.
999 It accepts the following parameters:
1003 This option string is in format:
1004 "c@var{chn} f=@var{cf} w=@var{w} g=@var{g} t=@var{f} | ..."
1005 Each equalizer band is separated by '|'.
1009 Set channel number to which equalization will be applied.
1010 If input doesn't have that channel the entry is ignored.
1013 Set central frequency for band.
1014 If input doesn't have that frequency the entry is ignored.
1017 Set band width in hertz.
1020 Set band gain in dB.
1023 Set filter type for band, optional, can be:
1027 Butterworth, this is default.
1038 With this option activated frequency response of anequalizer is displayed
1042 Set video stream size. Only useful if curves option is activated.
1045 Set max gain that will be displayed. Only useful if curves option is activated.
1046 Setting this to reasonable value allows to display gain which is derived from
1047 neighbour bands which are too close to each other and thus produce higher gain
1048 when both are activated.
1051 Set frequency scale used to draw frequency response in video output.
1052 Can be linear or logarithmic. Default is logarithmic.
1055 Set color for each channel curve which is going to be displayed in video stream.
1056 This is list of color names separated by space or by '|'.
1057 Unrecognised or missing colors will be replaced by white color.
1060 @subsection Examples
1064 Lower gain by 10 of central frequency 200Hz and width 100 Hz
1065 for first 2 channels using Chebyshev type 1 filter:
1067 anequalizer=c0 f=200 w=100 g=-10 t=1|c1 f=200 w=100 g=-10 t=1
1071 @subsection Commands
1073 This filter supports the following commands:
1076 Alter existing filter parameters.
1077 Syntax for the commands is : "@var{fN}|f=@var{freq}|w=@var{width}|g=@var{gain}"
1079 @var{fN} is existing filter number, starting from 0, if no such filter is available
1081 @var{freq} set new frequency parameter.
1082 @var{width} set new width parameter in herz.
1083 @var{gain} set new gain parameter in dB.
1085 Full filter invocation with asendcmd may look like this:
1086 asendcmd=c='4.0 anequalizer change 0|f=200|w=50|g=1',anequalizer=...
1091 Pass the audio source unchanged to the output.
1095 Pad the end of an audio stream with silence.
1097 This can be used together with @command{ffmpeg} @option{-shortest} to
1098 extend audio streams to the same length as the video stream.
1100 A description of the accepted options follows.
1104 Set silence packet size. Default value is 4096.
1107 Set the number of samples of silence to add to the end. After the
1108 value is reached, the stream is terminated. This option is mutually
1109 exclusive with @option{whole_len}.
1112 Set the minimum total number of samples in the output audio stream. If
1113 the value is longer than the input audio length, silence is added to
1114 the end, until the value is reached. This option is mutually exclusive
1115 with @option{pad_len}.
1118 If neither the @option{pad_len} nor the @option{whole_len} option is
1119 set, the filter will add silence to the end of the input stream
1122 @subsection Examples
1126 Add 1024 samples of silence to the end of the input:
1132 Make sure the audio output will contain at least 10000 samples, pad
1133 the input with silence if required:
1135 apad=whole_len=10000
1139 Use @command{ffmpeg} to pad the audio input with silence, so that the
1140 video stream will always result the shortest and will be converted
1141 until the end in the output file when using the @option{shortest}
1144 ffmpeg -i VIDEO -i AUDIO -filter_complex "[1:0]apad" -shortest OUTPUT
1149 Add a phasing effect to the input audio.
1151 A phaser filter creates series of peaks and troughs in the frequency spectrum.
1152 The position of the peaks and troughs are modulated so that they vary over time, creating a sweeping effect.
1154 A description of the accepted parameters follows.
1158 Set input gain. Default is 0.4.
1161 Set output gain. Default is 0.74
1164 Set delay in milliseconds. Default is 3.0.
1167 Set decay. Default is 0.4.
1170 Set modulation speed in Hz. Default is 0.5.
1173 Set modulation type. Default is triangular.
1175 It accepts the following values:
1184 Audio pulsator is something between an autopanner and a tremolo.
1185 But it can produce funny stereo effects as well. Pulsator changes the volume
1186 of the left and right channel based on a LFO (low frequency oscillator) with
1187 different waveforms and shifted phases.
1188 This filter have the ability to define an offset between left and right
1189 channel. An offset of 0 means that both LFO shapes match each other.
1190 The left and right channel are altered equally - a conventional tremolo.
1191 An offset of 50% means that the shape of the right channel is exactly shifted
1192 in phase (or moved backwards about half of the frequency) - pulsator acts as
1193 an autopanner. At 1 both curves match again. Every setting in between moves the
1194 phase shift gapless between all stages and produces some "bypassing" sounds with
1195 sine and triangle waveforms. The more you set the offset near 1 (starting from
1196 the 0.5) the faster the signal passes from the left to the right speaker.
1198 The filter accepts the following options:
1202 Set input gain. By default it is 1. Range is [0.015625 - 64].
1205 Set output gain. By default it is 1. Range is [0.015625 - 64].
1208 Set waveform shape the LFO will use. Can be one of: sine, triangle, square,
1209 sawup or sawdown. Default is sine.
1212 Set modulation. Define how much of original signal is affected by the LFO.
1215 Set left channel offset. Default is 0. Allowed range is [0 - 1].
1218 Set right channel offset. Default is 0.5. Allowed range is [0 - 1].
1221 Set pulse width. Default is 1. Allowed range is [0 - 2].
1224 Set possible timing mode. Can be one of: bpm, ms or hz. Default is hz.
1227 Set bpm. Default is 120. Allowed range is [30 - 300]. Only used if timing
1231 Set ms. Default is 500. Allowed range is [10 - 2000]. Only used if timing
1235 Set frequency in Hz. Default is 2. Allowed range is [0.01 - 100]. Only used
1236 if timing is set to hz.
1242 Resample the input audio to the specified parameters, using the
1243 libswresample library. If none are specified then the filter will
1244 automatically convert between its input and output.
1246 This filter is also able to stretch/squeeze the audio data to make it match
1247 the timestamps or to inject silence / cut out audio to make it match the
1248 timestamps, do a combination of both or do neither.
1250 The filter accepts the syntax
1251 [@var{sample_rate}:]@var{resampler_options}, where @var{sample_rate}
1252 expresses a sample rate and @var{resampler_options} is a list of
1253 @var{key}=@var{value} pairs, separated by ":". See the
1254 ffmpeg-resampler manual for the complete list of supported options.
1256 @subsection Examples
1260 Resample the input audio to 44100Hz:
1266 Stretch/squeeze samples to the given timestamps, with a maximum of 1000
1267 samples per second compensation:
1269 aresample=async=1000
1273 @section asetnsamples
1275 Set the number of samples per each output audio frame.
1277 The last output packet may contain a different number of samples, as
1278 the filter will flush all the remaining samples when the input audio
1281 The filter accepts the following options:
1285 @item nb_out_samples, n
1286 Set the number of frames per each output audio frame. The number is
1287 intended as the number of samples @emph{per each channel}.
1288 Default value is 1024.
1291 If set to 1, the filter will pad the last audio frame with zeroes, so
1292 that the last frame will contain the same number of samples as the
1293 previous ones. Default value is 1.
1296 For example, to set the number of per-frame samples to 1234 and
1297 disable padding for the last frame, use:
1299 asetnsamples=n=1234:p=0
1304 Set the sample rate without altering the PCM data.
1305 This will result in a change of speed and pitch.
1307 The filter accepts the following options:
1310 @item sample_rate, r
1311 Set the output sample rate. Default is 44100 Hz.
1316 Show a line containing various information for each input audio frame.
1317 The input audio is not modified.
1319 The shown line contains a sequence of key/value pairs of the form
1320 @var{key}:@var{value}.
1322 The following values are shown in the output:
1326 The (sequential) number of the input frame, starting from 0.
1329 The presentation timestamp of the input frame, in time base units; the time base
1330 depends on the filter input pad, and is usually 1/@var{sample_rate}.
1333 The presentation timestamp of the input frame in seconds.
1336 position of the frame in the input stream, -1 if this information in
1337 unavailable and/or meaningless (for example in case of synthetic audio)
1346 The sample rate for the audio frame.
1349 The number of samples (per channel) in the frame.
1352 The Adler-32 checksum (printed in hexadecimal) of the audio data. For planar
1353 audio, the data is treated as if all the planes were concatenated.
1355 @item plane_checksums
1356 A list of Adler-32 checksums for each data plane.
1362 Display time domain statistical information about the audio channels.
1363 Statistics are calculated and displayed for each audio channel and,
1364 where applicable, an overall figure is also given.
1366 It accepts the following option:
1369 Short window length in seconds, used for peak and trough RMS measurement.
1370 Default is @code{0.05} (50 milliseconds). Allowed range is @code{[0.1 - 10]}.
1374 Set metadata injection. All the metadata keys are prefixed with @code{lavfi.astats.X},
1375 where @code{X} is channel number starting from 1 or string @code{Overall}. Default is
1378 Available keys for each channel are:
1409 For example full key look like this @code{lavfi.astats.1.DC_offset} or
1410 this @code{lavfi.astats.Overall.Peak_count}.
1412 For description what each key means read below.
1415 Set number of frame after which stats are going to be recalculated.
1416 Default is disabled.
1419 A description of each shown parameter follows:
1423 Mean amplitude displacement from zero.
1426 Minimal sample level.
1429 Maximal sample level.
1431 @item Min difference
1432 Minimal difference between two consecutive samples.
1434 @item Max difference
1435 Maximal difference between two consecutive samples.
1437 @item Mean difference
1438 Mean difference between two consecutive samples.
1439 The average of each difference between two consecutive samples.
1443 Standard peak and RMS level measured in dBFS.
1447 Peak and trough values for RMS level measured over a short window.
1450 Standard ratio of peak to RMS level (note: not in dB).
1453 Flatness (i.e. consecutive samples with the same value) of the signal at its peak levels
1454 (i.e. either @var{Min level} or @var{Max level}).
1457 Number of occasions (not the number of samples) that the signal attained either
1458 @var{Min level} or @var{Max level}.
1461 Overall bit depth of audio. Number of bits used for each sample.
1466 Synchronize audio data with timestamps by squeezing/stretching it and/or
1467 dropping samples/adding silence when needed.
1469 This filter is not built by default, please use @ref{aresample} to do squeezing/stretching.
1471 It accepts the following parameters:
1475 Enable stretching/squeezing the data to make it match the timestamps. Disabled
1476 by default. When disabled, time gaps are covered with silence.
1479 The minimum difference between timestamps and audio data (in seconds) to trigger
1480 adding/dropping samples. The default value is 0.1. If you get an imperfect
1481 sync with this filter, try setting this parameter to 0.
1484 The maximum compensation in samples per second. Only relevant with compensate=1.
1485 The default value is 500.
1488 Assume that the first PTS should be this value. The time base is 1 / sample
1489 rate. This allows for padding/trimming at the start of the stream. By default,
1490 no assumption is made about the first frame's expected PTS, so no padding or
1491 trimming is done. For example, this could be set to 0 to pad the beginning with
1492 silence if an audio stream starts after the video stream or to trim any samples
1493 with a negative PTS due to encoder delay.
1501 The filter accepts exactly one parameter, the audio tempo. If not
1502 specified then the filter will assume nominal 1.0 tempo. Tempo must
1503 be in the [0.5, 2.0] range.
1505 @subsection Examples
1509 Slow down audio to 80% tempo:
1515 To speed up audio to 125% tempo:
1523 Trim the input so that the output contains one continuous subpart of the input.
1525 It accepts the following parameters:
1528 Timestamp (in seconds) of the start of the section to keep. I.e. the audio
1529 sample with the timestamp @var{start} will be the first sample in the output.
1532 Specify time of the first audio sample that will be dropped, i.e. the
1533 audio sample immediately preceding the one with the timestamp @var{end} will be
1534 the last sample in the output.
1537 Same as @var{start}, except this option sets the start timestamp in samples
1541 Same as @var{end}, except this option sets the end timestamp in samples instead
1545 The maximum duration of the output in seconds.
1548 The number of the first sample that should be output.
1551 The number of the first sample that should be dropped.
1554 @option{start}, @option{end}, and @option{duration} are expressed as time
1555 duration specifications; see
1556 @ref{time duration syntax,,the Time duration section in the ffmpeg-utils(1) manual,ffmpeg-utils}.
1558 Note that the first two sets of the start/end options and the @option{duration}
1559 option look at the frame timestamp, while the _sample options simply count the
1560 samples that pass through the filter. So start/end_pts and start/end_sample will
1561 give different results when the timestamps are wrong, inexact or do not start at
1562 zero. Also note that this filter does not modify the timestamps. If you wish
1563 to have the output timestamps start at zero, insert the asetpts filter after the
1566 If multiple start or end options are set, this filter tries to be greedy and
1567 keep all samples that match at least one of the specified constraints. To keep
1568 only the part that matches all the constraints at once, chain multiple atrim
1571 The defaults are such that all the input is kept. So it is possible to set e.g.
1572 just the end values to keep everything before the specified time.
1577 Drop everything except the second minute of input:
1579 ffmpeg -i INPUT -af atrim=60:120
1583 Keep only the first 1000 samples:
1585 ffmpeg -i INPUT -af atrim=end_sample=1000
1592 Apply a two-pole Butterworth band-pass filter with central
1593 frequency @var{frequency}, and (3dB-point) band-width width.
1594 The @var{csg} option selects a constant skirt gain (peak gain = Q)
1595 instead of the default: constant 0dB peak gain.
1596 The filter roll off at 6dB per octave (20dB per decade).
1598 The filter accepts the following options:
1602 Set the filter's central frequency. Default is @code{3000}.
1605 Constant skirt gain if set to 1. Defaults to 0.
1608 Set method to specify band-width of filter.
1621 Specify the band-width of a filter in width_type units.
1626 Apply a two-pole Butterworth band-reject filter with central
1627 frequency @var{frequency}, and (3dB-point) band-width @var{width}.
1628 The filter roll off at 6dB per octave (20dB per decade).
1630 The filter accepts the following options:
1634 Set the filter's central frequency. Default is @code{3000}.
1637 Set method to specify band-width of filter.
1650 Specify the band-width of a filter in width_type units.
1655 Boost or cut the bass (lower) frequencies of the audio using a two-pole
1656 shelving filter with a response similar to that of a standard
1657 hi-fi's tone-controls. This is also known as shelving equalisation (EQ).
1659 The filter accepts the following options:
1663 Give the gain at 0 Hz. Its useful range is about -20
1664 (for a large cut) to +20 (for a large boost).
1665 Beware of clipping when using a positive gain.
1668 Set the filter's central frequency and so can be used
1669 to extend or reduce the frequency range to be boosted or cut.
1670 The default value is @code{100} Hz.
1673 Set method to specify band-width of filter.
1686 Determine how steep is the filter's shelf transition.
1691 Apply a biquad IIR filter with the given coefficients.
1692 Where @var{b0}, @var{b1}, @var{b2} and @var{a0}, @var{a1}, @var{a2}
1693 are the numerator and denominator coefficients respectively.
1696 Bauer stereo to binaural transformation, which improves headphone listening of
1697 stereo audio records.
1699 It accepts the following parameters:
1703 Pre-defined crossfeed level.
1707 Default level (fcut=700, feed=50).
1710 Chu Moy circuit (fcut=700, feed=60).
1713 Jan Meier circuit (fcut=650, feed=95).
1718 Cut frequency (in Hz).
1727 Remap input channels to new locations.
1729 It accepts the following parameters:
1731 @item channel_layout
1732 The channel layout of the output stream.
1735 Map channels from input to output. The argument is a '|'-separated list of
1736 mappings, each in the @code{@var{in_channel}-@var{out_channel}} or
1737 @var{in_channel} form. @var{in_channel} can be either the name of the input
1738 channel (e.g. FL for front left) or its index in the input channel layout.
1739 @var{out_channel} is the name of the output channel or its index in the output
1740 channel layout. If @var{out_channel} is not given then it is implicitly an
1741 index, starting with zero and increasing by one for each mapping.
1744 If no mapping is present, the filter will implicitly map input channels to
1745 output channels, preserving indices.
1747 For example, assuming a 5.1+downmix input MOV file,
1749 ffmpeg -i in.mov -filter 'channelmap=map=DL-FL|DR-FR' out.wav
1751 will create an output WAV file tagged as stereo from the downmix channels of
1754 To fix a 5.1 WAV improperly encoded in AAC's native channel order
1756 ffmpeg -i in.wav -filter 'channelmap=1|2|0|5|3|4:5.1' out.wav
1759 @section channelsplit
1761 Split each channel from an input audio stream into a separate output stream.
1763 It accepts the following parameters:
1765 @item channel_layout
1766 The channel layout of the input stream. The default is "stereo".
1769 For example, assuming a stereo input MP3 file,
1771 ffmpeg -i in.mp3 -filter_complex channelsplit out.mkv
1773 will create an output Matroska file with two audio streams, one containing only
1774 the left channel and the other the right channel.
1776 Split a 5.1 WAV file into per-channel files:
1778 ffmpeg -i in.wav -filter_complex
1779 'channelsplit=channel_layout=5.1[FL][FR][FC][LFE][SL][SR]'
1780 -map '[FL]' front_left.wav -map '[FR]' front_right.wav -map '[FC]'
1781 front_center.wav -map '[LFE]' lfe.wav -map '[SL]' side_left.wav -map '[SR]'
1786 Add a chorus effect to the audio.
1788 Can make a single vocal sound like a chorus, but can also be applied to instrumentation.
1790 Chorus resembles an echo effect with a short delay, but whereas with echo the delay is
1791 constant, with chorus, it is varied using using sinusoidal or triangular modulation.
1792 The modulation depth defines the range the modulated delay is played before or after
1793 the delay. Hence the delayed sound will sound slower or faster, that is the delayed
1794 sound tuned around the original one, like in a chorus where some vocals are slightly
1797 It accepts the following parameters:
1800 Set input gain. Default is 0.4.
1803 Set output gain. Default is 0.4.
1806 Set delays. A typical delay is around 40ms to 60ms.
1818 @subsection Examples
1824 chorus=0.7:0.9:55:0.4:0.25:2
1830 chorus=0.6:0.9:50|60:0.4|0.32:0.25|0.4:2|1.3
1834 Fuller sounding chorus with three delays:
1836 chorus=0.5:0.9:50|60|40:0.4|0.32|0.3:0.25|0.4|0.3:2|2.3|1.3
1841 Compress or expand the audio's dynamic range.
1843 It accepts the following parameters:
1849 A list of times in seconds for each channel over which the instantaneous level
1850 of the input signal is averaged to determine its volume. @var{attacks} refers to
1851 increase of volume and @var{decays} refers to decrease of volume. For most
1852 situations, the attack time (response to the audio getting louder) should be
1853 shorter than the decay time, because the human ear is more sensitive to sudden
1854 loud audio than sudden soft audio. A typical value for attack is 0.3 seconds and
1855 a typical value for decay is 0.8 seconds.
1856 If specified number of attacks & decays is lower than number of channels, the last
1857 set attack/decay will be used for all remaining channels.
1860 A list of points for the transfer function, specified in dB relative to the
1861 maximum possible signal amplitude. Each key points list must be defined using
1862 the following syntax: @code{x0/y0|x1/y1|x2/y2|....} or
1863 @code{x0/y0 x1/y1 x2/y2 ....}
1865 The input values must be in strictly increasing order but the transfer function
1866 does not have to be monotonically rising. The point @code{0/0} is assumed but
1867 may be overridden (by @code{0/out-dBn}). Typical values for the transfer
1868 function are @code{-70/-70|-60/-20}.
1871 Set the curve radius in dB for all joints. It defaults to 0.01.
1874 Set the additional gain in dB to be applied at all points on the transfer
1875 function. This allows for easy adjustment of the overall gain.
1879 Set an initial volume, in dB, to be assumed for each channel when filtering
1880 starts. This permits the user to supply a nominal level initially, so that, for
1881 example, a very large gain is not applied to initial signal levels before the
1882 companding has begun to operate. A typical value for audio which is initially
1883 quiet is -90 dB. It defaults to 0.
1886 Set a delay, in seconds. The input audio is analyzed immediately, but audio is
1887 delayed before being fed to the volume adjuster. Specifying a delay
1888 approximately equal to the attack/decay times allows the filter to effectively
1889 operate in predictive rather than reactive mode. It defaults to 0.
1893 @subsection Examples
1897 Make music with both quiet and loud passages suitable for listening to in a
1900 compand=.3|.3:1|1:-90/-60|-60/-40|-40/-30|-20/-20:6:0:-90:0.2
1903 Another example for audio with whisper and explosion parts:
1905 compand=0|0:1|1:-90/-900|-70/-70|-30/-9|0/-3:6:0:0:0
1909 A noise gate for when the noise is at a lower level than the signal:
1911 compand=.1|.1:.2|.2:-900/-900|-50.1/-900|-50/-50:.01:0:-90:.1
1915 Here is another noise gate, this time for when the noise is at a higher level
1916 than the signal (making it, in some ways, similar to squelch):
1918 compand=.1|.1:.1|.1:-45.1/-45.1|-45/-900|0/-900:.01:45:-90:.1
1922 2:1 compression starting at -6dB:
1924 compand=points=-80/-80|-6/-6|0/-3.8|20/3.5
1928 2:1 compression starting at -9dB:
1930 compand=points=-80/-80|-9/-9|0/-5.3|20/2.9
1934 2:1 compression starting at -12dB:
1936 compand=points=-80/-80|-12/-12|0/-6.8|20/1.9
1940 2:1 compression starting at -18dB:
1942 compand=points=-80/-80|-18/-18|0/-9.8|20/0.7
1946 3:1 compression starting at -15dB:
1948 compand=points=-80/-80|-15/-15|0/-10.8|20/-5.2
1954 compand=points=-80/-105|-62/-80|-15.4/-15.4|0/-12|20/-7.6
1960 compand=attacks=0:points=-80/-169|-54/-80|-49.5/-64.6|-41.1/-41.1|-25.8/-15|-10.8/-4.5|0/0|20/8.3
1964 Hard limiter at -6dB:
1966 compand=attacks=0:points=-80/-80|-6/-6|20/-6
1970 Hard limiter at -12dB:
1972 compand=attacks=0:points=-80/-80|-12/-12|20/-12
1976 Hard noise gate at -35 dB:
1978 compand=attacks=0:points=-80/-115|-35.1/-80|-35/-35|20/20
1984 compand=attacks=0:points=-80/-80|-12.4/-12.4|-6/-8|0/-6.8|20/-2.8
1988 @section compensationdelay
1990 Compensation Delay Line is a metric based delay to compensate differing
1991 positions of microphones or speakers.
1993 For example, you have recorded guitar with two microphones placed in
1994 different location. Because the front of sound wave has fixed speed in
1995 normal conditions, the phasing of microphones can vary and depends on
1996 their location and interposition. The best sound mix can be achieved when
1997 these microphones are in phase (synchronized). Note that distance of
1998 ~30 cm between microphones makes one microphone to capture signal in
1999 antiphase to another microphone. That makes the final mix sounding moody.
2000 This filter helps to solve phasing problems by adding different delays
2001 to each microphone track and make them synchronized.
2003 The best result can be reached when you take one track as base and
2004 synchronize other tracks one by one with it.
2005 Remember that synchronization/delay tolerance depends on sample rate, too.
2006 Higher sample rates will give more tolerance.
2008 It accepts the following parameters:
2012 Set millimeters distance. This is compensation distance for fine tuning.
2016 Set cm distance. This is compensation distance for tightening distance setup.
2020 Set meters distance. This is compensation distance for hard distance setup.
2024 Set dry amount. Amount of unprocessed (dry) signal.
2028 Set wet amount. Amount of processed (wet) signal.
2032 Set temperature degree in Celsius. This is the temperature of the environment.
2037 Apply a DC shift to the audio.
2039 This can be useful to remove a DC offset (caused perhaps by a hardware problem
2040 in the recording chain) from the audio. The effect of a DC offset is reduced
2041 headroom and hence volume. The @ref{astats} filter can be used to determine if
2042 a signal has a DC offset.
2046 Set the DC shift, allowed range is [-1, 1]. It indicates the amount to shift
2050 Optional. It should have a value much less than 1 (e.g. 0.05 or 0.02) and is
2051 used to prevent clipping.
2055 Dynamic Audio Normalizer.
2057 This filter applies a certain amount of gain to the input audio in order
2058 to bring its peak magnitude to a target level (e.g. 0 dBFS). However, in
2059 contrast to more "simple" normalization algorithms, the Dynamic Audio
2060 Normalizer *dynamically* re-adjusts the gain factor to the input audio.
2061 This allows for applying extra gain to the "quiet" sections of the audio
2062 while avoiding distortions or clipping the "loud" sections. In other words:
2063 The Dynamic Audio Normalizer will "even out" the volume of quiet and loud
2064 sections, in the sense that the volume of each section is brought to the
2065 same target level. Note, however, that the Dynamic Audio Normalizer achieves
2066 this goal *without* applying "dynamic range compressing". It will retain 100%
2067 of the dynamic range *within* each section of the audio file.
2071 Set the frame length in milliseconds. In range from 10 to 8000 milliseconds.
2072 Default is 500 milliseconds.
2073 The Dynamic Audio Normalizer processes the input audio in small chunks,
2074 referred to as frames. This is required, because a peak magnitude has no
2075 meaning for just a single sample value. Instead, we need to determine the
2076 peak magnitude for a contiguous sequence of sample values. While a "standard"
2077 normalizer would simply use the peak magnitude of the complete file, the
2078 Dynamic Audio Normalizer determines the peak magnitude individually for each
2079 frame. The length of a frame is specified in milliseconds. By default, the
2080 Dynamic Audio Normalizer uses a frame length of 500 milliseconds, which has
2081 been found to give good results with most files.
2082 Note that the exact frame length, in number of samples, will be determined
2083 automatically, based on the sampling rate of the individual input audio file.
2086 Set the Gaussian filter window size. In range from 3 to 301, must be odd
2087 number. Default is 31.
2088 Probably the most important parameter of the Dynamic Audio Normalizer is the
2089 @code{window size} of the Gaussian smoothing filter. The filter's window size
2090 is specified in frames, centered around the current frame. For the sake of
2091 simplicity, this must be an odd number. Consequently, the default value of 31
2092 takes into account the current frame, as well as the 15 preceding frames and
2093 the 15 subsequent frames. Using a larger window results in a stronger
2094 smoothing effect and thus in less gain variation, i.e. slower gain
2095 adaptation. Conversely, using a smaller window results in a weaker smoothing
2096 effect and thus in more gain variation, i.e. faster gain adaptation.
2097 In other words, the more you increase this value, the more the Dynamic Audio
2098 Normalizer will behave like a "traditional" normalization filter. On the
2099 contrary, the more you decrease this value, the more the Dynamic Audio
2100 Normalizer will behave like a dynamic range compressor.
2103 Set the target peak value. This specifies the highest permissible magnitude
2104 level for the normalized audio input. This filter will try to approach the
2105 target peak magnitude as closely as possible, but at the same time it also
2106 makes sure that the normalized signal will never exceed the peak magnitude.
2107 A frame's maximum local gain factor is imposed directly by the target peak
2108 magnitude. The default value is 0.95 and thus leaves a headroom of 5%*.
2109 It is not recommended to go above this value.
2112 Set the maximum gain factor. In range from 1.0 to 100.0. Default is 10.0.
2113 The Dynamic Audio Normalizer determines the maximum possible (local) gain
2114 factor for each input frame, i.e. the maximum gain factor that does not
2115 result in clipping or distortion. The maximum gain factor is determined by
2116 the frame's highest magnitude sample. However, the Dynamic Audio Normalizer
2117 additionally bounds the frame's maximum gain factor by a predetermined
2118 (global) maximum gain factor. This is done in order to avoid excessive gain
2119 factors in "silent" or almost silent frames. By default, the maximum gain
2120 factor is 10.0, For most inputs the default value should be sufficient and
2121 it usually is not recommended to increase this value. Though, for input
2122 with an extremely low overall volume level, it may be necessary to allow even
2123 higher gain factors. Note, however, that the Dynamic Audio Normalizer does
2124 not simply apply a "hard" threshold (i.e. cut off values above the threshold).
2125 Instead, a "sigmoid" threshold function will be applied. This way, the
2126 gain factors will smoothly approach the threshold value, but never exceed that
2130 Set the target RMS. In range from 0.0 to 1.0. Default is 0.0 - disabled.
2131 By default, the Dynamic Audio Normalizer performs "peak" normalization.
2132 This means that the maximum local gain factor for each frame is defined
2133 (only) by the frame's highest magnitude sample. This way, the samples can
2134 be amplified as much as possible without exceeding the maximum signal
2135 level, i.e. without clipping. Optionally, however, the Dynamic Audio
2136 Normalizer can also take into account the frame's root mean square,
2137 abbreviated RMS. In electrical engineering, the RMS is commonly used to
2138 determine the power of a time-varying signal. It is therefore considered
2139 that the RMS is a better approximation of the "perceived loudness" than
2140 just looking at the signal's peak magnitude. Consequently, by adjusting all
2141 frames to a constant RMS value, a uniform "perceived loudness" can be
2142 established. If a target RMS value has been specified, a frame's local gain
2143 factor is defined as the factor that would result in exactly that RMS value.
2144 Note, however, that the maximum local gain factor is still restricted by the
2145 frame's highest magnitude sample, in order to prevent clipping.
2148 Enable channels coupling. By default is enabled.
2149 By default, the Dynamic Audio Normalizer will amplify all channels by the same
2150 amount. This means the same gain factor will be applied to all channels, i.e.
2151 the maximum possible gain factor is determined by the "loudest" channel.
2152 However, in some recordings, it may happen that the volume of the different
2153 channels is uneven, e.g. one channel may be "quieter" than the other one(s).
2154 In this case, this option can be used to disable the channel coupling. This way,
2155 the gain factor will be determined independently for each channel, depending
2156 only on the individual channel's highest magnitude sample. This allows for
2157 harmonizing the volume of the different channels.
2160 Enable DC bias correction. By default is disabled.
2161 An audio signal (in the time domain) is a sequence of sample values.
2162 In the Dynamic Audio Normalizer these sample values are represented in the
2163 -1.0 to 1.0 range, regardless of the original input format. Normally, the
2164 audio signal, or "waveform", should be centered around the zero point.
2165 That means if we calculate the mean value of all samples in a file, or in a
2166 single frame, then the result should be 0.0 or at least very close to that
2167 value. If, however, there is a significant deviation of the mean value from
2168 0.0, in either positive or negative direction, this is referred to as a
2169 DC bias or DC offset. Since a DC bias is clearly undesirable, the Dynamic
2170 Audio Normalizer provides optional DC bias correction.
2171 With DC bias correction enabled, the Dynamic Audio Normalizer will determine
2172 the mean value, or "DC correction" offset, of each input frame and subtract
2173 that value from all of the frame's sample values which ensures those samples
2174 are centered around 0.0 again. Also, in order to avoid "gaps" at the frame
2175 boundaries, the DC correction offset values will be interpolated smoothly
2176 between neighbouring frames.
2179 Enable alternative boundary mode. By default is disabled.
2180 The Dynamic Audio Normalizer takes into account a certain neighbourhood
2181 around each frame. This includes the preceding frames as well as the
2182 subsequent frames. However, for the "boundary" frames, located at the very
2183 beginning and at the very end of the audio file, not all neighbouring
2184 frames are available. In particular, for the first few frames in the audio
2185 file, the preceding frames are not known. And, similarly, for the last few
2186 frames in the audio file, the subsequent frames are not known. Thus, the
2187 question arises which gain factors should be assumed for the missing frames
2188 in the "boundary" region. The Dynamic Audio Normalizer implements two modes
2189 to deal with this situation. The default boundary mode assumes a gain factor
2190 of exactly 1.0 for the missing frames, resulting in a smooth "fade in" and
2191 "fade out" at the beginning and at the end of the input, respectively.
2194 Set the compress factor. In range from 0.0 to 30.0. Default is 0.0.
2195 By default, the Dynamic Audio Normalizer does not apply "traditional"
2196 compression. This means that signal peaks will not be pruned and thus the
2197 full dynamic range will be retained within each local neighbourhood. However,
2198 in some cases it may be desirable to combine the Dynamic Audio Normalizer's
2199 normalization algorithm with a more "traditional" compression.
2200 For this purpose, the Dynamic Audio Normalizer provides an optional compression
2201 (thresholding) function. If (and only if) the compression feature is enabled,
2202 all input frames will be processed by a soft knee thresholding function prior
2203 to the actual normalization process. Put simply, the thresholding function is
2204 going to prune all samples whose magnitude exceeds a certain threshold value.
2205 However, the Dynamic Audio Normalizer does not simply apply a fixed threshold
2206 value. Instead, the threshold value will be adjusted for each individual
2208 In general, smaller parameters result in stronger compression, and vice versa.
2209 Values below 3.0 are not recommended, because audible distortion may appear.
2214 Make audio easier to listen to on headphones.
2216 This filter adds `cues' to 44.1kHz stereo (i.e. audio CD format) audio
2217 so that when listened to on headphones the stereo image is moved from
2218 inside your head (standard for headphones) to outside and in front of
2219 the listener (standard for speakers).
2225 Apply a two-pole peaking equalisation (EQ) filter. With this
2226 filter, the signal-level at and around a selected frequency can
2227 be increased or decreased, whilst (unlike bandpass and bandreject
2228 filters) that at all other frequencies is unchanged.
2230 In order to produce complex equalisation curves, this filter can
2231 be given several times, each with a different central frequency.
2233 The filter accepts the following options:
2237 Set the filter's central frequency in Hz.
2240 Set method to specify band-width of filter.
2253 Specify the band-width of a filter in width_type units.
2256 Set the required gain or attenuation in dB.
2257 Beware of clipping when using a positive gain.
2260 @subsection Examples
2263 Attenuate 10 dB at 1000 Hz, with a bandwidth of 200 Hz:
2265 equalizer=f=1000:width_type=h:width=200:g=-10
2269 Apply 2 dB gain at 1000 Hz with Q 1 and attenuate 5 dB at 100 Hz with Q 2:
2271 equalizer=f=1000:width_type=q:width=1:g=2,equalizer=f=100:width_type=q:width=2:g=-5
2275 @section extrastereo
2277 Linearly increases the difference between left and right channels which
2278 adds some sort of "live" effect to playback.
2280 The filter accepts the following option:
2284 Sets the difference coefficient (default: 2.5). 0.0 means mono sound
2285 (average of both channels), with 1.0 sound will be unchanged, with
2286 -1.0 left and right channels will be swapped.
2289 Enable clipping. By default is enabled.
2293 Apply a flanging effect to the audio.
2295 The filter accepts the following options:
2299 Set base delay in milliseconds. Range from 0 to 30. Default value is 0.
2302 Set added swep delay in milliseconds. Range from 0 to 10. Default value is 2.
2305 Set percentage regeneration (delayed signal feedback). Range from -95 to 95.
2309 Set percentage of delayed signal mixed with original. Range from 0 to 100.
2310 Default value is 71.
2313 Set sweeps per second (Hz). Range from 0.1 to 10. Default value is 0.5.
2316 Set swept wave shape, can be @var{triangular} or @var{sinusoidal}.
2317 Default value is @var{sinusoidal}.
2320 Set swept wave percentage-shift for multi channel. Range from 0 to 100.
2321 Default value is 25.
2324 Set delay-line interpolation, @var{linear} or @var{quadratic}.
2325 Default is @var{linear}.
2330 Apply a high-pass filter with 3dB point frequency.
2331 The filter can be either single-pole, or double-pole (the default).
2332 The filter roll off at 6dB per pole per octave (20dB per pole per decade).
2334 The filter accepts the following options:
2338 Set frequency in Hz. Default is 3000.
2341 Set number of poles. Default is 2.
2344 Set method to specify band-width of filter.
2357 Specify the band-width of a filter in width_type units.
2358 Applies only to double-pole filter.
2359 The default is 0.707q and gives a Butterworth response.
2364 Join multiple input streams into one multi-channel stream.
2366 It accepts the following parameters:
2370 The number of input streams. It defaults to 2.
2372 @item channel_layout
2373 The desired output channel layout. It defaults to stereo.
2376 Map channels from inputs to output. The argument is a '|'-separated list of
2377 mappings, each in the @code{@var{input_idx}.@var{in_channel}-@var{out_channel}}
2378 form. @var{input_idx} is the 0-based index of the input stream. @var{in_channel}
2379 can be either the name of the input channel (e.g. FL for front left) or its
2380 index in the specified input stream. @var{out_channel} is the name of the output
2384 The filter will attempt to guess the mappings when they are not specified
2385 explicitly. It does so by first trying to find an unused matching input channel
2386 and if that fails it picks the first unused input channel.
2388 Join 3 inputs (with properly set channel layouts):
2390 ffmpeg -i INPUT1 -i INPUT2 -i INPUT3 -filter_complex join=inputs=3 OUTPUT
2393 Build a 5.1 output from 6 single-channel streams:
2395 ffmpeg -i fl -i fr -i fc -i sl -i sr -i lfe -filter_complex
2396 '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'
2402 Load a LADSPA (Linux Audio Developer's Simple Plugin API) plugin.
2404 To enable compilation of this filter you need to configure FFmpeg with
2405 @code{--enable-ladspa}.
2409 Specifies the name of LADSPA plugin library to load. If the environment
2410 variable @env{LADSPA_PATH} is defined, the LADSPA plugin is searched in
2411 each one of the directories specified by the colon separated list in
2412 @env{LADSPA_PATH}, otherwise in the standard LADSPA paths, which are in
2413 this order: @file{HOME/.ladspa/lib/}, @file{/usr/local/lib/ladspa/},
2414 @file{/usr/lib/ladspa/}.
2417 Specifies the plugin within the library. Some libraries contain only
2418 one plugin, but others contain many of them. If this is not set filter
2419 will list all available plugins within the specified library.
2422 Set the '|' separated list of controls which are zero or more floating point
2423 values that determine the behavior of the loaded plugin (for example delay,
2425 Controls need to be defined using the following syntax:
2426 c0=@var{value0}|c1=@var{value1}|c2=@var{value2}|..., where
2427 @var{valuei} is the value set on the @var{i}-th control.
2428 Alternatively they can be also defined using the following syntax:
2429 @var{value0}|@var{value1}|@var{value2}|..., where
2430 @var{valuei} is the value set on the @var{i}-th control.
2431 If @option{controls} is set to @code{help}, all available controls and
2432 their valid ranges are printed.
2434 @item sample_rate, s
2435 Specify the sample rate, default to 44100. Only used if plugin have
2439 Set the number of samples per channel per each output frame, default
2440 is 1024. Only used if plugin have zero inputs.
2443 Set the minimum duration of the sourced audio. See
2444 @ref{time duration syntax,,the Time duration section in the ffmpeg-utils(1) manual,ffmpeg-utils}
2445 for the accepted syntax.
2446 Note that the resulting duration may be greater than the specified duration,
2447 as the generated audio is always cut at the end of a complete frame.
2448 If not specified, or the expressed duration is negative, the audio is
2449 supposed to be generated forever.
2450 Only used if plugin have zero inputs.
2454 @subsection Examples
2458 List all available plugins within amp (LADSPA example plugin) library:
2464 List all available controls and their valid ranges for @code{vcf_notch}
2465 plugin from @code{VCF} library:
2467 ladspa=f=vcf:p=vcf_notch:c=help
2471 Simulate low quality audio equipment using @code{Computer Music Toolkit} (CMT)
2474 ladspa=file=cmt:plugin=lofi:controls=c0=22|c1=12|c2=12
2478 Add reverberation to the audio using TAP-plugins
2479 (Tom's Audio Processing plugins):
2481 ladspa=file=tap_reverb:tap_reverb
2485 Generate white noise, with 0.2 amplitude:
2487 ladspa=file=cmt:noise_source_white:c=c0=.2
2491 Generate 20 bpm clicks using plugin @code{C* Click - Metronome} from the
2492 @code{C* Audio Plugin Suite} (CAPS) library:
2494 ladspa=file=caps:Click:c=c1=20'
2498 Apply @code{C* Eq10X2 - Stereo 10-band equaliser} effect:
2500 ladspa=caps:Eq10X2:c=c0=-48|c9=-24|c3=12|c4=2
2504 Increase volume by 20dB using fast lookahead limiter from Steve Harris
2505 @code{SWH Plugins} collection:
2507 ladspa=fast_lookahead_limiter_1913:fastLookaheadLimiter:20|0|2
2511 Attenuate low frequencies using Multiband EQ from Steve Harris
2512 @code{SWH Plugins} collection:
2514 ladspa=mbeq_1197:mbeq:-24|-24|-24|0|0|0|0|0|0|0|0|0|0|0|0
2518 @subsection Commands
2520 This filter supports the following commands:
2523 Modify the @var{N}-th control value.
2525 If the specified value is not valid, it is ignored and prior one is kept.
2530 Apply a low-pass filter with 3dB point frequency.
2531 The filter can be either single-pole or double-pole (the default).
2532 The filter roll off at 6dB per pole per octave (20dB per pole per decade).
2534 The filter accepts the following options:
2538 Set frequency in Hz. Default is 500.
2541 Set number of poles. Default is 2.
2544 Set method to specify band-width of filter.
2557 Specify the band-width of a filter in width_type units.
2558 Applies only to double-pole filter.
2559 The default is 0.707q and gives a Butterworth response.
2565 Mix channels with specific gain levels. The filter accepts the output
2566 channel layout followed by a set of channels definitions.
2568 This filter is also designed to efficiently remap the channels of an audio
2571 The filter accepts parameters of the form:
2572 "@var{l}|@var{outdef}|@var{outdef}|..."
2576 output channel layout or number of channels
2579 output channel specification, of the form:
2580 "@var{out_name}=[@var{gain}*]@var{in_name}[+[@var{gain}*]@var{in_name}...]"
2583 output channel to define, either a channel name (FL, FR, etc.) or a channel
2584 number (c0, c1, etc.)
