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 Delay one or more audio channels.
325 Samples in delayed channel are filled with silence.
327 The filter accepts the following option:
331 Set list of delays in milliseconds for each channel separated by '|'.
332 At least one delay greater than 0 should be provided.
333 Unused delays will be silently ignored. If number of given delays is
334 smaller than number of channels all remaining channels will not be delayed.
341 Delay first channel by 1.5 seconds, the third channel by 0.5 seconds and leave
342 the second channel (and any other channels that may be present) unchanged.
350 Apply echoing to the input audio.
352 Echoes are reflected sound and can occur naturally amongst mountains
353 (and sometimes large buildings) when talking or shouting; digital echo
354 effects emulate this behaviour and are often used to help fill out the
355 sound of a single instrument or vocal. The time difference between the
356 original signal and the reflection is the @code{delay}, and the
357 loudness of the reflected signal is the @code{decay}.
358 Multiple echoes can have different delays and decays.
360 A description of the accepted parameters follows.
364 Set input gain of reflected signal. Default is @code{0.6}.
367 Set output gain of reflected signal. Default is @code{0.3}.
370 Set list of time intervals in milliseconds between original signal and reflections
371 separated by '|'. Allowed range for each @code{delay} is @code{(0 - 90000.0]}.
372 Default is @code{1000}.
375 Set list of loudnesses of reflected signals separated by '|'.
376 Allowed range for each @code{decay} is @code{(0 - 1.0]}.
377 Default is @code{0.5}.
384 Make it sound as if there are twice as many instruments as are actually playing:
386 aecho=0.8:0.88:60:0.4
390 If delay is very short, then it sound like a (metallic) robot playing music:
396 A longer delay will sound like an open air concert in the mountains:
398 aecho=0.8:0.9:1000:0.3
402 Same as above but with one more mountain:
404 aecho=0.8:0.9:1000|1800:0.3|0.25
410 Modify an audio signal according to the specified expressions.
412 This filter accepts one or more expressions (one for each channel),
413 which are evaluated and used to modify a corresponding audio signal.
415 It accepts the following parameters:
419 Set the '|'-separated expressions list for each separate channel. If
420 the number of input channels is greater than the number of
421 expressions, the last specified expression is used for the remaining
424 @item channel_layout, c
425 Set output channel layout. If not specified, the channel layout is
426 specified by the number of expressions. If set to @samp{same}, it will
427 use by default the same input channel layout.
430 Each expression in @var{exprs} can contain the following constants and functions:
434 channel number of the current expression
437 number of the evaluated sample, starting from 0
443 time of the evaluated sample expressed in seconds
446 @item nb_out_channels
447 input and output number of channels
450 the value of input channel with number @var{CH}
453 Note: this filter is slow. For faster processing you should use a
462 aeval=val(ch)/2:c=same
466 Invert phase of the second channel:
474 Apply fade-in/out effect to input audio.
476 A description of the accepted parameters follows.
480 Specify the effect type, can be either @code{in} for fade-in, or
481 @code{out} for a fade-out effect. Default is @code{in}.
483 @item start_sample, ss
484 Specify the number of the start sample for starting to apply the fade
485 effect. Default is 0.
488 Specify the number of samples for which the fade effect has to last. At
489 the end of the fade-in effect the output audio will have the same
490 volume as the input audio, at the end of the fade-out transition
491 the output audio will be silence. Default is 44100.
494 Specify the start time of the fade effect. Default is 0.
495 The value must be specified as a time duration; see
496 @ref{time duration syntax,,the Time duration section in the ffmpeg-utils(1) manual,ffmpeg-utils}
497 for the accepted syntax.
498 If set this option is used instead of @var{start_sample}.
501 Specify the duration of the fade effect. See
502 @ref{time duration syntax,,the Time duration section in the ffmpeg-utils(1) manual,ffmpeg-utils}
503 for the accepted syntax.
504 At the end of the fade-in effect the output audio will have the same
505 volume as the input audio, at the end of the fade-out transition
506 the output audio will be silence.