2587 multiplicative coefficient for the channel, 1 leaving the volume unchanged
2590 input channel to use, see out_name for details; it is not possible to mix
2591 named and numbered input channels
2594 If the `=' in a channel specification is replaced by `<', then the gains for
2595 that specification will be renormalized so that the total is 1, thus
2596 avoiding clipping noise.
2598 @subsection Mixing examples
2600 For example, if you want to down-mix from stereo to mono, but with a bigger
2601 factor for the left channel:
2603 pan=1c|c0=0.9*c0+0.1*c1
2606 A customized down-mix to stereo that works automatically for 3-, 4-, 5- and
2607 7-channels surround:
2609 pan=stereo| FL < FL + 0.5*FC + 0.6*BL + 0.6*SL | FR < FR + 0.5*FC + 0.6*BR + 0.6*SR
2612 Note that @command{ffmpeg} integrates a default down-mix (and up-mix) system
2613 that should be preferred (see "-ac" option) unless you have very specific
2616 @subsection Remapping examples
2618 The channel remapping will be effective if, and only if:
2621 @item gain coefficients are zeroes or ones,
2622 @item only one input per channel output,
2625 If all these conditions are satisfied, the filter will notify the user ("Pure
2626 channel mapping detected"), and use an optimized and lossless method to do the
2629 For example, if you have a 5.1 source and want a stereo audio stream by
2630 dropping the extra channels:
2632 pan="stereo| c0=FL | c1=FR"
2635 Given the same source, you can also switch front left and front right channels
2636 and keep the input channel layout:
2638 pan="5.1| c0=c1 | c1=c0 | c2=c2 | c3=c3 | c4=c4 | c5=c5"
2641 If the input is a stereo audio stream, you can mute the front left channel (and
2642 still keep the stereo channel layout) with:
2647 Still with a stereo audio stream input, you can copy the right channel in both
2648 front left and right:
2650 pan="stereo| c0=FR | c1=FR"
2655 ReplayGain scanner filter. This filter takes an audio stream as an input and
2656 outputs it unchanged.
2657 At end of filtering it displays @code{track_gain} and @code{track_peak}.
2661 Convert the audio sample format, sample rate and channel layout. It is
2662 not meant to be used directly.
2665 Apply time-stretching and pitch-shifting with librubberband.
2667 The filter accepts the following options:
2671 Set tempo scale factor.
2674 Set pitch scale factor.
2677 Set transients detector.
2678 Possible values are:
2687 Possible values are:
2696 Possible values are:
2703 Set processing window size.
2704 Possible values are:
2713 Possible values are:
2720 Enable formant preservation when shift pitching.
2721 Possible values are:
2729 Possible values are:
2738 Possible values are:
2745 @section sidechaincompress
2747 This filter acts like normal compressor but has the ability to compress
2748 detected signal using second input signal.
2749 It needs two input streams and returns one output stream.
2750 First input stream will be processed depending on second stream signal.
2751 The filtered signal then can be filtered with other filters in later stages of
2752 processing. See @ref{pan} and @ref{amerge} filter.
2754 The filter accepts the following options:
2758 Set input gain. Default is 1. Range is between 0.015625 and 64.
2761 If a signal of second stream raises above this level it will affect the gain
2762 reduction of first stream.
2763 By default is 0.125. Range is between 0.00097563 and 1.
2766 Set a ratio about which the signal is reduced. 1:2 means that if the level
2767 raised 4dB above the threshold, it will be only 2dB above after the reduction.
2768 Default is 2. Range is between 1 and 20.
2771 Amount of milliseconds the signal has to rise above the threshold before gain
2772 reduction starts. Default is 20. Range is between 0.01 and 2000.
2775 Amount of milliseconds the signal has to fall below the threshold before
2776 reduction is decreased again. Default is 250. Range is between 0.01 and 9000.
2779 Set the amount by how much signal will be amplified after processing.
2780 Default is 2. Range is from 1 and 64.
2783 Curve the sharp knee around the threshold to enter gain reduction more softly.
2784 Default is 2.82843. Range is between 1 and 8.
2787 Choose if the @code{average} level between all channels of side-chain stream
2788 or the louder(@code{maximum}) channel of side-chain stream affects the
2789 reduction. Default is @code{average}.
2792 Should the exact signal be taken in case of @code{peak} or an RMS one in case
2793 of @code{rms}. Default is @code{rms} which is mainly smoother.
2796 Set sidechain gain. Default is 1. Range is between 0.015625 and 64.
2799 How much to use compressed signal in output. Default is 1.
2800 Range is between 0 and 1.
2803 @subsection Examples
2807 Full ffmpeg example taking 2 audio inputs, 1st input to be compressed
2808 depending on the signal of 2nd input and later compressed signal to be
2809 merged with 2nd input:
2811 ffmpeg -i main.flac -i sidechain.flac -filter_complex "[1:a]asplit=2[sc][mix];[0:a][sc]sidechaincompress[compr];[compr][mix]amerge"
2815 @section sidechaingate
2817 A sidechain gate acts like a normal (wideband) gate but has the ability to
2818 filter the detected signal before sending it to the gain reduction stage.
2819 Normally a gate uses the full range signal to detect a level above the
2821 For example: If you cut all lower frequencies from your sidechain signal
2822 the gate will decrease the volume of your track only if not enough highs
2823 appear. With this technique you are able to reduce the resonation of a
2824 natural drum or remove "rumbling" of muted strokes from a heavily distorted
2826 It needs two input streams and returns one output stream.
2827 First input stream will be processed depending on second stream signal.
2829 The filter accepts the following options:
2833 Set input level before filtering.
2834 Default is 1. Allowed range is from 0.015625 to 64.
2837 Set the level of gain reduction when the signal is below the threshold.
2838 Default is 0.06125. Allowed range is from 0 to 1.
2841 If a signal rises above this level the gain reduction is released.
2842 Default is 0.125. Allowed range is from 0 to 1.
2845 Set a ratio about which the signal is reduced.
2846 Default is 2. Allowed range is from 1 to 9000.
2849 Amount of milliseconds the signal has to rise above the threshold before gain
2851 Default is 20 milliseconds. Allowed range is from 0.01 to 9000.
2854 Amount of milliseconds the signal has to fall below the threshold before the
2855 reduction is increased again. Default is 250 milliseconds.
2856 Allowed range is from 0.01 to 9000.
2859 Set amount of amplification of signal after processing.
2860 Default is 1. Allowed range is from 1 to 64.
2863 Curve the sharp knee around the threshold to enter gain reduction more softly.
2864 Default is 2.828427125. Allowed range is from 1 to 8.
2867 Choose if exact signal should be taken for detection or an RMS like one.
2868 Default is rms. Can be peak or rms.
2871 Choose if the average level between all channels or the louder channel affects
2873 Default is average. Can be average or maximum.
2876 Set sidechain gain. Default is 1. Range is from 0.015625 to 64.
2879 @section silencedetect
2881 Detect silence in an audio stream.
2883 This filter logs a message when it detects that the input audio volume is less
2884 or equal to a noise tolerance value for a duration greater or equal to the
2885 minimum detected noise duration.
2887 The printed times and duration are expressed in seconds.
2889 The filter accepts the following options:
2893 Set silence duration until notification (default is 2 seconds).
2896 Set noise tolerance. Can be specified in dB (in case "dB" is appended to the
2897 specified value) or amplitude ratio. Default is -60dB, or 0.001.
2900 @subsection Examples
2904 Detect 5 seconds of silence with -50dB noise tolerance:
2906 silencedetect=n=-50dB:d=5
2910 Complete example with @command{ffmpeg} to detect silence with 0.0001 noise
2911 tolerance in @file{silence.mp3}:
2913 ffmpeg -i silence.mp3 -af silencedetect=noise=0.0001 -f null -
2917 @section silenceremove
2919 Remove silence from the beginning, middle or end of the audio.
2921 The filter accepts the following options:
2925 This value is used to indicate if audio should be trimmed at beginning of
2926 the audio. A value of zero indicates no silence should be trimmed from the
2927 beginning. When specifying a non-zero value, it trims audio up until it
2928 finds non-silence. Normally, when trimming silence from beginning of audio
2929 the @var{start_periods} will be @code{1} but it can be increased to higher
2930 values to trim all audio up to specific count of non-silence periods.
2931 Default value is @code{0}.
2933 @item start_duration
2934 Specify the amount of time that non-silence must be detected before it stops
2935 trimming audio. By increasing the duration, bursts of noises can be treated
2936 as silence and trimmed off. Default value is @code{0}.
2938 @item start_threshold
2939 This indicates what sample value should be treated as silence. For digital
2940 audio, a value of @code{0} may be fine but for audio recorded from analog,
2941 you may wish to increase the value to account for background noise.
2942 Can be specified in dB (in case "dB" is appended to the specified value)
2943 or amplitude ratio. Default value is @code{0}.
2946 Set the count for trimming silence from the end of audio.
2947 To remove silence from the middle of a file, specify a @var{stop_periods}
2948 that is negative. This value is then treated as a positive value and is
2949 used to indicate the effect should restart processing as specified by
2950 @var{start_periods}, making it suitable for removing periods of silence
2951 in the middle of the audio.
2952 Default value is @code{0}.
2955 Specify a duration of silence that must exist before audio is not copied any
2956 more. By specifying a higher duration, silence that is wanted can be left in
2958 Default value is @code{0}.
2960 @item stop_threshold
2961 This is the same as @option{start_threshold} but for trimming silence from
2963 Can be specified in dB (in case "dB" is appended to the specified value)
2964 or amplitude ratio. Default value is @code{0}.
2967 This indicate that @var{stop_duration} length of audio should be left intact
2968 at the beginning of each period of silence.
2969 For example, if you want to remove long pauses between words but do not want
2970 to remove the pauses completely. Default value is @code{0}.
2973 Set how is silence detected. Can be @code{rms} or @code{peak}. Second is faster
2974 and works better with digital silence which is exactly 0.
2975 Default value is @code{rms}.
2978 Set ratio used to calculate size of window for detecting silence.
2979 Default value is @code{0.02}. Allowed range is from @code{0} to @code{10}.
2982 @subsection Examples
2986 The following example shows how this filter can be used to start a recording
2987 that does not contain the delay at the start which usually occurs between
2988 pressing the record button and the start of the performance:
2990 silenceremove=1:5:0.02
2994 Trim all silence encountered from begining to end where there is more than 1
2995 second of silence in audio:
2997 silenceremove=0:0:0:-1:1:-90dB
3003 SOFAlizer uses head-related transfer functions (HRTFs) to create virtual
3004 loudspeakers around the user for binaural listening via headphones (audio
3005 formats up to 9 channels supported).
3006 The HRTFs are stored in SOFA files (see @url{http://www.sofacoustics.org/} for a database).
3007 SOFAlizer is developed at the Acoustics Research Institute (ARI) of the
3008 Austrian Academy of Sciences.
3010 To enable compilation of this filter you need to configure FFmpeg with
3011 @code{--enable-netcdf}.
3013 The filter accepts the following options:
3017 Set the SOFA file used for rendering.
3020 Set gain applied to audio. Value is in dB. Default is 0.
3023 Set rotation of virtual loudspeakers in deg. Default is 0.
3026 Set elevation of virtual speakers in deg. Default is 0.
3029 Set distance in meters between loudspeakers and the listener with near-field
3030 HRTFs. Default is 1.
3033 Set processing type. Can be @var{time} or @var{freq}. @var{time} is
3034 processing audio in time domain which is slow but gives high quality output.
3035 @var{freq} is processing audio in frequency domain which is fast but gives
3036 mediocre output. Default is @var{freq}.
3039 @section stereotools
3041 This filter has some handy utilities to manage stereo signals, for converting
3042 M/S stereo recordings to L/R signal while having control over the parameters
3043 or spreading the stereo image of master track.
3045 The filter accepts the following options:
3049 Set input level before filtering for both channels. Defaults is 1.
3050 Allowed range is from 0.015625 to 64.
3053 Set output level after filtering for both channels. Defaults is 1.
3054 Allowed range is from 0.015625 to 64.
3057 Set input balance between both channels. Default is 0.
3058 Allowed range is from -1 to 1.
3061 Set output balance between both channels. Default is 0.
3062 Allowed range is from -1 to 1.
3065 Enable softclipping. Results in analog distortion instead of harsh digital 0dB
3066 clipping. Disabled by default.
3069 Mute the left channel. Disabled by default.
3072 Mute the right channel. Disabled by default.
3075 Change the phase of the left channel. Disabled by default.
3078 Change the phase of the right channel. Disabled by default.
3081 Set stereo mode. Available values are:
3085 Left/Right to Left/Right, this is default.
3088 Left/Right to Mid/Side.
3091 Mid/Side to Left/Right.
3094 Left/Right to Left/Left.
3097 Left/Right to Right/Right.
3100 Left/Right to Left + Right.
3103 Left/Right to Right/Left.
3107 Set level of side signal. Default is 1.
3108 Allowed range is from 0.015625 to 64.
3111 Set balance of side signal. Default is 0.
3112 Allowed range is from -1 to 1.
3115 Set level of the middle signal. Default is 1.
3116 Allowed range is from 0.015625 to 64.
3119 Set middle signal pan. Default is 0. Allowed range is from -1 to 1.
3122 Set stereo base between mono and inversed channels. Default is 0.
3123 Allowed range is from -1 to 1.
3126 Set delay in milliseconds how much to delay left from right channel and
3127 vice versa. Default is 0. Allowed range is from -20 to 20.
3130 Set S/C level. Default is 1. Allowed range is from 1 to 100.
3133 Set the stereo phase in degrees. Default is 0. Allowed range is from 0 to 360.
3136 @section stereowiden
3138 This filter enhance the stereo effect by suppressing signal common to both
3139 channels and by delaying the signal of left into right and vice versa,
3140 thereby widening the stereo effect.
3142 The filter accepts the following options:
3146 Time in milliseconds of the delay of left signal into right and vice versa.
3147 Default is 20 milliseconds.
3150 Amount of gain in delayed signal into right and vice versa. Gives a delay
3151 effect of left signal in right output and vice versa which gives widening
3152 effect. Default is 0.3.
3155 Cross feed of left into right with inverted phase. This helps in suppressing
3156 the mono. If the value is 1 it will cancel all the signal common to both
3157 channels. Default is 0.3.
3160 Set level of input signal of original channel. Default is 0.8.
3165 Boost or cut treble (upper) frequencies of the audio using a two-pole
3166 shelving filter with a response similar to that of a standard
3167 hi-fi's tone-controls. This is also known as shelving equalisation (EQ).
3169 The filter accepts the following options:
3173 Give the gain at whichever is the lower of ~22 kHz and the
3174 Nyquist frequency. Its useful range is about -20 (for a large cut)
3175 to +20 (for a large boost). Beware of clipping when using a positive gain.
3178 Set the filter's central frequency and so can be used
3179 to extend or reduce the frequency range to be boosted or cut.
3180 The default value is @code{3000} Hz.
3183 Set method to specify band-width of filter.
3196 Determine how steep is the filter's shelf transition.
3201 Sinusoidal amplitude modulation.
3203 The filter accepts the following options:
3207 Modulation frequency in Hertz. Modulation frequencies in the subharmonic range
3208 (20 Hz or lower) will result in a tremolo effect.
3209 This filter may also be used as a ring modulator by specifying
3210 a modulation frequency higher than 20 Hz.
3211 Range is 0.1 - 20000.0. Default value is 5.0 Hz.
3214 Depth of modulation as a percentage. Range is 0.0 - 1.0.
3215 Default value is 0.5.
3220 Sinusoidal phase modulation.
3222 The filter accepts the following options:
3226 Modulation frequency in Hertz.
3227 Range is 0.1 - 20000.0. Default value is 5.0 Hz.
3230 Depth of modulation as a percentage. Range is 0.0 - 1.0.
3231 Default value is 0.5.
3236 Adjust the input audio volume.
3238 It accepts the following parameters:
3242 Set audio volume expression.
3244 Output values are clipped to the maximum value.
3246 The output audio volume is given by the relation:
3248 @var{output_volume} = @var{volume} * @var{input_volume}
3251 The default value for @var{volume} is "1.0".
3254 This parameter represents the mathematical precision.
3256 It determines which input sample formats will be allowed, which affects the
3257 precision of the volume scaling.
3261 8-bit fixed-point; this limits input sample format to U8, S16, and S32.
3263 32-bit floating-point; this limits input sample format to FLT. (default)
3265 64-bit floating-point; this limits input sample format to DBL.
3269 Choose the behaviour on encountering ReplayGain side data in input frames.
3273 Remove ReplayGain side data, ignoring its contents (the default).
3276 Ignore ReplayGain side data, but leave it in the frame.
3279 Prefer the track gain, if present.
3282 Prefer the album gain, if present.
3285 @item replaygain_preamp
3286 Pre-amplification gain in dB to apply to the selected replaygain gain.
3288 Default value for @var{replaygain_preamp} is 0.0.
3291 Set when the volume expression is evaluated.
3293 It accepts the following values:
3296 only evaluate expression once during the filter initialization, or
3297 when the @samp{volume} command is sent
3300 evaluate expression for each incoming frame
3303 Default value is @samp{once}.
3306 The volume expression can contain the following parameters.
3310 frame number (starting at zero)
3313 @item nb_consumed_samples
3314 number of samples consumed by the filter
3316 number of samples in the current frame
3318 original frame position in the file
3324 PTS at start of stream
3326 time at start of stream
3332 last set volume value
3335 Note that when @option{eval} is set to @samp{once} only the
3336 @var{sample_rate} and @var{tb} variables are available, all other
3337 variables will evaluate to NAN.
3339 @subsection Commands
3341 This filter supports the following commands:
3344 Modify the volume expression.
3345 The command accepts the same syntax of the corresponding option.
3347 If the specified expression is not valid, it is kept at its current
3349 @item replaygain_noclip
3350 Prevent clipping by limiting the gain applied.
3352 Default value for @var{replaygain_noclip} is 1.
3356 @subsection Examples
3360 Halve the input audio volume:
3364 volume=volume=-6.0206dB
3367 In all the above example the named key for @option{volume} can be
3368 omitted, for example like in:
3374 Increase input audio power by 6 decibels using fixed-point precision:
3376 volume=volume=6dB:precision=fixed
3380 Fade volume after time 10 with an annihilation period of 5 seconds:
3382 volume='if(lt(t,10),1,max(1-(t-10)/5,0))':eval=frame
3386 @section volumedetect
3388 Detect the volume of the input video.
3390 The filter has no parameters. The input is not modified. Statistics about
3391 the volume will be printed in the log when the input stream end is reached.
3393 In particular it will show the mean volume (root mean square), maximum
3394 volume (on a per-sample basis), and the beginning of a histogram of the
3395 registered volume values (from the maximum value to a cumulated 1/1000 of
3398 All volumes are in decibels relative to the maximum PCM value.
3400 @subsection Examples
3402 Here is an excerpt of the output:
3404 [Parsed_volumedetect_0 @ 0xa23120] mean_volume: -27 dB
3405 [Parsed_volumedetect_0 @ 0xa23120] max_volume: -4 dB
3406 [Parsed_volumedetect_0 @ 0xa23120] histogram_4db: 6
3407 [Parsed_volumedetect_0 @ 0xa23120] histogram_5db: 62
3408 [Parsed_volumedetect_0 @ 0xa23120] histogram_6db: 286
3409 [Parsed_volumedetect_0 @ 0xa23120] histogram_7db: 1042
3410 [Parsed_volumedetect_0 @ 0xa23120] histogram_8db: 2551
3411 [Parsed_volumedetect_0 @ 0xa23120] histogram_9db: 4609
3412 [Parsed_volumedetect_0 @ 0xa23120] histogram_10db: 8409
3418 The mean square energy is approximately -27 dB, or 10^-2.7.
3420 The largest sample is at -4 dB, or more precisely between -4 dB and -5 dB.
3422 There are 6 samples at -4 dB, 62 at -5 dB, 286 at -6 dB, etc.
3425 In other words, raising the volume by +4 dB does not cause any clipping,
3426 raising it by +5 dB causes clipping for 6 samples, etc.
3428 @c man end AUDIO FILTERS
3430 @chapter Audio Sources
3431 @c man begin AUDIO SOURCES
3433 Below is a description of the currently available audio sources.
3437 Buffer audio frames, and make them available to the filter chain.
3439 This source is mainly intended for a programmatic use, in particular
3440 through the interface defined in @file{libavfilter/asrc_abuffer.h}.
3442 It accepts the following parameters:
3446 The timebase which will be used for timestamps of submitted frames. It must be
3447 either a floating-point number or in @var{numerator}/@var{denominator} form.
3450 The sample rate of the incoming audio buffers.
3453 The sample format of the incoming audio buffers.
3454 Either a sample format name or its corresponding integer representation from
3455 the enum AVSampleFormat in @file{libavutil/samplefmt.h}
3457 @item channel_layout
3458 The channel layout of the incoming audio buffers.
3459 Either a channel layout name from channel_layout_map in
3460 @file{libavutil/channel_layout.c} or its corresponding integer representation
3461 from the AV_CH_LAYOUT_* macros in @file{libavutil/channel_layout.h}
3464 The number of channels of the incoming audio buffers.
3465 If both @var{channels} and @var{channel_layout} are specified, then they
3470 @subsection Examples
3473 abuffer=sample_rate=44100:sample_fmt=s16p:channel_layout=stereo
3476 will instruct the source to accept planar 16bit signed stereo at 44100Hz.
3477 Since the sample format with name "s16p" corresponds to the number
3478 6 and the "stereo" channel layout corresponds to the value 0x3, this is
3481 abuffer=sample_rate=44100:sample_fmt=6:channel_layout=0x3
3486 Generate an audio signal specified by an expression.
3488 This source accepts in input one or more expressions (one for each
3489 channel), which are evaluated and used to generate a corresponding
3492 This source accepts the following options:
3496 Set the '|'-separated expressions list for each separate channel. In case the
3497 @option{channel_layout} option is not specified, the selected channel layout
3498 depends on the number of provided expressions. Otherwise the last
3499 specified expression is applied to the remaining output channels.
3501 @item channel_layout, c
3502 Set the channel layout. The number of channels in the specified layout
3503 must be equal to the number of specified expressions.
3506 Set the minimum duration of the sourced audio. See
3507 @ref{time duration syntax,,the Time duration section in the ffmpeg-utils(1) manual,ffmpeg-utils}
3508 for the accepted syntax.
3509 Note that the resulting duration may be greater than the specified
3510 duration, as the generated audio is always cut at the end of a
3513 If not specified, or the expressed duration is negative, the audio is
3514 supposed to be generated forever.
3517 Set the number of samples per channel per each output frame,
3520 @item sample_rate, s
3521 Specify the sample rate, default to 44100.
3524 Each expression in @var{exprs} can contain the following constants:
3528 number of the evaluated sample, starting from 0
3531 time of the evaluated sample expressed in seconds, starting from 0
3538 @subsection Examples
3548 Generate a sin signal with frequency of 440 Hz, set sample rate to
3551 aevalsrc="sin(440*2*PI*t):s=8000"
3555 Generate a two channels signal, specify the channel layout (Front
3556 Center + Back Center) explicitly:
3558 aevalsrc="sin(420*2*PI*t)|cos(430*2*PI*t):c=FC|BC"
3562 Generate white noise:
3564 aevalsrc="-2+random(0)"
3568 Generate an amplitude modulated signal:
3570 aevalsrc="sin(10*2*PI*t)*sin(880*2*PI*t)"
3574 Generate 2.5 Hz binaural beats on a 360 Hz carrier:
3576 aevalsrc="0.1*sin(2*PI*(360-2.5/2)*t) | 0.1*sin(2*PI*(360+2.5/2)*t)"
3583 The null audio source, return unprocessed audio frames. It is mainly useful
3584 as a template and to be employed in analysis / debugging tools, or as
3585 the source for filters which ignore the input data (for example the sox
3588 This source accepts the following options:
3592 @item channel_layout, cl
3594 Specifies the channel layout, and can be either an integer or a string
3595 representing a channel layout. The default value of @var{channel_layout}
3598 Check the channel_layout_map definition in
3599 @file{libavutil/channel_layout.c} for the mapping between strings and
3600 channel layout values.
3602 @item sample_rate, r
3603 Specifies the sample rate, and defaults to 44100.
3606 Set the number of samples per requested frames.
3610 @subsection Examples
3614 Set the sample rate to 48000 Hz and the channel layout to AV_CH_LAYOUT_MONO.
3616 anullsrc=r=48000:cl=4
3620 Do the same operation with a more obvious syntax:
3622 anullsrc=r=48000:cl=mono
3626 All the parameters need to be explicitly defined.
3630 Synthesize a voice utterance using the libflite library.
3632 To enable compilation of this filter you need to configure FFmpeg with
3633 @code{--enable-libflite}.
3635 Note that the flite library is not thread-safe.
3637 The filter accepts the following options:
3642 If set to 1, list the names of the available voices and exit
3643 immediately. Default value is 0.
3646 Set the maximum number of samples per frame. Default value is 512.
3649 Set the filename containing the text to speak.
3652 Set the text to speak.
3655 Set the voice to use for the speech synthesis. Default value is
3656 @code{kal}. See also the @var{list_voices} option.
3659 @subsection Examples
3663 Read from file @file{speech.txt}, and synthesize the text using the
3664 standard flite voice:
3666 flite=textfile=speech.txt
3670 Read the specified text selecting the @code{slt} voice:
3672 flite=text='So fare thee well, poor devil of a Sub-Sub, whose commentator I am':voice=slt
3676 Input text to ffmpeg:
3678 ffmpeg -f lavfi -i flite=text='So fare thee well, poor devil of a Sub-Sub, whose commentator I am':voice=slt
3682 Make @file{ffplay} speak the specified text, using @code{flite} and
3683 the @code{lavfi} device:
3685 ffplay -f lavfi flite=text='No more be grieved for which that thou hast done.'
3689 For more information about libflite, check:
3690 @url{http://www.speech.cs.cmu.edu/flite/}
3694 Generate a noise audio signal.
3696 The filter accepts the following options:
3699 @item sample_rate, r
3700 Specify the sample rate. Default value is 48000 Hz.
3703 Specify the amplitude (0.0 - 1.0) of the generated audio stream. Default value
3707 Specify the duration of the generated audio stream. Not specifying this option
3708 results in noise with an infinite length.
3710 @item color, colour, c
3711 Specify the color of noise. Available noise colors are white, pink, and brown.
3712 Default color is white.
3715 Specify a value used to seed the PRNG.
3718 Set the number of samples per each output frame, default is 1024.
3721 @subsection Examples
3726 Generate 60 seconds of pink noise, with a 44.1 kHz sampling rate and an amplitude of 0.5:
3728 anoisesrc=d=60:c=pink:r=44100:a=0.5
3734 Generate an audio signal made of a sine wave with amplitude 1/8.
3736 The audio signal is bit-exact.
3738 The filter accepts the following options:
3743 Set the carrier frequency. Default is 440 Hz.
3745 @item beep_factor, b
3746 Enable a periodic beep every second with frequency @var{beep_factor} times
3747 the carrier frequency. Default is 0, meaning the beep is disabled.
3749 @item sample_rate, r
3750 Specify the sample rate, default is 44100.
3753 Specify the duration of the generated audio stream.
3755 @item samples_per_frame
3756 Set the number of samples per output frame.
3758 The expression can contain the following constants:
3762 The (sequential) number of the output audio frame, starting from 0.
3765 The PTS (Presentation TimeStamp) of the output audio frame,
3766 expressed in @var{TB} units.
3769 The PTS of the output audio frame, expressed in seconds.
3772 The timebase of the output audio frames.
3775 Default is @code{1024}.
3778 @subsection Examples
3783 Generate a simple 440 Hz sine wave:
3789 Generate a 220 Hz sine wave with a 880 Hz beep each second, for 5 seconds:
3793 sine=frequency=220:beep_factor=4:duration=5
3797 Generate a 1 kHz sine wave following @code{1602,1601,1602,1601,1602} NTSC
3800 sine=1000:samples_per_frame='st(0,mod(n,5)); 1602-not(not(eq(ld(0),1)+eq(ld(0),3)))'
3804 @c man end AUDIO SOURCES
3806 @chapter Audio Sinks
3807 @c man begin AUDIO SINKS
3809 Below is a description of the currently available audio sinks.
3811 @section abuffersink
3813 Buffer audio frames, and make them available to the end of filter chain.
3815 This sink is mainly intended for programmatic use, in particular
3816 through the interface defined in @file{libavfilter/buffersink.h}
3817 or the options system.
3819 It accepts a pointer to an AVABufferSinkContext structure, which
3820 defines the incoming buffers' formats, to be passed as the opaque
3821 parameter to @code{avfilter_init_filter} for initialization.
3824 Null audio sink; do absolutely nothing with the input audio. It is
3825 mainly useful as a template and for use in analysis / debugging
3828 @c man end AUDIO SINKS
3830 @chapter Video Filters
3831 @c man begin VIDEO FILTERS
3833 When you configure your FFmpeg build, you can disable any of the
3834 existing filters using @code{--disable-filters}.
3835 The configure output will show the video filters included in your
3838 Below is a description of the currently available video filters.
3840 @section alphaextract
3842 Extract the alpha component from the input as a grayscale video. This
3843 is especially useful with the @var{alphamerge} filter.
3847 Add or replace the alpha component of the primary input with the
3848 grayscale value of a second input. This is intended for use with
3849 @var{alphaextract} to allow the transmission or storage of frame
3850 sequences that have alpha in a format that doesn't support an alpha
3853 For example, to reconstruct full frames from a normal YUV-encoded video
3854 and a separate video created with @var{alphaextract}, you might use:
3856 movie=in_alpha.mkv [alpha]; [in][alpha] alphamerge [out]
3859 Since this filter is designed for reconstruction, it operates on frame
3860 sequences without considering timestamps, and terminates when either
3861 input reaches end of stream. This will cause problems if your encoding
3862 pipeline drops frames. If you're trying to apply an image as an
3863 overlay to a video stream, consider the @var{overlay} filter instead.
3867 Same as the @ref{subtitles} filter, except that it doesn't require libavcodec
3868 and libavformat to work. On the other hand, it is limited to ASS (Advanced
3869 Substation Alpha) subtitles files.
3871 This filter accepts the following option in addition to the common options from
3872 the @ref{subtitles} filter:
3876 Set the shaping engine
3878 Available values are:
3881 The default libass shaping engine, which is the best available.
3883 Fast, font-agnostic shaper that can do only substitutions
3885 Slower shaper using OpenType for substitutions and positioning
3888 The default is @code{auto}.
3892 Apply an Adaptive Temporal Averaging Denoiser to the video input.
3894 The filter accepts the following options:
3898 Set threshold A for 1st plane. Default is 0.02.
3899 Valid range is 0 to 0.3.
3902 Set threshold B for 1st plane. Default is 0.04.
3903 Valid range is 0 to 5.
3906 Set threshold A for 2nd plane. Default is 0.02.
3907 Valid range is 0 to 0.3.
3910 Set threshold B for 2nd plane. Default is 0.04.
3911 Valid range is 0 to 5.
3914 Set threshold A for 3rd plane. Default is 0.02.
3915 Valid range is 0 to 0.3.
3918 Set threshold B for 3rd plane. Default is 0.04.
3919 Valid range is 0 to 5.
3921 Threshold A is designed to react on abrupt changes in the input signal and
3922 threshold B is designed to react on continuous changes in the input signal.
3925 Set number of frames filter will use for averaging. Default is 33. Must be odd
3926 number in range [5, 129].
3931 Compute the bounding box for the non-black pixels in the input frame
3934 This filter computes the bounding box containing all the pixels with a
3935 luminance value greater than the minimum allowed value.
3936 The parameters describing the bounding box are printed on the filter
3939 The filter accepts the following option:
3943 Set the minimal luminance value. Default is @code{16}.
3946 @section blackdetect
3948 Detect video intervals that are (almost) completely black. Can be
3949 useful to detect chapter transitions, commercials, or invalid
3950 recordings. Output lines contains the time for the start, end and
3951 duration of the detected black interval expressed in seconds.
3953 In order to display the output lines, you need to set the loglevel at
3954 least to the AV_LOG_INFO value.
3956 The filter accepts the following options:
3959 @item black_min_duration, d
3960 Set the minimum detected black duration expressed in seconds. It must
3961 be a non-negative floating point number.
3963 Default value is 2.0.
3965 @item picture_black_ratio_th, pic_th
3966 Set the threshold for considering a picture "black".
3967 Express the minimum value for the ratio:
3969 @var{nb_black_pixels} / @var{nb_pixels}
3972 for which a picture is considered black.
3973 Default value is 0.98.
3975 @item pixel_black_th, pix_th
3976 Set the threshold for considering a pixel "black".
3978 The threshold expresses the maximum pixel luminance value for which a
3979 pixel is considered "black". The provided value is scaled according to
3980 the following equation:
3982 @var{absolute_threshold} = @var{luminance_minimum_value} + @var{pixel_black_th} * @var{luminance_range_size}
3985 @var{luminance_range_size} and @var{luminance_minimum_value} depend on
3986 the input video format, the range is [0-255] for YUV full-range
3987 formats and [16-235] for YUV non full-range formats.
3989 Default value is 0.10.
3992 The following example sets the maximum pixel threshold to the minimum
3993 value, and detects only black intervals of 2 or more seconds:
3995 blackdetect=d=2:pix_th=0.00
4000 Detect frames that are (almost) completely black. Can be useful to
4001 detect chapter transitions or commercials. Output lines consist of
4002 the frame number of the detected frame, the percentage of blackness,
4003 the position in the file if known or -1 and the timestamp in seconds.
4005 In order to display the output lines, you need to set the loglevel at
4006 least to the AV_LOG_INFO value.
4008 It accepts the following parameters:
4013 The percentage of the pixels that have to be below the threshold; it defaults to
4016 @item threshold, thresh
4017 The threshold below which a pixel value is considered black; it defaults to
4022 @section blend, tblend
4024 Blend two video frames into each other.
4026 The @code{blend} filter takes two input streams and outputs one
4027 stream, the first input is the "top" layer and second input is
4028 "bottom" layer. Output terminates when shortest input terminates.
4030 The @code{tblend} (time blend) filter takes two consecutive frames
4031 from one single stream, and outputs the result obtained by blending
4032 the new frame on top of the old frame.
4034 A description of the accepted options follows.
4042 Set blend mode for specific pixel component or all pixel components in case
4043 of @var{all_mode}. Default value is @code{normal}.
4045 Available values for component modes are:
4083 Set blend opacity for specific pixel component or all pixel components in case
4084 of @var{all_opacity}. Only used in combination with pixel component blend modes.
4091 Set blend expression for specific pixel component or all pixel components in case
4092 of @var{all_expr}. Note that related mode options will be ignored if those are set.
4094 The expressions can use the following variables:
4098 The sequential number of the filtered frame, starting from @code{0}.
4102 the coordinates of the current sample
4106 the width and height of currently filtered plane
4110 Width and height scale depending on the currently filtered plane. It is the
4111 ratio between the corresponding luma plane number of pixels and the current
4112 plane ones. E.g. for YUV4:2:0 the values are @code{1,1} for the luma plane, and
4113 @code{0.5,0.5} for chroma planes.
4116 Time of the current frame, expressed in seconds.
4119 Value of pixel component at current location for first video frame (top layer).
4122 Value of pixel component at current location for second video frame (bottom layer).
4126 Force termination when the shortest input terminates. Default is
4127 @code{0}. This option is only defined for the @code{blend} filter.
4130 Continue applying the last bottom frame after the end of the stream. A value of
4131 @code{0} disable the filter after the last frame of the bottom layer is reached.
4132 Default is @code{1}. This option is only defined for the @code{blend} filter.
4135 @subsection Examples
4139 Apply transition from bottom layer to top layer in first 10 seconds:
4141 blend=all_expr='A*(if(gte(T,10),1,T/10))+B*(1-(if(gte(T,10),1,T/10)))'
4145 Apply 1x1 checkerboard effect:
4147 blend=all_expr='if(eq(mod(X,2),mod(Y,2)),A,B)'
4151 Apply uncover left effect:
4153 blend=all_expr='if(gte(N*SW+X,W),A,B)'
4157 Apply uncover down effect:
4159 blend=all_expr='if(gte(Y-N*SH,0),A,B)'
4163 Apply uncover up-left effect:
4165 blend=all_expr='if(gte(T*SH*40+Y,H)*gte((T*40*SW+X)*W/H,W),A,B)'
4169 Display differences between the current and the previous frame:
4171 tblend=all_mode=difference128
4177 Apply a boxblur algorithm to the input video.
4179 It accepts the following parameters:
4183 @item luma_radius, lr
4184 @item luma_power, lp
4185 @item chroma_radius, cr
4186 @item chroma_power, cp
4187 @item alpha_radius, ar
4188 @item alpha_power, ap
4192 A description of the accepted options follows.
4195 @item luma_radius, lr
4196 @item chroma_radius, cr
4197 @item alpha_radius, ar
4198 Set an expression for the box radius in pixels used for blurring the
4199 corresponding input plane.
4201 The radius value must be a non-negative number, and must not be
4202 greater than the value of the expression @code{min(w,h)/2} for the
4203 luma and alpha planes, and of @code{min(cw,ch)/2} for the chroma
4206 Default value for @option{luma_radius} is "2". If not specified,
4207 @option{chroma_radius} and @option{alpha_radius} default to the
4208 corresponding value set for @option{luma_radius}.
4210 The expressions can contain the following constants:
4214 The input width and height in pixels.
4218 The input chroma image width and height in pixels.
4222 The horizontal and vertical chroma subsample values. For example, for the
4223 pixel format "yuv422p", @var{hsub} is 2 and @var{vsub} is 1.
4226 @item luma_power, lp
4227 @item chroma_power, cp
4228 @item alpha_power, ap
4229 Specify how many times the boxblur filter is applied to the
4230 corresponding plane.
4232 Default value for @option{luma_power} is 2. If not specified,
4233 @option{chroma_power} and @option{alpha_power} default to the
4234 corresponding value set for @option{luma_power}.
4236 A value of 0 will disable the effect.
4239 @subsection Examples
4243 Apply a boxblur filter with the luma, chroma, and alpha radii
4246 boxblur=luma_radius=2:luma_power=1
4251 Set the luma radius to 2, and alpha and chroma radius to 0:
4253 boxblur=2:1:cr=0:ar=0
4257 Set the luma and chroma radii to a fraction of the video dimension:
4259 boxblur=luma_radius=min(h\,w)/10:luma_power=1:chroma_radius=min(cw\,ch)/10:chroma_power=1
4264 YUV colorspace color/chroma keying.
4266 The filter accepts the following options:
4270 The color which will be replaced with transparency.
4273 Similarity percentage with the key color.
4275 0.01 matches only the exact key color, while 1.0 matches everything.
4280 0.0 makes pixels either fully transparent, or not transparent at all.
4282 Higher values result in semi-transparent pixels, with a higher transparency
4283 the more similar the pixels color is to the key color.
4286 Signals that the color passed is already in YUV instead of RGB.
4288 Litteral colors like "green" or "red" don't make sense with this enabled anymore.
4289 This can be used to pass exact YUV values as hexadecimal numbers.
4292 @subsection Examples
4296 Make every green pixel in the input image transparent:
4298 ffmpeg -i input.png -vf chromakey=green out.png
4302 Overlay a greenscreen-video on top of a static black background.
4304 ffmpeg -f lavfi -i color=c=black:s=1280x720 -i video.mp4 -shortest -filter_complex "[1:v]chromakey=0x70de77:0.1:0.2[ckout];[0:v][ckout]overlay[out]" -map "[out]" output.mkv
4310 Visualize information exported by some codecs.
4312 Some codecs can export information through frames using side-data or other
4313 means. For example, some MPEG based codecs export motion vectors through the
4314 @var{export_mvs} flag in the codec @option{flags2} option.
4316 The filter accepts the following option:
4320 Set motion vectors to visualize.
4322 Available flags for @var{mv} are:
4326 forward predicted MVs of P-frames
4328 forward predicted MVs of B-frames
4330 backward predicted MVs of B-frames
4334 Display quantization parameters using the chroma planes
4337 @subsection Examples
4341 Visualizes multi-directionals MVs from P and B-Frames using @command{ffplay}:
4343 ffplay -flags2 +export_mvs input.mpg -vf codecview=mv=pf+bf+bb
4347 @section colorbalance
4348 Modify intensity of primary colors (red, green and blue) of input frames.
4350 The filter allows an input frame to be adjusted in the shadows, midtones or highlights
4351 regions for the red-cyan, green-magenta or blue-yellow balance.
4353 A positive adjustment value shifts the balance towards the primary color, a negative
4354 value towards the complementary color.
4356 The filter accepts the following options:
4362 Adjust red, green and blue shadows (darkest pixels).
4367 Adjust red, green and blue midtones (medium pixels).
4372 Adjust red, green and blue highlights (brightest pixels).
4374 Allowed ranges for options are @code{[-1.0, 1.0]}. Defaults are @code{0}.
4377 @subsection Examples
4381 Add red color cast to shadows:
4388 RGB colorspace color keying.
4390 The filter accepts the following options:
4394 The color which will be replaced with transparency.
4397 Similarity percentage with the key color.
4399 0.01 matches only the exact key color, while 1.0 matches everything.
4404 0.0 makes pixels either fully transparent, or not transparent at all.
4406 Higher values result in semi-transparent pixels, with a higher transparency
4407 the more similar the pixels color is to the key color.
4410 @subsection Examples
4414 Make every green pixel in the input image transparent:
4416 ffmpeg -i input.png -vf colorkey=green out.png
4420 Overlay a greenscreen-video on top of a static background image.
4422 ffmpeg -i background.png -i video.mp4 -filter_complex "[1:v]colorkey=0x3BBD1E:0.3:0.2[ckout];[0:v][ckout]overlay[out]" -map "[out]" output.flv
4426 @section colorlevels
4428 Adjust video input frames using levels.
4430 The filter accepts the following options:
4437 Adjust red, green, blue and alpha input black point.
4438 Allowed ranges for options are @code{[-1.0, 1.0]}. Defaults are @code{0}.
4444 Adjust red, green, blue and alpha input white point.
4445 Allowed ranges for options are @code{[-1.0, 1.0]}. Defaults are @code{1}.
4447 Input levels are used to lighten highlights (bright tones), darken shadows
4448 (dark tones), change the balance of bright and dark tones.
4454 Adjust red, green, blue and alpha output black point.
4455 Allowed ranges for options are @code{[0, 1.0]}. Defaults are @code{0}.
4461 Adjust red, green, blue and alpha output white point.
4462 Allowed ranges for options are @code{[0, 1.0]}. Defaults are @code{1}.
4464 Output levels allows manual selection of a constrained output level range.
4467 @subsection Examples
4471 Make video output darker:
4473 colorlevels=rimin=0.058:gimin=0.058:bimin=0.058
4479 colorlevels=rimin=0.039:gimin=0.039:bimin=0.039:rimax=0.96:gimax=0.96:bimax=0.96
4483 Make video output lighter:
4485 colorlevels=rimax=0.902:gimax=0.902:bimax=0.902
4489 Increase brightness:
4491 colorlevels=romin=0.5:gomin=0.5:bomin=0.5
4495 @section colorchannelmixer
4497 Adjust video input frames by re-mixing color channels.
4499 This filter modifies a color channel by adding the values associated to
4500 the other channels of the same pixels. For example if the value to
4501 modify is red, the output value will be:
4503 @var{red}=@var{red}*@var{rr} + @var{blue}*@var{rb} + @var{green}*@var{rg} + @var{alpha}*@var{ra}
4506 The filter accepts the following options:
4513 Adjust contribution of input red, green, blue and alpha channels for output red channel.
4514 Default is @code{1} for @var{rr}, and @code{0} for @var{rg}, @var{rb} and @var{ra}.
4520 Adjust contribution of input red, green, blue and alpha channels for output green channel.
4521 Default is @code{1} for @var{gg}, and @code{0} for @var{gr}, @var{gb} and @var{ga}.
4527 Adjust contribution of input red, green, blue and alpha channels for output blue channel.
4528 Default is @code{1} for @var{bb}, and @code{0} for @var{br}, @var{bg} and @var{ba}.
4534 Adjust contribution of input red, green, blue and alpha channels for output alpha channel.
4535 Default is @code{1} for @var{aa}, and @code{0} for @var{ar}, @var{ag} and @var{ab}.
4537 Allowed ranges for options are @code{[-2.0, 2.0]}.
4540 @subsection Examples
4544 Convert source to grayscale:
4546 colorchannelmixer=.3:.4:.3:0:.3:.4:.3:0:.3:.4:.3
4549 Simulate sepia tones:
4551 colorchannelmixer=.393:.769:.189:0:.349:.686:.168:0:.272:.534:.131
4555 @section colormatrix
4557 Convert color matrix.
4559 The filter accepts the following options:
4564 Specify the source and destination color matrix. Both values must be
4567 The accepted values are:
4583 For example to convert from BT.601 to SMPTE-240M, use the command:
4585 colormatrix=bt601:smpte240m
4590 Copy the input source unchanged to the output. This is mainly useful for
4595 Crop the input video to given dimensions.
4597 It accepts the following parameters:
4601 The width of the output video. It defaults to @code{iw}.
4602 This expression is evaluated only once during the filter
4603 configuration, or when the @samp{w} or @samp{out_w} command is sent.
4606 The height of the output video. It defaults to @code{ih}.
4607 This expression is evaluated only once during the filter
4608 configuration, or when the @samp{h} or @samp{out_h} command is sent.
4611 The horizontal position, in the input video, of the left edge of the output
4612 video. It defaults to @code{(in_w-out_w)/2}.
4613 This expression is evaluated per-frame.
4616 The vertical position, in the input video, of the top edge of the output video.
4617 It defaults to @code{(in_h-out_h)/2}.
4618 This expression is evaluated per-frame.
4621 If set to 1 will force the output display aspect ratio
4622 to be the same of the input, by changing the output sample aspect
4623 ratio. It defaults to 0.
4626 The @var{out_w}, @var{out_h}, @var{x}, @var{y} parameters are
4627 expressions containing the following constants:
4632 The computed values for @var{x} and @var{y}. They are evaluated for
4637 The input width and height.
4641 These are the same as @var{in_w} and @var{in_h}.
4645 The output (cropped) width and height.
4649 These are the same as @var{out_w} and @var{out_h}.
4652 same as @var{iw} / @var{ih}
4655 input sample aspect ratio
4658 input display aspect ratio, it is the same as (@var{iw} / @var{ih}) * @var{sar}
4662 horizontal and vertical chroma subsample values. For example for the
4663 pixel format "yuv422p" @var{hsub} is 2 and @var{vsub} is 1.
4666 The number of the input frame, starting from 0.
4669 the position in the file of the input frame, NAN if unknown
4672 The timestamp expressed in seconds. It's NAN if the input timestamp is unknown.
4676 The expression for @var{out_w} may depend on the value of @var{out_h},
4677 and the expression for @var{out_h} may depend on @var{out_w}, but they
4678 cannot depend on @var{x} and @var{y}, as @var{x} and @var{y} are
4679 evaluated after @var{out_w} and @var{out_h}.
4681 The @var{x} and @var{y} parameters specify the expressions for the
4682 position of the top-left corner of the output (non-cropped) area. They
4683 are evaluated for each frame. If the evaluated value is not valid, it
4684 is approximated to the nearest valid value.
4686 The expression for @var{x} may depend on @var{y}, and the expression
4687 for @var{y} may depend on @var{x}.
4689 @subsection Examples
4693 Crop area with size 100x100 at position (12,34).
4698 Using named options, the example above becomes:
4700 crop=w=100:h=100:x=12:y=34
4704 Crop the central input area with size 100x100:
4710 Crop the central input area with size 2/3 of the input video:
4712 crop=2/3*in_w:2/3*in_h
4716 Crop the input video central square:
4723 Delimit the rectangle with the top-left corner placed at position
4724 100:100 and the right-bottom corner corresponding to the right-bottom
4725 corner of the input image.
4727 crop=in_w-100:in_h-100:100:100
4731 Crop 10 pixels from the left and right borders, and 20 pixels from
4732 the top and bottom borders
4734 crop=in_w-2*10:in_h-2*20
4738 Keep only the bottom right quarter of the input image:
4740 crop=in_w/2:in_h/2:in_w/2:in_h/2
4744 Crop height for getting Greek harmony:
4746 crop=in_w:1/PHI*in_w
4750 Apply trembling effect:
4752 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)
4756 Apply erratic camera effect depending on timestamp:
4758 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)"
4762 Set x depending on the value of y:
4764 crop=in_w/2:in_h/2:y:10+10*sin(n/10)
4768 @subsection Commands
4770 This filter supports the following commands:
4776 Set width/height of the output video and the horizontal/vertical position
4778 The command accepts the same syntax of the corresponding option.
4780 If the specified expression is not valid, it is kept at its current
4786 Auto-detect the crop size.
4788 It calculates the necessary cropping parameters and prints the
4789 recommended parameters via the logging system. The detected dimensions
4790 correspond to the non-black area of the input video.
4792 It accepts the following parameters:
4797 Set higher black value threshold, which can be optionally specified
4798 from nothing (0) to everything (255 for 8bit based formats). An intensity
4799 value greater to the set value is considered non-black. It defaults to 24.
4800 You can also specify a value between 0.0 and 1.0 which will be scaled depending
4801 on the bitdepth of the pixel format.
4804 The value which the width/height should be divisible by. It defaults to
4805 16. The offset is automatically adjusted to center the video. Use 2 to
4806 get only even dimensions (needed for 4:2:2 video). 16 is best when
4807 encoding to most video codecs.
4809 @item reset_count, reset
4810 Set the counter that determines after how many frames cropdetect will
4811 reset the previously detected largest video area and start over to
4812 detect the current optimal crop area. Default value is 0.
4814 This can be useful when channel logos distort the video area. 0
4815 indicates 'never reset', and returns the largest area encountered during
4822 Apply color adjustments using curves.
4824 This filter is similar to the Adobe Photoshop and GIMP curves tools. Each
4825 component (red, green and blue) has its values defined by @var{N} key points
4826 tied from each other using a smooth curve. The x-axis represents the pixel
4827 values from the input frame, and the y-axis the new pixel values to be set for
4830 By default, a component curve is defined by the two points @var{(0;0)} and
4831 @var{(1;1)}. This creates a straight line where each original pixel value is
4832 "adjusted" to its own value, which means no change to the image.
4834 The filter allows you to redefine these two points and add some more. A new
4835 curve (using a natural cubic spline interpolation) will be define to pass
4836 smoothly through all these new coordinates. The new defined points needs to be
4837 strictly increasing over the x-axis, and their @var{x} and @var{y} values must
4838 be in the @var{[0;1]} interval. If the computed curves happened to go outside
4839 the vector spaces, the values will be clipped accordingly.
4841 If there is no key point defined in @code{x=0}, the filter will automatically
4842 insert a @var{(0;0)} point. In the same way, if there is no key point defined
4843 in @code{x=1}, the filter will automatically insert a @var{(1;1)} point.
4845 The filter accepts the following options:
4849 Select one of the available color presets. This option can be used in addition
4850 to the @option{r}, @option{g}, @option{b} parameters; in this case, the later
4851 options takes priority on the preset values.
4852 Available presets are:
4855 @item color_negative
4858 @item increase_contrast
4860 @item linear_contrast
4861 @item medium_contrast
4863 @item strong_contrast
4866 Default is @code{none}.
4868 Set the master key points. These points will define a second pass mapping. It
4869 is sometimes called a "luminance" or "value" mapping. It can be used with
4870 @option{r}, @option{g}, @option{b} or @option{all} since it acts like a
4871 post-processing LUT.
4873 Set the key points for the red component.
4875 Set the key points for the green component.
4877 Set the key points for the blue component.
4879 Set the key points for all components (not including master).
4880 Can be used in addition to the other key points component
4881 options. In this case, the unset component(s) will fallback on this
4882 @option{all} setting.
4884 Specify a Photoshop curves file (@code{.acv}) to import the settings from.
4887 To avoid some filtergraph syntax conflicts, each key points list need to be
4888 defined using the following syntax: @code{x0/y0 x1/y1 x2/y2 ...}.
4890 @subsection Examples
4894 Increase slightly the middle level of blue:
4896 curves=blue='0.5/0.58'
4902 curves=r='0/0.11 .42/.51 1/0.95':g='0.50/0.48':b='0/0.22 .49/.44 1/0.8'
4904 Here we obtain the following coordinates for each components:
4907 @code{(0;0.11) (0.42;0.51) (1;0.95)}
4909 @code{(0;0) (0.50;0.48) (1;1)}
4911 @code{(0;0.22) (0.49;0.44) (1;0.80)}
4915 The previous example can also be achieved with the associated built-in preset:
4917 curves=preset=vintage
4927 Use a Photoshop preset and redefine the points of the green component:
4929 curves=psfile='MyCurvesPresets/purple.acv':green='0.45/0.53'
4935 Denoise frames using 2D DCT (frequency domain filtering).
4937 This filter is not designed for real time.
4939 The filter accepts the following options:
4943 Set the noise sigma constant.
4945 This @var{sigma} defines a hard threshold of @code{3 * sigma}; every DCT
4946 coefficient (absolute value) below this threshold with be dropped.
4948 If you need a more advanced filtering, see @option{expr}.
4950 Default is @code{0}.
4953 Set number overlapping pixels for each block. Since the filter can be slow, you
4954 may want to reduce this value, at the cost of a less effective filter and the
4955 risk of various artefacts.
4957 If the overlapping value doesn't permit processing the whole input width or
4958 height, a warning will be displayed and according borders won't be denoised.
4960 Default value is @var{blocksize}-1, which is the best possible setting.
4963 Set the coefficient factor expression.
4965 For each coefficient of a DCT block, this expression will be evaluated as a
4966 multiplier value for the coefficient.
4968 If this is option is set, the @option{sigma} option will be ignored.
4970 The absolute value of the coefficient can be accessed through the @var{c}
4974 Set the @var{blocksize} using the number of bits. @code{1<<@var{n}} defines the
4975 @var{blocksize}, which is the width and height of the processed blocks.
4977 The default value is @var{3} (8x8) and can be raised to @var{4} for a
4978 @var{blocksize} of 16x16. Note that changing this setting has huge consequences
4979 on the speed processing. Also, a larger block size does not necessarily means a
4983 @subsection Examples
4985 Apply a denoise with a @option{sigma} of @code{4.5}:
4990 The same operation can be achieved using the expression system:
4992 dctdnoiz=e='gte(c, 4.5*3)'
4995 Violent denoise using a block size of @code{16x16}:
5002 Remove banding artifacts from input video.
5003 It works by replacing banded pixels with average value of referenced pixels.
5005 The filter accepts the following options:
5012 Set banding detection threshold for each plane. Default is 0.02.
5013 Valid range is 0.00003 to 0.5.
5014 If difference between current pixel and reference pixel is less than threshold,
5015 it will be considered as banded.
5018 Banding detection range in pixels. Default is 16. If positive, random number
5019 in range 0 to set value will be used. If negative, exact absolute value
5021 The range defines square of four pixels around current pixel.
5024 Set direction in radians from which four pixel will be compared. If positive,
5025 random direction from 0 to set direction will be picked. If negative, exact of
5026 absolute value will be picked. For example direction 0, -PI or -2*PI radians
5027 will pick only pixels on same row and -PI/2 will pick only pixels on same
5031 If enabled, current pixel is compared with average value of all four
5032 surrounding pixels. The default is enabled. If disabled current pixel is
5033 compared with all four surrounding pixels. The pixel is considered banded
5034 if only all four differences with surrounding pixels are less than threshold.
5040 Drop duplicated frames at regular intervals.
5042 The filter accepts the following options:
5046 Set the number of frames from which one will be dropped. Setting this to
5047 @var{N} means one frame in every batch of @var{N} frames will be dropped.
5048 Default is @code{5}.
5051 Set the threshold for duplicate detection. If the difference metric for a frame
5052 is less than or equal to this value, then it is declared as duplicate. Default
5056 Set scene change threshold. Default is @code{15}.
5060 Set the size of the x and y-axis blocks used during metric calculations.
5061 Larger blocks give better noise suppression, but also give worse detection of
5062 small movements. Must be a power of two. Default is @code{32}.
5065 Mark main input as a pre-processed input and activate clean source input
5066 stream. This allows the input to be pre-processed with various filters to help
5067 the metrics calculation while keeping the frame selection lossless. When set to
5068 @code{1}, the first stream is for the pre-processed input, and the second
5069 stream is the clean source from where the kept frames are chosen. Default is
5073 Set whether or not chroma is considered in the metric calculations. Default is
5079 Apply deflate effect to the video.
5081 This filter replaces the pixel by the local(3x3) average by taking into account
5082 only values lower than the pixel.
5084 It accepts the following options:
5091 Limit the maximum change for each plane, default is 65535.
5092 If 0, plane will remain unchanged.
5097 Remove judder produced by partially interlaced telecined content.
5099 Judder can be introduced, for instance, by @ref{pullup} filter. If the original
5100 source was partially telecined content then the output of @code{pullup,dejudder}
5101 will have a variable frame rate. May change the recorded frame rate of the
5102 container. Aside from that change, this filter will not affect constant frame
5105 The option available in this filter is:
5109 Specify the length of the window over which the judder repeats.
5111 Accepts any integer greater than 1. Useful values are:
5115 If the original was telecined from 24 to 30 fps (Film to NTSC).
5118 If the original was telecined from 25 to 30 fps (PAL to NTSC).
5121 If a mixture of the two.
5124 The default is @samp{4}.
5129 Suppress a TV station logo by a simple interpolation of the surrounding
5130 pixels. Just set a rectangle covering the logo and watch it disappear
5131 (and sometimes something even uglier appear - your mileage may vary).
5133 It accepts the following parameters:
5138 Specify the top left corner coordinates of the logo. They must be
5143 Specify the width and height of the logo to clear. They must be
5147 Specify the thickness of the fuzzy edge of the rectangle (added to
5148 @var{w} and @var{h}). The default value is 1. This option is
5149 deprecated, setting higher values should no longer be necessary and
5153 When set to 1, a green rectangle is drawn on the screen to simplify
5154 finding the right @var{x}, @var{y}, @var{w}, and @var{h} parameters.
5155 The default value is 0.
5157 The rectangle is drawn on the outermost pixels which will be (partly)
5158 replaced with interpolated values. The values of the next pixels
5159 immediately outside this rectangle in each direction will be used to
5160 compute the interpolated pixel values inside the rectangle.
5164 @subsection Examples
5168 Set a rectangle covering the area with top left corner coordinates 0,0
5169 and size 100x77, and a band of size 10:
5171 delogo=x=0:y=0:w=100:h=77:band=10
5178 Attempt to fix small changes in horizontal and/or vertical shift. This
5179 filter helps remove camera shake from hand-holding a camera, bumping a
5180 tripod, moving on a vehicle, etc.
5182 The filter accepts the following options:
5190 Specify a rectangular area where to limit the search for motion
5192 If desired the search for motion vectors can be limited to a
5193 rectangular area of the frame defined by its top left corner, width
5194 and height. These parameters have the same meaning as the drawbox
5195 filter which can be used to visualise the position of the bounding
5198 This is useful when simultaneous movement of subjects within the frame
5199 might be confused for camera motion by the motion vector search.
5201 If any or all of @var{x}, @var{y}, @var{w} and @var{h} are set to -1
5202 then the full frame is used. This allows later options to be set
5203 without specifying the bounding box for the motion vector search.
5205 Default - search the whole frame.
5209 Specify the maximum extent of movement in x and y directions in the
5210 range 0-64 pixels. Default 16.
5213 Specify how to generate pixels to fill blanks at the edge of the
5214 frame. Available values are:
5217 Fill zeroes at blank locations
5219 Original image at blank locations
5221 Extruded edge value at blank locations
5223 Mirrored edge at blank locations
5225 Default value is @samp{mirror}.
5228 Specify the blocksize to use for motion search. Range 4-128 pixels,
5232 Specify the contrast threshold for blocks. Only blocks with more than
5233 the specified contrast (difference between darkest and lightest
5234 pixels) will be considered. Range 1-255, default 125.
5237 Specify the search strategy. Available values are:
5240 Set exhaustive search
5242 Set less exhaustive search.
5244 Default value is @samp{exhaustive}.
5247 If set then a detailed log of the motion search is written to the
5251 If set to 1, specify using OpenCL capabilities, only available if
5252 FFmpeg was configured with @code{--enable-opencl}. Default value is 0.
5258 Apply an exact inverse of the telecine operation. It requires a predefined
5259 pattern specified using the pattern option which must be the same as that passed
5260 to the telecine filter.
5262 This filter accepts the following options:
5271 The default value is @code{top}.
5275 A string of numbers representing the pulldown pattern you wish to apply.
5276 The default value is @code{23}.
5279 A number representing position of the first frame with respect to the telecine
5280 pattern. This is to be used if the stream is cut. The default value is @code{0}.
5285 Apply dilation effect to the video.
5287 This filter replaces the pixel by the local(3x3) maximum.
5289 It accepts the following options:
5296 Limit the maximum change for each plane, default is 65535.
5297 If 0, plane will remain unchanged.
5300 Flag which specifies the pixel to refer to. Default is 255 i.e. all eight
5303 Flags to local 3x3 coordinates maps like this:
5312 Displace pixels as indicated by second and third input stream.
5314 It takes three input streams and outputs one stream, the first input is the
5315 source, and second and third input are displacement maps.
5317 The second input specifies how much to displace pixels along the
5318 x-axis, while the third input specifies how much to displace pixels
5320 If one of displacement map streams terminates, last frame from that
5321 displacement map will be used.
5323 Note that once generated, displacements maps can be reused over and over again.
5325 A description of the accepted options follows.
5329 Set displace behavior for pixels that are out of range.
5331 Available values are:
5334 Missing pixels are replaced by black pixels.
5337 Adjacent pixels will spread out to replace missing pixels.
5340 Out of range pixels are wrapped so they point to pixels of other side.
5342 Default is @samp{smear}.
5346 @subsection Examples
5350 Add ripple effect to rgb input of video size hd720:
5352 ffmpeg -i INPUT -f lavfi -i nullsrc=s=hd720,lutrgb=128:128:128 -f lavfi -i nullsrc=s=hd720,geq='r=128+30*sin(2*PI*X/400+T):g=128+30*sin(2*PI*X/400+T):b=128+30*sin(2*PI*X/400+T)' -lavfi '[0][1][2]displace' OUTPUT
5356 Add wave effect to rgb input of video size hd720:
5358 ffmpeg -i INPUT -f lavfi -i nullsrc=hd720,geq='r=128+80*(sin(sqrt((X-W/2)*(X-W/2)+(Y-H/2)*(Y-H/2))/220*2*PI+T)):g=128+80*(sin(sqrt((X-W/2)*(X-W/2)+(Y-H/2)*(Y-H/2))/220*2*PI+T)):b=128+80*(sin(sqrt((X-W/2)*(X-W/2)+(Y-H/2)*(Y-H/2))/220*2*PI+T))' -lavfi '[1]split[x][y],[0][x][y]displace' OUTPUT
5364 Draw a colored box on the input image.
5366 It accepts the following parameters:
5371 The expressions which specify the top left corner coordinates of the box. It defaults to 0.
5375 The expressions which specify the width and height of the box; if 0 they are interpreted as
5376 the input width and height. It defaults to 0.
5379 Specify the color of the box to write. For the general syntax of this option,
5380 check the "Color" section in the ffmpeg-utils manual. If the special
5381 value @code{invert} is used, the box edge color is the same as the
5382 video with inverted luma.
5385 The expression which sets the thickness of the box edge. Default value is @code{3}.
5387 See below for the list of accepted constants.
5390 The parameters for @var{x}, @var{y}, @var{w} and @var{h} and @var{t} are expressions containing the
5391 following constants:
5395 The input display aspect ratio, it is the same as (@var{w} / @var{h}) * @var{sar}.
5399 horizontal and vertical chroma subsample values. For example for the
5400 pixel format "yuv422p" @var{hsub} is 2 and @var{vsub} is 1.
5404 The input width and height.
5407 The input sample aspect ratio.
5411 The x and y offset coordinates where the box is drawn.
5415 The width and height of the drawn box.
5418 The thickness of the drawn box.
5420 These constants allow the @var{x}, @var{y}, @var{w}, @var{h} and @var{t} expressions to refer to
5421 each other, so you may for example specify @code{y=x/dar} or @code{h=w/dar}.
5425 @subsection Examples
5429 Draw a black box around the edge of the input image:
5435 Draw a box with color red and an opacity of 50%:
5437 drawbox=10:20:200:60:red@@0.5
5440 The previous example can be specified as:
5442 drawbox=x=10:y=20:w=200:h=60:color=red@@0.5
5446 Fill the box with pink color:
5448 drawbox=x=10:y=10:w=100:h=100:color=pink@@0.5:t=max
5452 Draw a 2-pixel red 2.40:1 mask:
5454 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
5458 @section drawgraph, adrawgraph
5460 Draw a graph using input video or audio metadata.
5462 It accepts the following parameters:
5466 Set 1st frame metadata key from which metadata values will be used to draw a graph.
5469 Set 1st foreground color expression.
5472 Set 2nd frame metadata key from which metadata values will be used to draw a graph.
5475 Set 2nd foreground color expression.
5478 Set 3rd frame metadata key from which metadata values will be used to draw a graph.
5481 Set 3rd foreground color expression.
5484 Set 4th frame metadata key from which metadata values will be used to draw a graph.
5487 Set 4th foreground color expression.
5490 Set minimal value of metadata value.
5493 Set maximal value of metadata value.
5496 Set graph background color. Default is white.
5501 Available values for mode is:
5508 Default is @code{line}.
5513 Available values for slide is:
5516 Draw new frame when right border is reached.
5519 Replace old columns with new ones.
5522 Scroll from right to left.
5525 Scroll from left to right.
5528 Default is @code{frame}.
5531 Set size of graph video. For the syntax of this option, check the
5532 @ref{video size syntax,,"Video size" section in the ffmpeg-utils manual,ffmpeg-utils}.
5533 The default value is @code{900x256}.
5535 The foreground color expressions can use the following variables:
5538 Minimal value of metadata value.
5541 Maximal value of metadata value.
5544 Current metadata key value.
5547 The color is defined as 0xAABBGGRR.
5550 Example using metadata from @ref{signalstats} filter:
5552 signalstats,drawgraph=lavfi.signalstats.YAVG:min=0:max=255
5555 Example using metadata from @ref{ebur128} filter:
5557 ebur128=metadata=1,adrawgraph=lavfi.r128.M:min=-120:max=5
5562 Draw a grid on the input image.
5564 It accepts the following parameters:
5569 The expressions which specify the coordinates of some point of grid intersection (meant to configure offset). Both default to 0.
5573 The expressions which specify the width and height of the grid cell, if 0 they are interpreted as the
5574 input width and height, respectively, minus @code{thickness}, so image gets
5575 framed. Default to 0.
5578 Specify the color of the grid. For the general syntax of this option,
5579 check the "Color" section in the ffmpeg-utils manual. If the special
5580 value @code{invert} is used, the grid color is the same as the
5581 video with inverted luma.
5584 The expression which sets the thickness of the grid line. Default value is @code{1}.
5586 See below for the list of accepted constants.
5589 The parameters for @var{x}, @var{y}, @var{w} and @var{h} and @var{t} are expressions containing the
5590 following constants:
5594 The input display aspect ratio, it is the same as (@var{w} / @var{h}) * @var{sar}.
5598 horizontal and vertical chroma subsample values. For example for the
5599 pixel format "yuv422p" @var{hsub} is 2 and @var{vsub} is 1.
5603 The input grid cell width and height.
5606 The input sample aspect ratio.
5610 The x and y coordinates of some point of grid intersection (meant to configure offset).
5614 The width and height of the drawn cell.
5617 The thickness of the drawn cell.
5619 These constants allow the @var{x}, @var{y}, @var{w}, @var{h} and @var{t} expressions to refer to
5620 each other, so you may for example specify @code{y=x/dar} or @code{h=w/dar}.
5624 @subsection Examples
5628 Draw a grid with cell 100x100 pixels, thickness 2 pixels, with color red and an opacity of 50%:
5630 drawgrid=width=100:height=100:thickness=2:color=red@@0.5
5634 Draw a white 3x3 grid with an opacity of 50%:
5636 drawgrid=w=iw/3:h=ih/3:t=2:c=white@@0.5
5643 Draw a text string or text from a specified file on top of a video, using the
5644 libfreetype library.
5646 To enable compilation of this filter, you need to configure FFmpeg with
5647 @code{--enable-libfreetype}.
5648 To enable default font fallback and the @var{font} option you need to
5649 configure FFmpeg with @code{--enable-libfontconfig}.
5650 To enable the @var{text_shaping} option, you need to configure FFmpeg with
5651 @code{--enable-libfribidi}.
5655 It accepts the following parameters:
5660 Used to draw a box around text using the background color.
5661 The value must be either 1 (enable) or 0 (disable).
5662 The default value of @var{box} is 0.
5665 Set the width of the border to be drawn around the box using @var{boxcolor}.
5666 The default value of @var{boxborderw} is 0.
5669 The color to be used for drawing box around text. For the syntax of this
5670 option, check the "Color" section in the ffmpeg-utils manual.
5672 The default value of @var{boxcolor} is "white".
5675 Set the width of the border to be drawn around the text using @var{bordercolor}.
5676 The default value of @var{borderw} is 0.
5679 Set the color to be used for drawing border around text. For the syntax of this
5680 option, check the "Color" section in the ffmpeg-utils manual.
5682 The default value of @var{bordercolor} is "black".
5685 Select how the @var{text} is expanded. Can be either @code{none},
5686 @code{strftime} (deprecated) or
5687 @code{normal} (default). See the @ref{drawtext_expansion, Text expansion} section
5691 If true, check and fix text coords to avoid clipping.
5694 The color to be used for drawing fonts. For the syntax of this option, check
5695 the "Color" section in the ffmpeg-utils manual.
5697 The default value of @var{fontcolor} is "black".
5699 @item fontcolor_expr
5700 String which is expanded the same way as @var{text} to obtain dynamic
5701 @var{fontcolor} value. By default this option has empty value and is not
5702 processed. When this option is set, it overrides @var{fontcolor} option.
5705 The font family to be used for drawing text. By default Sans.
5708 The font file to be used for drawing text. The path must be included.
5709 This parameter is mandatory if the fontconfig support is disabled.
5712 This option does not exist, please see the timeline system
5715 Draw the text applying alpha blending. The value can
5716 be either a number between 0.0 and 1.0
5717 The expression accepts the same variables @var{x, y} do.
5718 The default value is 1.
5719 Please see fontcolor_expr
5722 The font size to be used for drawing text.
5723 The default value of @var{fontsize} is 16.
5726 If set to 1, attempt to shape the text (for example, reverse the order of
5727 right-to-left text and join Arabic characters) before drawing it.
5728 Otherwise, just draw the text exactly as given.
5729 By default 1 (if supported).
5732 The flags to be used for loading the fonts.
5734 The flags map the corresponding flags supported by libfreetype, and are
5735 a combination of the following values:
5742 @item vertical_layout
5743 @item force_autohint
5746 @item ignore_global_advance_width
5748 @item ignore_transform
5754 Default value is "default".
5756 For more information consult the documentation for the FT_LOAD_*
5760 The color to be used for drawing a shadow behind the drawn text. For the
5761 syntax of this option, check the "Color" section in the ffmpeg-utils manual.
5763 The default value of @var{shadowcolor} is "black".
5767 The x and y offsets for the text shadow position with respect to the
5768 position of the text. They can be either positive or negative
5769 values. The default value for both is "0".
5772 The starting frame number for the n/frame_num variable. The default value
5776 The size in number of spaces to use for rendering the tab.
5780 Set the initial timecode representation in "hh:mm:ss[:;.]ff"
5781 format. It can be used with or without text parameter. @var{timecode_rate}
5782 option must be specified.
5784 @item timecode_rate, rate, r
5785 Set the timecode frame rate (timecode only).
5788 The text string to be drawn. The text must be a sequence of UTF-8
5790 This parameter is mandatory if no file is specified with the parameter
5794 A text file containing text to be drawn. The text must be a sequence
5795 of UTF-8 encoded characters.
5797 This parameter is mandatory if no text string is specified with the
5798 parameter @var{text}.
5800 If both @var{text} and @var{textfile} are specified, an error is thrown.
5803 If set to 1, the @var{textfile} will be reloaded before each frame.
5804 Be sure to update it atomically, or it may be read partially, or even fail.
5808 The expressions which specify the offsets where text will be drawn
5809 within the video frame. They are relative to the top/left border of the
5812 The default value of @var{x} and @var{y} is "0".
5814 See below for the list of accepted constants and functions.
5817 The parameters for @var{x} and @var{y} are expressions containing the
5818 following constants and functions:
5822 input display aspect ratio, it is the same as (@var{w} / @var{h}) * @var{sar}
5826 horizontal and vertical chroma subsample values. For example for the
5827 pixel format "yuv422p" @var{hsub} is 2 and @var{vsub} is 1.
5830 the height of each text line
5838 @item max_glyph_a, ascent
5839 the maximum distance from the baseline to the highest/upper grid
5840 coordinate used to place a glyph outline point, for all the rendered
5842 It is a positive value, due to the grid's orientation with the Y axis
5845 @item max_glyph_d, descent
5846 the maximum distance from the baseline to the lowest grid coordinate
5847 used to place a glyph outline point, for all the rendered glyphs.
5848 This is a negative value, due to the grid's orientation, with the Y axis
5852 maximum glyph height, that is the maximum height for all the glyphs
5853 contained in the rendered text, it is equivalent to @var{ascent} -
5857 maximum glyph width, that is the maximum width for all the glyphs
5858 contained in the rendered text
5861 the number of input frame, starting from 0
5863 @item rand(min, max)
5864 return a random number included between @var{min} and @var{max}
5867 The input sample aspect ratio.
5870 timestamp expressed in seconds, NAN if the input timestamp is unknown
5873 the height of the rendered text
5876 the width of the rendered text
5880 the x and y offset coordinates where the text is drawn.
5882 These parameters allow the @var{x} and @var{y} expressions to refer
5883 each other, so you can for example specify @code{y=x/dar}.
5886 @anchor{drawtext_expansion}
5887 @subsection Text expansion
5889 If @option{expansion} is set to @code{strftime},
5890 the filter recognizes strftime() sequences in the provided text and
5891 expands them accordingly. Check the documentation of strftime(). This
5892 feature is deprecated.
5894 If @option{expansion} is set to @code{none}, the text is printed verbatim.
5896 If @option{expansion} is set to @code{normal} (which is the default),
5897 the following expansion mechanism is used.
5899 The backslash character @samp{\}, followed by any character, always expands to
5900 the second character.
5902 Sequence of the form @code{%@{...@}} are expanded. The text between the
5903 braces is a function name, possibly followed by arguments separated by ':'.
5904 If the arguments contain special characters or delimiters (':' or '@}'),
5905 they should be escaped.
5907 Note that they probably must also be escaped as the value for the
5908 @option{text} option in the filter argument string and as the filter
5909 argument in the filtergraph description, and possibly also for the shell,
5910 that makes up to four levels of escaping; using a text file avoids these
5913 The following functions are available:
5918 The expression evaluation result.
5920 It must take one argument specifying the expression to be evaluated,
5921 which accepts the same constants and functions as the @var{x} and
5922 @var{y} values. Note that not all constants should be used, for
5923 example the text size is not known when evaluating the expression, so
5924 the constants @var{text_w} and @var{text_h} will have an undefined
5927 @item expr_int_format, eif
5928 Evaluate the expression's value and output as formatted integer.
5930 The first argument is the expression to be evaluated, just as for the @var{expr} function.
5931 The second argument specifies the output format. Allowed values are @samp{x},
5932 @samp{X}, @samp{d} and @samp{u}. They are treated exactly as in the
5933 @code{printf} function.
5934 The third parameter is optional and sets the number of positions taken by the output.
5935 It can be used to add padding with zeros from the left.
5938 The time at which the filter is running, expressed in UTC.
5939 It can accept an argument: a strftime() format string.
5942 The time at which the filter is running, expressed in the local time zone.
5943 It can accept an argument: a strftime() format string.
5946 Frame metadata. It must take one argument specifying metadata key.
5949 The frame number, starting from 0.
5952 A 1 character description of the current picture type.
5955 The timestamp of the current frame.
5956 It can take up to three arguments.
5958 The first argument is the format of the timestamp; it defaults to @code{flt}
5959 for seconds as a decimal number with microsecond accuracy; @code{hms} stands
5960 for a formatted @var{[-]HH:MM:SS.mmm} timestamp with millisecond accuracy.
5961 @code{gmtime} stands for the timestamp of the frame formatted as UTC time;
5962 @code{localtime} stands for the timestamp of the frame formatted as
5963 local time zone time.
5965 The second argument is an offset added to the timestamp.
5967 If the format is set to @code{localtime} or @code{gmtime},
5968 a third argument may be supplied: a strftime() format string.
5969 By default, @var{YYYY-MM-DD HH:MM:SS} format will be used.
5972 @subsection Examples
5976 Draw "Test Text" with font FreeSerif, using the default values for the
5977 optional parameters.
5980 drawtext="fontfile=/usr/share/fonts/truetype/freefont/FreeSerif.ttf: text='Test Text'"
5984 Draw 'Test Text' with font FreeSerif of size 24 at position x=100
5985 and y=50 (counting from the top-left corner of the screen), text is
5986 yellow with a red box around it. Both the text and the box have an
5990 drawtext="fontfile=/usr/share/fonts/truetype/freefont/FreeSerif.ttf: text='Test Text':\
5991 x=100: y=50: fontsize=24: fontcolor=yellow@@0.2: box=1: boxcolor=red@@0.2"
5994 Note that the double quotes are not necessary if spaces are not used
5995 within the parameter list.
5998 Show the text at the center of the video frame:
6000 drawtext="fontsize=30:fontfile=FreeSerif.ttf:text='hello world':x=(w-text_w)/2:y=(h-text_h)/2"
6004 Show a text line sliding from right to left in the last row of the video
6005 frame. The file @file{LONG_LINE} is assumed to contain a single line
6008 drawtext="fontsize=15:fontfile=FreeSerif.ttf:text=LONG_LINE:y=h-line_h:x=-50*t"
6012 Show the content of file @file{CREDITS} off the bottom of the frame and scroll up.
6014 drawtext="fontsize=20:fontfile=FreeSerif.ttf:textfile=CREDITS:y=h-20*t"
6018 Draw a single green letter "g", at the center of the input video.
6019 The glyph baseline is placed at half screen height.
6021 drawtext="fontsize=60:fontfile=FreeSerif.ttf:fontcolor=green:text=g:x=(w-max_glyph_w)/2:y=h/2-ascent"
6025 Show text for 1 second every 3 seconds:
6027 drawtext="fontfile=FreeSerif.ttf:fontcolor=white:x=100:y=x/dar:enable=lt(mod(t\,3)\,1):text='blink'"
6031 Use fontconfig to set the font. Note that the colons need to be escaped.
6033 drawtext='fontfile=Linux Libertine O-40\:style=Semibold:text=FFmpeg'
6037 Print the date of a real-time encoding (see strftime(3)):
6039 drawtext='fontfile=FreeSans.ttf:text=%@{localtime\:%a %b %d %Y@}'
6043 Show text fading in and out (appearing/disappearing):
6046 DS=1.0 # display start
6047 DE=10.0 # display end
6048 FID=1.5 # fade in duration
6049 FOD=5 # fade out duration
6050 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 @}"
6055 For more information about libfreetype, check:
6056 @url{http://www.freetype.org/}.
6058 For more information about fontconfig, check:
6059 @url{http://freedesktop.org/software/fontconfig/fontconfig-user.html}.
6061 For more information about libfribidi, check:
6062 @url{http://fribidi.org/}.
6066 Detect and draw edges. The filter uses the Canny Edge Detection algorithm.
6068 The filter accepts the following options:
6073 Set low and high threshold values used by the Canny thresholding
6076 The high threshold selects the "strong" edge pixels, which are then
6077 connected through 8-connectivity with the "weak" edge pixels selected
6078 by the low threshold.
6080 @var{low} and @var{high} threshold values must be chosen in the range
6081 [0,1], and @var{low} should be lesser or equal to @var{high}.
6083 Default value for @var{low} is @code{20/255}, and default value for @var{high}
6087 Define the drawing mode.
6091 Draw white/gray wires on black background.
6094 Mix the colors to create a paint/cartoon effect.
6097 Default value is @var{wires}.
6100 @subsection Examples
6104 Standard edge detection with custom values for the hysteresis thresholding:
6106 edgedetect=low=0.1:high=0.4
6110 Painting effect without thresholding:
6112 edgedetect=mode=colormix:high=0
6117 Set brightness, contrast, saturation and approximate gamma adjustment.
6119 The filter accepts the following options:
6123 Set the contrast expression. The value must be a float value in range
6124 @code{-2.0} to @code{2.0}. The default value is "1".
6127 Set the brightness expression. The value must be a float value in
6128 range @code{-1.0} to @code{1.0}. The default value is "0".
6131 Set the saturation expression. The value must be a float in
6132 range @code{0.0} to @code{3.0}. The default value is "1".
6135 Set the gamma expression. The value must be a float in range
6136 @code{0.1} to @code{10.0}. The default value is "1".
6139 Set the gamma expression for red. The value must be a float in
6140 range @code{0.1} to @code{10.0}. The default value is "1".
6143 Set the gamma expression for green. The value must be a float in range
6144 @code{0.1} to @code{10.0}. The default value is "1".
6147 Set the gamma expression for blue. The value must be a float in range
6148 @code{0.1} to @code{10.0}. The default value is "1".
6151 Set the gamma weight expression. It can be used to reduce the effect
6152 of a high gamma value on bright image areas, e.g. keep them from
6153 getting overamplified and just plain white. The value must be a float
6154 in range @code{0.0} to @code{1.0}. A value of @code{0.0} turns the
6155 gamma correction all the way down while @code{1.0} leaves it at its
6156 full strength. Default is "1".
6159 Set when the expressions for brightness, contrast, saturation and
6160 gamma expressions are evaluated.
6162 It accepts the following values:
6165 only evaluate expressions once during the filter initialization or
6166 when a command is processed
6169 evaluate expressions for each incoming frame
6172 Default value is @samp{init}.
6175 The expressions accept the following parameters:
6178 frame count of the input frame starting from 0
6181 byte position of the corresponding packet in the input file, NAN if
6185 frame rate of the input video, NAN if the input frame rate is unknown
6188 timestamp expressed in seconds, NAN if the input timestamp is unknown
6191 @subsection Commands
6192 The filter supports the following commands:
6196 Set the contrast expression.
6199 Set the brightness expression.
6202 Set the saturation expression.
6205 Set the gamma expression.
6208 Set the gamma_r expression.
6211 Set gamma_g expression.
6214 Set gamma_b expression.
6217 Set gamma_weight expression.
6219 The command accepts the same syntax of the corresponding option.
6221 If the specified expression is not valid, it is kept at its current
6228 Apply erosion effect to the video.
6230 This filter replaces the pixel by the local(3x3) minimum.
6232 It accepts the following options:
6239 Limit the maximum change for each plane, default is 65535.
6240 If 0, plane will remain unchanged.
6243 Flag which specifies the pixel to refer to. Default is 255 i.e. all eight
6246 Flags to local 3x3 coordinates maps like this:
6253 @section extractplanes
6255 Extract color channel components from input video stream into
6256 separate grayscale video streams.