507 By default the duration is determined by @var{nb_samples}.
508 If set this option is used instead of @var{nb_samples}.
511 Set curve for fade transition.
513 It accepts the following values:
516 select triangular, linear slope (default)
518 select quarter of sine wave
520 select half of sine wave
522 select exponential sine wave
526 select inverted parabola
542 Fade in first 15 seconds of audio:
548 Fade out last 25 seconds of a 900 seconds audio:
550 afade=t=out:st=875:d=25
557 Set output format constraints for the input audio. The framework will
558 negotiate the most appropriate format to minimize conversions.
560 It accepts the following parameters:
564 A '|'-separated list of requested sample formats.
567 A '|'-separated list of requested sample rates.
569 @item channel_layouts
570 A '|'-separated list of requested channel layouts.
572 See @ref{channel layout syntax,,the Channel Layout section in the ffmpeg-utils(1) manual,ffmpeg-utils}
573 for the required syntax.
576 If a parameter is omitted, all values are allowed.
578 Force the output to either unsigned 8-bit or signed 16-bit stereo
580 aformat=sample_fmts=u8|s16:channel_layouts=stereo
585 Apply a two-pole all-pass filter with central frequency (in Hz)
586 @var{frequency}, and filter-width @var{width}.
587 An all-pass filter changes the audio's frequency to phase relationship
588 without changing its frequency to amplitude relationship.
590 The filter accepts the following options:
597 Set method to specify band-width of filter.
610 Specify the band-width of a filter in width_type units.
615 Merge two or more audio streams into a single multi-channel stream.
617 The filter accepts the following options:
622 Set the number of inputs. Default is 2.
626 If the channel layouts of the inputs are disjoint, and therefore compatible,
627 the channel layout of the output will be set accordingly and the channels
628 will be reordered as necessary. If the channel layouts of the inputs are not
629 disjoint, the output will have all the channels of the first input then all
630 the channels of the second input, in that order, and the channel layout of
631 the output will be the default value corresponding to the total number of
634 For example, if the first input is in 2.1 (FL+FR+LF) and the second input
635 is FC+BL+BR, then the output will be in 5.1, with the channels in the
636 following order: a1, a2, b1, a3, b2, b3 (a1 is the first channel of the
637 first input, b1 is the first channel of the second input).
639 On the other hand, if both input are in stereo, the output channels will be
640 in the default order: a1, a2, b1, b2, and the channel layout will be
641 arbitrarily set to 4.0, which may or may not be the expected value.
643 All inputs must have the same sample rate, and format.
645 If inputs do not have the same duration, the output will stop with the
652 Merge two mono files into a stereo stream:
654 amovie=left.wav [l] ; amovie=right.mp3 [r] ; [l] [r] amerge
658 Multiple merges assuming 1 video stream and 6 audio streams in @file{input.mkv}:
660 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
666 Mixes multiple audio inputs into a single output.
668 Note that this filter only supports float samples (the @var{amerge}
669 and @var{pan} audio filters support many formats). If the @var{amix}
670 input has integer samples then @ref{aresample} will be automatically
671 inserted to perform the conversion to float samples.
675 ffmpeg -i INPUT1 -i INPUT2 -i INPUT3 -filter_complex amix=inputs=3:duration=first:dropout_transition=3 OUTPUT
677 will mix 3 input audio streams to a single output with the same duration as the
678 first input and a dropout transition time of 3 seconds.
680 It accepts the following parameters:
684 The number of inputs. If unspecified, it defaults to 2.
687 How to determine the end-of-stream.
691 The duration of the longest input. (default)
694 The duration of the shortest input.
697 The duration of the first input.
701 @item dropout_transition
702 The transition time, in seconds, for volume renormalization when an input
703 stream ends. The default value is 2 seconds.
709 Pass the audio source unchanged to the output.
713 Pad the end of an audio stream with silence.
715 This can be used together with @command{ffmpeg} @option{-shortest} to
716 extend audio streams to the same length as the video stream.
718 A description of the accepted options follows.
722 Set silence packet size. Default value is 4096.
725 Set the number of samples of silence to add to the end. After the
726 value is reached, the stream is terminated. This option is mutually
727 exclusive with @option{whole_len}.
730 Set the minimum total number of samples in the output audio stream. If
731 the value is longer than the input audio length, silence is added to
732 the end, until the value is reached. This option is mutually exclusive
733 with @option{pad_len}.
736 If neither the @option{pad_len} nor the @option{whole_len} option is
737 set, the filter will add silence to the end of the input stream
744 Add 1024 samples of silence to the end of the input:
750 Make sure the audio output will contain at least 10000 samples, pad
751 the input with silence if required:
757 Use @command{ffmpeg} to pad the audio input with silence, so that the
758 video stream will always result the shortest and will be converted
759 until the end in the output file when using the @option{shortest}
762 ffmpeg -i VIDEO -i AUDIO -filter_complex "[1:0]apad" -shortest OUTPUT
767 Add a phasing effect to the input audio.
769 A phaser filter creates series of peaks and troughs in the frequency spectrum.
770 The position of the peaks and troughs are modulated so that they vary over time, creating a sweeping effect.
772 A description of the accepted parameters follows.
776 Set input gain. Default is 0.4.
779 Set output gain. Default is 0.74
782 Set delay in milliseconds. Default is 3.0.
785 Set decay. Default is 0.4.
788 Set modulation speed in Hz. Default is 0.5.
791 Set modulation type. Default is triangular.
793 It accepts the following values:
803 Resample the input audio to the specified parameters, using the
804 libswresample library. If none are specified then the filter will
805 automatically convert between its input and output.
807 This filter is also able to stretch/squeeze the audio data to make it match
808 the timestamps or to inject silence / cut out audio to make it match the
809 timestamps, do a combination of both or do neither.
811 The filter accepts the syntax
812 [@var{sample_rate}:]@var{resampler_options}, where @var{sample_rate}
813 expresses a sample rate and @var{resampler_options} is a list of
814 @var{key}=@var{value} pairs, separated by ":". See the
815 ffmpeg-resampler manual for the complete list of supported options.
821 Resample the input audio to 44100Hz:
827 Stretch/squeeze samples to the given timestamps, with a maximum of 1000
828 samples per second compensation:
834 @section asetnsamples
836 Set the number of samples per each output audio frame.
838 The last output packet may contain a different number of samples, as
839 the filter will flush all the remaining samples when the input audio
842 The filter accepts the following options:
846 @item nb_out_samples, n
847 Set the number of frames per each output audio frame. The number is
848 intended as the number of samples @emph{per each channel}.
849 Default value is 1024.
852 If set to 1, the filter will pad the last audio frame with zeroes, so
853 that the last frame will contain the same number of samples as the
854 previous ones. Default value is 1.
857 For example, to set the number of per-frame samples to 1234 and
858 disable padding for the last frame, use:
860 asetnsamples=n=1234:p=0
865 Set the sample rate without altering the PCM data.
866 This will result in a change of speed and pitch.
868 The filter accepts the following options:
872 Set the output sample rate. Default is 44100 Hz.
877 Show a line containing various information for each input audio frame.
878 The input audio is not modified.
880 The shown line contains a sequence of key/value pairs of the form
881 @var{key}:@var{value}.
883 The following values are shown in the output:
887 The (sequential) number of the input frame, starting from 0.
890 The presentation timestamp of the input frame, in time base units; the time base
891 depends on the filter input pad, and is usually 1/@var{sample_rate}.
894 The presentation timestamp of the input frame in seconds.
897 position of the frame in the input stream, -1 if this information in
898 unavailable and/or meaningless (for example in case of synthetic audio)
907 The sample rate for the audio frame.
910 The number of samples (per channel) in the frame.
913 The Adler-32 checksum (printed in hexadecimal) of the audio data. For planar
914 audio, the data is treated as if all the planes were concatenated.
916 @item plane_checksums
917 A list of Adler-32 checksums for each data plane.
923 Display time domain statistical information about the audio channels.
924 Statistics are calculated and displayed for each audio channel and,
925 where applicable, an overall figure is also given.