6258 The filter accepts the following option:
6262 Set plane(s) to extract.
6264 Available values for planes are:
6275 Choosing planes not available in the input will result in an error.
6276 That means you cannot select @code{r}, @code{g}, @code{b} planes
6277 with @code{y}, @code{u}, @code{v} planes at same time.
6280 @subsection Examples
6284 Extract luma, u and v color channel component from input video frame
6285 into 3 grayscale outputs:
6287 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
6293 Apply a posterize effect using the ELBG (Enhanced LBG) algorithm.
6295 For each input image, the filter will compute the optimal mapping from
6296 the input to the output given the codebook length, that is the number
6297 of distinct output colors.
6299 This filter accepts the following options.
6302 @item codebook_length, l
6303 Set codebook length. The value must be a positive integer, and
6304 represents the number of distinct output colors. Default value is 256.
6307 Set the maximum number of iterations to apply for computing the optimal
6308 mapping. The higher the value the better the result and the higher the
6309 computation time. Default value is 1.
6312 Set a random seed, must be an integer included between 0 and
6313 UINT32_MAX. If not specified, or if explicitly set to -1, the filter
6314 will try to use a good random seed on a best effort basis.
6317 Set pal8 output pixel format. This option does not work with codebook
6318 length greater than 256.
6323 Apply a fade-in/out effect to the input video.
6325 It accepts the following parameters:
6329 The effect type can be either "in" for a fade-in, or "out" for a fade-out
6331 Default is @code{in}.
6333 @item start_frame, s
6334 Specify the number of the frame to start applying the fade
6335 effect at. Default is 0.
6338 The number of frames that the fade effect lasts. At the end of the
6339 fade-in effect, the output video will have the same intensity as the input video.
6340 At the end of the fade-out transition, the output video will be filled with the
6341 selected @option{color}.
6345 If set to 1, fade only alpha channel, if one exists on the input.
6348 @item start_time, st
6349 Specify the timestamp (in seconds) of the frame to start to apply the fade
6350 effect. If both start_frame and start_time are specified, the fade will start at
6351 whichever comes last. Default is 0.
6354 The number of seconds for which the fade effect has to last. At the end of the
6355 fade-in effect the output video will have the same intensity as the input video,
6356 at the end of the fade-out transition the output video will be filled with the
6357 selected @option{color}.
6358 If both duration and nb_frames are specified, duration is used. Default is 0
6359 (nb_frames is used by default).
6362 Specify the color of the fade. Default is "black".
6365 @subsection Examples
6369 Fade in the first 30 frames of video:
6374 The command above is equivalent to:
6380 Fade out the last 45 frames of a 200-frame video:
6383 fade=type=out:start_frame=155:nb_frames=45
6387 Fade in the first 25 frames and fade out the last 25 frames of a 1000-frame video:
6389 fade=in:0:25, fade=out:975:25
6393 Make the first 5 frames yellow, then fade in from frame 5-24:
6395 fade=in:5:20:color=yellow
6399 Fade in alpha over first 25 frames of video:
6401 fade=in:0:25:alpha=1
6405 Make the first 5.5 seconds black, then fade in for 0.5 seconds:
6407 fade=t=in:st=5.5:d=0.5
6413 Apply arbitrary expressions to samples in frequency domain
6417 Adjust the dc value (gain) of the luma plane of the image. The filter
6418 accepts an integer value in range @code{0} to @code{1000}. The default
6419 value is set to @code{0}.
6422 Adjust the dc value (gain) of the 1st chroma plane of the image. The
6423 filter accepts an integer value in range @code{0} to @code{1000}. The
6424 default value is set to @code{0}.
6427 Adjust the dc value (gain) of the 2nd chroma plane of the image. The
6428 filter accepts an integer value in range @code{0} to @code{1000}. The
6429 default value is set to @code{0}.
6432 Set the frequency domain weight expression for the luma plane.
6435 Set the frequency domain weight expression for the 1st chroma plane.
6438 Set the frequency domain weight expression for the 2nd chroma plane.
6440 The filter accepts the following variables:
6443 The coordinates of the current sample.
6447 The width and height of the image.
6450 @subsection Examples
6456 fftfilt=dc_Y=128:weight_Y='squish(1-(Y+X)/100)'
6462 fftfilt=dc_Y=0:weight_Y='squish((Y+X)/100-1)'
6468 fftfilt=dc_Y=0:weight_Y='1+squish(1-(Y+X)/100)'
6474 fftfilt=dc_Y=0:weight_Y='exp(-4 * ((Y+X)/(W+H)))'
6481 Extract a single field from an interlaced image using stride
6482 arithmetic to avoid wasting CPU time. The output frames are marked as
6485 The filter accepts the following options:
6489 Specify whether to extract the top (if the value is @code{0} or
6490 @code{top}) or the bottom field (if the value is @code{1} or
6496 Field matching filter for inverse telecine. It is meant to reconstruct the
6497 progressive frames from a telecined stream. The filter does not drop duplicated
6498 frames, so to achieve a complete inverse telecine @code{fieldmatch} needs to be
6499 followed by a decimation filter such as @ref{decimate} in the filtergraph.
6501 The separation of the field matching and the decimation is notably motivated by
6502 the possibility of inserting a de-interlacing filter fallback between the two.
6503 If the source has mixed telecined and real interlaced content,
6504 @code{fieldmatch} will not be able to match fields for the interlaced parts.
6505 But these remaining combed frames will be marked as interlaced, and thus can be
6506 de-interlaced by a later filter such as @ref{yadif} before decimation.
6508 In addition to the various configuration options, @code{fieldmatch} can take an
6509 optional second stream, activated through the @option{ppsrc} option. If
6510 enabled, the frames reconstruction will be based on the fields and frames from
6511 this second stream. This allows the first input to be pre-processed in order to
6512 help the various algorithms of the filter, while keeping the output lossless
6513 (assuming the fields are matched properly). Typically, a field-aware denoiser,
6514 or brightness/contrast adjustments can help.
6516 Note that this filter uses the same algorithms as TIVTC/TFM (AviSynth project)
6517 and VIVTC/VFM (VapourSynth project). The later is a light clone of TFM from
6518 which @code{fieldmatch} is based on. While the semantic and usage are very
6519 close, some behaviour and options names can differ.
6521 The @ref{decimate} filter currently only works for constant frame rate input.
6522 If your input has mixed telecined (30fps) and progressive content with a lower
6523 framerate like 24fps use the following filterchain to produce the necessary cfr
6524 stream: @code{dejudder,fps=30000/1001,fieldmatch,decimate}.
6526 The filter accepts the following options:
6530 Specify the assumed field order of the input stream. Available values are:
6534 Auto detect parity (use FFmpeg's internal parity value).
6536 Assume bottom field first.
6538 Assume top field first.
6541 Note that it is sometimes recommended not to trust the parity announced by the
6544 Default value is @var{auto}.
6547 Set the matching mode or strategy to use. @option{pc} mode is the safest in the
6548 sense that it won't risk creating jerkiness due to duplicate frames when
6549 possible, but if there are bad edits or blended fields it will end up
6550 outputting combed frames when a good match might actually exist. On the other
6551 hand, @option{pcn_ub} mode is the most risky in terms of creating jerkiness,
6552 but will almost always find a good frame if there is one. The other values are
6553 all somewhere in between @option{pc} and @option{pcn_ub} in terms of risking
6554 jerkiness and creating duplicate frames versus finding good matches in sections
6555 with bad edits, orphaned fields, blended fields, etc.
6557 More details about p/c/n/u/b are available in @ref{p/c/n/u/b meaning} section.
6559 Available values are:
6563 2-way matching (p/c)
6565 2-way matching, and trying 3rd match if still combed (p/c + n)
6567 2-way matching, and trying 3rd match (same order) if still combed (p/c + u)
6569 2-way matching, trying 3rd match if still combed, and trying 4th/5th matches if
6570 still combed (p/c + n + u/b)
6572 3-way matching (p/c/n)
6574 3-way matching, and trying 4th/5th matches if all 3 of the original matches are
6575 detected as combed (p/c/n + u/b)
6578 The parenthesis at the end indicate the matches that would be used for that
6579 mode assuming @option{order}=@var{tff} (and @option{field} on @var{auto} or
6582 In terms of speed @option{pc} mode is by far the fastest and @option{pcn_ub} is
6585 Default value is @var{pc_n}.
6588 Mark the main input stream as a pre-processed input, and enable the secondary
6589 input stream as the clean source to pick the fields from. See the filter
6590 introduction for more details. It is similar to the @option{clip2} feature from
6593 Default value is @code{0} (disabled).
6596 Set the field to match from. It is recommended to set this to the same value as
6597 @option{order} unless you experience matching failures with that setting. In
6598 certain circumstances changing the field that is used to match from can have a
6599 large impact on matching performance. Available values are:
6603 Automatic (same value as @option{order}).
6605 Match from the bottom field.
6607 Match from the top field.
6610 Default value is @var{auto}.
6613 Set whether or not chroma is included during the match comparisons. In most
6614 cases it is recommended to leave this enabled. You should set this to @code{0}
6615 only if your clip has bad chroma problems such as heavy rainbowing or other
6616 artifacts. Setting this to @code{0} could also be used to speed things up at
6617 the cost of some accuracy.
6619 Default value is @code{1}.
6623 These define an exclusion band which excludes the lines between @option{y0} and
6624 @option{y1} from being included in the field matching decision. An exclusion
6625 band can be used to ignore subtitles, a logo, or other things that may
6626 interfere with the matching. @option{y0} sets the starting scan line and
6627 @option{y1} sets the ending line; all lines in between @option{y0} and
6628 @option{y1} (including @option{y0} and @option{y1}) will be ignored. Setting
6629 @option{y0} and @option{y1} to the same value will disable the feature.
6630 @option{y0} and @option{y1} defaults to @code{0}.
6633 Set the scene change detection threshold as a percentage of maximum change on
6634 the luma plane. Good values are in the @code{[8.0, 14.0]} range. Scene change
6635 detection is only relevant in case @option{combmatch}=@var{sc}. The range for
6636 @option{scthresh} is @code{[0.0, 100.0]}.
6638 Default value is @code{12.0}.
6641 When @option{combatch} is not @var{none}, @code{fieldmatch} will take into
6642 account the combed scores of matches when deciding what match to use as the
6643 final match. Available values are:
6647 No final matching based on combed scores.
6649 Combed scores are only used when a scene change is detected.
6651 Use combed scores all the time.
6654 Default is @var{sc}.
6657 Force @code{fieldmatch} to calculate the combed metrics for certain matches and
6658 print them. This setting is known as @option{micout} in TFM/VFM vocabulary.
6659 Available values are:
6663 No forced calculation.
6665 Force p/c/n calculations.
6667 Force p/c/n/u/b calculations.
6670 Default value is @var{none}.
6673 This is the area combing threshold used for combed frame detection. This
6674 essentially controls how "strong" or "visible" combing must be to be detected.
6675 Larger values mean combing must be more visible and smaller values mean combing
6676 can be less visible or strong and still be detected. Valid settings are from
6677 @code{-1} (every pixel will be detected as combed) to @code{255} (no pixel will
6678 be detected as combed). This is basically a pixel difference value. A good
6679 range is @code{[8, 12]}.
6681 Default value is @code{9}.
6684 Sets whether or not chroma is considered in the combed frame decision. Only
6685 disable this if your source has chroma problems (rainbowing, etc.) that are
6686 causing problems for the combed frame detection with chroma enabled. Actually,
6687 using @option{chroma}=@var{0} is usually more reliable, except for the case
6688 where there is chroma only combing in the source.
6690 Default value is @code{0}.
6694 Respectively set the x-axis and y-axis size of the window used during combed
6695 frame detection. This has to do with the size of the area in which
6696 @option{combpel} pixels are required to be detected as combed for a frame to be
6697 declared combed. See the @option{combpel} parameter description for more info.
6698 Possible values are any number that is a power of 2 starting at 4 and going up
6701 Default value is @code{16}.
6704 The number of combed pixels inside any of the @option{blocky} by
6705 @option{blockx} size blocks on the frame for the frame to be detected as
6706 combed. While @option{cthresh} controls how "visible" the combing must be, this
6707 setting controls "how much" combing there must be in any localized area (a
6708 window defined by the @option{blockx} and @option{blocky} settings) on the
6709 frame. Minimum value is @code{0} and maximum is @code{blocky x blockx} (at
6710 which point no frames will ever be detected as combed). This setting is known
6711 as @option{MI} in TFM/VFM vocabulary.
6713 Default value is @code{80}.
6716 @anchor{p/c/n/u/b meaning}
6717 @subsection p/c/n/u/b meaning
6719 @subsubsection p/c/n
6721 We assume the following telecined stream:
6724 Top fields: 1 2 2 3 4
6725 Bottom fields: 1 2 3 4 4
6728 The numbers correspond to the progressive frame the fields relate to. Here, the
6729 first two frames are progressive, the 3rd and 4th are combed, and so on.
6731 When @code{fieldmatch} is configured to run a matching from bottom
6732 (@option{field}=@var{bottom}) this is how this input stream get transformed:
6737 B 1 2 3 4 4 <-- matching reference
6746 As a result of the field matching, we can see that some frames get duplicated.
6747 To perform a complete inverse telecine, you need to rely on a decimation filter
6748 after this operation. See for instance the @ref{decimate} filter.
6750 The same operation now matching from top fields (@option{field}=@var{top})
6755 T 1 2 2 3 4 <-- matching reference
6765 In these examples, we can see what @var{p}, @var{c} and @var{n} mean;
6766 basically, they refer to the frame and field of the opposite parity:
6769 @item @var{p} matches the field of the opposite parity in the previous frame
6770 @item @var{c} matches the field of the opposite parity in the current frame
6771 @item @var{n} matches the field of the opposite parity in the next frame
6776 The @var{u} and @var{b} matching are a bit special in the sense that they match
6777 from the opposite parity flag. In the following examples, we assume that we are
6778 currently matching the 2nd frame (Top:2, bottom:2). According to the match, a
6779 'x' is placed above and below each matched fields.
6781 With bottom matching (@option{field}=@var{bottom}):
6786 Top 1 2 2 1 2 2 1 2 2 1 2 2 1 2 2
6787 Bottom 1 2 3 1 2 3 1 2 3 1 2 3 1 2 3
6795 With top matching (@option{field}=@var{top}):
6800 Top 1 2 2 1 2 2 1 2 2 1 2 2 1 2 2
6801 Bottom 1 2 3 1 2 3 1 2 3 1 2 3 1 2 3
6809 @subsection Examples
6811 Simple IVTC of a top field first telecined stream:
6813 fieldmatch=order=tff:combmatch=none, decimate
6816 Advanced IVTC, with fallback on @ref{yadif} for still combed frames:
6818 fieldmatch=order=tff:combmatch=full, yadif=deint=interlaced, decimate
6823 Transform the field order of the input video.
6825 It accepts the following parameters:
6830 The output field order. Valid values are @var{tff} for top field first or @var{bff}
6831 for bottom field first.
6834 The default value is @samp{tff}.
6836 The transformation is done by shifting the picture content up or down
6837 by one line, and filling the remaining line with appropriate picture content.
6838 This method is consistent with most broadcast field order converters.
6840 If the input video is not flagged as being interlaced, or it is already
6841 flagged as being of the required output field order, then this filter does
6842 not alter the incoming video.
6844 It is very useful when converting to or from PAL DV material,
6845 which is bottom field first.
6849 ffmpeg -i in.vob -vf "fieldorder=bff" out.dv
6852 @section fifo, afifo
6854 Buffer input images and send them when they are requested.
6856 It is mainly useful when auto-inserted by the libavfilter
6859 It does not take parameters.
6863 Find a rectangular object
6865 It accepts the following options:
6869 Filepath of the object image, needs to be in gray8.
6872 Detection threshold, default is 0.5.
6875 Number of mipmaps, default is 3.
6877 @item xmin, ymin, xmax, ymax
6878 Specifies the rectangle in which to search.
6881 @subsection Examples
6885 Generate a representative palette of a given video using @command{ffmpeg}:
6887 ffmpeg -i file.ts -vf find_rect=newref.pgm,cover_rect=cover.jpg:mode=cover new.mkv
6893 Cover a rectangular object
6895 It accepts the following options:
6899 Filepath of the optional cover image, needs to be in yuv420.
6904 It accepts the following values:
6907 cover it by the supplied image
6909 cover it by interpolating the surrounding pixels
6912 Default value is @var{blur}.
6915 @subsection Examples
6919 Generate a representative palette of a given video using @command{ffmpeg}:
6921 ffmpeg -i file.ts -vf find_rect=newref.pgm,cover_rect=cover.jpg:mode=cover new.mkv
6928 Convert the input video to one of the specified pixel formats.
6929 Libavfilter will try to pick one that is suitable as input to
6932 It accepts the following parameters:
6936 A '|'-separated list of pixel format names, such as
6937 "pix_fmts=yuv420p|monow|rgb24".
6941 @subsection Examples
6945 Convert the input video to the @var{yuv420p} format
6947 format=pix_fmts=yuv420p
6950 Convert the input video to any of the formats in the list
6952 format=pix_fmts=yuv420p|yuv444p|yuv410p
6959 Convert the video to specified constant frame rate by duplicating or dropping
6960 frames as necessary.
6962 It accepts the following parameters:
6966 The desired output frame rate. The default is @code{25}.
6971 Possible values are:
6974 zero round towards 0
6978 round towards -infinity
6980 round towards +infinity
6984 The default is @code{near}.
6987 Assume the first PTS should be the given value, in seconds. This allows for
6988 padding/trimming at the start of stream. By default, no assumption is made
6989 about the first frame's expected PTS, so no padding or trimming is done.
6990 For example, this could be set to 0 to pad the beginning with duplicates of
6991 the first frame if a video stream starts after the audio stream or to trim any
6992 frames with a negative PTS.
6996 Alternatively, the options can be specified as a flat string:
6997 @var{fps}[:@var{round}].
6999 See also the @ref{setpts} filter.
7001 @subsection Examples
7005 A typical usage in order to set the fps to 25:
7011 Sets the fps to 24, using abbreviation and rounding method to round to nearest:
7013 fps=fps=film:round=near
7019 Pack two different video streams into a stereoscopic video, setting proper
7020 metadata on supported codecs. The two views should have the same size and
7021 framerate and processing will stop when the shorter video ends. Please note
7022 that you may conveniently adjust view properties with the @ref{scale} and
7025 It accepts the following parameters:
7029 The desired packing format. Supported values are:
7034 The views are next to each other (default).
7037 The views are on top of each other.
7040 The views are packed by line.
7043 The views are packed by column.
7046 The views are temporally interleaved.
7055 # Convert left and right views into a frame-sequential video
7056 ffmpeg -i LEFT -i RIGHT -filter_complex framepack=frameseq OUTPUT
7058 # Convert views into a side-by-side video with the same output resolution as the input
7059 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
7064 Change the frame rate by interpolating new video output frames from the source
7067 This filter is not designed to function correctly with interlaced media. If
7068 you wish to change the frame rate of interlaced media then you are required
7069 to deinterlace before this filter and re-interlace after this filter.
7071 A description of the accepted options follows.
7075 Specify the output frames per second. This option can also be specified
7076 as a value alone. The default is @code{50}.
7079 Specify the start of a range where the output frame will be created as a
7080 linear interpolation of two frames. The range is [@code{0}-@code{255}],
7081 the default is @code{15}.
7084 Specify the end of a range where the output frame will be created as a
7085 linear interpolation of two frames. The range is [@code{0}-@code{255}],
7086 the default is @code{240}.
7089 Specify the level at which a scene change is detected as a value between
7090 0 and 100 to indicate a new scene; a low value reflects a low
7091 probability for the current frame to introduce a new scene, while a higher
7092 value means the current frame is more likely to be one.
7093 The default is @code{7}.
7096 Specify flags influencing the filter process.
7098 Available value for @var{flags} is:
7101 @item scene_change_detect, scd
7102 Enable scene change detection using the value of the option @var{scene}.
7103 This flag is enabled by default.
7109 Select one frame every N-th frame.
7111 This filter accepts the following option:
7114 Select frame after every @code{step} frames.
7115 Allowed values are positive integers higher than 0. Default value is @code{1}.
7121 Apply a frei0r effect to the input video.
7123 To enable the compilation of this filter, you need to install the frei0r
7124 header and configure FFmpeg with @code{--enable-frei0r}.
7126 It accepts the following parameters:
7131 The name of the frei0r effect to load. If the environment variable
7132 @env{FREI0R_PATH} is defined, the frei0r effect is searched for in each of the
7133 directories specified by the colon-separated list in @env{FREIOR_PATH}.
7134 Otherwise, the standard frei0r paths are searched, in this order:
7135 @file{HOME/.frei0r-1/lib/}, @file{/usr/local/lib/frei0r-1/},
7136 @file{/usr/lib/frei0r-1/}.
7139 A '|'-separated list of parameters to pass to the frei0r effect.
7143 A frei0r effect parameter can be a boolean (its value is either
7144 "y" or "n"), a double, a color (specified as
7145 @var{R}/@var{G}/@var{B}, where @var{R}, @var{G}, and @var{B} are floating point
7146 numbers between 0.0 and 1.0, inclusive) or by a color description specified in the "Color"
7147 section in the ffmpeg-utils manual), a position (specified as @var{X}/@var{Y}, where
7148 @var{X} and @var{Y} are floating point numbers) and/or a string.
7150 The number and types of parameters depend on the loaded effect. If an
7151 effect parameter is not specified, the default value is set.
7153 @subsection Examples
7157 Apply the distort0r effect, setting the first two double parameters:
7159 frei0r=filter_name=distort0r:filter_params=0.5|0.01
7163 Apply the colordistance effect, taking a color as the first parameter:
7165 frei0r=colordistance:0.2/0.3/0.4
7166 frei0r=colordistance:violet
7167 frei0r=colordistance:0x112233
7171 Apply the perspective effect, specifying the top left and top right image
7174 frei0r=perspective:0.2/0.2|0.8/0.2
7178 For more information, see
7179 @url{http://frei0r.dyne.org}
7183 Apply fast and simple postprocessing. It is a faster version of @ref{spp}.
7185 It splits (I)DCT into horizontal/vertical passes. Unlike the simple post-
7186 processing filter, one of them is performed once per block, not per pixel.
7187 This allows for much higher speed.
7189 The filter accepts the following options:
7193 Set quality. This option defines the number of levels for averaging. It accepts
7194 an integer in the range 4-5. Default value is @code{4}.
7197 Force a constant quantization parameter. It accepts an integer in range 0-63.
7198 If not set, the filter will use the QP from the video stream (if available).
7201 Set filter strength. It accepts an integer in range -15 to 32. Lower values mean
7202 more details but also more artifacts, while higher values make the image smoother
7203 but also blurrier. Default value is @code{0} − PSNR optimal.
7206 Enable the use of the QP from the B-Frames if set to @code{1}. Using this
7207 option may cause flicker since the B-Frames have often larger QP. Default is
7208 @code{0} (not enabled).
7214 The filter accepts the following options:
7218 Set the luminance expression.
7220 Set the chrominance blue expression.
7222 Set the chrominance red expression.
7224 Set the alpha expression.
7226 Set the red expression.
7228 Set the green expression.
7230 Set the blue expression.
7233 The colorspace is selected according to the specified options. If one
7234 of the @option{lum_expr}, @option{cb_expr}, or @option{cr_expr}
7235 options is specified, the filter will automatically select a YCbCr
7236 colorspace. If one of the @option{red_expr}, @option{green_expr}, or
7237 @option{blue_expr} options is specified, it will select an RGB
7240 If one of the chrominance expression is not defined, it falls back on the other
7241 one. If no alpha expression is specified it will evaluate to opaque value.
7242 If none of chrominance expressions are specified, they will evaluate
7243 to the luminance expression.
7245 The expressions can use the following variables and functions:
7249 The sequential number of the filtered frame, starting from @code{0}.
7253 The coordinates of the current sample.
7257 The width and height of the image.
7261 Width and height scale depending on the currently filtered plane. It is the
7262 ratio between the corresponding luma plane number of pixels and the current
7263 plane ones. E.g. for YUV4:2:0 the values are @code{1,1} for the luma plane, and
7264 @code{0.5,0.5} for chroma planes.
7267 Time of the current frame, expressed in seconds.
7270 Return the value of the pixel at location (@var{x},@var{y}) of the current
7274 Return the value of the pixel at location (@var{x},@var{y}) of the luminance
7278 Return the value of the pixel at location (@var{x},@var{y}) of the
7279 blue-difference chroma plane. Return 0 if there is no such plane.
7282 Return the value of the pixel at location (@var{x},@var{y}) of the
7283 red-difference chroma plane. Return 0 if there is no such plane.
7288 Return the value of the pixel at location (@var{x},@var{y}) of the
7289 red/green/blue component. Return 0 if there is no such component.
7292 Return the value of the pixel at location (@var{x},@var{y}) of the alpha
7293 plane. Return 0 if there is no such plane.
7296 For functions, if @var{x} and @var{y} are outside the area, the value will be
7297 automatically clipped to the closer edge.
7299 @subsection Examples
7303 Flip the image horizontally:
7309 Generate a bidimensional sine wave, with angle @code{PI/3} and a
7310 wavelength of 100 pixels:
7312 geq=128 + 100*sin(2*(PI/100)*(cos(PI/3)*(X-50*T) + sin(PI/3)*Y)):128:128
7316 Generate a fancy enigmatic moving light:
7318 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
7322 Generate a quick emboss effect:
7324 format=gray,geq=lum_expr='(p(X,Y)+(256-p(X-4,Y-4)))/2'
7328 Modify RGB components depending on pixel position:
7330 geq=r='X/W*r(X,Y)':g='(1-X/W)*g(X,Y)':b='(H-Y)/H*b(X,Y)'
7334 Create a radial gradient that is the same size as the input (also see
7335 the @ref{vignette} filter):
7337 geq=lum=255*gauss((X/W-0.5)*3)*gauss((Y/H-0.5)*3)/gauss(0)/gauss(0),format=gray
7341 Create a linear gradient to use as a mask for another filter, then
7342 compose with @ref{overlay}. In this example the video will gradually
7343 become more blurry from the top to the bottom of the y-axis as defined
7344 by the linear gradient:
7346 ffmpeg -i input.mp4 -filter_complex "geq=lum=255*(Y/H),format=gray[grad];[0:v]boxblur=4[blur];[blur][grad]alphamerge[alpha];[0:v][alpha]overlay" output.mp4
7352 Fix the banding artifacts that are sometimes introduced into nearly flat
7353 regions by truncation to 8bit color depth.
7354 Interpolate the gradients that should go where the bands are, and
7357 It is designed for playback only. Do not use it prior to
7358 lossy compression, because compression tends to lose the dither and
7359 bring back the bands.
7361 It accepts the following parameters:
7366 The maximum amount by which the filter will change any one pixel. This is also
7367 the threshold for detecting nearly flat regions. Acceptable values range from
7368 .51 to 64; the default value is 1.2. Out-of-range values will be clipped to the
7372 The neighborhood to fit the gradient to. A larger radius makes for smoother
7373 gradients, but also prevents the filter from modifying the pixels near detailed
7374 regions. Acceptable values are 8-32; the default value is 16. Out-of-range
7375 values will be clipped to the valid range.
7379 Alternatively, the options can be specified as a flat string:
7380 @var{strength}[:@var{radius}]
7382 @subsection Examples
7386 Apply the filter with a @code{3.5} strength and radius of @code{8}:
7392 Specify radius, omitting the strength (which will fall-back to the default
7403 Apply a Hald CLUT to a video stream.
7405 First input is the video stream to process, and second one is the Hald CLUT.
7406 The Hald CLUT input can be a simple picture or a complete video stream.
7408 The filter accepts the following options:
7412 Force termination when the shortest input terminates. Default is @code{0}.
7414 Continue applying the last CLUT after the end of the stream. A value of
7415 @code{0} disable the filter after the last frame of the CLUT is reached.
7416 Default is @code{1}.
7419 @code{haldclut} also has the same interpolation options as @ref{lut3d} (both
7420 filters share the same internals).
7422 More information about the Hald CLUT can be found on Eskil Steenberg's website
7423 (Hald CLUT author) at @url{http://www.quelsolaar.com/technology/clut.html}.
7425 @subsection Workflow examples
7427 @subsubsection Hald CLUT video stream
7429 Generate an identity Hald CLUT stream altered with various effects:
7431 ffmpeg -f lavfi -i @ref{haldclutsrc}=8 -vf "hue=H=2*PI*t:s=sin(2*PI*t)+1, curves=cross_process" -t 10 -c:v ffv1 clut.nut
7434 Note: make sure you use a lossless codec.
7436 Then use it with @code{haldclut} to apply it on some random stream:
7438 ffmpeg -f lavfi -i mandelbrot -i clut.nut -filter_complex '[0][1] haldclut' -t 20 mandelclut.mkv
7441 The Hald CLUT will be applied to the 10 first seconds (duration of
7442 @file{clut.nut}), then the latest picture of that CLUT stream will be applied
7443 to the remaining frames of the @code{mandelbrot} stream.
7445 @subsubsection Hald CLUT with preview
7447 A Hald CLUT is supposed to be a squared image of @code{Level*Level*Level} by
7448 @code{Level*Level*Level} pixels. For a given Hald CLUT, FFmpeg will select the
7449 biggest possible square starting at the top left of the picture. The remaining
7450 padding pixels (bottom or right) will be ignored. This area can be used to add
7451 a preview of the Hald CLUT.
7453 Typically, the following generated Hald CLUT will be supported by the
7454 @code{haldclut} filter:
7457 ffmpeg -f lavfi -i @ref{haldclutsrc}=8 -vf "
7458 pad=iw+320 [padded_clut];
7459 smptebars=s=320x256, split [a][b];
7460 [padded_clut][a] overlay=W-320:h, curves=color_negative [main];
7461 [main][b] overlay=W-320" -frames:v 1 clut.png
7464 It contains the original and a preview of the effect of the CLUT: SMPTE color
7465 bars are displayed on the right-top, and below the same color bars processed by
7468 Then, the effect of this Hald CLUT can be visualized with:
7470 ffplay input.mkv -vf "movie=clut.png, [in] haldclut"
7475 Flip the input video horizontally.
7477 For example, to horizontally flip the input video with @command{ffmpeg}:
7479 ffmpeg -i in.avi -vf "hflip" out.avi
7483 This filter applies a global color histogram equalization on a
7486 It can be used to correct video that has a compressed range of pixel
7487 intensities. The filter redistributes the pixel intensities to
7488 equalize their distribution across the intensity range. It may be
7489 viewed as an "automatically adjusting contrast filter". This filter is
7490 useful only for correcting degraded or poorly captured source
7493 The filter accepts the following options:
7497 Determine the amount of equalization to be applied. As the strength
7498 is reduced, the distribution of pixel intensities more-and-more
7499 approaches that of the input frame. The value must be a float number
7500 in the range [0,1] and defaults to 0.200.
7503 Set the maximum intensity that can generated and scale the output
7504 values appropriately. The strength should be set as desired and then
7505 the intensity can be limited if needed to avoid washing-out. The value
7506 must be a float number in the range [0,1] and defaults to 0.210.
7509 Set the antibanding level. If enabled the filter will randomly vary
7510 the luminance of output pixels by a small amount to avoid banding of
7511 the histogram. Possible values are @code{none}, @code{weak} or
7512 @code{strong}. It defaults to @code{none}.
7517 Compute and draw a color distribution histogram for the input video.
7519 The computed histogram is a representation of the color component
7520 distribution in an image.
7522 Standard histogram displays the color components distribution in an image.
7523 Displays color graph for each color component. Shows distribution of
7524 the Y, U, V, A or R, G, B components, depending on input format, in the
7525 current frame. Below each graph a color component scale meter is shown.
7527 The filter accepts the following options:
7531 Set height of level. Default value is @code{200}.
7532 Allowed range is [50, 2048].
7535 Set height of color scale. Default value is @code{12}.
7536 Allowed range is [0, 40].
7540 It accepts the following values:
7543 Per color component graphs are placed below each other.
7546 Presents information identical to that in the @code{parade}, except
7547 that the graphs representing color components are superimposed directly
7550 Default is @code{parade}.
7553 Set mode. Can be either @code{linear}, or @code{logarithmic}.
7554 Default is @code{linear}.
7557 Set what color components to display.
7558 Default is @code{7}.
7561 @subsection Examples
7566 Calculate and draw histogram:
7568 ffplay -i input -vf histogram
7576 This is a high precision/quality 3d denoise filter. It aims to reduce
7577 image noise, producing smooth images and making still images really
7578 still. It should enhance compressibility.
7580 It accepts the following optional parameters:
7584 A non-negative floating point number which specifies spatial luma strength.
7587 @item chroma_spatial
7588 A non-negative floating point number which specifies spatial chroma strength.
7589 It defaults to 3.0*@var{luma_spatial}/4.0.
7592 A floating point number which specifies luma temporal strength. It defaults to
7593 6.0*@var{luma_spatial}/4.0.
7596 A floating point number which specifies chroma temporal strength. It defaults to
7597 @var{luma_tmp}*@var{chroma_spatial}/@var{luma_spatial}.
7602 Apply a high-quality magnification filter designed for pixel art. This filter
7603 was originally created by Maxim Stepin.
7605 It accepts the following option:
7609 Set the scaling dimension: @code{2} for @code{hq2x}, @code{3} for
7610 @code{hq3x} and @code{4} for @code{hq4x}.
7611 Default is @code{3}.
7615 Stack input videos horizontally.
7617 All streams must be of same pixel format and of same height.
7619 Note that this filter is faster than using @ref{overlay} and @ref{pad} filter
7620 to create same output.
7622 The filter accept the following option:
7626 Set number of input streams. Default is 2.
7629 If set to 1, force the output to terminate when the shortest input
7630 terminates. Default value is 0.
7635 Modify the hue and/or the saturation of the input.
7637 It accepts the following parameters:
7641 Specify the hue angle as a number of degrees. It accepts an expression,
7642 and defaults to "0".
7645 Specify the saturation in the [-10,10] range. It accepts an expression and
7649 Specify the hue angle as a number of radians. It accepts an
7650 expression, and defaults to "0".
7653 Specify the brightness in the [-10,10] range. It accepts an expression and
7657 @option{h} and @option{H} are mutually exclusive, and can't be
7658 specified at the same time.
7660 The @option{b}, @option{h}, @option{H} and @option{s} option values are
7661 expressions containing the following constants:
7665 frame count of the input frame starting from 0
7668 presentation timestamp of the input frame expressed in time base units
7671 frame rate of the input video, NAN if the input frame rate is unknown
7674 timestamp expressed in seconds, NAN if the input timestamp is unknown
7677 time base of the input video
7680 @subsection Examples
7684 Set the hue to 90 degrees and the saturation to 1.0:
7690 Same command but expressing the hue in radians:
7696 Rotate hue and make the saturation swing between 0
7697 and 2 over a period of 1 second:
7699 hue="H=2*PI*t: s=sin(2*PI*t)+1"
7703 Apply a 3 seconds saturation fade-in effect starting at 0:
7708 The general fade-in expression can be written as:
7710 hue="s=min(0\, max((t-START)/DURATION\, 1))"
7714 Apply a 3 seconds saturation fade-out effect starting at 5 seconds:
7716 hue="s=max(0\, min(1\, (8-t)/3))"
7719 The general fade-out expression can be written as:
7721 hue="s=max(0\, min(1\, (START+DURATION-t)/DURATION))"
7726 @subsection Commands
7728 This filter supports the following commands:
7734 Modify the hue and/or the saturation and/or brightness of the input video.
7735 The command accepts the same syntax of the corresponding option.
7737 If the specified expression is not valid, it is kept at its current
7743 Detect video interlacing type.
7745 This filter tries to detect if the input frames as interlaced, progressive,
7746 top or bottom field first. It will also try and detect fields that are
7747 repeated between adjacent frames (a sign of telecine).
7749 Single frame detection considers only immediately adjacent frames when classifying each frame.
7750 Multiple frame detection incorporates the classification history of previous frames.
7752 The filter will log these metadata values:
7755 @item single.current_frame
7756 Detected type of current frame using single-frame detection. One of:
7757 ``tff'' (top field first), ``bff'' (bottom field first),
7758 ``progressive'', or ``undetermined''
7761 Cumulative number of frames detected as top field first using single-frame detection.
7764 Cumulative number of frames detected as top field first using multiple-frame detection.
7767 Cumulative number of frames detected as bottom field first using single-frame detection.
7769 @item multiple.current_frame
7770 Detected type of current frame using multiple-frame detection. One of:
7771 ``tff'' (top field first), ``bff'' (bottom field first),
7772 ``progressive'', or ``undetermined''
7775 Cumulative number of frames detected as bottom field first using multiple-frame detection.
7777 @item single.progressive
7778 Cumulative number of frames detected as progressive using single-frame detection.
7780 @item multiple.progressive
7781 Cumulative number of frames detected as progressive using multiple-frame detection.
7783 @item single.undetermined
7784 Cumulative number of frames that could not be classified using single-frame detection.
7786 @item multiple.undetermined
7787 Cumulative number of frames that could not be classified using multiple-frame detection.
7789 @item repeated.current_frame
7790 Which field in the current frame is repeated from the last. One of ``neither'', ``top'', or ``bottom''.
7792 @item repeated.neither
7793 Cumulative number of frames with no repeated field.
7796 Cumulative number of frames with the top field repeated from the previous frame's top field.
7798 @item repeated.bottom
7799 Cumulative number of frames with the bottom field repeated from the previous frame's bottom field.
7802 The filter accepts the following options:
7806 Set interlacing threshold.
7808 Set progressive threshold.
7810 Threshold for repeated field detection.
7812 Number of frames after which a given frame's contribution to the
7813 statistics is halved (i.e., it contributes only 0.5 to it's
7814 classification). The default of 0 means that all frames seen are given
7815 full weight of 1.0 forever.
7816 @item analyze_interlaced_flag
7817 When this is not 0 then idet will use the specified number of frames to determine
7818 if the interlaced flag is accurate, it will not count undetermined frames.
7819 If the flag is found to be accurate it will be used without any further
7820 computations, if it is found to be inaccurate it will be cleared without any
7821 further computations. This allows inserting the idet filter as a low computational
7822 method to clean up the interlaced flag
7827 Deinterleave or interleave fields.
7829 This filter allows one to process interlaced images fields without
7830 deinterlacing them. Deinterleaving splits the input frame into 2
7831 fields (so called half pictures). Odd lines are moved to the top
7832 half of the output image, even lines to the bottom half.
7833 You can process (filter) them independently and then re-interleave them.
7835 The filter accepts the following options:
7839 @item chroma_mode, c
7841 Available values for @var{luma_mode}, @var{chroma_mode} and
7842 @var{alpha_mode} are:
7848 @item deinterleave, d
7849 Deinterleave fields, placing one above the other.
7852 Interleave fields. Reverse the effect of deinterleaving.
7854 Default value is @code{none}.
7857 @item chroma_swap, cs
7858 @item alpha_swap, as
7859 Swap luma/chroma/alpha fields. Exchange even & odd lines. Default value is @code{0}.
7864 Apply inflate effect to the video.
7866 This filter replaces the pixel by the local(3x3) average by taking into account
7867 only values higher than the pixel.
7869 It accepts the following options:
7876 Limit the maximum change for each plane, default is 65535.
7877 If 0, plane will remain unchanged.
7882 Simple interlacing filter from progressive contents. This interleaves upper (or
7883 lower) lines from odd frames with lower (or upper) lines from even frames,
7884 halving the frame rate and preserving image height.
7887 Original Original New Frame
7888 Frame 'j' Frame 'j+1' (tff)
7889 ========== =========== ==================
7890 Line 0 --------------------> Frame 'j' Line 0
7891 Line 1 Line 1 ----> Frame 'j+1' Line 1
7892 Line 2 ---------------------> Frame 'j' Line 2
7893 Line 3 Line 3 ----> Frame 'j+1' Line 3
7895 New Frame + 1 will be generated by Frame 'j+2' and Frame 'j+3' and so on
7898 It accepts the following optional parameters:
7902 This determines whether the interlaced frame is taken from the even
7903 (tff - default) or odd (bff) lines of the progressive frame.
7906 Enable (default) or disable the vertical lowpass filter to avoid twitter
7907 interlacing and reduce moire patterns.
7912 Deinterlace input video by applying Donald Graft's adaptive kernel
7913 deinterling. Work on interlaced parts of a video to produce
7916 The description of the accepted parameters follows.
7920 Set the threshold which affects the filter's tolerance when
7921 determining if a pixel line must be processed. It must be an integer
7922 in the range [0,255] and defaults to 10. A value of 0 will result in
7923 applying the process on every pixels.
7926 Paint pixels exceeding the threshold value to white if set to 1.
7930 Set the fields order. Swap fields if set to 1, leave fields alone if
7934 Enable additional sharpening if set to 1. Default is 0.
7937 Enable twoway sharpening if set to 1. Default is 0.
7940 @subsection Examples
7944 Apply default values:
7946 kerndeint=thresh=10:map=0:order=0:sharp=0:twoway=0
7950 Enable additional sharpening:
7956 Paint processed pixels in white:
7962 @section lenscorrection
7964 Correct radial lens distortion
7966 This filter can be used to correct for radial distortion as can result from the use
7967 of wide angle lenses, and thereby re-rectify the image. To find the right parameters
7968 one can use tools available for example as part of opencv or simply trial-and-error.
7969 To use opencv use the calibration sample (under samples/cpp) from the opencv sources
7970 and extract the k1 and k2 coefficients from the resulting matrix.
7972 Note that effectively the same filter is available in the open-source tools Krita and
7973 Digikam from the KDE project.
7975 In contrast to the @ref{vignette} filter, which can also be used to compensate lens errors,
7976 this filter corrects the distortion of the image, whereas @ref{vignette} corrects the
7977 brightness distribution, so you may want to use both filters together in certain
7978 cases, though you will have to take care of ordering, i.e. whether vignetting should
7979 be applied before or after lens correction.