927 It accepts the following option:
930 Short window length in seconds, used for peak and trough RMS measurement.
931 Default is @code{0.05} (50 milliseconds). Allowed range is @code{[0.1 - 10]}.
934 A description of each shown parameter follows:
938 Mean amplitude displacement from zero.
941 Minimal sample level.
944 Maximal sample level.
948 Standard peak and RMS level measured in dBFS.
952 Peak and trough values for RMS level measured over a short window.
955 Standard ratio of peak to RMS level (note: not in dB).
958 Flatness (i.e. consecutive samples with the same value) of the signal at its peak levels
959 (i.e. either @var{Min level} or @var{Max level}).
962 Number of occasions (not the number of samples) that the signal attained either
963 @var{Min level} or @var{Max level}.
968 Forward two audio streams and control the order the buffers are forwarded.
970 The filter accepts the following options:
974 Set the expression deciding which stream should be
975 forwarded next: if the result is negative, the first stream is forwarded; if
976 the result is positive or zero, the second stream is forwarded. It can use
977 the following variables:
981 number of buffers forwarded so far on each stream
983 number of samples forwarded so far on each stream
985 current timestamp of each stream
988 The default value is @code{t1-t2}, which means to always forward the stream
989 that has a smaller timestamp.
994 Stress-test @code{amerge} by randomly sending buffers on the wrong
995 input, while avoiding too much of a desynchronization:
997 amovie=file.ogg [a] ; amovie=file.mp3 [b] ;
998 [a] [b] astreamsync=(2*random(1))-1+tanh(5*(t1-t2)) [a2] [b2] ;
1004 Synchronize audio data with timestamps by squeezing/stretching it and/or
1005 dropping samples/adding silence when needed.
1007 This filter is not built by default, please use @ref{aresample} to do squeezing/stretching.
1009 It accepts the following parameters:
1013 Enable stretching/squeezing the data to make it match the timestamps. Disabled
1014 by default. When disabled, time gaps are covered with silence.
1017 The minimum difference between timestamps and audio data (in seconds) to trigger
1018 adding/dropping samples. The default value is 0.1. If you get an imperfect
1019 sync with this filter, try setting this parameter to 0.
1022 The maximum compensation in samples per second. Only relevant with compensate=1.
1023 The default value is 500.
1026 Assume that the first PTS should be this value. The time base is 1 / sample
1027 rate. This allows for padding/trimming at the start of the stream. By default,
1028 no assumption is made about the first frame's expected PTS, so no padding or
1029 trimming is done. For example, this could be set to 0 to pad the beginning with
1030 silence if an audio stream starts after the video stream or to trim any samples
1031 with a negative PTS due to encoder delay.
1039 The filter accepts exactly one parameter, the audio tempo. If not
1040 specified then the filter will assume nominal 1.0 tempo. Tempo must
1041 be in the [0.5, 2.0] range.
1043 @subsection Examples
1047 Slow down audio to 80% tempo:
1053 To speed up audio to 125% tempo:
1061 Trim the input so that the output contains one continuous subpart of the input.
1063 It accepts the following parameters:
1066 Timestamp (in seconds) of the start of the section to keep. I.e. the audio
1067 sample with the timestamp @var{start} will be the first sample in the output.
1070 Specify time of the first audio sample that will be dropped, i.e. the
1071 audio sample immediately preceding the one with the timestamp @var{end} will be
1072 the last sample in the output.
1075 Same as @var{start}, except this option sets the start timestamp in samples
1079 Same as @var{end}, except this option sets the end timestamp in samples instead
1083 The maximum duration of the output in seconds.
1086 The number of the first sample that should be output.
1089 The number of the first sample that should be dropped.
1092 @option{start}, @option{end}, and @option{duration} are expressed as time
1093 duration specifications; see
1094 @ref{time duration syntax,,the Time duration section in the ffmpeg-utils(1) manual,ffmpeg-utils}.
1096 Note that the first two sets of the start/end options and the @option{duration}
1097 option look at the frame timestamp, while the _sample options simply count the
1098 samples that pass through the filter. So start/end_pts and start/end_sample will
1099 give different results when the timestamps are wrong, inexact or do not start at
1100 zero. Also note that this filter does not modify the timestamps. If you wish
1101 to have the output timestamps start at zero, insert the asetpts filter after the
1104 If multiple start or end options are set, this filter tries to be greedy and
1105 keep all samples that match at least one of the specified constraints. To keep
1106 only the part that matches all the constraints at once, chain multiple atrim
1109 The defaults are such that all the input is kept. So it is possible to set e.g.
1110 just the end values to keep everything before the specified time.
1115 Drop everything except the second minute of input:
1117 ffmpeg -i INPUT -af atrim=60:120
1121 Keep only the first 1000 samples:
1123 ffmpeg -i INPUT -af atrim=end_sample=1000
1130 Apply a two-pole Butterworth band-pass filter with central
1131 frequency @var{frequency}, and (3dB-point) band-width width.
1132 The @var{csg} option selects a constant skirt gain (peak gain = Q)
1133 instead of the default: constant 0dB peak gain.
1134 The filter roll off at 6dB per octave (20dB per decade).
1136 The filter accepts the following options:
1140 Set the filter's central frequency. Default is @code{3000}.
1143 Constant skirt gain if set to 1. Defaults to 0.
1146 Set method to specify band-width of filter.
1159 Specify the band-width of a filter in width_type units.
1164 Apply a two-pole Butterworth band-reject filter with central
1165 frequency @var{frequency}, and (3dB-point) band-width @var{width}.
1166 The filter roll off at 6dB per octave (20dB per decade).
1168 The filter accepts the following options:
1172 Set the filter's central frequency. Default is @code{3000}.
1175 Set method to specify band-width of filter.
1188 Specify the band-width of a filter in width_type units.
1193 Boost or cut the bass (lower) frequencies of the audio using a two-pole
1194 shelving filter with a response similar to that of a standard
1195 hi-fi's tone-controls. This is also known as shelving equalisation (EQ).
1197 The filter accepts the following options:
1201 Give the gain at 0 Hz. Its useful range is about -20
1202 (for a large cut) to +20 (for a large boost).
1203 Beware of clipping when using a positive gain.
1206 Set the filter's central frequency and so can be used
1207 to extend or reduce the frequency range to be boosted or cut.
1208 The default value is @code{100} Hz.
1211 Set method to specify band-width of filter.
1224 Determine how steep is the filter's shelf transition.
1229 Apply a biquad IIR filter with the given coefficients.
1230 Where @var{b0}, @var{b1}, @var{b2} and @var{a0}, @var{a1}, @var{a2}
1231 are the numerator and denominator coefficients respectively.
1234 Bauer stereo to binaural transformation, which improves headphone listening of
1235 stereo audio records.
1237 It accepts the following parameters:
1241 Pre-defined crossfeed level.
1245 Default level (fcut=700, feed=50).
1248 Chu Moy circuit (fcut=700, feed=60).
1251 Jan Meier circuit (fcut=650, feed=95).
1256 Cut frequency (in Hz).
1265 Remap input channels to new locations.
1267 It accepts the following parameters:
1269 @item channel_layout
1270 The channel layout of the output stream.
1273 Map channels from input to output. The argument is a '|'-separated list of
1274 mappings, each in the @code{@var{in_channel}-@var{out_channel}} or
1275 @var{in_channel} form. @var{in_channel} can be either the name of the input
1276 channel (e.g. FL for front left) or its index in the input channel layout.
1277 @var{out_channel} is the name of the output channel or its index in the output
1278 channel layout. If @var{out_channel} is not given then it is implicitly an
1279 index, starting with zero and increasing by one for each mapping.
1282 If no mapping is present, the filter will implicitly map input channels to
1283 output channels, preserving indices.
1285 For example, assuming a 5.1+downmix input MOV file,
1287 ffmpeg -i in.mov -filter 'channelmap=map=DL-FL|DR-FR' out.wav
1289 will create an output WAV file tagged as stereo from the downmix channels of
1292 To fix a 5.1 WAV improperly encoded in AAC's native channel order
1294 ffmpeg -i in.wav -filter 'channelmap=1|2|0|5|3|4:channel_layout=5.1' out.wav
1297 @section channelsplit
1299 Split each channel from an input audio stream into a separate output stream.
1301 It accepts the following parameters:
1303 @item channel_layout
1304 The channel layout of the input stream. The default is "stereo".
1307 For example, assuming a stereo input MP3 file,
1309 ffmpeg -i in.mp3 -filter_complex channelsplit out.mkv
1311 will create an output Matroska file with two audio streams, one containing only
1312 the left channel and the other the right channel.