7983 The filter accepts the following options:
7987 Relative x-coordinate of the focal point of the image, and thereby the center of the
7988 distortion. This value has a range [0,1] and is expressed as fractions of the image
7991 Relative y-coordinate of the focal point of the image, and thereby the center of the
7992 distortion. This value has a range [0,1] and is expressed as fractions of the image
7995 Coefficient of the quadratic correction term. 0.5 means no correction.
7997 Coefficient of the double quadratic correction term. 0.5 means no correction.
8000 The formula that generates the correction is:
8002 @var{r_src} = @var{r_tgt} * (1 + @var{k1} * (@var{r_tgt} / @var{r_0})^2 + @var{k2} * (@var{r_tgt} / @var{r_0})^4)
8004 where @var{r_0} is halve of the image diagonal and @var{r_src} and @var{r_tgt} are the
8005 distances from the focal point in the source and target images, respectively.
8010 Apply a 3D LUT to an input video.
8012 The filter accepts the following options:
8016 Set the 3D LUT file name.
8018 Currently supported formats:
8030 Select interpolation mode.
8032 Available values are:
8036 Use values from the nearest defined point.
8038 Interpolate values using the 8 points defining a cube.
8040 Interpolate values using a tetrahedron.
8044 @section lut, lutrgb, lutyuv
8046 Compute a look-up table for binding each pixel component input value
8047 to an output value, and apply it to the input video.
8049 @var{lutyuv} applies a lookup table to a YUV input video, @var{lutrgb}
8050 to an RGB input video.
8052 These filters accept the following parameters:
8055 set first pixel component expression
8057 set second pixel component expression
8059 set third pixel component expression
8061 set fourth pixel component expression, corresponds to the alpha component
8064 set red component expression
8066 set green component expression
8068 set blue component expression
8070 alpha component expression
8073 set Y/luminance component expression
8075 set U/Cb component expression
8077 set V/Cr component expression
8080 Each of them specifies the expression to use for computing the lookup table for
8081 the corresponding pixel component values.
8083 The exact component associated to each of the @var{c*} options depends on the
8086 The @var{lut} filter requires either YUV or RGB pixel formats in input,
8087 @var{lutrgb} requires RGB pixel formats in input, and @var{lutyuv} requires YUV.
8089 The expressions can contain the following constants and functions:
8094 The input width and height.
8097 The input value for the pixel component.
8100 The input value, clipped to the @var{minval}-@var{maxval} range.
8103 The maximum value for the pixel component.
8106 The minimum value for the pixel component.
8109 The negated value for the pixel component value, clipped to the
8110 @var{minval}-@var{maxval} range; it corresponds to the expression
8111 "maxval-clipval+minval".
8114 The computed value in @var{val}, clipped to the
8115 @var{minval}-@var{maxval} range.
8117 @item gammaval(gamma)
8118 The computed gamma correction value of the pixel component value,
8119 clipped to the @var{minval}-@var{maxval} range. It corresponds to the
8121 "pow((clipval-minval)/(maxval-minval)\,@var{gamma})*(maxval-minval)+minval"
8125 All expressions default to "val".
8127 @subsection Examples
8133 lutrgb="r=maxval+minval-val:g=maxval+minval-val:b=maxval+minval-val"
8134 lutyuv="y=maxval+minval-val:u=maxval+minval-val:v=maxval+minval-val"
8137 The above is the same as:
8139 lutrgb="r=negval:g=negval:b=negval"
8140 lutyuv="y=negval:u=negval:v=negval"
8150 Remove chroma components, turning the video into a graytone image:
8152 lutyuv="u=128:v=128"
8156 Apply a luma burning effect:
8162 Remove green and blue components:
8168 Set a constant alpha channel value on input:
8170 format=rgba,lutrgb=a="maxval-minval/2"
8174 Correct luminance gamma by a factor of 0.5:
8176 lutyuv=y=gammaval(0.5)
8180 Discard least significant bits of luma:
8182 lutyuv=y='bitand(val, 128+64+32)'
8186 @section maskedmerge
8188 Merge the first input stream with the second input stream using per pixel
8189 weights in the third input stream.
8191 A value of 0 in the third stream pixel component means that pixel component
8192 from first stream is returned unchanged, while maximum value (eg. 255 for
8193 8-bit videos) means that pixel component from second stream is returned
8194 unchanged. Intermediate values define the amount of merging between both
8195 input stream's pixel components.
8197 This filter accepts the following options:
8200 Set which planes will be processed as bitmap, unprocessed planes will be
8201 copied from first stream.
8202 By default value 0xf, all planes will be processed.
8207 Apply motion-compensation deinterlacing.
8209 It needs one field per frame as input and must thus be used together
8210 with yadif=1/3 or equivalent.
8212 This filter accepts the following options:
8215 Set the deinterlacing mode.
8217 It accepts one of the following values:
8222 use iterative motion estimation
8224 like @samp{slow}, but use multiple reference frames.
8226 Default value is @samp{fast}.
8229 Set the picture field parity assumed for the input video. It must be
8230 one of the following values:
8234 assume top field first
8236 assume bottom field first
8239 Default value is @samp{bff}.
8242 Set per-block quantization parameter (QP) used by the internal
8245 Higher values should result in a smoother motion vector field but less
8246 optimal individual vectors. Default value is 1.
8249 @section mergeplanes
8251 Merge color channel components from several video streams.
8253 The filter accepts up to 4 input streams, and merge selected input
8254 planes to the output video.
8256 This filter accepts the following options:
8259 Set input to output plane mapping. Default is @code{0}.
8261 The mappings is specified as a bitmap. It should be specified as a
8262 hexadecimal number in the form 0xAa[Bb[Cc[Dd]]]. 'Aa' describes the
8263 mapping for the first plane of the output stream. 'A' sets the number of
8264 the input stream to use (from 0 to 3), and 'a' the plane number of the
8265 corresponding input to use (from 0 to 3). The rest of the mappings is
8266 similar, 'Bb' describes the mapping for the output stream second
8267 plane, 'Cc' describes the mapping for the output stream third plane and
8268 'Dd' describes the mapping for the output stream fourth plane.
8271 Set output pixel format. Default is @code{yuva444p}.
8274 @subsection Examples
8278 Merge three gray video streams of same width and height into single video stream:
8280 [a0][a1][a2]mergeplanes=0x001020:yuv444p
8284 Merge 1st yuv444p stream and 2nd gray video stream into yuva444p video stream:
8286 [a0][a1]mergeplanes=0x00010210:yuva444p
8290 Swap Y and A plane in yuva444p stream:
8292 format=yuva444p,mergeplanes=0x03010200:yuva444p
8296 Swap U and V plane in yuv420p stream:
8298 format=yuv420p,mergeplanes=0x000201:yuv420p
8302 Cast a rgb24 clip to yuv444p:
8304 format=rgb24,mergeplanes=0x000102:yuv444p
8310 Drop frames that do not differ greatly from the previous frame in
8311 order to reduce frame rate.
8313 The main use of this filter is for very-low-bitrate encoding
8314 (e.g. streaming over dialup modem), but it could in theory be used for
8315 fixing movies that were inverse-telecined incorrectly.
8317 A description of the accepted options follows.
8321 Set the maximum number of consecutive frames which can be dropped (if
8322 positive), or the minimum interval between dropped frames (if
8323 negative). If the value is 0, the frame is dropped unregarding the
8324 number of previous sequentially dropped frames.
8331 Set the dropping threshold values.
8333 Values for @option{hi} and @option{lo} are for 8x8 pixel blocks and
8334 represent actual pixel value differences, so a threshold of 64
8335 corresponds to 1 unit of difference for each pixel, or the same spread
8336 out differently over the block.
8338 A frame is a candidate for dropping if no 8x8 blocks differ by more
8339 than a threshold of @option{hi}, and if no more than @option{frac} blocks (1
8340 meaning the whole image) differ by more than a threshold of @option{lo}.
8342 Default value for @option{hi} is 64*12, default value for @option{lo} is
8343 64*5, and default value for @option{frac} is 0.33.
8351 It accepts an integer in input; if non-zero it negates the
8352 alpha component (if available). The default value in input is 0.
8356 Force libavfilter not to use any of the specified pixel formats for the
8357 input to the next filter.
8359 It accepts the following parameters:
8363 A '|'-separated list of pixel format names, such as
8364 apix_fmts=yuv420p|monow|rgb24".
8368 @subsection Examples
8372 Force libavfilter to use a format different from @var{yuv420p} for the
8373 input to the vflip filter:
8375 noformat=pix_fmts=yuv420p,vflip
8379 Convert the input video to any of the formats not contained in the list:
8381 noformat=yuv420p|yuv444p|yuv410p
8387 Add noise on video input frame.
8389 The filter accepts the following options:
8397 Set noise seed for specific pixel component or all pixel components in case
8398 of @var{all_seed}. Default value is @code{123457}.
8400 @item all_strength, alls
8401 @item c0_strength, c0s
8402 @item c1_strength, c1s
8403 @item c2_strength, c2s
8404 @item c3_strength, c3s
8405 Set noise strength for specific pixel component or all pixel components in case
8406 @var{all_strength}. Default value is @code{0}. Allowed range is [0, 100].
8408 @item all_flags, allf
8413 Set pixel component flags or set flags for all components if @var{all_flags}.
8414 Available values for component flags are:
8417 averaged temporal noise (smoother)
8419 mix random noise with a (semi)regular pattern
8421 temporal noise (noise pattern changes between frames)
8423 uniform noise (gaussian otherwise)
8427 @subsection Examples
8429 Add temporal and uniform noise to input video:
8431 noise=alls=20:allf=t+u
8436 Pass the video source unchanged to the output.
8439 Optical Character Recognition
8441 This filter uses Tesseract for optical character recognition.
8443 It accepts the following options:
8447 Set datapath to tesseract data. Default is to use whatever was
8448 set at installation.
8451 Set language, default is "eng".
8454 Set character whitelist.
8457 Set character blacklist.
8460 The filter exports recognized text as the frame metadata @code{lavfi.ocr.text}.
8464 Apply a video transform using libopencv.
8466 To enable this filter, install the libopencv library and headers and
8467 configure FFmpeg with @code{--enable-libopencv}.
8469 It accepts the following parameters:
8474 The name of the libopencv filter to apply.
8477 The parameters to pass to the libopencv filter. If not specified, the default
8482 Refer to the official libopencv documentation for more precise
8484 @url{http://docs.opencv.org/master/modules/imgproc/doc/filtering.html}
8486 Several libopencv filters are supported; see the following subsections.
8491 Dilate an image by using a specific structuring element.
8492 It corresponds to the libopencv function @code{cvDilate}.
8494 It accepts the parameters: @var{struct_el}|@var{nb_iterations}.
8496 @var{struct_el} represents a structuring element, and has the syntax:
8497 @var{cols}x@var{rows}+@var{anchor_x}x@var{anchor_y}/@var{shape}
8499 @var{cols} and @var{rows} represent the number of columns and rows of
8500 the structuring element, @var{anchor_x} and @var{anchor_y} the anchor
8501 point, and @var{shape} the shape for the structuring element. @var{shape}
8502 must be "rect", "cross", "ellipse", or "custom".
8504 If the value for @var{shape} is "custom", it must be followed by a
8505 string of the form "=@var{filename}". The file with name
8506 @var{filename} is assumed to represent a binary image, with each
8507 printable character corresponding to a bright pixel. When a custom
8508 @var{shape} is used, @var{cols} and @var{rows} are ignored, the number
8509 or columns and rows of the read file are assumed instead.
8511 The default value for @var{struct_el} is "3x3+0x0/rect".
8513 @var{nb_iterations} specifies the number of times the transform is
8514 applied to the image, and defaults to 1.
8518 # Use the default values
8521 # Dilate using a structuring element with a 5x5 cross, iterating two times
8522 ocv=filter_name=dilate:filter_params=5x5+2x2/cross|2
8524 # Read the shape from the file diamond.shape, iterating two times.
8525 # The file diamond.shape may contain a pattern of characters like this
8531 # The specified columns and rows are ignored
8532 # but the anchor point coordinates are not
8533 ocv=dilate:0x0+2x2/custom=diamond.shape|2
8538 Erode an image by using a specific structuring element.
8539 It corresponds to the libopencv function @code{cvErode}.
8541 It accepts the parameters: @var{struct_el}:@var{nb_iterations},
8542 with the same syntax and semantics as the @ref{dilate} filter.
8546 Smooth the input video.
8548 The filter takes the following parameters:
8549 @var{type}|@var{param1}|@var{param2}|@var{param3}|@var{param4}.
8551 @var{type} is the type of smooth filter to apply, and must be one of
8552 the following values: "blur", "blur_no_scale", "median", "gaussian",
8553 or "bilateral". The default value is "gaussian".
8555 The meaning of @var{param1}, @var{param2}, @var{param3}, and @var{param4}
8556 depend on the smooth type. @var{param1} and
8557 @var{param2} accept integer positive values or 0. @var{param3} and
8558 @var{param4} accept floating point values.
8560 The default value for @var{param1} is 3. The default value for the
8561 other parameters is 0.
8563 These parameters correspond to the parameters assigned to the
8564 libopencv function @code{cvSmooth}.
8569 Overlay one video on top of another.
8571 It takes two inputs and has one output. The first input is the "main"
8572 video on which the second input is overlaid.
8574 It accepts the following parameters:
8576 A description of the accepted options follows.
8581 Set the expression for the x and y coordinates of the overlaid video
8582 on the main video. Default value is "0" for both expressions. In case
8583 the expression is invalid, it is set to a huge value (meaning that the
8584 overlay will not be displayed within the output visible area).
8587 The action to take when EOF is encountered on the secondary input; it accepts
8588 one of the following values:
8592 Repeat the last frame (the default).
8596 Pass the main input through.
8600 Set when the expressions for @option{x}, and @option{y} are evaluated.
8602 It accepts the following values:
8605 only evaluate expressions once during the filter initialization or
8606 when a command is processed
8609 evaluate expressions for each incoming frame
8612 Default value is @samp{frame}.
8615 If set to 1, force the output to terminate when the shortest input
8616 terminates. Default value is 0.
8619 Set the format for the output video.
8621 It accepts the following values:
8636 Default value is @samp{yuv420}.
8638 @item rgb @emph{(deprecated)}
8639 If set to 1, force the filter to accept inputs in the RGB
8640 color space. Default value is 0. This option is deprecated, use
8641 @option{format} instead.
8644 If set to 1, force the filter to draw the last overlay frame over the
8645 main input until the end of the stream. A value of 0 disables this
8646 behavior. Default value is 1.
8649 The @option{x}, and @option{y} expressions can contain the following
8655 The main input width and height.
8659 The overlay input width and height.
8663 The computed values for @var{x} and @var{y}. They are evaluated for
8668 horizontal and vertical chroma subsample values of the output
8669 format. For example for the pixel format "yuv422p" @var{hsub} is 2 and
8673 the number of input frame, starting from 0
8676 the position in the file of the input frame, NAN if unknown
8679 The timestamp, expressed in seconds. It's NAN if the input timestamp is unknown.
8683 Note that the @var{n}, @var{pos}, @var{t} variables are available only
8684 when evaluation is done @emph{per frame}, and will evaluate to NAN
8685 when @option{eval} is set to @samp{init}.
8687 Be aware that frames are taken from each input video in timestamp
8688 order, hence, if their initial timestamps differ, it is a good idea
8689 to pass the two inputs through a @var{setpts=PTS-STARTPTS} filter to
8690 have them begin in the same zero timestamp, as the example for
8691 the @var{movie} filter does.
8693 You can chain together more overlays but you should test the
8694 efficiency of such approach.
8696 @subsection Commands
8698 This filter supports the following commands:
8702 Modify the x and y of the overlay input.
8703 The command accepts the same syntax of the corresponding option.
8705 If the specified expression is not valid, it is kept at its current
8709 @subsection Examples
8713 Draw the overlay at 10 pixels from the bottom right corner of the main
8716 overlay=main_w-overlay_w-10:main_h-overlay_h-10
8719 Using named options the example above becomes:
8721 overlay=x=main_w-overlay_w-10:y=main_h-overlay_h-10
8725 Insert a transparent PNG logo in the bottom left corner of the input,
8726 using the @command{ffmpeg} tool with the @code{-filter_complex} option:
8728 ffmpeg -i input -i logo -filter_complex 'overlay=10:main_h-overlay_h-10' output
8732 Insert 2 different transparent PNG logos (second logo on bottom
8733 right corner) using the @command{ffmpeg} tool:
8735 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
8739 Add a transparent color layer on top of the main video; @code{WxH}
8740 must specify the size of the main input to the overlay filter:
8742 color=color=red@@.3:size=WxH [over]; [in][over] overlay [out]
8746 Play an original video and a filtered version (here with the deshake
8747 filter) side by side using the @command{ffplay} tool:
8749 ffplay input.avi -vf 'split[a][b]; [a]pad=iw*2:ih[src]; [b]deshake[filt]; [src][filt]overlay=w'
8752 The above command is the same as:
8754 ffplay input.avi -vf 'split[b], pad=iw*2[src], [b]deshake, [src]overlay=w'
8758 Make a sliding overlay appearing from the left to the right top part of the
8759 screen starting since time 2:
8761 overlay=x='if(gte(t,2), -w+(t-2)*20, NAN)':y=0
8765 Compose output by putting two input videos side to side:
8767 ffmpeg -i left.avi -i right.avi -filter_complex "
8768 nullsrc=size=200x100 [background];
8769 [0:v] setpts=PTS-STARTPTS, scale=100x100 [left];
8770 [1:v] setpts=PTS-STARTPTS, scale=100x100 [right];
8771 [background][left] overlay=shortest=1 [background+left];
8772 [background+left][right] overlay=shortest=1:x=100 [left+right]
8777 Mask 10-20 seconds of a video by applying the delogo filter to a section
8779 ffmpeg -i test.avi -codec:v:0 wmv2 -ar 11025 -b:v 9000k
8780 -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]'
8785 Chain several overlays in cascade:
8787 nullsrc=s=200x200 [bg];
8788 testsrc=s=100x100, split=4 [in0][in1][in2][in3];
8789 [in0] lutrgb=r=0, [bg] overlay=0:0 [mid0];
8790 [in1] lutrgb=g=0, [mid0] overlay=100:0 [mid1];
8791 [in2] lutrgb=b=0, [mid1] overlay=0:100 [mid2];
8792 [in3] null, [mid2] overlay=100:100 [out0]
8799 Apply Overcomplete Wavelet denoiser.
8801 The filter accepts the following options:
8807 Larger depth values will denoise lower frequency components more, but
8808 slow down filtering.
8810 Must be an int in the range 8-16, default is @code{8}.
8812 @item luma_strength, ls
8815 Must be a double value in the range 0-1000, default is @code{1.0}.
8817 @item chroma_strength, cs
8818 Set chroma strength.
8820 Must be a double value in the range 0-1000, default is @code{1.0}.
8826 Add paddings to the input image, and place the original input at the
8827 provided @var{x}, @var{y} coordinates.
8829 It accepts the following parameters:
8834 Specify an expression for the size of the output image with the
8835 paddings added. If the value for @var{width} or @var{height} is 0, the
8836 corresponding input size is used for the output.
8838 The @var{width} expression can reference the value set by the
8839 @var{height} expression, and vice versa.
8841 The default value of @var{width} and @var{height} is 0.
8845 Specify the offsets to place the input image at within the padded area,
8846 with respect to the top/left border of the output image.
8848 The @var{x} expression can reference the value set by the @var{y}
8849 expression, and vice versa.
8851 The default value of @var{x} and @var{y} is 0.
8854 Specify the color of the padded area. For the syntax of this option,
8855 check the "Color" section in the ffmpeg-utils manual.
8857 The default value of @var{color} is "black".
8860 The value for the @var{width}, @var{height}, @var{x}, and @var{y}
8861 options are expressions containing the following constants:
8866 The input video width and height.
8870 These are the same as @var{in_w} and @var{in_h}.
8874 The output width and height (the size of the padded area), as
8875 specified by the @var{width} and @var{height} expressions.
8879 These are the same as @var{out_w} and @var{out_h}.
8883 The x and y offsets as specified by the @var{x} and @var{y}
8884 expressions, or NAN if not yet specified.
8887 same as @var{iw} / @var{ih}
8890 input sample aspect ratio
8893 input display aspect ratio, it is the same as (@var{iw} / @var{ih}) * @var{sar}
8897 The horizontal and vertical chroma subsample values. For example for the
8898 pixel format "yuv422p" @var{hsub} is 2 and @var{vsub} is 1.
8901 @subsection Examples
8905 Add paddings with the color "violet" to the input video. The output video
8906 size is 640x480, and the top-left corner of the input video is placed at
8909 pad=640:480:0:40:violet
8912 The example above is equivalent to the following command:
8914 pad=width=640:height=480:x=0:y=40:color=violet
8918 Pad the input to get an output with dimensions increased by 3/2,
8919 and put the input video at the center of the padded area:
8921 pad="3/2*iw:3/2*ih:(ow-iw)/2:(oh-ih)/2"
8925 Pad the input to get a squared output with size equal to the maximum
8926 value between the input width and height, and put the input video at
8927 the center of the padded area:
8929 pad="max(iw\,ih):ow:(ow-iw)/2:(oh-ih)/2"
8933 Pad the input to get a final w/h ratio of 16:9:
8935 pad="ih*16/9:ih:(ow-iw)/2:(oh-ih)/2"
8939 In case of anamorphic video, in order to set the output display aspect
8940 correctly, it is necessary to use @var{sar} in the expression,
8941 according to the relation:
8943 (ih * X / ih) * sar = output_dar
8944 X = output_dar / sar
8947 Thus the previous example needs to be modified to:
8949 pad="ih*16/9/sar:ih:(ow-iw)/2:(oh-ih)/2"
8953 Double the output size and put the input video in the bottom-right
8954 corner of the output padded area:
8956 pad="2*iw:2*ih:ow-iw:oh-ih"
8963 Generate one palette for a whole video stream.
8965 It accepts the following options:
8969 Set the maximum number of colors to quantize in the palette.
8970 Note: the palette will still contain 256 colors; the unused palette entries
8973 @item reserve_transparent
8974 Create a palette of 255 colors maximum and reserve the last one for
8975 transparency. Reserving the transparency color is useful for GIF optimization.
8976 If not set, the maximum of colors in the palette will be 256. You probably want
8977 to disable this option for a standalone image.
8981 Set statistics mode.
8983 It accepts the following values:
8986 Compute full frame histograms.
8988 Compute histograms only for the part that differs from previous frame. This
8989 might be relevant to give more importance to the moving part of your input if
8990 the background is static.
8993 Default value is @var{full}.
8996 The filter also exports the frame metadata @code{lavfi.color_quant_ratio}
8997 (@code{nb_color_in / nb_color_out}) which you can use to evaluate the degree of
8998 color quantization of the palette. This information is also visible at
8999 @var{info} logging level.
9001 @subsection Examples
9005 Generate a representative palette of a given video using @command{ffmpeg}:
9007 ffmpeg -i input.mkv -vf palettegen palette.png
9013 Use a palette to downsample an input video stream.
9015 The filter takes two inputs: one video stream and a palette. The palette must
9016 be a 256 pixels image.
9018 It accepts the following options:
9022 Select dithering mode. Available algorithms are:
9025 Ordered 8x8 bayer dithering (deterministic)
9027 Dithering as defined by Paul Heckbert in 1982 (simple error diffusion).
9028 Note: this dithering is sometimes considered "wrong" and is included as a
9030 @item floyd_steinberg
9031 Floyd and Steingberg dithering (error diffusion)
9033 Frankie Sierra dithering v2 (error diffusion)
9035 Frankie Sierra dithering v2 "Lite" (error diffusion)
9038 Default is @var{sierra2_4a}.
9041 When @var{bayer} dithering is selected, this option defines the scale of the
9042 pattern (how much the crosshatch pattern is visible). A low value means more
9043 visible pattern for less banding, and higher value means less visible pattern
9044 at the cost of more banding.
9046 The option must be an integer value in the range [0,5]. Default is @var{2}.
9049 If set, define the zone to process
9053 Only the changing rectangle will be reprocessed. This is similar to GIF
9054 cropping/offsetting compression mechanism. This option can be useful for speed
9055 if only a part of the image is changing, and has use cases such as limiting the
9056 scope of the error diffusal @option{dither} to the rectangle that bounds the
9057 moving scene (it leads to more deterministic output if the scene doesn't change
9058 much, and as a result less moving noise and better GIF compression).
9061 Default is @var{none}.
9064 @subsection Examples
9068 Use a palette (generated for example with @ref{palettegen}) to encode a GIF
9069 using @command{ffmpeg}:
9071 ffmpeg -i input.mkv -i palette.png -lavfi paletteuse output.gif
9075 @section perspective
9077 Correct perspective of video not recorded perpendicular to the screen.
9079 A description of the accepted parameters follows.
9090 Set coordinates expression for top left, top right, bottom left and bottom right corners.
9091 Default values are @code{0:0:W:0:0:H:W:H} with which perspective will remain unchanged.
9092 If the @code{sense} option is set to @code{source}, then the specified points will be sent
9093 to the corners of the destination. If the @code{sense} option is set to @code{destination},
9094 then the corners of the source will be sent to the specified coordinates.
9096 The expressions can use the following variables:
9101 the width and height of video frame.
9105 Set interpolation for perspective correction.
9107 It accepts the following values:
9113 Default value is @samp{linear}.
9116 Set interpretation of coordinate options.
9118 It accepts the following values:
9122 Send point in the source specified by the given coordinates to
9123 the corners of the destination.
9125 @item 1, destination
9127 Send the corners of the source to the point in the destination specified
9128 by the given coordinates.
9130 Default value is @samp{source}.
9136 Delay interlaced video by one field time so that the field order changes.
9138 The intended use is to fix PAL movies that have been captured with the
9139 opposite field order to the film-to-video transfer.
9141 A description of the accepted parameters follows.
9147 It accepts the following values:
9150 Capture field order top-first, transfer bottom-first.
9151 Filter will delay the bottom field.
9154 Capture field order bottom-first, transfer top-first.
9155 Filter will delay the top field.
9158 Capture and transfer with the same field order. This mode only exists
9159 for the documentation of the other options to refer to, but if you
9160 actually select it, the filter will faithfully do nothing.
9163 Capture field order determined automatically by field flags, transfer
9165 Filter selects among @samp{t} and @samp{b} modes on a frame by frame
9166 basis using field flags. If no field information is available,
9167 then this works just like @samp{u}.
9170 Capture unknown or varying, transfer opposite.
9171 Filter selects among @samp{t} and @samp{b} on a frame by frame basis by
9172 analyzing the images and selecting the alternative that produces best
9173 match between the fields.
9176 Capture top-first, transfer unknown or varying.
9177 Filter selects among @samp{t} and @samp{p} using image analysis.
9180 Capture bottom-first, transfer unknown or varying.
9181 Filter selects among @samp{b} and @samp{p} using image analysis.
9184 Capture determined by field flags, transfer unknown or varying.
9185 Filter selects among @samp{t}, @samp{b} and @samp{p} using field flags and
9186 image analysis. If no field information is available, then this works just
9187 like @samp{U}. This is the default mode.
9190 Both capture and transfer unknown or varying.
9191 Filter selects among @samp{t}, @samp{b} and @samp{p} using image analysis only.
9195 @section pixdesctest
9197 Pixel format descriptor test filter, mainly useful for internal
9198 testing. The output video should be equal to the input video.
9202 format=monow, pixdesctest
9205 can be used to test the monowhite pixel format descriptor definition.
9209 Enable the specified chain of postprocessing subfilters using libpostproc. This
9210 library should be automatically selected with a GPL build (@code{--enable-gpl}).
9211 Subfilters must be separated by '/' and can be disabled by prepending a '-'.
9212 Each subfilter and some options have a short and a long name that can be used
9213 interchangeably, i.e. dr/dering are the same.
9215 The filters accept the following options:
9219 Set postprocessing subfilters string.
9222 All subfilters share common options to determine their scope:
9226 Honor the quality commands for this subfilter.
9229 Do chrominance filtering, too (default).
9232 Do luminance filtering only (no chrominance).
9235 Do chrominance filtering only (no luminance).
9238 These options can be appended after the subfilter name, separated by a '|'.
9240 Available subfilters are:
9243 @item hb/hdeblock[|difference[|flatness]]
9244 Horizontal deblocking filter
9247 Difference factor where higher values mean more deblocking (default: @code{32}).
9249 Flatness threshold where lower values mean more deblocking (default: @code{39}).
9252 @item vb/vdeblock[|difference[|flatness]]
9253 Vertical deblocking filter
9256 Difference factor where higher values mean more deblocking (default: @code{32}).
9258 Flatness threshold where lower values mean more deblocking (default: @code{39}).
9261 @item ha/hadeblock[|difference[|flatness]]
9262 Accurate horizontal deblocking filter
9265 Difference factor where higher values mean more deblocking (default: @code{32}).
9267 Flatness threshold where lower values mean more deblocking (default: @code{39}).
9270 @item va/vadeblock[|difference[|flatness]]
9271 Accurate vertical deblocking filter
9274 Difference factor where higher values mean more deblocking (default: @code{32}).
9276 Flatness threshold where lower values mean more deblocking (default: @code{39}).
9280 The horizontal and vertical deblocking filters share the difference and
9281 flatness values so you cannot set different horizontal and vertical
9286 Experimental horizontal deblocking filter
9289 Experimental vertical deblocking filter
9294 @item tn/tmpnoise[|threshold1[|threshold2[|threshold3]]], temporal noise reducer
9297 larger -> stronger filtering
9299 larger -> stronger filtering
9301 larger -> stronger filtering
9304 @item al/autolevels[:f/fullyrange], automatic brightness / contrast correction
9307 Stretch luminance to @code{0-255}.
9310 @item lb/linblenddeint
9311 Linear blend deinterlacing filter that deinterlaces the given block by
9312 filtering all lines with a @code{(1 2 1)} filter.
9314 @item li/linipoldeint
9315 Linear interpolating deinterlacing filter that deinterlaces the given block by
9316 linearly interpolating every second line.
9318 @item ci/cubicipoldeint
9319 Cubic interpolating deinterlacing filter deinterlaces the given block by
9320 cubically interpolating every second line.
9322 @item md/mediandeint
9323 Median deinterlacing filter that deinterlaces the given block by applying a
9324 median filter to every second line.
9326 @item fd/ffmpegdeint
9327 FFmpeg deinterlacing filter that deinterlaces the given block by filtering every
9328 second line with a @code{(-1 4 2 4 -1)} filter.
9331 Vertically applied FIR lowpass deinterlacing filter that deinterlaces the given
9332 block by filtering all lines with a @code{(-1 2 6 2 -1)} filter.
9334 @item fq/forceQuant[|quantizer]
9335 Overrides the quantizer table from the input with the constant quantizer you
9343 Default pp filter combination (@code{hb|a,vb|a,dr|a})
9346 Fast pp filter combination (@code{h1|a,v1|a,dr|a})
9349 High quality pp filter combination (@code{ha|a|128|7,va|a,dr|a})
9352 @subsection Examples
9356 Apply horizontal and vertical deblocking, deringing and automatic
9357 brightness/contrast:
9363 Apply default filters without brightness/contrast correction:
9369 Apply default filters and temporal denoiser:
9371 pp=default/tmpnoise|1|2|3
9375 Apply deblocking on luminance only, and switch vertical deblocking on or off
9376 automatically depending on available CPU time:
9383 Apply Postprocessing filter 7. It is variant of the @ref{spp} filter,
9384 similar to spp = 6 with 7 point DCT, where only the center sample is
9387 The filter accepts the following options:
9391 Force a constant quantization parameter. It accepts an integer in range
9392 0 to 63. If not set, the filter will use the QP from the video stream
9396 Set thresholding mode. Available modes are:
9400 Set hard thresholding.
9402 Set soft thresholding (better de-ringing effect, but likely blurrier).
9404 Set medium thresholding (good results, default).
9410 Obtain the average, maximum and minimum PSNR (Peak Signal to Noise
9411 Ratio) between two input videos.
9413 This filter takes in input two input videos, the first input is
9414 considered the "main" source and is passed unchanged to the
9415 output. The second input is used as a "reference" video for computing
9418 Both video inputs must have the same resolution and pixel format for
9419 this filter to work correctly. Also it assumes that both inputs
9420 have the same number of frames, which are compared one by one.
9422 The obtained average PSNR is printed through the logging system.
9424 The filter stores the accumulated MSE (mean squared error) of each
9425 frame, and at the end of the processing it is averaged across all frames
9426 equally, and the following formula is applied to obtain the PSNR:
9429 PSNR = 10*log10(MAX^2/MSE)
9432 Where MAX is the average of the maximum values of each component of the
9435 The description of the accepted parameters follows.
9439 If specified the filter will use the named file to save the PSNR of
9440 each individual frame. When filename equals "-" the data is sent to
9444 The file printed if @var{stats_file} is selected, contains a sequence of
9445 key/value pairs of the form @var{key}:@var{value} for each compared
9448 A description of each shown parameter follows:
9452 sequential number of the input frame, starting from 1
9455 Mean Square Error pixel-by-pixel average difference of the compared
9456 frames, averaged over all the image components.
9458 @item mse_y, mse_u, mse_v, mse_r, mse_g, mse_g, mse_a
9459 Mean Square Error pixel-by-pixel average difference of the compared
9460 frames for the component specified by the suffix.
9462 @item psnr_y, psnr_u, psnr_v, psnr_r, psnr_g, psnr_b, psnr_a
9463 Peak Signal to Noise ratio of the compared frames for the component
9464 specified by the suffix.
9469 movie=ref_movie.mpg, setpts=PTS-STARTPTS [main];
9470 [main][ref] psnr="stats_file=stats.log" [out]
9473 On this example the input file being processed is compared with the
9474 reference file @file{ref_movie.mpg}. The PSNR of each individual frame
9475 is stored in @file{stats.log}.
9480 Pulldown reversal (inverse telecine) filter, capable of handling mixed
9481 hard-telecine, 24000/1001 fps progressive, and 30000/1001 fps progressive
9484 The pullup filter is designed to take advantage of future context in making
9485 its decisions. This filter is stateless in the sense that it does not lock
9486 onto a pattern to follow, but it instead looks forward to the following
9487 fields in order to identify matches and rebuild progressive frames.
9489 To produce content with an even framerate, insert the fps filter after
9490 pullup, use @code{fps=24000/1001} if the input frame rate is 29.97fps,
9491 @code{fps=24} for 30fps and the (rare) telecined 25fps input.
9493 The filter accepts the following options:
9500 These options set the amount of "junk" to ignore at the left, right, top, and
9501 bottom of the image, respectively. Left and right are in units of 8 pixels,
9502 while top and bottom are in units of 2 lines.
9503 The default is 8 pixels on each side.
9506 Set the strict breaks. Setting this option to 1 will reduce the chances of
9507 filter generating an occasional mismatched frame, but it may also cause an
9508 excessive number of frames to be dropped during high motion sequences.
9509 Conversely, setting it to -1 will make filter match fields more easily.
9510 This may help processing of video where there is slight blurring between
9511 the fields, but may also cause there to be interlaced frames in the output.
9512 Default value is @code{0}.
9515 Set the metric plane to use. It accepts the following values:
9521 Use chroma blue plane.
9524 Use chroma red plane.
9527 This option may be set to use chroma plane instead of the default luma plane
9528 for doing filter's computations. This may improve accuracy on very clean
9529 source material, but more likely will decrease accuracy, especially if there
9530 is chroma noise (rainbow effect) or any grayscale video.
9531 The main purpose of setting @option{mp} to a chroma plane is to reduce CPU
9532 load and make pullup usable in realtime on slow machines.
9535 For best results (without duplicated frames in the output file) it is
9536 necessary to change the output frame rate. For example, to inverse
9537 telecine NTSC input:
9539 ffmpeg -i input -vf pullup -r 24000/1001 ...
9544 Change video quantization parameters (QP).
9546 The filter accepts the following option:
9550 Set expression for quantization parameter.
9553 The expression is evaluated through the eval API and can contain, among others,
9554 the following constants:
9558 1 if index is not 129, 0 otherwise.
9561 Sequentional index starting from -129 to 128.
9564 @subsection Examples
9576 Flush video frames from internal cache of frames into a random order.
9577 No frame is discarded.
9578 Inspired by @ref{frei0r} nervous filter.
9582 Set size in number of frames of internal cache, in range from @code{2} to
9583 @code{512}. Default is @code{30}.
9586 Set seed for random number generator, must be an integer included between
9587 @code{0} and @code{UINT32_MAX}. If not specified, or if explicitly set to
9588 less than @code{0}, the filter will try to use a good random seed on a
9592 @section removegrain
9594 The removegrain filter is a spatial denoiser for progressive video.
9598 Set mode for the first plane.
9601 Set mode for the second plane.
9604 Set mode for the third plane.
9607 Set mode for the fourth plane.
9610 Range of mode is from 0 to 24. Description of each mode follows:
9614 Leave input plane unchanged. Default.
9617 Clips the pixel with the minimum and maximum of the 8 neighbour pixels.
9620 Clips the pixel with the second minimum and maximum of the 8 neighbour pixels.
9623 Clips the pixel with the third minimum and maximum of the 8 neighbour pixels.
9626 Clips the pixel with the fourth minimum and maximum of the 8 neighbour pixels.
9627 This is equivalent to a median filter.
9630 Line-sensitive clipping giving the minimal change.
9633 Line-sensitive clipping, intermediate.
9636 Line-sensitive clipping, intermediate.
9639 Line-sensitive clipping, intermediate.
9642 Line-sensitive clipping on a line where the neighbours pixels are the closest.
9645 Replaces the target pixel with the closest neighbour.
9648 [1 2 1] horizontal and vertical kernel blur.
9654 Bob mode, interpolates top field from the line where the neighbours
9655 pixels are the closest.
9658 Bob mode, interpolates bottom field from the line where the neighbours
9659 pixels are the closest.
9662 Bob mode, interpolates top field. Same as 13 but with a more complicated
9663 interpolation formula.
9666 Bob mode, interpolates bottom field. Same as 14 but with a more complicated
9667 interpolation formula.
9670 Clips the pixel with the minimum and maximum of respectively the maximum and
9671 minimum of each pair of opposite neighbour pixels.
9674 Line-sensitive clipping using opposite neighbours whose greatest distance from
9675 the current pixel is minimal.
9678 Replaces the pixel with the average of its 8 neighbours.
9681 Averages the 9 pixels ([1 1 1] horizontal and vertical blur).
9684 Clips pixels using the averages of opposite neighbour.
9687 Same as mode 21 but simpler and faster.
9690 Small edge and halo removal, but reputed useless.
9698 Suppress a TV station logo, using an image file to determine which
9699 pixels comprise the logo. It works by filling in the pixels that
9700 comprise the logo with neighboring pixels.
9702 The filter accepts the following options:
9706 Set the filter bitmap file, which can be any image format supported by
9707 libavformat. The width and height of the image file must match those of the
9708 video stream being processed.
9711 Pixels in the provided bitmap image with a value of zero are not
9712 considered part of the logo, non-zero pixels are considered part of
9713 the logo. If you use white (255) for the logo and black (0) for the
9714 rest, you will be safe. For making the filter bitmap, it is
9715 recommended to take a screen capture of a black frame with the logo
9716 visible, and then using a threshold filter followed by the erode
9717 filter once or twice.
9719 If needed, little splotches can be fixed manually. Remember that if
9720 logo pixels are not covered, the filter quality will be much
9721 reduced. Marking too many pixels as part of the logo does not hurt as
9722 much, but it will increase the amount of blurring needed to cover over
9723 the image and will destroy more information than necessary, and extra
9724 pixels will slow things down on a large logo.
9726 @section repeatfields
9728 This filter uses the repeat_field flag from the Video ES headers and hard repeats
9729 fields based on its value.
9731 @section reverse, areverse
9735 Warning: This filter requires memory to buffer the entire clip, so trimming
9738 @subsection Examples
9742 Take the first 5 seconds of a clip, and reverse it.
9750 Rotate video by an arbitrary angle expressed in radians.
9752 The filter accepts the following options:
9754 A description of the optional parameters follows.
9757 Set an expression for the angle by which to rotate the input video
9758 clockwise, expressed as a number of radians. A negative value will
9759 result in a counter-clockwise rotation. By default it is set to "0".
9761 This expression is evaluated for each frame.
9764 Set the output width expression, default value is "iw".
9765 This expression is evaluated just once during configuration.
9768 Set the output height expression, default value is "ih".
9769 This expression is evaluated just once during configuration.
9772 Enable bilinear interpolation if set to 1, a value of 0 disables
9773 it. Default value is 1.
9776 Set the color used to fill the output area not covered by the rotated
9777 image. For the general syntax of this option, check the "Color" section in the
9778 ffmpeg-utils manual. If the special value "none" is selected then no
9779 background is printed (useful for example if the background is never shown).
9781 Default value is "black".
9784 The expressions for the angle and the output size can contain the
9785 following constants and functions:
9789 sequential number of the input frame, starting from 0. It is always NAN
9790 before the first frame is filtered.
9793 time in seconds of the input frame, it is set to 0 when the filter is
9794 configured. It is always NAN before the first frame is filtered.
9798 horizontal and vertical chroma subsample values. For example for the
9799 pixel format "yuv422p" @var{hsub} is 2 and @var{vsub} is 1.
9803 the input video width and height
9807 the output width and height, that is the size of the padded area as
9808 specified by the @var{width} and @var{height} expressions
9812 the minimal width/height required for completely containing the input
9813 video rotated by @var{a} radians.
9815 These are only available when computing the @option{out_w} and
9816 @option{out_h} expressions.