1314 Split a 5.1 WAV file into per-channel files:
1316 ffmpeg -i in.wav -filter_complex
1317 'channelsplit=channel_layout=5.1[FL][FR][FC][LFE][SL][SR]'
1318 -map '[FL]' front_left.wav -map '[FR]' front_right.wav -map '[FC]'
1319 front_center.wav -map '[LFE]' lfe.wav -map '[SL]' side_left.wav -map '[SR]'
1324 Add a chorus effect to the audio.
1326 Can make a single vocal sound like a chorus, but can also be applied to instrumentation.
1328 Chorus resembles an echo effect with a short delay, but whereas with echo the delay is
1329 constant, with chorus, it is varied using using sinusoidal or triangular modulation.
1330 The modulation depth defines the range the modulated delay is played before or after
1331 the delay. Hence the delayed sound will sound slower or faster, that is the delayed
1332 sound tuned around the original one, like in a chorus where some vocals are slightly
1335 It accepts the following parameters:
1338 Set input gain. Default is 0.4.
1341 Set output gain. Default is 0.4.
1344 Set delays. A typical delay is around 40ms to 60ms.
1356 @subsection Examples
1362 chorus=0.7:0.9:55:0.4:0.25:2
1368 chorus=0.6:0.9:50|60:0.4|0.32:0.25|0.4:2|1.3
1372 Fuller sounding chorus with three delays:
1374 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
1379 Compress or expand the audio's dynamic range.
1381 It accepts the following parameters:
1387 A list of times in seconds for each channel over which the instantaneous level
1388 of the input signal is averaged to determine its volume. @var{attacks} refers to
1389 increase of volume and @var{decays} refers to decrease of volume. For most
1390 situations, the attack time (response to the audio getting louder) should be
1391 shorter than the decay time, because the human ear is more sensitive to sudden
1392 loud audio than sudden soft audio. A typical value for attack is 0.3 seconds and
1393 a typical value for decay is 0.8 seconds.
1396 A list of points for the transfer function, specified in dB relative to the
1397 maximum possible signal amplitude. Each key points list must be defined using
1398 the following syntax: @code{x0/y0|x1/y1|x2/y2|....} or
1399 @code{x0/y0 x1/y1 x2/y2 ....}
1401 The input values must be in strictly increasing order but the transfer function
1402 does not have to be monotonically rising. The point @code{0/0} is assumed but
1403 may be overridden (by @code{0/out-dBn}). Typical values for the transfer
1404 function are @code{-70/-70|-60/-20}.
1407 Set the curve radius in dB for all joints. It defaults to 0.01.
1410 Set the additional gain in dB to be applied at all points on the transfer
1411 function. This allows for easy adjustment of the overall gain.
1415 Set an initial volume, in dB, to be assumed for each channel when filtering
1416 starts. This permits the user to supply a nominal level initially, so that, for
1417 example, a very large gain is not applied to initial signal levels before the
1418 companding has begun to operate. A typical value for audio which is initially
1419 quiet is -90 dB. It defaults to 0.
1422 Set a delay, in seconds. The input audio is analyzed immediately, but audio is
1423 delayed before being fed to the volume adjuster. Specifying a delay
1424 approximately equal to the attack/decay times allows the filter to effectively
1425 operate in predictive rather than reactive mode. It defaults to 0.
1429 @subsection Examples
1433 Make music with both quiet and loud passages suitable for listening to in a
1436 compand=.3|.3:1|1:-90/-60|-60/-40|-40/-30|-20/-20:6:0:-90:0.2
1440 A noise gate for when the noise is at a lower level than the signal:
1442 compand=.1|.1:.2|.2:-900/-900|-50.1/-900|-50/-50:.01:0:-90:.1
1446 Here is another noise gate, this time for when the noise is at a higher level
1447 than the signal (making it, in some ways, similar to squelch):
1449 compand=.1|.1:.1|.1:-45.1/-45.1|-45/-900|0/-900:.01:45:-90:.1
1454 Apply a DC shift to the audio.
1456 This can be useful to remove a DC offset (caused perhaps by a hardware problem
1457 in the recording chain) from the audio. The effect of a DC offset is reduced
1458 headroom and hence volume. The @ref{astats} filter can be used to determine if
1459 a signal has a DC offset.