9819 @subsection Examples
9823 Rotate the input by PI/6 radians clockwise:
9829 Rotate the input by PI/6 radians counter-clockwise:
9835 Rotate the input by 45 degrees clockwise:
9841 Apply a constant rotation with period T, starting from an angle of PI/3:
9843 rotate=PI/3+2*PI*t/T
9847 Make the input video rotation oscillating with a period of T
9848 seconds and an amplitude of A radians:
9850 rotate=A*sin(2*PI/T*t)
9854 Rotate the video, output size is chosen so that the whole rotating
9855 input video is always completely contained in the output:
9857 rotate='2*PI*t:ow=hypot(iw,ih):oh=ow'
9861 Rotate the video, reduce the output size so that no background is ever
9864 rotate=2*PI*t:ow='min(iw,ih)/sqrt(2)':oh=ow:c=none
9868 @subsection Commands
9870 The filter supports the following commands:
9874 Set the angle expression.
9875 The command accepts the same syntax of the corresponding option.
9877 If the specified expression is not valid, it is kept at its current
9883 Apply Shape Adaptive Blur.
9885 The filter accepts the following options:
9888 @item luma_radius, lr
9889 Set luma blur filter strength, must be a value in range 0.1-4.0, default
9890 value is 1.0. A greater value will result in a more blurred image, and
9891 in slower processing.
9893 @item luma_pre_filter_radius, lpfr
9894 Set luma pre-filter radius, must be a value in the 0.1-2.0 range, default
9897 @item luma_strength, ls
9898 Set luma maximum difference between pixels to still be considered, must
9899 be a value in the 0.1-100.0 range, default value is 1.0.
9901 @item chroma_radius, cr
9902 Set chroma blur filter strength, must be a value in range 0.1-4.0. A
9903 greater value will result in a more blurred image, and in slower
9906 @item chroma_pre_filter_radius, cpfr
9907 Set chroma pre-filter radius, must be a value in the 0.1-2.0 range.
9909 @item chroma_strength, cs
9910 Set chroma maximum difference between pixels to still be considered,
9911 must be a value in the 0.1-100.0 range.
9914 Each chroma option value, if not explicitly specified, is set to the
9915 corresponding luma option value.
9920 Scale (resize) the input video, using the libswscale library.
9922 The scale filter forces the output display aspect ratio to be the same
9923 of the input, by changing the output sample aspect ratio.
9925 If the input image format is different from the format requested by
9926 the next filter, the scale filter will convert the input to the
9930 The filter accepts the following options, or any of the options
9931 supported by the libswscale scaler.
9933 See @ref{scaler_options,,the ffmpeg-scaler manual,ffmpeg-scaler} for
9934 the complete list of scaler options.
9939 Set the output video dimension expression. Default value is the input
9942 If the value is 0, the input width is used for the output.
9944 If one of the values is -1, the scale filter will use a value that
9945 maintains the aspect ratio of the input image, calculated from the
9946 other specified dimension. If both of them are -1, the input size is
9949 If one of the values is -n with n > 1, the scale filter will also use a value
9950 that maintains the aspect ratio of the input image, calculated from the other
9951 specified dimension. After that it will, however, make sure that the calculated
9952 dimension is divisible by n and adjust the value if necessary.
9954 See below for the list of accepted constants for use in the dimension
9958 Specify when to evaluate @var{width} and @var{height} expression. It accepts the following values:
9962 Only evaluate expressions once during the filter initialization or when a command is processed.
9965 Evaluate expressions for each incoming frame.
9969 Default value is @samp{init}.
9973 Set the interlacing mode. It accepts the following values:
9977 Force interlaced aware scaling.
9980 Do not apply interlaced scaling.
9983 Select interlaced aware scaling depending on whether the source frames
9984 are flagged as interlaced or not.
9987 Default value is @samp{0}.
9990 Set libswscale scaling flags. See
9991 @ref{sws_flags,,the ffmpeg-scaler manual,ffmpeg-scaler} for the
9992 complete list of values. If not explicitly specified the filter applies
9996 @item param0, param1
9997 Set libswscale input parameters for scaling algorithms that need them. See
9998 @ref{sws_params,,the ffmpeg-scaler manual,ffmpeg-scaler} for the
9999 complete documentation. If not explicitly specified the filter applies
10005 Set the video size. For the syntax of this option, check the
10006 @ref{video size syntax,,"Video size" section in the ffmpeg-utils manual,ffmpeg-utils}.
10008 @item in_color_matrix
10009 @item out_color_matrix
10010 Set in/output YCbCr color space type.
10012 This allows the autodetected value to be overridden as well as allows forcing
10013 a specific value used for the output and encoder.
10015 If not specified, the color space type depends on the pixel format.
10021 Choose automatically.
10024 Format conforming to International Telecommunication Union (ITU)
10025 Recommendation BT.709.
10028 Set color space conforming to the United States Federal Communications
10029 Commission (FCC) Code of Federal Regulations (CFR) Title 47 (2003) 73.682 (a).
10032 Set color space conforming to:
10036 ITU Radiocommunication Sector (ITU-R) Recommendation BT.601
10039 ITU-R Rec. BT.470-6 (1998) Systems B, B1, and G
10042 Society of Motion Picture and Television Engineers (SMPTE) ST 170:2004
10047 Set color space conforming to SMPTE ST 240:1999.
10052 Set in/output YCbCr sample range.
10054 This allows the autodetected value to be overridden as well as allows forcing
10055 a specific value used for the output and encoder. If not specified, the
10056 range depends on the pixel format. Possible values:
10060 Choose automatically.
10063 Set full range (0-255 in case of 8-bit luma).
10066 Set "MPEG" range (16-235 in case of 8-bit luma).
10069 @item force_original_aspect_ratio
10070 Enable decreasing or increasing output video width or height if necessary to
10071 keep the original aspect ratio. Possible values:
10075 Scale the video as specified and disable this feature.
10078 The output video dimensions will automatically be decreased if needed.
10081 The output video dimensions will automatically be increased if needed.
10085 One useful instance of this option is that when you know a specific device's
10086 maximum allowed resolution, you can use this to limit the output video to
10087 that, while retaining the aspect ratio. For example, device A allows
10088 1280x720 playback, and your video is 1920x800. Using this option (set it to
10089 decrease) and specifying 1280x720 to the command line makes the output
10092 Please note that this is a different thing than specifying -1 for @option{w}
10093 or @option{h}, you still need to specify the output resolution for this option
10098 The values of the @option{w} and @option{h} options are expressions
10099 containing the following constants:
10104 The input width and height
10108 These are the same as @var{in_w} and @var{in_h}.
10112 The output (scaled) width and height
10116 These are the same as @var{out_w} and @var{out_h}
10119 The same as @var{iw} / @var{ih}
10122 input sample aspect ratio
10125 The input display aspect ratio. Calculated from @code{(iw / ih) * sar}.
10129 horizontal and vertical input chroma subsample values. For example for the
10130 pixel format "yuv422p" @var{hsub} is 2 and @var{vsub} is 1.
10134 horizontal and vertical output chroma subsample values. For example for the
10135 pixel format "yuv422p" @var{hsub} is 2 and @var{vsub} is 1.
10138 @subsection Examples
10142 Scale the input video to a size of 200x100
10147 This is equivalent to:
10158 Specify a size abbreviation for the output size:
10163 which can also be written as:
10169 Scale the input to 2x:
10171 scale=w=2*iw:h=2*ih
10175 The above is the same as:
10177 scale=2*in_w:2*in_h
10181 Scale the input to 2x with forced interlaced scaling:
10183 scale=2*iw:2*ih:interl=1
10187 Scale the input to half size:
10189 scale=w=iw/2:h=ih/2
10193 Increase the width, and set the height to the same size:
10199 Seek Greek harmony:
10206 Increase the height, and set the width to 3/2 of the height:
10208 scale=w=3/2*oh:h=3/5*ih
10212 Increase the size, making the size a multiple of the chroma
10215 scale="trunc(3/2*iw/hsub)*hsub:trunc(3/2*ih/vsub)*vsub"
10219 Increase the width to a maximum of 500 pixels,
10220 keeping the same aspect ratio as the input:
10222 scale=w='min(500\, iw*3/2):h=-1'
10226 @subsection Commands
10228 This filter supports the following commands:
10232 Set the output video dimension expression.
10233 The command accepts the same syntax of the corresponding option.
10235 If the specified expression is not valid, it is kept at its current
10241 Scale (resize) the input video, based on a reference video.
10243 See the scale filter for available options, scale2ref supports the same but
10244 uses the reference video instead of the main input as basis.
10246 @subsection Examples
10250 Scale a subtitle stream to match the main video in size before overlaying
10252 'scale2ref[b][a];[a][b]overlay'
10256 @section selectivecolor
10258 Adjust cyan, magenta, yellow and black (CMYK) to certain ranges of colors (such
10259 as "reds", "yellows", "greens", "cyans", ...). The adjustment range is defined
10260 by the "purity" of the color (that is, how saturated it already is).
10262 This filter is similar to the Adobe Photoshop Selective Color tool.
10264 The filter accepts the following options:
10267 @item correction_method
10268 Select color correction method.
10270 Available values are:
10273 Specified adjustments are applied "as-is" (added/subtracted to original pixel
10276 Specified adjustments are relative to the original component value.
10278 Default is @code{absolute}.
10280 Adjustments for red pixels (pixels where the red component is the maximum)
10282 Adjustments for yellow pixels (pixels where the blue component is the minimum)
10284 Adjustments for green pixels (pixels where the green component is the maximum)
10286 Adjustments for cyan pixels (pixels where the red component is the minimum)
10288 Adjustments for blue pixels (pixels where the blue component is the maximum)
10290 Adjustments for magenta pixels (pixels where the green component is the minimum)
10292 Adjustments for white pixels (pixels where all components are greater than 128)
10294 Adjustments for all pixels except pure black and pure white
10296 Adjustments for black pixels (pixels where all components are lesser than 128)
10298 Specify a Photoshop selective color file (@code{.asv}) to import the settings from.
10301 All the adjustment settings (@option{reds}, @option{yellows}, ...) accept up to
10302 4 space separated floating point adjustment values in the [-1,1] range,
10303 respectively to adjust the amount of cyan, magenta, yellow and black for the
10304 pixels of its range.
10306 @subsection Examples
10310 Increase cyan by 50% and reduce yellow by 33% in every green areas, and
10311 increase magenta by 27% in blue areas:
10313 selectivecolor=greens=.5 0 -.33 0:blues=0 .27
10317 Use a Photoshop selective color preset:
10319 selectivecolor=psfile=MySelectiveColorPresets/Misty.asv
10323 @section separatefields
10325 The @code{separatefields} takes a frame-based video input and splits
10326 each frame into its components fields, producing a new half height clip
10327 with twice the frame rate and twice the frame count.
10329 This filter use field-dominance information in frame to decide which
10330 of each pair of fields to place first in the output.
10331 If it gets it wrong use @ref{setfield} filter before @code{separatefields} filter.
10333 @section setdar, setsar
10335 The @code{setdar} filter sets the Display Aspect Ratio for the filter
10338 This is done by changing the specified Sample (aka Pixel) Aspect
10339 Ratio, according to the following equation:
10341 @var{DAR} = @var{HORIZONTAL_RESOLUTION} / @var{VERTICAL_RESOLUTION} * @var{SAR}
10344 Keep in mind that the @code{setdar} filter does not modify the pixel
10345 dimensions of the video frame. Also, the display aspect ratio set by
10346 this filter may be changed by later filters in the filterchain,
10347 e.g. in case of scaling or if another "setdar" or a "setsar" filter is
10350 The @code{setsar} filter sets the Sample (aka Pixel) Aspect Ratio for
10351 the filter output video.
10353 Note that as a consequence of the application of this filter, the
10354 output display aspect ratio will change according to the equation
10357 Keep in mind that the sample aspect ratio set by the @code{setsar}
10358 filter may be changed by later filters in the filterchain, e.g. if
10359 another "setsar" or a "setdar" filter is applied.
10361 It accepts the following parameters:
10364 @item r, ratio, dar (@code{setdar} only), sar (@code{setsar} only)
10365 Set the aspect ratio used by the filter.
10367 The parameter can be a floating point number string, an expression, or
10368 a string of the form @var{num}:@var{den}, where @var{num} and
10369 @var{den} are the numerator and denominator of the aspect ratio. If
10370 the parameter is not specified, it is assumed the value "0".
10371 In case the form "@var{num}:@var{den}" is used, the @code{:} character
10375 Set the maximum integer value to use for expressing numerator and
10376 denominator when reducing the expressed aspect ratio to a rational.
10377 Default value is @code{100}.
10381 The parameter @var{sar} is an expression containing
10382 the following constants:
10386 These are approximated values for the mathematical constants e
10387 (Euler's number), pi (Greek pi), and phi (the golden ratio).
10390 The input width and height.
10393 These are the same as @var{w} / @var{h}.
10396 The input sample aspect ratio.
10399 The input display aspect ratio. It is the same as
10400 (@var{w} / @var{h}) * @var{sar}.
10403 Horizontal and vertical chroma subsample values. For example, for the
10404 pixel format "yuv422p" @var{hsub} is 2 and @var{vsub} is 1.
10407 @subsection Examples
10412 To change the display aspect ratio to 16:9, specify one of the following:
10420 To change the sample aspect ratio to 10:11, specify:
10426 To set a display aspect ratio of 16:9, and specify a maximum integer value of
10427 1000 in the aspect ratio reduction, use the command:
10429 setdar=ratio=16/9:max=1000
10437 Force field for the output video frame.
10439 The @code{setfield} filter marks the interlace type field for the
10440 output frames. It does not change the input frame, but only sets the
10441 corresponding property, which affects how the frame is treated by
10442 following filters (e.g. @code{fieldorder} or @code{yadif}).
10444 The filter accepts the following options:
10449 Available values are:
10453 Keep the same field property.
10456 Mark the frame as bottom-field-first.
10459 Mark the frame as top-field-first.
10462 Mark the frame as progressive.
10468 Show a line containing various information for each input video frame.
10469 The input video is not modified.
10471 The shown line contains a sequence of key/value pairs of the form
10472 @var{key}:@var{value}.
10474 The following values are shown in the output:
10478 The (sequential) number of the input frame, starting from 0.
10481 The Presentation TimeStamp of the input frame, expressed as a number of
10482 time base units. The time base unit depends on the filter input pad.
10485 The Presentation TimeStamp of the input frame, expressed as a number of
10489 The position of the frame in the input stream, or -1 if this information is
10490 unavailable and/or meaningless (for example in case of synthetic video).
10493 The pixel format name.
10496 The sample aspect ratio of the input frame, expressed in the form
10497 @var{num}/@var{den}.
10500 The size of the input frame. For the syntax of this option, check the
10501 @ref{video size syntax,,"Video size" section in the ffmpeg-utils manual,ffmpeg-utils}.
10504 The type of interlaced mode ("P" for "progressive", "T" for top field first, "B"
10505 for bottom field first).
10508 This is 1 if the frame is a key frame, 0 otherwise.
10511 The picture type of the input frame ("I" for an I-frame, "P" for a
10512 P-frame, "B" for a B-frame, or "?" for an unknown type).
10513 Also refer to the documentation of the @code{AVPictureType} enum and of
10514 the @code{av_get_picture_type_char} function defined in
10515 @file{libavutil/avutil.h}.
10518 The Adler-32 checksum (printed in hexadecimal) of all the planes of the input frame.
10520 @item plane_checksum
10521 The Adler-32 checksum (printed in hexadecimal) of each plane of the input frame,
10522 expressed in the form "[@var{c0} @var{c1} @var{c2} @var{c3}]".
10525 @section showpalette
10527 Displays the 256 colors palette of each frame. This filter is only relevant for
10528 @var{pal8} pixel format frames.
10530 It accepts the following option:
10534 Set the size of the box used to represent one palette color entry. Default is
10535 @code{30} (for a @code{30x30} pixel box).
10538 @section shuffleframes
10540 Reorder and/or duplicate video frames.
10542 It accepts the following parameters:
10546 Set the destination indexes of input frames.
10547 This is space or '|' separated list of indexes that maps input frames to output
10548 frames. Number of indexes also sets maximal value that each index may have.
10551 The first frame has the index 0. The default is to keep the input unchanged.
10553 Swap second and third frame of every three frames of the input:
10555 ffmpeg -i INPUT -vf "shuffleframes=0 2 1" OUTPUT
10558 @section shuffleplanes
10560 Reorder and/or duplicate video planes.
10562 It accepts the following parameters:
10567 The index of the input plane to be used as the first output plane.
10570 The index of the input plane to be used as the second output plane.
10573 The index of the input plane to be used as the third output plane.
10576 The index of the input plane to be used as the fourth output plane.
10580 The first plane has the index 0. The default is to keep the input unchanged.
10582 Swap the second and third planes of the input:
10584 ffmpeg -i INPUT -vf shuffleplanes=0:2:1:3 OUTPUT
10587 @anchor{signalstats}
10588 @section signalstats
10589 Evaluate various visual metrics that assist in determining issues associated
10590 with the digitization of analog video media.
10592 By default the filter will log these metadata values:
10596 Display the minimal Y value contained within the input frame. Expressed in
10600 Display the Y value at the 10% percentile within the input frame. Expressed in
10604 Display the average Y value within the input frame. Expressed in range of
10608 Display the Y value at the 90% percentile within the input frame. Expressed in
10612 Display the maximum Y value contained within the input frame. Expressed in
10616 Display the minimal U value contained within the input frame. Expressed in
10620 Display the U value at the 10% percentile within the input frame. Expressed in
10624 Display the average U value within the input frame. Expressed in range of
10628 Display the U value at the 90% percentile within the input frame. Expressed in
10632 Display the maximum U value contained within the input frame. Expressed in
10636 Display the minimal V value contained within the input frame. Expressed in
10640 Display the V value at the 10% percentile within the input frame. Expressed in
10644 Display the average V value within the input frame. Expressed in range of
10648 Display the V value at the 90% percentile within the input frame. Expressed in
10652 Display the maximum V value contained within the input frame. Expressed in
10656 Display the minimal saturation value contained within the input frame.
10657 Expressed in range of [0-~181.02].
10660 Display the saturation value at the 10% percentile within the input frame.
10661 Expressed in range of [0-~181.02].
10664 Display the average saturation value within the input frame. Expressed in range
10668 Display the saturation value at the 90% percentile within the input frame.
10669 Expressed in range of [0-~181.02].
10672 Display the maximum saturation value contained within the input frame.
10673 Expressed in range of [0-~181.02].
10676 Display the median value for hue within the input frame. Expressed in range of
10680 Display the average value for hue within the input frame. Expressed in range of
10684 Display the average of sample value difference between all values of the Y
10685 plane in the current frame and corresponding values of the previous input frame.
10686 Expressed in range of [0-255].
10689 Display the average of sample value difference between all values of the U
10690 plane in the current frame and corresponding values of the previous input frame.
10691 Expressed in range of [0-255].
10694 Display the average of sample value difference between all values of the V
10695 plane in the current frame and corresponding values of the previous input frame.
10696 Expressed in range of [0-255].
10699 The filter accepts the following options:
10705 @option{stat} specify an additional form of image analysis.
10706 @option{out} output video with the specified type of pixel highlighted.
10708 Both options accept the following values:
10712 Identify @var{temporal outliers} pixels. A @var{temporal outlier} is a pixel
10713 unlike the neighboring pixels of the same field. Examples of temporal outliers
10714 include the results of video dropouts, head clogs, or tape tracking issues.
10717 Identify @var{vertical line repetition}. Vertical line repetition includes
10718 similar rows of pixels within a frame. In born-digital video vertical line
10719 repetition is common, but this pattern is uncommon in video digitized from an
10720 analog source. When it occurs in video that results from the digitization of an
10721 analog source it can indicate concealment from a dropout compensator.
10724 Identify pixels that fall outside of legal broadcast range.
10728 Set the highlight color for the @option{out} option. The default color is
10732 @subsection Examples
10736 Output data of various video metrics:
10738 ffprobe -f lavfi movie=example.mov,signalstats="stat=tout+vrep+brng" -show_frames
10742 Output specific data about the minimum and maximum values of the Y plane per frame:
10744 ffprobe -f lavfi movie=example.mov,signalstats -show_entries frame_tags=lavfi.signalstats.YMAX,lavfi.signalstats.YMIN
10748 Playback video while highlighting pixels that are outside of broadcast range in red.
10750 ffplay example.mov -vf signalstats="out=brng:color=red"
10754 Playback video with signalstats metadata drawn over the frame.
10756 ffplay example.mov -vf signalstats=stat=brng+vrep+tout,drawtext=fontfile=FreeSerif.ttf:textfile=signalstat_drawtext.txt
10759 The contents of signalstat_drawtext.txt used in the command are:
10762 Y (%@{metadata:lavfi.signalstats.YMIN@}-%@{metadata:lavfi.signalstats.YMAX@})
10763 U (%@{metadata:lavfi.signalstats.UMIN@}-%@{metadata:lavfi.signalstats.UMAX@})
10764 V (%@{metadata:lavfi.signalstats.VMIN@}-%@{metadata:lavfi.signalstats.VMAX@})
10765 saturation maximum: %@{metadata:lavfi.signalstats.SATMAX@}
10773 Blur the input video without impacting the outlines.
10775 It accepts the following options:
10778 @item luma_radius, lr
10779 Set the luma radius. The option value must be a float number in
10780 the range [0.1,5.0] that specifies the variance of the gaussian filter
10781 used to blur the image (slower if larger). Default value is 1.0.
10783 @item luma_strength, ls
10784 Set the luma strength. The option value must be a float number
10785 in the range [-1.0,1.0] that configures the blurring. A value included
10786 in [0.0,1.0] will blur the image whereas a value included in
10787 [-1.0,0.0] will sharpen the image. Default value is 1.0.
10789 @item luma_threshold, lt
10790 Set the luma threshold used as a coefficient to determine
10791 whether a pixel should be blurred or not. The option value must be an
10792 integer in the range [-30,30]. A value of 0 will filter all the image,
10793 a value included in [0,30] will filter flat areas and a value included
10794 in [-30,0] will filter edges. Default value is 0.
10796 @item chroma_radius, cr
10797 Set the chroma radius. The option value must be a float number in
10798 the range [0.1,5.0] that specifies the variance of the gaussian filter
10799 used to blur the image (slower if larger). Default value is 1.0.
10801 @item chroma_strength, cs
10802 Set the chroma strength. The option value must be a float number
10803 in the range [-1.0,1.0] that configures the blurring. A value included
10804 in [0.0,1.0] will blur the image whereas a value included in
10805 [-1.0,0.0] will sharpen the image. Default value is 1.0.
10807 @item chroma_threshold, ct
10808 Set the chroma threshold used as a coefficient to determine
10809 whether a pixel should be blurred or not. The option value must be an
10810 integer in the range [-30,30]. A value of 0 will filter all the image,
10811 a value included in [0,30] will filter flat areas and a value included
10812 in [-30,0] will filter edges. Default value is 0.
10815 If a chroma option is not explicitly set, the corresponding luma value
10820 Obtain the SSIM (Structural SImilarity Metric) between two input videos.
10822 This filter takes in input two input videos, the first input is
10823 considered the "main" source and is passed unchanged to the
10824 output. The second input is used as a "reference" video for computing
10827 Both video inputs must have the same resolution and pixel format for
10828 this filter to work correctly. Also it assumes that both inputs
10829 have the same number of frames, which are compared one by one.
10831 The filter stores the calculated SSIM of each frame.
10833 The description of the accepted parameters follows.
10836 @item stats_file, f
10837 If specified the filter will use the named file to save the SSIM of
10838 each individual frame. When filename equals "-" the data is sent to
10842 The file printed if @var{stats_file} is selected, contains a sequence of
10843 key/value pairs of the form @var{key}:@var{value} for each compared
10846 A description of each shown parameter follows:
10850 sequential number of the input frame, starting from 1
10852 @item Y, U, V, R, G, B
10853 SSIM of the compared frames for the component specified by the suffix.
10856 SSIM of the compared frames for the whole frame.
10859 Same as above but in dB representation.
10864 movie=ref_movie.mpg, setpts=PTS-STARTPTS [main];
10865 [main][ref] ssim="stats_file=stats.log" [out]
10868 On this example the input file being processed is compared with the
10869 reference file @file{ref_movie.mpg}. The SSIM of each individual frame
10870 is stored in @file{stats.log}.
10872 Another example with both psnr and ssim at same time:
10874 ffmpeg -i main.mpg -i ref.mpg -lavfi "ssim;[0:v][1:v]psnr" -f null -
10879 Convert between different stereoscopic image formats.
10881 The filters accept the following options:
10885 Set stereoscopic image format of input.
10887 Available values for input image formats are:
10890 side by side parallel (left eye left, right eye right)
10893 side by side crosseye (right eye left, left eye right)
10896 side by side parallel with half width resolution
10897 (left eye left, right eye right)
10900 side by side crosseye with half width resolution
10901 (right eye left, left eye right)
10904 above-below (left eye above, right eye below)
10907 above-below (right eye above, left eye below)
10910 above-below with half height resolution
10911 (left eye above, right eye below)
10914 above-below with half height resolution
10915 (right eye above, left eye below)
10918 alternating frames (left eye first, right eye second)
10921 alternating frames (right eye first, left eye second)
10924 interleaved rows (left eye has top row, right eye starts on next row)
10927 interleaved rows (right eye has top row, left eye starts on next row)
10930 interleaved columns, left eye first
10933 interleaved columns, right eye first
10935 Default value is @samp{sbsl}.
10939 Set stereoscopic image format of output.
10943 side by side parallel (left eye left, right eye right)
10946 side by side crosseye (right eye left, left eye right)
10949 side by side parallel with half width resolution
10950 (left eye left, right eye right)
10953 side by side crosseye with half width resolution
10954 (right eye left, left eye right)
10957 above-below (left eye above, right eye below)
10960 above-below (right eye above, left eye below)
10963 above-below with half height resolution
10964 (left eye above, right eye below)
10967 above-below with half height resolution
10968 (right eye above, left eye below)
10971 alternating frames (left eye first, right eye second)
10974 alternating frames (right eye first, left eye second)
10977 interleaved rows (left eye has top row, right eye starts on next row)
10980 interleaved rows (right eye has top row, left eye starts on next row)
10983 anaglyph red/blue gray
10984 (red filter on left eye, blue filter on right eye)
10987 anaglyph red/green gray
10988 (red filter on left eye, green filter on right eye)
10991 anaglyph red/cyan gray
10992 (red filter on left eye, cyan filter on right eye)
10995 anaglyph red/cyan half colored
10996 (red filter on left eye, cyan filter on right eye)
10999 anaglyph red/cyan color
11000 (red filter on left eye, cyan filter on right eye)
11003 anaglyph red/cyan color optimized with the least squares projection of dubois
11004 (red filter on left eye, cyan filter on right eye)
11007 anaglyph green/magenta gray
11008 (green filter on left eye, magenta filter on right eye)
11011 anaglyph green/magenta half colored
11012 (green filter on left eye, magenta filter on right eye)
11015 anaglyph green/magenta colored
11016 (green filter on left eye, magenta filter on right eye)
11019 anaglyph green/magenta color optimized with the least squares projection of dubois
11020 (green filter on left eye, magenta filter on right eye)
11023 anaglyph yellow/blue gray
11024 (yellow filter on left eye, blue filter on right eye)
11027 anaglyph yellow/blue half colored
11028 (yellow filter on left eye, blue filter on right eye)
11031 anaglyph yellow/blue colored
11032 (yellow filter on left eye, blue filter on right eye)
11035 anaglyph yellow/blue color optimized with the least squares projection of dubois
11036 (yellow filter on left eye, blue filter on right eye)
11039 mono output (left eye only)
11042 mono output (right eye only)
11045 checkerboard, left eye first
11048 checkerboard, right eye first
11051 interleaved columns, left eye first
11054 interleaved columns, right eye first
11057 Default value is @samp{arcd}.
11060 @subsection Examples
11064 Convert input video from side by side parallel to anaglyph yellow/blue dubois:
11070 Convert input video from above below (left eye above, right eye below) to side by side crosseye.
11079 Apply a simple postprocessing filter that compresses and decompresses the image
11080 at several (or - in the case of @option{quality} level @code{6} - all) shifts
11081 and average the results.
11083 The filter accepts the following options:
11087 Set quality. This option defines the number of levels for averaging. It accepts
11088 an integer in the range 0-6. If set to @code{0}, the filter will have no
11089 effect. A value of @code{6} means the higher quality. For each increment of
11090 that value the speed drops by a factor of approximately 2. Default value is
11094 Force a constant quantization parameter. If not set, the filter will use the QP
11095 from the video stream (if available).
11098 Set thresholding mode. Available modes are:
11102 Set hard thresholding (default).
11104 Set soft thresholding (better de-ringing effect, but likely blurrier).
11107 @item use_bframe_qp
11108 Enable the use of the QP from the B-Frames if set to @code{1}. Using this
11109 option may cause flicker since the B-Frames have often larger QP. Default is
11110 @code{0} (not enabled).
11116 Draw subtitles on top of input video using the libass library.
11118 To enable compilation of this filter you need to configure FFmpeg with
11119 @code{--enable-libass}. This filter also requires a build with libavcodec and
11120 libavformat to convert the passed subtitles file to ASS (Advanced Substation
11121 Alpha) subtitles format.
11123 The filter accepts the following options:
11127 Set the filename of the subtitle file to read. It must be specified.
11129 @item original_size
11130 Specify the size of the original video, the video for which the ASS file
11131 was composed. For the syntax of this option, check the
11132 @ref{video size syntax,,"Video size" section in the ffmpeg-utils manual,ffmpeg-utils}.
11133 Due to a misdesign in ASS aspect ratio arithmetic, this is necessary to
11134 correctly scale the fonts if the aspect ratio has been changed.
11137 Set a directory path containing fonts that can be used by the filter.
11138 These fonts will be used in addition to whatever the font provider uses.
11141 Set subtitles input character encoding. @code{subtitles} filter only. Only
11142 useful if not UTF-8.
11144 @item stream_index, si
11145 Set subtitles stream index. @code{subtitles} filter only.
11148 Override default style or script info parameters of the subtitles. It accepts a
11149 string containing ASS style format @code{KEY=VALUE} couples separated by ",".
11152 If the first key is not specified, it is assumed that the first value
11153 specifies the @option{filename}.
11155 For example, to render the file @file{sub.srt} on top of the input
11156 video, use the command:
11161 which is equivalent to:
11163 subtitles=filename=sub.srt
11166 To render the default subtitles stream from file @file{video.mkv}, use:
11168 subtitles=video.mkv
11171 To render the second subtitles stream from that file, use:
11173 subtitles=video.mkv:si=1
11176 To make the subtitles stream from @file{sub.srt} appear in transparent green
11177 @code{DejaVu Serif}, use:
11179 subtitles=sub.srt:force_style='FontName=DejaVu Serif,PrimaryColour=&HAA00FF00'
11182 @section super2xsai
11184 Scale the input by 2x and smooth using the Super2xSaI (Scale and
11185 Interpolate) pixel art scaling algorithm.
11187 Useful for enlarging pixel art images without reducing sharpness.
11194 Apply telecine process to the video.
11196 This filter accepts the following options:
11205 The default value is @code{top}.
11209 A string of numbers representing the pulldown pattern you wish to apply.
11210 The default value is @code{23}.
11214 Some typical patterns:
11219 24p: 2332 (preferred)
11226 24p: 222222222223 ("Euro pulldown")
11232 Select the most representative frame in a given sequence of consecutive frames.
11234 The filter accepts the following options:
11238 Set the frames batch size to analyze; in a set of @var{n} frames, the filter
11239 will pick one of them, and then handle the next batch of @var{n} frames until
11240 the end. Default is @code{100}.
11243 Since the filter keeps track of the whole frames sequence, a bigger @var{n}
11244 value will result in a higher memory usage, so a high value is not recommended.
11246 @subsection Examples
11250 Extract one picture each 50 frames:
11256 Complete example of a thumbnail creation with @command{ffmpeg}:
11258 ffmpeg -i in.avi -vf thumbnail,scale=300:200 -frames:v 1 out.png
11264 Tile several successive frames together.
11266 The filter accepts the following options:
11271 Set the grid size (i.e. the number of lines and columns). For the syntax of
11272 this option, check the
11273 @ref{video size syntax,,"Video size" section in the ffmpeg-utils manual,ffmpeg-utils}.
11276 Set the maximum number of frames to render in the given area. It must be less
11277 than or equal to @var{w}x@var{h}. The default value is @code{0}, meaning all
11278 the area will be used.
11281 Set the outer border margin in pixels.
11284 Set the inner border thickness (i.e. the number of pixels between frames). For
11285 more advanced padding options (such as having different values for the edges),
11286 refer to the pad video filter.
11289 Specify the color of the unused area. For the syntax of this option, check the
11290 "Color" section in the ffmpeg-utils manual. The default value of @var{color}
11294 @subsection Examples
11298 Produce 8x8 PNG tiles of all keyframes (@option{-skip_frame nokey}) in a movie:
11300 ffmpeg -skip_frame nokey -i file.avi -vf 'scale=128:72,tile=8x8' -an -vsync 0 keyframes%03d.png
11302 The @option{-vsync 0} is necessary to prevent @command{ffmpeg} from
11303 duplicating each output frame to accommodate the originally detected frame
11307 Display @code{5} pictures in an area of @code{3x2} frames,
11308 with @code{7} pixels between them, and @code{2} pixels of initial margin, using
11309 mixed flat and named options:
11311 tile=3x2:nb_frames=5:padding=7:margin=2
11315 @section tinterlace
11317 Perform various types of temporal field interlacing.
11319 Frames are counted starting from 1, so the first input frame is
11322 The filter accepts the following options:
11327 Specify the mode of the interlacing. This option can also be specified
11328 as a value alone. See below for a list of values for this option.
11330 Available values are:
11334 Move odd frames into the upper field, even into the lower field,
11335 generating a double height frame at half frame rate.
11339 Frame 1 Frame 2 Frame 3 Frame 4
11341 11111 22222 33333 44444
11342 11111 22222 33333 44444
11343 11111 22222 33333 44444
11344 11111 22222 33333 44444
11358 Only output even frames, odd frames are dropped, generating a frame with
11359 unchanged height at half frame rate.
11364 Frame 1 Frame 2 Frame 3 Frame 4
11366 11111 22222 33333 44444
11367 11111 22222 33333 44444
11368 11111 22222 33333 44444
11369 11111 22222 33333 44444
11379 Only output odd frames, even frames are dropped, generating a frame with
11380 unchanged height at half frame rate.
11385 Frame 1 Frame 2 Frame 3 Frame 4
11387 11111 22222 33333 44444
11388 11111 22222 33333 44444
11389 11111 22222 33333 44444
11390 11111 22222 33333 44444
11400 Expand each frame to full height, but pad alternate lines with black,
11401 generating a frame with double height at the same input frame rate.
11406 Frame 1 Frame 2 Frame 3 Frame 4
11408 11111 22222 33333 44444
11409 11111 22222 33333 44444
11410 11111 22222 33333 44444
11411 11111 22222 33333 44444
11414 11111 ..... 33333 .....
11415 ..... 22222 ..... 44444
11416 11111 ..... 33333 .....
11417 ..... 22222 ..... 44444
11418 11111 ..... 33333 .....
11419 ..... 22222 ..... 44444
11420 11111 ..... 33333 .....
11421 ..... 22222 ..... 44444
11425 @item interleave_top, 4
11426 Interleave the upper field from odd frames with the lower field from
11427 even frames, generating a frame with unchanged height at half frame rate.
11432 Frame 1 Frame 2 Frame 3 Frame 4
11434 11111<- 22222 33333<- 44444
11435 11111 22222<- 33333 44444<-
11436 11111<- 22222 33333<- 44444
11437 11111 22222<- 33333 44444<-
11447 @item interleave_bottom, 5
11448 Interleave the lower field from odd frames with the upper field from
11449 even frames, generating a frame with unchanged height at half frame rate.
11454 Frame 1 Frame 2 Frame 3 Frame 4
11456 11111 22222<- 33333 44444<-
11457 11111<- 22222 33333<- 44444
11458 11111 22222<- 33333 44444<-
11459 11111<- 22222 33333<- 44444
11469 @item interlacex2, 6
11470 Double frame rate with unchanged height. Frames are inserted each
11471 containing the second temporal field from the previous input frame and
11472 the first temporal field from the next input frame. This mode relies on
11473 the top_field_first flag. Useful for interlaced video displays with no
11474 field synchronisation.
11479 Frame 1 Frame 2 Frame 3 Frame 4
11481 11111 22222 33333 44444
11482 11111 22222 33333 44444
11483 11111 22222 33333 44444
11484 11111 22222 33333 44444
11487 11111 22222 22222 33333 33333 44444 44444
11488 11111 11111 22222 22222 33333 33333 44444
11489 11111 22222 22222 33333 33333 44444 44444
11490 11111 11111 22222 22222 33333 33333 44444
11494 Move odd frames into the upper field, even into the lower field,
11495 generating a double height frame at same frame rate.
11499 Frame 1 Frame 2 Frame 3 Frame 4
11501 11111 22222 33333 44444
11502 11111 22222 33333 44444
11503 11111 22222 33333 44444
11504 11111 22222 33333 44444
11507 11111 33333 33333 55555
11508 22222 22222 44444 44444
11509 11111 33333 33333 55555
11510 22222 22222 44444 44444
11511 11111 33333 33333 55555
11512 22222 22222 44444 44444
11513 11111 33333 33333 55555
11514 22222 22222 44444 44444
11519 Numeric values are deprecated but are accepted for backward
11520 compatibility reasons.
11522 Default mode is @code{merge}.
11525 Specify flags influencing the filter process.
11527 Available value for @var{flags} is:
11530 @item low_pass_filter, vlfp
11531 Enable vertical low-pass filtering in the filter.
11532 Vertical low-pass filtering is required when creating an interlaced
11533 destination from a progressive source which contains high-frequency
11534 vertical detail. Filtering will reduce interlace 'twitter' and Moire
11537 Vertical low-pass filtering can only be enabled for @option{mode}
11538 @var{interleave_top} and @var{interleave_bottom}.
11545 Transpose rows with columns in the input video and optionally flip it.
11547 It accepts the following parameters:
11552 Specify the transposition direction.
11554 Can assume the following values:
11556 @item 0, 4, cclock_flip
11557 Rotate by 90 degrees counterclockwise and vertically flip (default), that is:
11565 Rotate by 90 degrees clockwise, that is:
11573 Rotate by 90 degrees counterclockwise, that is:
11580 @item 3, 7, clock_flip
11581 Rotate by 90 degrees clockwise and vertically flip, that is:
11589 For values between 4-7, the transposition is only done if the input
11590 video geometry is portrait and not landscape. These values are
11591 deprecated, the @code{passthrough} option should be used instead.
11593 Numerical values are deprecated, and should be dropped in favor of
11594 symbolic constants.
11597 Do not apply the transposition if the input geometry matches the one
11598 specified by the specified value. It accepts the following values:
11601 Always apply transposition.
11603 Preserve portrait geometry (when @var{height} >= @var{width}).
11605 Preserve landscape geometry (when @var{width} >= @var{height}).
11608 Default value is @code{none}.
11611 For example to rotate by 90 degrees clockwise and preserve portrait
11614 transpose=dir=1:passthrough=portrait
11617 The command above can also be specified as:
11619 transpose=1:portrait
11623 Trim the input so that the output contains one continuous subpart of the input.
11625 It accepts the following parameters:
11628 Specify the time of the start of the kept section, i.e. the frame with the
11629 timestamp @var{start} will be the first frame in the output.
11632 Specify the time of the first frame that will be dropped, i.e. the frame
11633 immediately preceding the one with the timestamp @var{end} will be the last
11634 frame in the output.
11637 This is the same as @var{start}, except this option sets the start timestamp
11638 in timebase units instead of seconds.
11641 This is the same as @var{end}, except this option sets the end timestamp
11642 in timebase units instead of seconds.
11645 The maximum duration of the output in seconds.
11648 The number of the first frame that should be passed to the output.
11651 The number of the first frame that should be dropped.
11654 @option{start}, @option{end}, and @option{duration} are expressed as time
11655 duration specifications; see
11656 @ref{time duration syntax,,the Time duration section in the ffmpeg-utils(1) manual,ffmpeg-utils}
11657 for the accepted syntax.
11659 Note that the first two sets of the start/end options and the @option{duration}
11660 option look at the frame timestamp, while the _frame variants simply count the
11661 frames that pass through the filter. Also note that this filter does not modify
11662 the timestamps. If you wish for the output timestamps to start at zero, insert a
11663 setpts filter after the trim filter.
11665 If multiple start or end options are set, this filter tries to be greedy and
11666 keep all the frames that match at least one of the specified constraints. To keep
11667 only the part that matches all the constraints at once, chain multiple trim
11670 The defaults are such that all the input is kept. So it is possible to set e.g.
11671 just the end values to keep everything before the specified time.
11676 Drop everything except the second minute of input:
11678 ffmpeg -i INPUT -vf trim=60:120
11682 Keep only the first second:
11684 ffmpeg -i INPUT -vf trim=duration=1
11693 Sharpen or blur the input video.
11695 It accepts the following parameters:
11698 @item luma_msize_x, lx
11699 Set the luma matrix horizontal size. It must be an odd integer between
11700 3 and 63. The default value is 5.
11702 @item luma_msize_y, ly
11703 Set the luma matrix vertical size. It must be an odd integer between 3
11704 and 63. The default value is 5.
11706 @item luma_amount, la
11707 Set the luma effect strength. It must be a floating point number, reasonable
11708 values lay between -1.5 and 1.5.
11710 Negative values will blur the input video, while positive values will
11711 sharpen it, a value of zero will disable the effect.
11713 Default value is 1.0.
11715 @item chroma_msize_x, cx
11716 Set the chroma matrix horizontal size. It must be an odd integer
11717 between 3 and 63. The default value is 5.
11719 @item chroma_msize_y, cy
11720 Set the chroma matrix vertical size. It must be an odd integer
11721 between 3 and 63. The default value is 5.
11723 @item chroma_amount, ca
11724 Set the chroma effect strength. It must be a floating point number, reasonable
11725 values lay between -1.5 and 1.5.
11727 Negative values will blur the input video, while positive values will
11728 sharpen it, a value of zero will disable the effect.
11730 Default value is 0.0.
11733 If set to 1, specify using OpenCL capabilities, only available if
11734 FFmpeg was configured with @code{--enable-opencl}. Default value is 0.
11738 All parameters are optional and default to the equivalent of the
11739 string '5:5:1.0:5:5:0.0'.
11741 @subsection Examples
11745 Apply strong luma sharpen effect:
11747 unsharp=luma_msize_x=7:luma_msize_y=7:luma_amount=2.5
11751 Apply a strong blur of both luma and chroma parameters:
11753 unsharp=7:7:-2:7:7:-2
11759 Apply ultra slow/simple postprocessing filter that compresses and decompresses
11760 the image at several (or - in the case of @option{quality} level @code{8} - all)
11761 shifts and average the results.
11763 The way this differs from the behavior of spp is that uspp actually encodes &
11764 decodes each case with libavcodec Snow, whereas spp uses a simplified intra only 8x8
11765 DCT similar to MJPEG.
11767 The filter accepts the following options:
11771 Set quality. This option defines the number of levels for averaging. It accepts
11772 an integer in the range 0-8. If set to @code{0}, the filter will have no
11773 effect. A value of @code{8} means the higher quality. For each increment of
11774 that value the speed drops by a factor of approximately 2. Default value is
11778 Force a constant quantization parameter. If not set, the filter will use the QP
11779 from the video stream (if available).
11782 @section vectorscope
11784 Display 2 color component values in the two dimensional graph (which is called
11787 This filter accepts the following options:
11791 Set vectorscope mode.
11793 It accepts the following values:
11796 Gray values are displayed on graph, higher brightness means more pixels have
11797 same component color value on location in graph. This is the default mode.
11800 Gray values are displayed on graph. Surrounding pixels values which are not
11801 present in video frame are drawn in gradient of 2 color components which are
11802 set by option @code{x} and @code{y}.
11805 Actual color components values present in video frame are displayed on graph.
11808 Similar as color2 but higher frequency of same values @code{x} and @code{y}
11809 on graph increases value of another color component, which is luminance by
11810 default values of @code{x} and @code{y}.
11813 Actual colors present in video frame are displayed on graph. If two different
11814 colors map to same position on graph then color with higher value of component
11815 not present in graph is picked.
11819 Set which color component will be represented on X-axis. Default is @code{1}.
11822 Set which color component will be represented on Y-axis. Default is @code{2}.
11825 Set intensity, used by modes: gray, color and color3 for increasing brightness
11826 of color component which represents frequency of (X, Y) location in graph.
11831 No envelope, this is default.
11834 Instant envelope, even darkest single pixel will be clearly highlighted.
11837 Hold maximum and minimum values presented in graph over time. This way you
11838 can still spot out of range values without constantly looking at vectorscope.
11841 Peak and instant envelope combined together.
11845 @anchor{vidstabdetect}
11846 @section vidstabdetect
11848 Analyze video stabilization/deshaking. Perform pass 1 of 2, see
11849 @ref{vidstabtransform} for pass 2.
11851 This filter generates a file with relative translation and rotation
11852 transform information about subsequent frames, which is then used by
11853 the @ref{vidstabtransform} filter.
11855 To enable compilation of this filter you need to configure FFmpeg with
11856 @code{--enable-libvidstab}.
11858 This filter accepts the following options:
11862 Set the path to the file used to write the transforms information.
11863 Default value is @file{transforms.trf}.
11866 Set how shaky the video is and how quick the camera is. It accepts an
11867 integer in the range 1-10, a value of 1 means little shakiness, a
11868 value of 10 means strong shakiness. Default value is 5.
11871 Set the accuracy of the detection process. It must be a value in the
11872 range 1-15. A value of 1 means low accuracy, a value of 15 means high
11873 accuracy. Default value is 15.
11876 Set stepsize of the search process. The region around minimum is
11877 scanned with 1 pixel resolution. Default value is 6.
11880 Set minimum contrast. Below this value a local measurement field is
11881 discarded. Must be a floating point value in the range 0-1. Default
11885 Set reference frame number for tripod mode.
11887 If enabled, the motion of the frames is compared to a reference frame
11888 in the filtered stream, identified by the specified number. The idea
11889 is to compensate all movements in a more-or-less static scene and keep
11890 the camera view absolutely still.
11892 If set to 0, it is disabled. The frames are counted starting from 1.
11895 Show fields and transforms in the resulting frames. It accepts an
11896 integer in the range 0-2. Default value is 0, which disables any
11900 @subsection Examples
11904 Use default values:
11910 Analyze strongly shaky movie and put the results in file
11911 @file{mytransforms.trf}:
11913 vidstabdetect=shakiness=10:accuracy=15:result="mytransforms.trf"
11917 Visualize the result of internal transformations in the resulting
11920 vidstabdetect=show=1
11924 Analyze a video with medium shakiness using @command{ffmpeg}:
11926 ffmpeg -i input -vf vidstabdetect=shakiness=5:show=1 dummy.avi
11930 @anchor{vidstabtransform}
11931 @section vidstabtransform
11933 Video stabilization/deshaking: pass 2 of 2,
11934 see @ref{vidstabdetect} for pass 1.
11936 Read a file with transform information for each frame and
11937 apply/compensate them. Together with the @ref{vidstabdetect}
11938 filter this can be used to deshake videos. See also
11939 @url{http://public.hronopik.de/vid.stab}. It is important to also use
11940 the @ref{unsharp} filter, see below.
11942 To enable compilation of this filter you need to configure FFmpeg with
11943 @code{--enable-libvidstab}.
11945 @subsection Options
11949 Set path to the file used to read the transforms. Default value is
11950 @file{transforms.trf}.
11953 Set the number of frames (value*2 + 1) used for lowpass filtering the
11954 camera movements. Default value is 10.
11956 For example a number of 10 means that 21 frames are used (10 in the
11957 past and 10 in the future) to smoothen the motion in the video. A
11958 larger value leads to a smoother video, but limits the acceleration of
11959 the camera (pan/tilt movements). 0 is a special case where a static
11960 camera is simulated.
11963 Set the camera path optimization algorithm.
11965 Accepted values are:
11968 gaussian kernel low-pass filter on camera motion (default)
11970 averaging on transformations
11974 Set maximal number of pixels to translate frames. Default value is -1,
11978 Set maximal angle in radians (degree*PI/180) to rotate frames. Default
11979 value is -1, meaning no limit.
11982 Specify how to deal with borders that may be visible due to movement
11985 Available values are:
11988 keep image information from previous frame (default)
11990 fill the border black
11994 Invert transforms if set to 1. Default value is 0.
11997 Consider transforms as relative to previous frame if set to 1,
11998 absolute if set to 0. Default value is 0.
12001 Set percentage to zoom. A positive value will result in a zoom-in
12002 effect, a negative value in a zoom-out effect. Default value is 0 (no
12006 Set optimal zooming to avoid borders.
12008 Accepted values are:
12013 optimal static zoom value is determined (only very strong movements
12014 will lead to visible borders) (default)
12016 optimal adaptive zoom value is determined (no borders will be
12017 visible), see @option{zoomspeed}
12020 Note that the value given at zoom is added to the one calculated here.
12023 Set percent to zoom maximally each frame (enabled when
12024 @option{optzoom} is set to 2). Range is from 0 to 5, default value is
12028 Specify type of interpolation.
12030 Available values are:
12035 linear only horizontal
12037 linear in both directions (default)
12039 cubic in both directions (slow)
12043 Enable virtual tripod mode if set to 1, which is equivalent to
12044 @code{relative=0:smoothing=0}. Default value is 0.
12046 Use also @code{tripod} option of @ref{vidstabdetect}.
12049 Increase log verbosity if set to 1. Also the detected global motions
12050 are written to the temporary file @file{global_motions.trf}. Default
12054 @subsection Examples
12058 Use @command{ffmpeg} for a typical stabilization with default values:
12060 ffmpeg -i inp.mpeg -vf vidstabtransform,unsharp=5:5:0.8:3:3:0.4 inp_stabilized.mpeg
12063 Note the use of the @ref{unsharp} filter which is always recommended.
12066 Zoom in a bit more and load transform data from a given file:
12068 vidstabtransform=zoom=5:input="mytransforms.trf"
12072 Smoothen the video even more:
12074 vidstabtransform=smoothing=30
12080 Flip the input video vertically.
12082 For example, to vertically flip a video with @command{ffmpeg}:
12084 ffmpeg -i in.avi -vf "vflip" out.avi
12090 Make or reverse a natural vignetting effect.
12092 The filter accepts the following options:
12096 Set lens angle expression as a number of radians.
12098 The value is clipped in the @code{[0,PI/2]} range.
12100 Default value: @code{"PI/5"}
12104 Set center coordinates expressions. Respectively @code{"w/2"} and @code{"h/2"}
12108 Set forward/backward mode.
12110 Available modes are:
12113 The larger the distance from the central point, the darker the image becomes.
12116 The larger the distance from the central point, the brighter the image becomes.
12117 This can be used to reverse a vignette effect, though there is no automatic
12118 detection to extract the lens @option{angle} and other settings (yet). It can
12119 also be used to create a burning effect.
12122 Default value is @samp{forward}.
12125 Set evaluation mode for the expressions (@option{angle}, @option{x0}, @option{y0}).
12127 It accepts the following values:
12130 Evaluate expressions only once during the filter initialization.
12133 Evaluate expressions for each incoming frame. This is way slower than the
12134 @samp{init} mode since it requires all the scalers to be re-computed, but it
12135 allows advanced dynamic expressions.
12138 Default value is @samp{init}.
12141 Set dithering to reduce the circular banding effects. Default is @code{1}
12145 Set vignette aspect. This setting allows one to adjust the shape of the vignette.
12146 Setting this value to the SAR of the input will make a rectangular vignetting
12147 following the dimensions of the video.
12149 Default is @code{1/1}.
12152 @subsection Expressions
12154 The @option{alpha}, @option{x0} and @option{y0} expressions can contain the
12155 following parameters.
12160 input width and height
12163 the number of input frame, starting from 0
12166 the PTS (Presentation TimeStamp) time of the filtered video frame, expressed in
12167 @var{TB} units, NAN if undefined
12170 frame rate of the input video, NAN if the input frame rate is unknown
12173 the PTS (Presentation TimeStamp) of the filtered video frame,
12174 expressed in seconds, NAN if undefined
12177 time base of the input video
12181 @subsection Examples
12185 Apply simple strong vignetting effect:
12191 Make a flickering vignetting:
12193 vignette='PI/4+random(1)*PI/50':eval=frame
12199 Stack input videos vertically.
12201 All streams must be of same pixel format and of same width.
12203 Note that this filter is faster than using @ref{overlay} and @ref{pad} filter
12204 to create same output.
12206 The filter accept the following option:
12210 Set number of input streams. Default is 2.
12213 If set to 1, force the output to terminate when the shortest input
12214 terminates. Default value is 0.
12219 Deinterlace the input video ("w3fdif" stands for "Weston 3 Field
12220 Deinterlacing Filter").
12222 Based on the process described by Martin Weston for BBC R&D, and
12223 implemented based on the de-interlace algorithm written by Jim
12224 Easterbrook for BBC R&D, the Weston 3 field deinterlacing filter
12225 uses filter coefficients calculated by BBC R&D.
12227 There are two sets of filter coefficients, so called "simple":
12228 and "complex". Which set of filter coefficients is used can
12229 be set by passing an optional parameter:
12233 Set the interlacing filter coefficients. Accepts one of the following values:
12237 Simple filter coefficient set.
12239 More-complex filter coefficient set.
12241 Default value is @samp{complex}.
12244 Specify which frames to deinterlace. Accept one of the following values:
12248 Deinterlace all frames,
12250 Only deinterlace frames marked as interlaced.
12253 Default value is @samp{all}.
12257 Video waveform monitor.
12259 The waveform monitor plots color component intensity. By default luminance
12260 only. Each column of the waveform corresponds to a column of pixels in the
12263 It accepts the following options:
12267 Can be either @code{row}, or @code{column}. Default is @code{column}.
12268 In row mode, the graph on the left side represents color component value 0 and
12269 the right side represents value = 255. In column mode, the top side represents
12270 color component value = 0 and bottom side represents value = 255.
12273 Set intensity. Smaller values are useful to find out how many values of the same
12274 luminance are distributed across input rows/columns.
12275 Default value is @code{0.04}. Allowed range is [0, 1].
12278 Set mirroring mode. @code{0} means unmirrored, @code{1} means mirrored.
12279 In mirrored mode, higher values will be represented on the left
12280 side for @code{row} mode and at the top for @code{column} mode. Default is
12281 @code{1} (mirrored).
12285 It accepts the following values:
12288 Presents information identical to that in the @code{parade}, except
12289 that the graphs representing color components are superimposed directly
12292 This display mode makes it easier to spot relative differences or similarities
12293 in overlapping areas of the color components that are supposed to be identical,
12294 such as neutral whites, grays, or blacks.
12297 Display separate graph for the color components side by side in
12298 @code{row} mode or one below the other in @code{column} mode.
12300 Using this display mode makes it easy to spot color casts in the highlights
12301 and shadows of an image, by comparing the contours of the top and the bottom
12302 graphs of each waveform. Since whites, grays, and blacks are characterized
12303 by exactly equal amounts of red, green, and blue, neutral areas of the picture
12304 should display three waveforms of roughly equal width/height. If not, the
12305 correction is easy to perform by making level adjustments the three waveforms.
12307 Default is @code{parade}.
12309 @item components, c
12310 Set which color components to display. Default is 1, which means only luminance
12311 or red color component if input is in RGB colorspace. If is set for example to
12312 7 it will display all 3 (if) available color components.
12317 No envelope, this is default.
12320 Instant envelope, minimum and maximum values presented in graph will be easily
12321 visible even with small @code{step} value.
12324 Hold minimum and maximum values presented in graph across time. This way you
12325 can still spot out of range values without constantly looking at waveforms.
12328 Peak and instant envelope combined together.
12334 No filtering, this is default.
12337 Luma and chroma combined together.
12340 Similar as above, but shows difference between blue and red chroma.
12343 Displays only chroma.
12346 Similar as above, but shows difference between blue and red chroma.
12349 Displays actual color value on waveform.
12354 Apply the xBR high-quality magnification filter which is designed for pixel
12355 art. It follows a set of edge-detection rules, see
12356 @url{http://www.libretro.com/forums/viewtopic.php?f=6&t=134}.
12358 It accepts the following option:
12362 Set the scaling dimension: @code{2} for @code{2xBR}, @code{3} for
12363 @code{3xBR} and @code{4} for @code{4xBR}.
12364 Default is @code{3}.
12370 Deinterlace the input video ("yadif" means "yet another deinterlacing
12373 It accepts the following parameters:
12379 The interlacing mode to adopt. It accepts one of the following values:
12382 @item 0, send_frame
12383 Output one frame for each frame.
12384 @item 1, send_field
12385 Output one frame for each field.
12386 @item 2, send_frame_nospatial
12387 Like @code{send_frame}, but it skips the spatial interlacing check.
12388 @item 3, send_field_nospatial
12389 Like @code{send_field}, but it skips the spatial interlacing check.
12392 The default value is @code{send_frame}.
12395 The picture field parity assumed for the input interlaced video. It accepts one
12396 of the following values:
12400 Assume the top field is first.
12402 Assume the bottom field is first.
12404 Enable automatic detection of field parity.
12407 The default value is @code{auto}.
12408 If the interlacing is unknown or the decoder does not export this information,
12409 top field first will be assumed.
12412 Specify which frames to deinterlace. Accept one of the following
12417 Deinterlace all frames.
12418 @item 1, interlaced
12419 Only deinterlace frames marked as interlaced.
12422 The default value is @code{all}.
12427 Apply Zoom & Pan effect.
12429 This filter accepts the following options:
12433 Set the zoom expression. Default is 1.
12437 Set the x and y expression. Default is 0.
12440 Set the duration expression in number of frames.
12441 This sets for how many number of frames effect will last for
12442 single input image.
12445 Set the output image size, default is 'hd720'.
12448 Each expression can contain the following constants:
12467 Output frame count.
12471 Last calculated 'x' and 'y' position from 'x' and 'y' expression
12472 for current input frame.
12476 'x' and 'y' of last output frame of previous input frame or 0 when there was
12477 not yet such frame (first input frame).
12480 Last calculated zoom from 'z' expression for current input frame.
12483 Last calculated zoom of last output frame of previous input frame.
12486 Number of output frames for current input frame. Calculated from 'd' expression
12487 for each input frame.
12490 number of output frames created for previous input frame
12493 Rational number: input width / input height
12496 sample aspect ratio
12499 display aspect ratio
12503 @subsection Examples
12507 Zoom-in up to 1.5 and pan at same time to some spot near center of picture:
12509 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
12513 Zoom-in up to 1.5 and pan always at center of picture:
12515 zoompan=z='min(zoom+0.0015,1.5)':d=700:x='iw/2-(iw/zoom/2)':y='ih/2-(ih/zoom/2)'
12520 Scale (resize) the input video, using the z.lib library:
12521 https://github.com/sekrit-twc/zimg.
12523 The zscale filter forces the output display aspect ratio to be the same
12524 as the input, by changing the output sample aspect ratio.
12526 If the input image format is different from the format requested by
12527 the next filter, the zscale filter will convert the input to the
12530 @subsection Options
12531 The filter accepts the following options.
12536 Set the output video dimension expression. Default value is the input
12539 If the @var{width} or @var{w} is 0, the input width is used for the output.
12540 If the @var{height} or @var{h} is 0, the input height is used for the output.
12542 If one of the values is -1, the zscale filter will use a value that
12543 maintains the aspect ratio of the input image, calculated from the
12544 other specified dimension. If both of them are -1, the input size is
12547 If one of the values is -n with n > 1, the zscale filter will also use a value
12548 that maintains the aspect ratio of the input image, calculated from the other
12549 specified dimension. After that it will, however, make sure that the calculated
12550 dimension is divisible by n and adjust the value if necessary.
12552 See below for the list of accepted constants for use in the dimension
12556 Set the video size. For the syntax of this option, check the
12557 @ref{video size syntax,,"Video size" section in the ffmpeg-utils manual,ffmpeg-utils}.
12560 Set the dither type.
12562 Possible values are:
12567 @item error_diffusion
12573 Set the resize filter type.
12575 Possible values are:
12585 Default is bilinear.
12588 Set the color range.
12590 Possible values are:
12597 Default is same as input.
12600 Set the color primaries.
12602 Possible values are:
12612 Default is same as input.
12615 Set the transfer characteristics.
12617 Possible values are:
12628 Default is same as input.
12631 Set the colorspace matrix.
12633 Possible value are:
12644 Default is same as input.
12647 The values of the @option{w} and @option{h} options are expressions
12648 containing the following constants:
12653 The input width and height
12657 These are the same as @var{in_w} and @var{in_h}.
12661 The output (scaled) width and height
12665 These are the same as @var{out_w} and @var{out_h}
12668 The same as @var{iw} / @var{ih}
12671 input sample aspect ratio
12674 The input display aspect ratio. Calculated from @code{(iw / ih) * sar}.
12678 horizontal and vertical input chroma subsample values. For example for the
12679 pixel format "yuv422p" @var{hsub} is 2 and @var{vsub} is 1.
12683 horizontal and vertical output chroma subsample values. For example for the
12684 pixel format "yuv422p" @var{hsub} is 2 and @var{vsub} is 1.
12690 @c man end VIDEO FILTERS
12692 @chapter Video Sources
12693 @c man begin VIDEO SOURCES
12695 Below is a description of the currently available video sources.
12699 Buffer video frames, and make them available to the filter chain.
12701 This source is mainly intended for a programmatic use, in particular
12702 through the interface defined in @file{libavfilter/vsrc_buffer.h}.
12704 It accepts the following parameters:
12709 Specify the size (width and height) of the buffered video frames. For the
12710 syntax of this option, check the
12711 @ref{video size syntax,,"Video size" section in the ffmpeg-utils manual,ffmpeg-utils}.
12714 The input video width.
12717 The input video height.
12720 A string representing the pixel format of the buffered video frames.
12721 It may be a number corresponding to a pixel format, or a pixel format
12725 Specify the timebase assumed by the timestamps of the buffered frames.
12728 Specify the frame rate expected for the video stream.
12730 @item pixel_aspect, sar
12731 The sample (pixel) aspect ratio of the input video.
12734 Specify the optional parameters to be used for the scale filter which
12735 is automatically inserted when an input change is detected in the
12736 input size or format.
12741 buffer=width=320:height=240:pix_fmt=yuv410p:time_base=1/24:sar=1
12744 will instruct the source to accept video frames with size 320x240 and
12745 with format "yuv410p", assuming 1/24 as the timestamps timebase and
12746 square pixels (1:1 sample aspect ratio).
12747 Since the pixel format with name "yuv410p" corresponds to the number 6
12748 (check the enum AVPixelFormat definition in @file{libavutil/pixfmt.h}),
12749 this example corresponds to:
12751 buffer=size=320x240:pixfmt=6:time_base=1/24:pixel_aspect=1/1
12754 Alternatively, the options can be specified as a flat string, but this
12755 syntax is deprecated:
12757 @var{width}:@var{height}:@var{pix_fmt}:@var{time_base.num}:@var{time_base.den}:@var{pixel_aspect.num}:@var{pixel_aspect.den}[:@var{sws_param}]
12761 Create a pattern generated by an elementary cellular automaton.
12763 The initial state of the cellular automaton can be defined through the
12764 @option{filename}, and @option{pattern} options. If such options are
12765 not specified an initial state is created randomly.
12767 At each new frame a new row in the video is filled with the result of
12768 the cellular automaton next generation. The behavior when the whole
12769 frame is filled is defined by the @option{scroll} option.
12771 This source accepts the following options:
12775 Read the initial cellular automaton state, i.e. the starting row, from
12776 the specified file.
12777 In the file, each non-whitespace character is considered an alive
12778 cell, a newline will terminate the row, and further characters in the
12779 file will be ignored.
12782 Read the initial cellular automaton state, i.e. the starting row, from
12783 the specified string.
12785 Each non-whitespace character in the string is considered an alive
12786 cell, a newline will terminate the row, and further characters in the
12787 string will be ignored.
12790 Set the video rate, that is the number of frames generated per second.
12793 @item random_fill_ratio, ratio
12794 Set the random fill ratio for the initial cellular automaton row. It
12795 is a floating point number value ranging from 0 to 1, defaults to
12798 This option is ignored when a file or a pattern is specified.
12800 @item random_seed, seed
12801 Set the seed for filling randomly the initial row, must be an integer
12802 included between 0 and UINT32_MAX. If not specified, or if explicitly
12803 set to -1, the filter will try to use a good random seed on a best
12807 Set the cellular automaton rule, it is a number ranging from 0 to 255.
12808 Default value is 110.
12811 Set the size of the output video. For the syntax of this option, check the
12812 @ref{video size syntax,,"Video size" section in the ffmpeg-utils manual,ffmpeg-utils}.
12814 If @option{filename} or @option{pattern} is specified, the size is set
12815 by default to the width of the specified initial state row, and the
12816 height is set to @var{width} * PHI.
12818 If @option{size} is set, it must contain the width of the specified
12819 pattern string, and the specified pattern will be centered in the
12822 If a filename or a pattern string is not specified, the size value
12823 defaults to "320x518" (used for a randomly generated initial state).
12826 If set to 1, scroll the output upward when all the rows in the output
12827 have been already filled. If set to 0, the new generated row will be
12828 written over the top row just after the bottom row is filled.
12831 @item start_full, full
12832 If set to 1, completely fill the output with generated rows before
12833 outputting the first frame.
12834 This is the default behavior, for disabling set the value to 0.
12837 If set to 1, stitch the left and right row edges together.
12838 This is the default behavior, for disabling set the value to 0.
12841 @subsection Examples
12845 Read the initial state from @file{pattern}, and specify an output of
12848 cellauto=f=pattern:s=200x400
12852 Generate a random initial row with a width of 200 cells, with a fill
12855 cellauto=ratio=2/3:s=200x200
12859 Create a pattern generated by rule 18 starting by a single alive cell
12860 centered on an initial row with width 100:
12862 cellauto=p=@@:s=100x400:full=0:rule=18
12866 Specify a more elaborated initial pattern:
12868 cellauto=p='@@@@ @@ @@@@':s=100x400:full=0:rule=18
12873 @section mandelbrot
12875 Generate a Mandelbrot set fractal, and progressively zoom towards the
12876 point specified with @var{start_x} and @var{start_y}.
12878 This source accepts the following options:
12883 Set the terminal pts value. Default value is 400.
12886 Set the terminal scale value.
12887 Must be a floating point value. Default value is 0.3.
12890 Set the inner coloring mode, that is the algorithm used to draw the
12891 Mandelbrot fractal internal region.
12893 It shall assume one of the following values:
12898 Show time until convergence.
12900 Set color based on point closest to the origin of the iterations.
12905 Default value is @var{mincol}.
12908 Set the bailout value. Default value is 10.0.
12911 Set the maximum of iterations performed by the rendering
12912 algorithm. Default value is 7189.
12915 Set outer coloring mode.
12916 It shall assume one of following values:
12918 @item iteration_count
12919 Set iteration cound mode.
12920 @item normalized_iteration_count
12921 set normalized iteration count mode.
12923 Default value is @var{normalized_iteration_count}.
12926 Set frame rate, expressed as number of frames per second. Default
12930 Set frame size. For the syntax of this option, check the "Video
12931 size" section in the ffmpeg-utils manual. Default value is "640x480".
12934 Set the initial scale value. Default value is 3.0.
12937 Set the initial x position. Must be a floating point value between
12938 -100 and 100. Default value is -0.743643887037158704752191506114774.
12941 Set the initial y position. Must be a floating point value between
12942 -100 and 100. Default value is -0.131825904205311970493132056385139.
12947 Generate various test patterns, as generated by the MPlayer test filter.
12949 The size of the generated video is fixed, and is 256x256.
12950 This source is useful in particular for testing encoding features.
12952 This source accepts the following options:
12957 Specify the frame rate of the sourced video, as the number of frames
12958 generated per second. It has to be a string in the format
12959 @var{frame_rate_num}/@var{frame_rate_den}, an integer number, a floating point
12960 number or a valid video frame rate abbreviation. The default value is
12964 Set the duration of the sourced video. See
12965 @ref{time duration syntax,,the Time duration section in the ffmpeg-utils(1) manual,ffmpeg-utils}
12966 for the accepted syntax.
12968 If not specified, or the expressed duration is negative, the video is
12969 supposed to be generated forever.
12973 Set the number or the name of the test to perform. Supported tests are:
12989 Default value is "all", which will cycle through the list of all tests.
12994 mptestsrc=t=dc_luma
12997 will generate a "dc_luma" test pattern.
12999 @section frei0r_src
13001 Provide a frei0r source.
13003 To enable compilation of this filter you need to install the frei0r
13004 header and configure FFmpeg with @code{--enable-frei0r}.
13006 This source accepts the following parameters:
13011 The size of the video to generate. For the syntax of this option, check the
13012 @ref{video size syntax,,"Video size" section in the ffmpeg-utils manual,ffmpeg-utils}.
13015 The framerate of the generated video. It may be a string of the form
13016 @var{num}/@var{den} or a frame rate abbreviation.
13019 The name to the frei0r source to load. For more information regarding frei0r and
13020 how to set the parameters, read the @ref{frei0r} section in the video filters
13023 @item filter_params
13024 A '|'-separated list of parameters to pass to the frei0r source.
13028 For example, to generate a frei0r partik0l source with size 200x200
13029 and frame rate 10 which is overlaid on the overlay filter main input:
13031 frei0r_src=size=200x200:framerate=10:filter_name=partik0l:filter_params=1234 [overlay]; [in][overlay] overlay
13036 Generate a life pattern.
13038 This source is based on a generalization of John Conway's life game.
13040 The sourced input represents a life grid, each pixel represents a cell
13041 which can be in one of two possible states, alive or dead. Every cell
13042 interacts with its eight neighbours, which are the cells that are
13043 horizontally, vertically, or diagonally adjacent.
13045 At each interaction the grid evolves according to the adopted rule,
13046 which specifies the number of neighbor alive cells which will make a
13047 cell stay alive or born. The @option{rule} option allows one to specify
13050 This source accepts the following options:
13054 Set the file from which to read the initial grid state. In the file,
13055 each non-whitespace character is considered an alive cell, and newline
13056 is used to delimit the end of each row.
13058 If this option is not specified, the initial grid is generated
13062 Set the video rate, that is the number of frames generated per second.
13065 @item random_fill_ratio, ratio
13066 Set the random fill ratio for the initial random grid. It is a
13067 floating point number value ranging from 0 to 1, defaults to 1/PHI.
13068 It is ignored when a file is specified.
13070 @item random_seed, seed
13071 Set the seed for filling the initial random grid, must be an integer
13072 included between 0 and UINT32_MAX. If not specified, or if explicitly
13073 set to -1, the filter will try to use a good random seed on a best
13079 A rule can be specified with a code of the kind "S@var{NS}/B@var{NB}",
13080 where @var{NS} and @var{NB} are sequences of numbers in the range 0-8,
13081 @var{NS} specifies the number of alive neighbor cells which make a
13082 live cell stay alive, and @var{NB} the number of alive neighbor cells
13083 which make a dead cell to become alive (i.e. to "born").
13084 "s" and "b" can be used in place of "S" and "B", respectively.
13086 Alternatively a rule can be specified by an 18-bits integer. The 9
13087 high order bits are used to encode the next cell state if it is alive
13088 for each number of neighbor alive cells, the low order bits specify
13089 the rule for "borning" new cells. Higher order bits encode for an
13090 higher number of neighbor cells.
13091 For example the number 6153 = @code{(12<<9)+9} specifies a stay alive
13092 rule of 12 and a born rule of 9, which corresponds to "S23/B03".
13094 Default value is "S23/B3", which is the original Conway's game of life
13095 rule, and will keep a cell alive if it has 2 or 3 neighbor alive
13096 cells, and will born a new cell if there are three alive cells around
13100 Set the size of the output video. For the syntax of this option, check the
13101 @ref{video size syntax,,"Video size" section in the ffmpeg-utils manual,ffmpeg-utils}.
13103 If @option{filename} is specified, the size is set by default to the
13104 same size of the input file. If @option{size} is set, it must contain
13105 the size specified in the input file, and the initial grid defined in
13106 that file is centered in the larger resulting area.
13108 If a filename is not specified, the size value defaults to "320x240"
13109 (used for a randomly generated initial grid).
13112 If set to 1, stitch the left and right grid edges together, and the
13113 top and bottom edges also. Defaults to 1.
13116 Set cell mold speed. If set, a dead cell will go from @option{death_color} to
13117 @option{mold_color} with a step of @option{mold}. @option{mold} can have a
13118 value from 0 to 255.
13121 Set the color of living (or new born) cells.
13124 Set the color of dead cells. If @option{mold} is set, this is the first color
13125 used to represent a dead cell.
13128 Set mold color, for definitely dead and moldy cells.
13130 For the syntax of these 3 color options, check the "Color" section in the
13131 ffmpeg-utils manual.
13134 @subsection Examples
13138 Read a grid from @file{pattern}, and center it on a grid of size
13141 life=f=pattern:s=300x300
13145 Generate a random grid of size 200x200, with a fill ratio of 2/3:
13147 life=ratio=2/3:s=200x200
13151 Specify a custom rule for evolving a randomly generated grid:
13157 Full example with slow death effect (mold) using @command{ffplay}:
13159 ffplay -f lavfi life=s=300x200:mold=10:r=60:ratio=0.1:death_color=#C83232:life_color=#00ff00,scale=1200:800:flags=16
13166 @anchor{haldclutsrc}
13168 @anchor{rgbtestsrc}
13170 @anchor{smptehdbars}
13172 @section allrgb, allyuv, color, haldclutsrc, nullsrc, rgbtestsrc, smptebars, smptehdbars, testsrc
13174 The @code{allrgb} source returns frames of size 4096x4096 of all rgb colors.
13176 The @code{allyuv} source returns frames of size 4096x4096 of all yuv colors.
13178 The @code{color} source provides an uniformly colored input.
13180 The @code{haldclutsrc} source provides an identity Hald CLUT. See also
13181 @ref{haldclut} filter.
13183 The @code{nullsrc} source returns unprocessed video frames. It is
13184 mainly useful to be employed in analysis / debugging tools, or as the
13185 source for filters which ignore the input data.
13187 The @code{rgbtestsrc} source generates an RGB test pattern useful for
13188 detecting RGB vs BGR issues. You should see a red, green and blue
13189 stripe from top to bottom.
13191 The @code{smptebars} source generates a color bars pattern, based on
13192 the SMPTE Engineering Guideline EG 1-1990.
13194 The @code{smptehdbars} source generates a color bars pattern, based on
13195 the SMPTE RP 219-2002.
13197 The @code{testsrc} source generates a test video pattern, showing a
13198 color pattern, a scrolling gradient and a timestamp. This is mainly
13199 intended for testing purposes.
13201 The sources accept the following parameters:
13206 Specify the color of the source, only available in the @code{color}
13207 source. For the syntax of this option, check the "Color" section in the
13208 ffmpeg-utils manual.
13211 Specify the level of the Hald CLUT, only available in the @code{haldclutsrc}
13212 source. A level of @code{N} generates a picture of @code{N*N*N} by @code{N*N*N}
13213 pixels to be used as identity matrix for 3D lookup tables. Each component is
13214 coded on a @code{1/(N*N)} scale.
13217 Specify the size of the sourced video. For the syntax of this option, check the
13218 @ref{video size syntax,,"Video size" section in the ffmpeg-utils manual,ffmpeg-utils}.
13219 The default value is @code{320x240}.
13221 This option is not available with the @code{haldclutsrc} filter.
13224 Specify the frame rate of the sourced video, as the number of frames
13225 generated per second. It has to be a string in the format
13226 @var{frame_rate_num}/@var{frame_rate_den}, an integer number, a floating point
13227 number or a valid video frame rate abbreviation. The default value is
13231 Set the sample aspect ratio of the sourced video.
13234 Set the duration of the sourced video. See
13235 @ref{time duration syntax,,the Time duration section in the ffmpeg-utils(1) manual,ffmpeg-utils}
13236 for the accepted syntax.
13238 If not specified, or the expressed duration is negative, the video is
13239 supposed to be generated forever.
13242 Set the number of decimals to show in the timestamp, only available in the
13243 @code{testsrc} source.
13245 The displayed timestamp value will correspond to the original
13246 timestamp value multiplied by the power of 10 of the specified
13247 value. Default value is 0.
13250 For example the following:
13252 testsrc=duration=5.3:size=qcif:rate=10
13255 will generate a video with a duration of 5.3 seconds, with size
13256 176x144 and a frame rate of 10 frames per second.
13258 The following graph description will generate a red source
13259 with an opacity of 0.2, with size "qcif" and a frame rate of 10
13262 color=c=red@@0.2:s=qcif:r=10
13265 If the input content is to be ignored, @code{nullsrc} can be used. The
13266 following command generates noise in the luminance plane by employing
13267 the @code{geq} filter:
13269 nullsrc=s=256x256, geq=random(1)*255:128:128
13272 @subsection Commands
13274 The @code{color} source supports the following commands:
13278 Set the color of the created image. Accepts the same syntax of the
13279 corresponding @option{color} option.
13282 @c man end VIDEO SOURCES
13284 @chapter Video Sinks
13285 @c man begin VIDEO SINKS
13287 Below is a description of the currently available video sinks.
13289 @section buffersink
13291 Buffer video frames, and make them available to the end of the filter
13294 This sink is mainly intended for programmatic use, in particular
13295 through the interface defined in @file{libavfilter/buffersink.h}
13296 or the options system.
13298 It accepts a pointer to an AVBufferSinkContext structure, which
13299 defines the incoming buffers' formats, to be passed as the opaque
13300 parameter to @code{avfilter_init_filter} for initialization.
13304 Null video sink: do absolutely nothing with the input video. It is
13305 mainly useful as a template and for use in analysis / debugging
13308 @c man end VIDEO SINKS
13310 @chapter Multimedia Filters
13311 @c man begin MULTIMEDIA FILTERS
13313 Below is a description of the currently available multimedia filters.
13315 @section ahistogram
13317 Convert input audio to a video output, displaying the volume histogram.
13319 The filter accepts the following options:
13323 Specify how histogram is calculated.
13325 It accepts the following values:
13328 Use single histogram for all channels.
13330 Use separate histogram for each channel.
13332 Default is @code{single}.
13335 Set frame rate, expressed as number of frames per second. Default
13339 Specify the video size for the output. For the syntax of this option, check the
13340 @ref{video size syntax,,"Video size" section in the ffmpeg-utils manual,ffmpeg-utils}.
13341 Default value is @code{hd720}.
13346 It accepts the following values:
13357 reverse logarithmic
13359 Default is @code{log}.
13362 Set amplitude scale.
13364 It accepts the following values:
13371 Default is @code{log}.
13374 Set how much frames to accumulate in histogram.
13375 Defauls is 1. Setting this to -1 accumulates all frames.
13378 Set histogram ratio of window height.
13381 Set sonogram sliding.
13383 It accepts the following values:
13386 replace old rows with new ones.
13388 scroll from top to bottom.
13390 Default is @code{replace}.
13393 @section aphasemeter
13395 Convert input audio to a video output, displaying the audio phase.
13397 The filter accepts the following options:
13401 Set the output frame rate. Default value is @code{25}.
13404 Set the video size for the output. For the syntax of this option, check the
13405 @ref{video size syntax,,"Video size" section in the ffmpeg-utils manual,ffmpeg-utils}.
13406 Default value is @code{800x400}.
13411 Specify the red, green, blue contrast. Default values are @code{2},
13412 @code{7} and @code{1}.
13413 Allowed range is @code{[0, 255]}.
13416 Set color which will be used for drawing median phase. If color is
13417 @code{none} which is default, no median phase value will be drawn.
13420 The filter also exports the frame metadata @code{lavfi.aphasemeter.phase} which
13421 represents mean phase of current audio frame. Value is in range @code{[-1, 1]}.
13422 The @code{-1} means left and right channels are completely out of phase and
13423 @code{1} means channels are in phase.
13425 @section avectorscope
13427 Convert input audio to a video output, representing the audio vector
13430 The filter is used to measure the difference between channels of stereo
13431 audio stream. A monoaural signal, consisting of identical left and right
13432 signal, results in straight vertical line. Any stereo separation is visible
13433 as a deviation from this line, creating a Lissajous figure.
13434 If the straight (or deviation from it) but horizontal line appears this
13435 indicates that the left and right channels are out of phase.
13437 The filter accepts the following options:
13441 Set the vectorscope mode.
13443 Available values are:
13446 Lissajous rotated by 45 degrees.
13449 Same as above but not rotated.
13452 Shape resembling half of circle.
13455 Default value is @samp{lissajous}.
13458 Set the video size for the output. For the syntax of this option, check the
13459 @ref{video size syntax,,"Video size" section in the ffmpeg-utils manual,ffmpeg-utils}.
13460 Default value is @code{400x400}.
13463 Set the output frame rate. Default value is @code{25}.
13469 Specify the red, green, blue and alpha contrast. Default values are @code{40},
13470 @code{160}, @code{80} and @code{255}.
13471 Allowed range is @code{[0, 255]}.
13477 Specify the red, green, blue and alpha fade. Default values are @code{15},
13478 @code{10}, @code{5} and @code{5}.
13479 Allowed range is @code{[0, 255]}.
13482 Set the zoom factor. Default value is @code{1}. Allowed range is @code{[1, 10]}.
13485 Set the vectorscope drawing mode.
13487 Available values are:
13490 Draw dot for each sample.
13493 Draw line between previous and current sample.
13496 Default value is @samp{dot}.
13499 @subsection Examples
13503 Complete example using @command{ffplay}:
13505 ffplay -f lavfi 'amovie=input.mp3, asplit [a][out1];
13506 [a] avectorscope=zoom=1.3:rc=2:gc=200:bc=10:rf=1:gf=8:bf=7 [out0]'
13512 Concatenate audio and video streams, joining them together one after the
13515 The filter works on segments of synchronized video and audio streams. All
13516 segments must have the same number of streams of each type, and that will
13517 also be the number of streams at output.
13519 The filter accepts the following options:
13524 Set the number of segments. Default is 2.
13527 Set the number of output video streams, that is also the number of video
13528 streams in each segment. Default is 1.
13531 Set the number of output audio streams, that is also the number of audio
13532 streams in each segment. Default is 0.
13535 Activate unsafe mode: do not fail if segments have a different format.
13539 The filter has @var{v}+@var{a} outputs: first @var{v} video outputs, then
13540 @var{a} audio outputs.
13542 There are @var{n}x(@var{v}+@var{a}) inputs: first the inputs for the first
13543 segment, in the same order as the outputs, then the inputs for the second
13546 Related streams do not always have exactly the same duration, for various
13547 reasons including codec frame size or sloppy authoring. For that reason,
13548 related synchronized streams (e.g. a video and its audio track) should be
13549 concatenated at once. The concat filter will use the duration of the longest
13550 stream in each segment (except the last one), and if necessary pad shorter
13551 audio streams with silence.
13553 For this filter to work correctly, all segments must start at timestamp 0.
13555 All corresponding streams must have the same parameters in all segments; the
13556 filtering system will automatically select a common pixel format for video
13557 streams, and a common sample format, sample rate and channel layout for
13558 audio streams, but other settings, such as resolution, must be converted
13559 explicitly by the user.
13561 Different frame rates are acceptable but will result in variable frame rate
13562 at output; be sure to configure the output file to handle it.