1463 Set the DC shift, allowed range is [-1, 1]. It indicates the amount to shift
1467 Optional. It should have a value much less than 1 (e.g. 0.05 or 0.02) and is
1468 used to prevent clipping.
1473 Make audio easier to listen to on headphones.
1475 This filter adds `cues' to 44.1kHz stereo (i.e. audio CD format) audio
1476 so that when listened to on headphones the stereo image is moved from
1477 inside your head (standard for headphones) to outside and in front of
1478 the listener (standard for speakers).
1484 Apply a two-pole peaking equalisation (EQ) filter. With this
1485 filter, the signal-level at and around a selected frequency can
1486 be increased or decreased, whilst (unlike bandpass and bandreject
1487 filters) that at all other frequencies is unchanged.
1489 In order to produce complex equalisation curves, this filter can
1490 be given several times, each with a different central frequency.
1492 The filter accepts the following options:
1496 Set the filter's central frequency in Hz.
1499 Set method to specify band-width of filter.
1512 Specify the band-width of a filter in width_type units.
1515 Set the required gain or attenuation in dB.
1516 Beware of clipping when using a positive gain.
1519 @subsection Examples
1522 Attenuate 10 dB at 1000 Hz, with a bandwidth of 200 Hz:
1524 equalizer=f=1000:width_type=h:width=200:g=-10
1528 Apply 2 dB gain at 1000 Hz with Q 1 and attenuate 5 dB at 100 Hz with Q 2:
1530 equalizer=f=1000:width_type=q:width=1:g=2,equalizer=f=100:width_type=q:width=2:g=-5
1535 Apply a flanging effect to the audio.
1537 The filter accepts the following options:
1541 Set base delay in milliseconds. Range from 0 to 30. Default value is 0.
1544 Set added swep delay in milliseconds. Range from 0 to 10. Default value is 2.
1547 Set percentage regeneration (delayed signal feedback). Range from -95 to 95.
1551 Set percentage of delayed signal mixed with original. Range from 0 to 100.
1552 Default value is 71.
1555 Set sweeps per second (Hz). Range from 0.1 to 10. Default value is 0.5.
1558 Set swept wave shape, can be @var{triangular} or @var{sinusoidal}.
1559 Default value is @var{sinusoidal}.
1562 Set swept wave percentage-shift for multi channel. Range from 0 to 100.
1563 Default value is 25.
1566 Set delay-line interpolation, @var{linear} or @var{quadratic}.
1567 Default is @var{linear}.
1572 Apply a high-pass filter with 3dB point frequency.
1573 The filter can be either single-pole, or double-pole (the default).
1574 The filter roll off at 6dB per pole per octave (20dB per pole per decade).
1576 The filter accepts the following options:
1580 Set frequency in Hz. Default is 3000.
1583 Set number of poles. Default is 2.
1586 Set method to specify band-width of filter.
1599 Specify the band-width of a filter in width_type units.
1600 Applies only to double-pole filter.
1601 The default is 0.707q and gives a Butterworth response.
1606 Join multiple input streams into one multi-channel stream.
1608 It accepts the following parameters:
1612 The number of input streams. It defaults to 2.
1614 @item channel_layout
1615 The desired output channel layout. It defaults to stereo.
1618 Map channels from inputs to output. The argument is a '|'-separated list of
1619 mappings, each in the @code{@var{input_idx}.@var{in_channel}-@var{out_channel}}
1620 form. @var{input_idx} is the 0-based index of the input stream. @var{in_channel}
1621 can be either the name of the input channel (e.g. FL for front left) or its
1622 index in the specified input stream. @var{out_channel} is the name of the output
1626 The filter will attempt to guess the mappings when they are not specified
1627 explicitly. It does so by first trying to find an unused matching input channel
1628 and if that fails it picks the first unused input channel.