13564 @subsection Examples
13568 Concatenate an opening, an episode and an ending, all in bilingual version
13569 (video in stream 0, audio in streams 1 and 2):
13571 ffmpeg -i opening.mkv -i episode.mkv -i ending.mkv -filter_complex \
13572 '[0:0] [0:1] [0:2] [1:0] [1:1] [1:2] [2:0] [2:1] [2:2]
13573 concat=n=3:v=1:a=2 [v] [a1] [a2]' \
13574 -map '[v]' -map '[a1]' -map '[a2]' output.mkv
13578 Concatenate two parts, handling audio and video separately, using the
13579 (a)movie sources, and adjusting the resolution:
13581 movie=part1.mp4, scale=512:288 [v1] ; amovie=part1.mp4 [a1] ;
13582 movie=part2.mp4, scale=512:288 [v2] ; amovie=part2.mp4 [a2] ;
13583 [v1] [v2] concat [outv] ; [a1] [a2] concat=v=0:a=1 [outa]
13585 Note that a desync will happen at the stitch if the audio and video streams
13586 do not have exactly the same duration in the first file.
13593 EBU R128 scanner filter. This filter takes an audio stream as input and outputs
13594 it unchanged. By default, it logs a message at a frequency of 10Hz with the
13595 Momentary loudness (identified by @code{M}), Short-term loudness (@code{S}),
13596 Integrated loudness (@code{I}) and Loudness Range (@code{LRA}).
13598 The filter also has a video output (see the @var{video} option) with a real
13599 time graph to observe the loudness evolution. The graphic contains the logged
13600 message mentioned above, so it is not printed anymore when this option is set,
13601 unless the verbose logging is set. The main graphing area contains the
13602 short-term loudness (3 seconds of analysis), and the gauge on the right is for
13603 the momentary loudness (400 milliseconds).
13605 More information about the Loudness Recommendation EBU R128 on
13606 @url{http://tech.ebu.ch/loudness}.
13608 The filter accepts the following options:
13613 Activate the video output. The audio stream is passed unchanged whether this
13614 option is set or no. The video stream will be the first output stream if
13615 activated. Default is @code{0}.
13618 Set the video size. This option is for video only. For the syntax of this
13620 @ref{video size syntax,,"Video size" section in the ffmpeg-utils manual,ffmpeg-utils}.
13621 Default and minimum resolution is @code{640x480}.
13624 Set the EBU scale meter. Default is @code{9}. Common values are @code{9} and
13625 @code{18}, respectively for EBU scale meter +9 and EBU scale meter +18. Any
13626 other integer value between this range is allowed.
13629 Set metadata injection. If set to @code{1}, the audio input will be segmented
13630 into 100ms output frames, each of them containing various loudness information
13631 in metadata. All the metadata keys are prefixed with @code{lavfi.r128.}.
13633 Default is @code{0}.
13636 Force the frame logging level.
13638 Available values are:
13641 information logging level
13643 verbose logging level
13646 By default, the logging level is set to @var{info}. If the @option{video} or
13647 the @option{metadata} options are set, it switches to @var{verbose}.
13652 Available modes can be cumulated (the option is a @code{flag} type). Possible
13656 Disable any peak mode (default).
13658 Enable sample-peak mode.
13660 Simple peak mode looking for the higher sample value. It logs a message
13661 for sample-peak (identified by @code{SPK}).
13663 Enable true-peak mode.
13665 If enabled, the peak lookup is done on an over-sampled version of the input
13666 stream for better peak accuracy. It logs a message for true-peak.
13667 (identified by @code{TPK}) and true-peak per frame (identified by @code{FTPK}).
13668 This mode requires a build with @code{libswresample}.
13672 Treat mono input files as "dual mono". If a mono file is intended for playback
13673 on a stereo system, its EBU R128 measurement will be perceptually incorrect.
13674 If set to @code{true}, this option will compensate for this effect.
13675 Multi-channel input files are not affected by this option.
13678 Set a specific pan law to be used for the measurement of dual mono files.
13679 This parameter is optional, and has a default value of -3.01dB.
13682 @subsection Examples
13686 Real-time graph using @command{ffplay}, with a EBU scale meter +18:
13688 ffplay -f lavfi -i "amovie=input.mp3,ebur128=video=1:meter=18 [out0][out1]"
13692 Run an analysis with @command{ffmpeg}:
13694 ffmpeg -nostats -i input.mp3 -filter_complex ebur128 -f null -
13698 @section interleave, ainterleave
13700 Temporally interleave frames from several inputs.
13702 @code{interleave} works with video inputs, @code{ainterleave} with audio.
13704 These filters read frames from several inputs and send the oldest
13705 queued frame to the output.
13707 Input streams must have a well defined, monotonically increasing frame
13710 In order to submit one frame to output, these filters need to enqueue
13711 at least one frame for each input, so they cannot work in case one
13712 input is not yet terminated and will not receive incoming frames.
13714 For example consider the case when one input is a @code{select} filter
13715 which always drop input frames. The @code{interleave} filter will keep
13716 reading from that input, but it will never be able to send new frames
13717 to output until the input will send an end-of-stream signal.
13719 Also, depending on inputs synchronization, the filters will drop
13720 frames in case one input receives more frames than the other ones, and
13721 the queue is already filled.
13723 These filters accept the following options:
13727 Set the number of different inputs, it is 2 by default.
13730 @subsection Examples
13734 Interleave frames belonging to different streams using @command{ffmpeg}:
13736 ffmpeg -i bambi.avi -i pr0n.mkv -filter_complex "[0:v][1:v] interleave" out.avi
13740 Add flickering blur effect:
13742 select='if(gt(random(0), 0.2), 1, 2)':n=2 [tmp], boxblur=2:2, [tmp] interleave
13746 @section perms, aperms
13748 Set read/write permissions for the output frames.
13750 These filters are mainly aimed at developers to test direct path in the
13751 following filter in the filtergraph.
13753 The filters accept the following options:
13757 Select the permissions mode.
13759 It accepts the following values:
13762 Do nothing. This is the default.
13764 Set all the output frames read-only.
13766 Set all the output frames directly writable.
13768 Make the frame read-only if writable, and writable if read-only.
13770 Set each output frame read-only or writable randomly.
13774 Set the seed for the @var{random} mode, must be an integer included between
13775 @code{0} and @code{UINT32_MAX}. If not specified, or if explicitly set to
13776 @code{-1}, the filter will try to use a good random seed on a best effort
13780 Note: in case of auto-inserted filter between the permission filter and the
13781 following one, the permission might not be received as expected in that
13782 following filter. Inserting a @ref{format} or @ref{aformat} filter before the
13783 perms/aperms filter can avoid this problem.
13785 @section realtime, arealtime
13787 Slow down filtering to match real time approximatively.
13789 These filters will pause the filtering for a variable amount of time to
13790 match the output rate with the input timestamps.
13791 They are similar to the @option{re} option to @code{ffmpeg}.
13793 They accept the following options:
13797 Time limit for the pauses. Any pause longer than that will be considered
13798 a timestamp discontinuity and reset the timer. Default is 2 seconds.
13801 @section select, aselect
13803 Select frames to pass in output.
13805 This filter accepts the following options:
13810 Set expression, which is evaluated for each input frame.
13812 If the expression is evaluated to zero, the frame is discarded.
13814 If the evaluation result is negative or NaN, the frame is sent to the
13815 first output; otherwise it is sent to the output with index
13816 @code{ceil(val)-1}, assuming that the input index starts from 0.
13818 For example a value of @code{1.2} corresponds to the output with index
13819 @code{ceil(1.2)-1 = 2-1 = 1}, that is the second output.
13822 Set the number of outputs. The output to which to send the selected
13823 frame is based on the result of the evaluation. Default value is 1.
13826 The expression can contain the following constants:
13830 The (sequential) number of the filtered frame, starting from 0.
13833 The (sequential) number of the selected frame, starting from 0.
13835 @item prev_selected_n
13836 The sequential number of the last selected frame. It's NAN if undefined.
13839 The timebase of the input timestamps.
13842 The PTS (Presentation TimeStamp) of the filtered video frame,
13843 expressed in @var{TB} units. It's NAN if undefined.
13846 The PTS of the filtered video frame,
13847 expressed in seconds. It's NAN if undefined.
13850 The PTS of the previously filtered video frame. It's NAN if undefined.
13852 @item prev_selected_pts
13853 The PTS of the last previously filtered video frame. It's NAN if undefined.
13855 @item prev_selected_t
13856 The PTS of the last previously selected video frame. It's NAN if undefined.
13859 The PTS of the first video frame in the video. It's NAN if undefined.
13862 The time of the first video frame in the video. It's NAN if undefined.
13864 @item pict_type @emph{(video only)}
13865 The type of the filtered frame. It can assume one of the following
13877 @item interlace_type @emph{(video only)}
13878 The frame interlace type. It can assume one of the following values:
13881 The frame is progressive (not interlaced).
13883 The frame is top-field-first.
13885 The frame is bottom-field-first.
13888 @item consumed_sample_n @emph{(audio only)}
13889 the number of selected samples before the current frame
13891 @item samples_n @emph{(audio only)}
13892 the number of samples in the current frame
13894 @item sample_rate @emph{(audio only)}
13895 the input sample rate
13898 This is 1 if the filtered frame is a key-frame, 0 otherwise.
13901 the position in the file of the filtered frame, -1 if the information
13902 is not available (e.g. for synthetic video)
13904 @item scene @emph{(video only)}
13905 value between 0 and 1 to indicate a new scene; a low value reflects a low
13906 probability for the current frame to introduce a new scene, while a higher
13907 value means the current frame is more likely to be one (see the example below)
13909 @item concatdec_select
13910 The concat demuxer can select only part of a concat input file by setting an
13911 inpoint and an outpoint, but the output packets may not be entirely contained
13912 in the selected interval. By using this variable, it is possible to skip frames
13913 generated by the concat demuxer which are not exactly contained in the selected
13916 This works by comparing the frame pts against the @var{lavf.concat.start_time}
13917 and the @var{lavf.concat.duration} packet metadata values which are also
13918 present in the decoded frames.
13920 The @var{concatdec_select} variable is -1 if the frame pts is at least
13921 start_time and either the duration metadata is missing or the frame pts is less
13922 than start_time + duration, 0 otherwise, and NaN if the start_time metadata is
13925 That basically means that an input frame is selected if its pts is within the
13926 interval set by the concat demuxer.
13930 The default value of the select expression is "1".
13932 @subsection Examples
13936 Select all frames in input:
13941 The example above is the same as:
13953 Select only I-frames:
13955 select='eq(pict_type\,I)'
13959 Select one frame every 100:
13961 select='not(mod(n\,100))'
13965 Select only frames contained in the 10-20 time interval:
13967 select=between(t\,10\,20)
13971 Select only I frames contained in the 10-20 time interval:
13973 select=between(t\,10\,20)*eq(pict_type\,I)
13977 Select frames with a minimum distance of 10 seconds:
13979 select='isnan(prev_selected_t)+gte(t-prev_selected_t\,10)'
13983 Use aselect to select only audio frames with samples number > 100:
13985 aselect='gt(samples_n\,100)'
13989 Create a mosaic of the first scenes:
13991 ffmpeg -i video.avi -vf select='gt(scene\,0.4)',scale=160:120,tile -frames:v 1 preview.png
13994 Comparing @var{scene} against a value between 0.3 and 0.5 is generally a sane
13998 Send even and odd frames to separate outputs, and compose them:
14000 select=n=2:e='mod(n, 2)+1' [odd][even]; [odd] pad=h=2*ih [tmp]; [tmp][even] overlay=y=h
14004 Select useful frames from an ffconcat file which is using inpoints and
14005 outpoints but where the source files are not intra frame only.
14007 ffmpeg -copyts -vsync 0 -segment_time_metadata 1 -i input.ffconcat -vf select=concatdec_select -af aselect=concatdec_select output.avi
14011 @section sendcmd, asendcmd
14013 Send commands to filters in the filtergraph.
14015 These filters read commands to be sent to other filters in the
14018 @code{sendcmd} must be inserted between two video filters,
14019 @code{asendcmd} must be inserted between two audio filters, but apart
14020 from that they act the same way.
14022 The specification of commands can be provided in the filter arguments
14023 with the @var{commands} option, or in a file specified by the
14024 @var{filename} option.
14026 These filters accept the following options:
14029 Set the commands to be read and sent to the other filters.
14031 Set the filename of the commands to be read and sent to the other
14035 @subsection Commands syntax
14037 A commands description consists of a sequence of interval
14038 specifications, comprising a list of commands to be executed when a
14039 particular event related to that interval occurs. The occurring event
14040 is typically the current frame time entering or leaving a given time
14043 An interval is specified by the following syntax:
14045 @var{START}[-@var{END}] @var{COMMANDS};
14048 The time interval is specified by the @var{START} and @var{END} times.
14049 @var{END} is optional and defaults to the maximum time.
14051 The current frame time is considered within the specified interval if
14052 it is included in the interval [@var{START}, @var{END}), that is when
14053 the time is greater or equal to @var{START} and is lesser than
14056 @var{COMMANDS} consists of a sequence of one or more command
14057 specifications, separated by ",", relating to that interval. The
14058 syntax of a command specification is given by:
14060 [@var{FLAGS}] @var{TARGET} @var{COMMAND} @var{ARG}
14063 @var{FLAGS} is optional and specifies the type of events relating to
14064 the time interval which enable sending the specified command, and must
14065 be a non-null sequence of identifier flags separated by "+" or "|" and
14066 enclosed between "[" and "]".
14068 The following flags are recognized:
14071 The command is sent when the current frame timestamp enters the
14072 specified interval. In other words, the command is sent when the
14073 previous frame timestamp was not in the given interval, and the
14077 The command is sent when the current frame timestamp leaves the
14078 specified interval. In other words, the command is sent when the
14079 previous frame timestamp was in the given interval, and the
14083 If @var{FLAGS} is not specified, a default value of @code{[enter]} is
14086 @var{TARGET} specifies the target of the command, usually the name of
14087 the filter class or a specific filter instance name.
14089 @var{COMMAND} specifies the name of the command for the target filter.
14091 @var{ARG} is optional and specifies the optional list of argument for
14092 the given @var{COMMAND}.
14094 Between one interval specification and another, whitespaces, or
14095 sequences of characters starting with @code{#} until the end of line,
14096 are ignored and can be used to annotate comments.
14098 A simplified BNF description of the commands specification syntax
14101 @var{COMMAND_FLAG} ::= "enter" | "leave"
14102 @var{COMMAND_FLAGS} ::= @var{COMMAND_FLAG} [(+|"|")@var{COMMAND_FLAG}]
14103 @var{COMMAND} ::= ["[" @var{COMMAND_FLAGS} "]"] @var{TARGET} @var{COMMAND} [@var{ARG}]
14104 @var{COMMANDS} ::= @var{COMMAND} [,@var{COMMANDS}]
14105 @var{INTERVAL} ::= @var{START}[-@var{END}] @var{COMMANDS}
14106 @var{INTERVALS} ::= @var{INTERVAL}[;@var{INTERVALS}]
14109 @subsection Examples
14113 Specify audio tempo change at second 4:
14115 asendcmd=c='4.0 atempo tempo 1.5',atempo
14119 Specify a list of drawtext and hue commands in a file.
14121 # show text in the interval 5-10
14122 5.0-10.0 [enter] drawtext reinit 'fontfile=FreeSerif.ttf:text=hello world',
14123 [leave] drawtext reinit 'fontfile=FreeSerif.ttf:text=';
14125 # desaturate the image in the interval 15-20
14126 15.0-20.0 [enter] hue s 0,
14127 [enter] drawtext reinit 'fontfile=FreeSerif.ttf:text=nocolor',
14129 [leave] drawtext reinit 'fontfile=FreeSerif.ttf:text=color';
14131 # apply an exponential saturation fade-out effect, starting from time 25
14132 25 [enter] hue s exp(25-t)
14135 A filtergraph allowing to read and process the above command list
14136 stored in a file @file{test.cmd}, can be specified with:
14138 sendcmd=f=test.cmd,drawtext=fontfile=FreeSerif.ttf:text='',hue
14143 @section setpts, asetpts
14145 Change the PTS (presentation timestamp) of the input frames.
14147 @code{setpts} works on video frames, @code{asetpts} on audio frames.
14149 This filter accepts the following options:
14154 The expression which is evaluated for each frame to construct its timestamp.
14158 The expression is evaluated through the eval API and can contain the following
14163 frame rate, only defined for constant frame-rate video
14166 The presentation timestamp in input
14169 The count of the input frame for video or the number of consumed samples,
14170 not including the current frame for audio, starting from 0.
14172 @item NB_CONSUMED_SAMPLES
14173 The number of consumed samples, not including the current frame (only
14176 @item NB_SAMPLES, S
14177 The number of samples in the current frame (only audio)
14179 @item SAMPLE_RATE, SR
14180 The audio sample rate.
14183 The PTS of the first frame.
14186 the time in seconds of the first frame
14189 State whether the current frame is interlaced.
14192 the time in seconds of the current frame
14195 original position in the file of the frame, or undefined if undefined
14196 for the current frame
14199 The previous input PTS.
14202 previous input time in seconds
14205 The previous output PTS.
14208 previous output time in seconds
14211 The wallclock (RTC) time in microseconds. This is deprecated, use time(0)
14215 The wallclock (RTC) time at the start of the movie in microseconds.
14218 The timebase of the input timestamps.
14222 @subsection Examples
14226 Start counting PTS from zero
14228 setpts=PTS-STARTPTS
14232 Apply fast motion effect:
14238 Apply slow motion effect:
14244 Set fixed rate of 25 frames per second:
14250 Set fixed rate 25 fps with some jitter:
14252 setpts='1/(25*TB) * (N + 0.05 * sin(N*2*PI/25))'
14256 Apply an offset of 10 seconds to the input PTS:
14262 Generate timestamps from a "live source" and rebase onto the current timebase:
14264 setpts='(RTCTIME - RTCSTART) / (TB * 1000000)'
14268 Generate timestamps by counting samples:
14275 @section settb, asettb
14277 Set the timebase to use for the output frames timestamps.
14278 It is mainly useful for testing timebase configuration.
14280 It accepts the following parameters:
14285 The expression which is evaluated into the output timebase.
14289 The value for @option{tb} is an arithmetic expression representing a
14290 rational. The expression can contain the constants "AVTB" (the default
14291 timebase), "intb" (the input timebase) and "sr" (the sample rate,
14292 audio only). Default value is "intb".
14294 @subsection Examples
14298 Set the timebase to 1/25:
14304 Set the timebase to 1/10:
14310 Set the timebase to 1001/1000:
14316 Set the timebase to 2*intb:
14322 Set the default timebase value:
14329 Convert input audio to a video output representing frequency spectrum
14330 logarithmically using Brown-Puckette constant Q transform algorithm with
14331 direct frequency domain coefficient calculation (but the transform itself
14332 is not really constant Q, instead the Q factor is actually variable/clamped),
14333 with musical tone scale, from E0 to D#10.
14335 The filter accepts the following options:
14339 Specify the video size for the output. It must be even. For the syntax of this option,
14340 check the @ref{video size syntax,,"Video size" section in the ffmpeg-utils manual,ffmpeg-utils}.
14341 Default value is @code{1920x1080}.
14344 Set the output frame rate. Default value is @code{25}.
14347 Set the bargraph height. It must be even. Default value is @code{-1} which
14348 computes the bargraph height automatically.
14351 Set the axis height. It must be even. Default value is @code{-1} which computes
14352 the axis height automatically.
14355 Set the sonogram height. It must be even. Default value is @code{-1} which
14356 computes the sonogram height automatically.
14359 Set the fullhd resolution. This option is deprecated, use @var{size}, @var{s}
14360 instead. Default value is @code{1}.
14362 @item sono_v, volume
14363 Specify the sonogram volume expression. It can contain variables:
14366 the @var{bar_v} evaluated expression
14367 @item frequency, freq, f
14368 the frequency where it is evaluated
14369 @item timeclamp, tc
14370 the value of @var{timeclamp} option
14374 @item a_weighting(f)
14375 A-weighting of equal loudness
14376 @item b_weighting(f)
14377 B-weighting of equal loudness
14378 @item c_weighting(f)
14379 C-weighting of equal loudness.
14381 Default value is @code{16}.
14383 @item bar_v, volume2
14384 Specify the bargraph volume expression. It can contain variables:
14387 the @var{sono_v} evaluated expression
14388 @item frequency, freq, f
14389 the frequency where it is evaluated
14390 @item timeclamp, tc
14391 the value of @var{timeclamp} option
14395 @item a_weighting(f)
14396 A-weighting of equal loudness
14397 @item b_weighting(f)
14398 B-weighting of equal loudness
14399 @item c_weighting(f)
14400 C-weighting of equal loudness.
14402 Default value is @code{sono_v}.
14404 @item sono_g, gamma
14405 Specify the sonogram gamma. Lower gamma makes the spectrum more contrast,
14406 higher gamma makes the spectrum having more range. Default value is @code{3}.
14407 Acceptable range is @code{[1, 7]}.
14409 @item bar_g, gamma2
14410 Specify the bargraph gamma. Default value is @code{1}. Acceptable range is
14413 @item timeclamp, tc
14414 Specify the transform timeclamp. At low frequency, there is trade-off between
14415 accuracy in time domain and frequency domain. If timeclamp is lower,
14416 event in time domain is represented more accurately (such as fast bass drum),
14417 otherwise event in frequency domain is represented more accurately
14418 (such as bass guitar). Acceptable range is @code{[0.1, 1]}. Default value is @code{0.17}.
14421 Specify the transform base frequency. Default value is @code{20.01523126408007475},
14422 which is frequency 50 cents below E0. Acceptable range is @code{[10, 100000]}.
14425 Specify the transform end frequency. Default value is @code{20495.59681441799654},
14426 which is frequency 50 cents above D#10. Acceptable range is @code{[10, 100000]}.
14429 This option is deprecated and ignored.
14432 Specify the transform length in time domain. Use this option to control accuracy
14433 trade-off between time domain and frequency domain at every frequency sample.
14434 It can contain variables:
14436 @item frequency, freq, f
14437 the frequency where it is evaluated
14438 @item timeclamp, tc
14439 the value of @var{timeclamp} option.
14441 Default value is @code{384*tc/(384+tc*f)}.
14444 Specify the transform count for every video frame. Default value is @code{6}.
14445 Acceptable range is @code{[1, 30]}.
14448 Specify the transform count for every single pixel. Default value is @code{0},
14449 which makes it computed automatically. Acceptable range is @code{[0, 10]}.
14452 Specify font file for use with freetype to draw the axis. If not specified,
14453 use embedded font. Note that drawing with font file or embedded font is not
14454 implemented with custom @var{basefreq} and @var{endfreq}, use @var{axisfile}
14458 Specify font color expression. This is arithmetic expression that should return
14459 integer value 0xRRGGBB. It can contain variables:
14461 @item frequency, freq, f
14462 the frequency where it is evaluated
14463 @item timeclamp, tc
14464 the value of @var{timeclamp} option
14469 midi number of frequency f, some midi numbers: E0(16), C1(24), C2(36), A4(69)
14470 @item r(x), g(x), b(x)
14471 red, green, and blue value of intensity x.
14473 Default value is @code{st(0, (midi(f)-59.5)/12);
14474 st(1, if(between(ld(0),0,1), 0.5-0.5*cos(2*PI*ld(0)), 0));
14475 r(1-ld(1)) + b(ld(1))}.
14478 Specify image file to draw the axis. This option override @var{fontfile} and
14479 @var{fontcolor} option.
14482 Enable/disable drawing text to the axis. If it is set to @code{0}, drawing to
14483 the axis is disabled, ignoring @var{fontfile} and @var{axisfile} option.
14484 Default value is @code{1}.
14488 @subsection Examples
14492 Playing audio while showing the spectrum:
14494 ffplay -f lavfi 'amovie=a.mp3, asplit [a][out1]; [a] showcqt [out0]'
14498 Same as above, but with frame rate 30 fps:
14500 ffplay -f lavfi 'amovie=a.mp3, asplit [a][out1]; [a] showcqt=fps=30:count=5 [out0]'
14504 Playing at 1280x720:
14506 ffplay -f lavfi 'amovie=a.mp3, asplit [a][out1]; [a] showcqt=s=1280x720:count=4 [out0]'
14510 Disable sonogram display:
14516 A1 and its harmonics: A1, A2, (near)E3, A3:
14518 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),
14519 asplit[a][out1]; [a] showcqt [out0]'
14523 Same as above, but with more accuracy in frequency domain:
14525 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),
14526 asplit[a][out1]; [a] showcqt=timeclamp=0.5 [out0]'
14532 bar_v=10:sono_v=bar_v*a_weighting(f)
14536 Custom gamma, now spectrum is linear to the amplitude.
14542 Custom tlength equation:
14544 tc=0.33:tlength='st(0,0.17); 384*tc / (384 / ld(0) + tc*f /(1-ld(0))) + 384*tc / (tc*f / ld(0) + 384 /(1-ld(0)))'
14548 Custom fontcolor and fontfile, C-note is colored green, others are colored blue:
14550 fontcolor='if(mod(floor(midi(f)+0.5),12), 0x0000FF, g(1))':fontfile=myfont.ttf
14554 Custom frequency range with custom axis using image file:
14556 axisfile=myaxis.png:basefreq=40:endfreq=10000
14562 Convert input audio to video output representing the audio power spectrum.
14563 Audio amplitude is on Y-axis while frequency is on X-axis.
14565 The filter accepts the following options:
14569 Specify size of video. For the syntax of this option, check the
14570 @ref{video size syntax,,"Video size" section in the ffmpeg-utils manual,ffmpeg-utils}.
14571 Default is @code{1024x512}.
14575 This set how each frequency bin will be represented.
14577 It accepts the following values:
14583 Default is @code{bar}.
14586 Set amplitude scale.
14588 It accepts the following values:
14602 Default is @code{log}.
14605 Set frequency scale.
14607 It accepts the following values:
14616 Reverse logarithmic scale.
14618 Default is @code{lin}.
14623 It accepts the following values:
14639 Default is @code{w2048}
14642 Set windowing function.
14644 It accepts the following values:
14662 Default is @code{hanning}.
14665 Set window overlap. In range @code{[0, 1]}. Default is @code{1},
14666 which means optimal overlap for selected window function will be picked.
14669 Set time averaging. Setting this to 0 will display current maximal peaks.
14670 Default is @code{1}, which means time averaging is disabled.
14673 Specify list of colors separated by space or by '|' which will be used to
14674 draw channel frequencies. Unrecognized or missing colors will be replaced
14678 Set channel display mode.
14680 It accepts the following values:
14685 Default is @code{combined}.
14689 @anchor{showspectrum}
14690 @section showspectrum
14692 Convert input audio to a video output, representing the audio frequency
14695 The filter accepts the following options:
14699 Specify the video size for the output. For the syntax of this option, check the
14700 @ref{video size syntax,,"Video size" section in the ffmpeg-utils manual,ffmpeg-utils}.
14701 Default value is @code{640x512}.
14704 Specify how the spectrum should slide along the window.
14706 It accepts the following values:
14709 the samples start again on the left when they reach the right
14711 the samples scroll from right to left
14713 the samples scroll from left to right
14715 frames are only produced when the samples reach the right
14718 Default value is @code{replace}.
14721 Specify display mode.
14723 It accepts the following values:
14726 all channels are displayed in the same row
14728 all channels are displayed in separate rows
14731 Default value is @samp{combined}.
14734 Specify display color mode.
14736 It accepts the following values:
14739 each channel is displayed in a separate color
14741 each channel is displayed using the same color scheme
14743 each channel is displayed using the rainbow color scheme
14745 each channel is displayed using the moreland color scheme
14747 each channel is displayed using the nebulae color scheme
14749 each channel is displayed using the fire color scheme
14751 each channel is displayed using the fiery color scheme
14753 each channel is displayed using the fruit color scheme
14755 each channel is displayed using the cool color scheme
14758 Default value is @samp{channel}.
14761 Specify scale used for calculating intensity color values.
14763 It accepts the following values:
14768 square root, default
14779 Default value is @samp{sqrt}.
14782 Set saturation modifier for displayed colors. Negative values provide
14783 alternative color scheme. @code{0} is no saturation at all.
14784 Saturation must be in [-10.0, 10.0] range.
14785 Default value is @code{1}.
14788 Set window function.
14790 It accepts the following values:
14810 Default value is @code{hann}.
14813 Set orientation of time vs frequency axis. Can be @code{vertical} or
14814 @code{horizontal}. Default is @code{vertical}.
14817 Set ratio of overlap window. Default value is @code{0}.
14818 When value is @code{1} overlap is set to recommended size for specific
14819 window function currently used.
14822 Set scale gain for calculating intensity color values.
14823 Default value is @code{1}.
14826 Set which data to display. Can be @code{magnitude}, default or @code{phase}.
14829 The usage is very similar to the showwaves filter; see the examples in that
14832 @subsection Examples
14836 Large window with logarithmic color scaling:
14838 showspectrum=s=1280x480:scale=log
14842 Complete example for a colored and sliding spectrum per channel using @command{ffplay}:
14844 ffplay -f lavfi 'amovie=input.mp3, asplit [a][out1];
14845 [a] showspectrum=mode=separate:color=intensity:slide=1:scale=cbrt [out0]'
14849 @section showspectrumpic
14851 Convert input audio to a single video frame, representing the audio frequency
14854 The filter accepts the following options:
14858 Specify the video size for the output. For the syntax of this option, check the
14859 @ref{video size syntax,,"Video size" section in the ffmpeg-utils manual,ffmpeg-utils}.
14860 Default value is @code{4096x2048}.
14863 Specify display mode.
14865 It accepts the following values:
14868 all channels are displayed in the same row
14870 all channels are displayed in separate rows
14872 Default value is @samp{combined}.
14875 Specify display color mode.
14877 It accepts the following values:
14880 each channel is displayed in a separate color
14882 each channel is displayed using the same color scheme
14884 each channel is displayed using the rainbow color scheme
14886 each channel is displayed using the moreland color scheme
14888 each channel is displayed using the nebulae color scheme
14890 each channel is displayed using the fire color scheme
14892 each channel is displayed using the fiery color scheme
14894 each channel is displayed using the fruit color scheme
14896 each channel is displayed using the cool color scheme
14898 Default value is @samp{intensity}.
14901 Specify scale used for calculating intensity color values.
14903 It accepts the following values:
14908 square root, default
14918 Default value is @samp{log}.
14921 Set saturation modifier for displayed colors. Negative values provide
14922 alternative color scheme. @code{0} is no saturation at all.
14923 Saturation must be in [-10.0, 10.0] range.
14924 Default value is @code{1}.
14927 Set window function.
14929 It accepts the following values:
14948 Default value is @code{hann}.
14951 Set orientation of time vs frequency axis. Can be @code{vertical} or
14952 @code{horizontal}. Default is @code{vertical}.
14955 Set scale gain for calculating intensity color values.
14956 Default value is @code{1}.
14959 Draw time and frequency axes and legends. Default is enabled.
14962 @subsection Examples
14966 Extract an audio spectrogram of a whole audio track
14967 in a 1024x1024 picture using @command{ffmpeg}:
14969 ffmpeg -i audio.flac -lavfi showspectrumpic=s=1024x1024 spectrogram.png
14973 @section showvolume
14975 Convert input audio volume to a video output.
14977 The filter accepts the following options:
14984 Set border width, allowed range is [0, 5]. Default is 1.
14987 Set channel width, allowed range is [80, 1080]. Default is 400.
14990 Set channel height, allowed range is [1, 100]. Default is 20.
14993 Set fade, allowed range is [0.001, 1]. Default is 0.95.
14996 Set volume color expression.
14998 The expression can use the following variables:
15002 Current max volume of channel in dB.
15005 Current channel number, starting from 0.
15009 If set, displays channel names. Default is enabled.
15012 If set, displays volume values. Default is enabled.
15017 Convert input audio to a video output, representing the samples waves.
15019 The filter accepts the following options:
15023 Specify the video size for the output. For the syntax of this option, check the
15024 @ref{video size syntax,,"Video size" section in the ffmpeg-utils manual,ffmpeg-utils}.
15025 Default value is @code{600x240}.
15030 Available values are:
15033 Draw a point for each sample.
15036 Draw a vertical line for each sample.
15039 Draw a point for each sample and a line between them.
15042 Draw a centered vertical line for each sample.
15045 Default value is @code{point}.
15048 Set the number of samples which are printed on the same column. A
15049 larger value will decrease the frame rate. Must be a positive
15050 integer. This option can be set only if the value for @var{rate}
15051 is not explicitly specified.
15054 Set the (approximate) output frame rate. This is done by setting the
15055 option @var{n}. Default value is "25".
15057 @item split_channels
15058 Set if channels should be drawn separately or overlap. Default value is 0.
15062 @subsection Examples
15066 Output the input file audio and the corresponding video representation
15069 amovie=a.mp3,asplit[out0],showwaves[out1]
15073 Create a synthetic signal and show it with showwaves, forcing a
15074 frame rate of 30 frames per second:
15076 aevalsrc=sin(1*2*PI*t)*sin(880*2*PI*t):cos(2*PI*200*t),asplit[out0],showwaves=r=30[out1]
15080 @section showwavespic
15082 Convert input audio to a single video frame, representing the samples waves.
15084 The filter accepts the following options:
15088 Specify the video size for the output. For the syntax of this option, check the
15089 @ref{video size syntax,,"Video size" section in the ffmpeg-utils manual,ffmpeg-utils}.
15090 Default value is @code{600x240}.
15092 @item split_channels
15093 Set if channels should be drawn separately or overlap. Default value is 0.
15096 @subsection Examples
15100 Extract a channel split representation of the wave form of a whole audio track
15101 in a 1024x800 picture using @command{ffmpeg}:
15103 ffmpeg -i audio.flac -lavfi showwavespic=split_channels=1:s=1024x800 waveform.png
15107 Colorize the waveform with colorchannelmixer. This example will make
15108 the waveform a green color approximately RGB(66,217,150). Additional
15109 channels will be shades of this color.
15111 ffmpeg -i audio.mp3 -filter_complex "showwavespic,colorchannelmixer=rr=66/255:gg=217/255:bb=150/255" waveform.png
15115 @section spectrumsynth
15117 Sythesize audio from 2 input video spectrums, first input stream represents
15118 magnitude across time and second represents phase across time.
15119 The filter will transform from frequency domain as displayed in videos back
15120 to time domain as presented in audio output.
15122 This filter is primarly created for reversing processed @ref{showspectrum}
15123 filter outputs, but can synthesize sound from other spectrograms too.
15124 But in such case results are going to be poor if the phase data is not
15125 available, because in such cases phase data need to be recreated, usually
15126 its just recreated from random noise.
15127 For best results use gray only output (@code{channel} color mode in
15128 @ref{showspectrum} filter) and @code{log} scale for magnitude video and
15129 @code{lin} scale for phase video. To produce phase, for 2nd video, use
15130 @code{data} option. Inputs videos should generally use @code{fullframe}
15131 slide mode as that saves resources needed for decoding video.
15133 The filter accepts the following options:
15137 Specify sample rate of output audio, the sample rate of audio from which
15138 spectrum was generated may differ.
15141 Set number of channels represented in input video spectrums.
15144 Set scale which was used when generating magnitude input spectrum.
15145 Can be @code{lin} or @code{log}. Default is @code{log}.
15148 Set slide which was used when generating inputs spectrums.
15149 Can be @code{replace}, @code{scroll}, @code{fullframe} or @code{rscroll}.
15150 Default is @code{fullframe}.
15153 Set window function used for resynthesis.
15156 Set window overlap. In range @code{[0, 1]}. Default is @code{1},
15157 which means optimal overlap for selected window function will be picked.
15160 Set orientation of input videos. Can be @code{vertical} or @code{horizontal}.
15161 Default is @code{vertical}.
15164 @subsection Examples
15168 First create magnitude and phase videos from audio, assuming audio is stereo with 44100 sample rate,
15169 then resynthesize videos back to audio with spectrumsynth:
15171 ffmpeg -i input.flac -lavfi showspectrum=mode=separate:scale=log:overlap=0.875:color=channel:slide=fullframe:data=magnitude -an -c:v rawvideo magnitude.nut
15172 ffmpeg -i input.flac -lavfi showspectrum=mode=separate:scale=lin:overlap=0.875:color=channel:slide=fullframe:data=phase -an -c:v rawvideo phase.nut
15173 ffmpeg -i magnitude.nut -i phase.nut -lavfi spectrumsynth=channels=2:sample_rate=44100:win_fun=hann:overlap=0.875:slide=fullframe output.flac
15177 @section split, asplit
15179 Split input into several identical outputs.
15181 @code{asplit} works with audio input, @code{split} with video.
15183 The filter accepts a single parameter which specifies the number of outputs. If
15184 unspecified, it defaults to 2.
15186 @subsection Examples
15190 Create two separate outputs from the same input:
15192 [in] split [out0][out1]
15196 To create 3 or more outputs, you need to specify the number of
15199 [in] asplit=3 [out0][out1][out2]
15203 Create two separate outputs from the same input, one cropped and
15206 [in] split [splitout1][splitout2];
15207 [splitout1] crop=100:100:0:0 [cropout];
15208 [splitout2] pad=200:200:100:100 [padout];
15212 Create 5 copies of the input audio with @command{ffmpeg}:
15214 ffmpeg -i INPUT -filter_complex asplit=5 OUTPUT
15220 Receive commands sent through a libzmq client, and forward them to
15221 filters in the filtergraph.
15223 @code{zmq} and @code{azmq} work as a pass-through filters. @code{zmq}
15224 must be inserted between two video filters, @code{azmq} between two
15227 To enable these filters you need to install the libzmq library and
15228 headers and configure FFmpeg with @code{--enable-libzmq}.
15230 For more information about libzmq see:
15231 @url{http://www.zeromq.org/}
15233 The @code{zmq} and @code{azmq} filters work as a libzmq server, which
15234 receives messages sent through a network interface defined by the
15235 @option{bind_address} option.
15237 The received message must be in the form:
15239 @var{TARGET} @var{COMMAND} [@var{ARG}]
15242 @var{TARGET} specifies the target of the command, usually the name of
15243 the filter class or a specific filter instance name.
15245 @var{COMMAND} specifies the name of the command for the target filter.
15247 @var{ARG} is optional and specifies the optional argument list for the
15248 given @var{COMMAND}.
15250 Upon reception, the message is processed and the corresponding command
15251 is injected into the filtergraph. Depending on the result, the filter
15252 will send a reply to the client, adopting the format:
15254 @var{ERROR_CODE} @var{ERROR_REASON}
15258 @var{MESSAGE} is optional.
15260 @subsection Examples
15262 Look at @file{tools/zmqsend} for an example of a zmq client which can
15263 be used to send commands processed by these filters.
15265 Consider the following filtergraph generated by @command{ffplay}
15267 ffplay -dumpgraph 1 -f lavfi "
15268 color=s=100x100:c=red [l];
15269 color=s=100x100:c=blue [r];
15270 nullsrc=s=200x100, zmq [bg];
15271 [bg][l] overlay [bg+l];
15272 [bg+l][r] overlay=x=100 "
15275 To change the color of the left side of the video, the following
15276 command can be used:
15278 echo Parsed_color_0 c yellow | tools/zmqsend
15281 To change the right side:
15283 echo Parsed_color_1 c pink | tools/zmqsend
15286 @c man end MULTIMEDIA FILTERS
15288 @chapter Multimedia Sources
15289 @c man begin MULTIMEDIA SOURCES
15291 Below is a description of the currently available multimedia sources.
15295 This is the same as @ref{movie} source, except it selects an audio
15301 Read audio and/or video stream(s) from a movie container.
15303 It accepts the following parameters:
15307 The name of the resource to read (not necessarily a file; it can also be a
15308 device or a stream accessed through some protocol).
15310 @item format_name, f
15311 Specifies the format assumed for the movie to read, and can be either
15312 the name of a container or an input device. If not specified, the
15313 format is guessed from @var{movie_name} or by probing.
15315 @item seek_point, sp
15316 Specifies the seek point in seconds. The frames will be output
15317 starting from this seek point. The parameter is evaluated with
15318 @code{av_strtod}, so the numerical value may be suffixed by an IS
15319 postfix. The default value is "0".
15322 Specifies the streams to read. Several streams can be specified,
15323 separated by "+". The source will then have as many outputs, in the
15324 same order. The syntax is explained in the ``Stream specifiers''
15325 section in the ffmpeg manual. Two special names, "dv" and "da" specify
15326 respectively the default (best suited) video and audio stream. Default
15327 is "dv", or "da" if the filter is called as "amovie".
15329 @item stream_index, si
15330 Specifies the index of the video stream to read. If the value is -1,
15331 the most suitable video stream will be automatically selected. The default
15332 value is "-1". Deprecated. If the filter is called "amovie", it will select
15333 audio instead of video.
15336 Specifies how many times to read the stream in sequence.
15337 If the value is less than 1, the stream will be read again and again.
15338 Default value is "1".
15340 Note that when the movie is looped the source timestamps are not
15341 changed, so it will generate non monotonically increasing timestamps.
15344 It allows overlaying a second video on top of the main input of
15345 a filtergraph, as shown in this graph:
15347 input -----------> deltapts0 --> overlay --> output
15350 movie --> scale--> deltapts1 -------+
15352 @subsection Examples
15356 Skip 3.2 seconds from the start of the AVI file in.avi, and overlay it
15357 on top of the input labelled "in":
15359 movie=in.avi:seek_point=3.2, scale=180:-1, setpts=PTS-STARTPTS [over];
15360 [in] setpts=PTS-STARTPTS [main];
15361 [main][over] overlay=16:16 [out]
15365 Read from a video4linux2 device, and overlay it on top of the input
15368 movie=/dev/video0:f=video4linux2, scale=180:-1, setpts=PTS-STARTPTS [over];
15369 [in] setpts=PTS-STARTPTS [main];
15370 [main][over] overlay=16:16 [out]
15374 Read the first video stream and the audio stream with id 0x81 from
15375 dvd.vob; the video is connected to the pad named "video" and the audio is
15376 connected to the pad named "audio":
15378 movie=dvd.vob:s=v:0+#0x81 [video] [audio]
15382 @c man end MULTIMEDIA SOURCES