1630 Join 3 inputs (with properly set channel layouts):
1632 ffmpeg -i INPUT1 -i INPUT2 -i INPUT3 -filter_complex join=inputs=3 OUTPUT
1635 Build a 5.1 output from 6 single-channel streams:
1637 ffmpeg -i fl -i fr -i fc -i sl -i sr -i lfe -filter_complex
1638 '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'
1644 Load a LADSPA (Linux Audio Developer's Simple Plugin API) plugin.
1646 To enable compilation of this filter you need to configure FFmpeg with
1647 @code{--enable-ladspa}.
1651 Specifies the name of LADSPA plugin library to load. If the environment
1652 variable @env{LADSPA_PATH} is defined, the LADSPA plugin is searched in
1653 each one of the directories specified by the colon separated list in
1654 @env{LADSPA_PATH}, otherwise in the standard LADSPA paths, which are in
1655 this order: @file{HOME/.ladspa/lib/}, @file{/usr/local/lib/ladspa/},
1656 @file{/usr/lib/ladspa/}.
1659 Specifies the plugin within the library. Some libraries contain only
1660 one plugin, but others contain many of them. If this is not set filter
1661 will list all available plugins within the specified library.
1664 Set the '|' separated list of controls which are zero or more floating point
1665 values that determine the behavior of the loaded plugin (for example delay,
1667 Controls need to be defined using the following syntax:
1668 c0=@var{value0}|c1=@var{value1}|c2=@var{value2}|..., where
1669 @var{valuei} is the value set on the @var{i}-th control.
1670 If @option{controls} is set to @code{help}, all available controls and
1671 their valid ranges are printed.
1673 @item sample_rate, s
1674 Specify the sample rate, default to 44100. Only used if plugin have
1678 Set the number of samples per channel per each output frame, default
1679 is 1024. Only used if plugin have zero inputs.
1682 Set the minimum duration of the sourced audio. See
1683 @ref{time duration syntax,,the Time duration section in the ffmpeg-utils(1) manual,ffmpeg-utils}
1684 for the accepted syntax.
1685 Note that the resulting duration may be greater than the specified duration,
1686 as the generated audio is always cut at the end of a complete frame.
1687 If not specified, or the expressed duration is negative, the audio is
1688 supposed to be generated forever.
1689 Only used if plugin have zero inputs.
1693 @subsection Examples
1697 List all available plugins within amp (LADSPA example plugin) library:
1703 List all available controls and their valid ranges for @code{vcf_notch}
1704 plugin from @code{VCF} library:
1706 ladspa=f=vcf:p=vcf_notch:c=help
1710 Simulate low quality audio equipment using @code{Computer Music Toolkit} (CMT)
1713 ladspa=file=cmt:plugin=lofi:controls=c0=22|c1=12|c2=12
1717 Add reverberation to the audio using TAP-plugins
1718 (Tom's Audio Processing plugins):
1720 ladspa=file=tap_reverb:tap_reverb
1724 Generate white noise, with 0.2 amplitude:
1726 ladspa=file=cmt:noise_source_white:c=c0=.2
1730 Generate 20 bpm clicks using plugin @code{C* Click - Metronome} from the
1731 @code{C* Audio Plugin Suite} (CAPS) library:
1733 ladspa=file=caps:Click:c=c1=20'
1737 Apply @code{C* Eq10X2 - Stereo 10-band equaliser} effect:
1739 ladspa=caps:Eq10X2:c=c0=-48|c9=-24|c3=12|c4=2
1743 @subsection Commands
1745 This filter supports the following commands:
1748 Modify the @var{N}-th control value.
1750 If the specified value is not valid, it is ignored and prior one is kept.
1755 Apply a low-pass filter with 3dB point frequency.
1756 The filter can be either single-pole or double-pole (the default).
1757 The filter roll off at 6dB per pole per octave (20dB per pole per decade).
1759 The filter accepts the following options:
1763 Set frequency in Hz. Default is 500.
1766 Set number of poles. Default is 2.
1769 Set method to specify band-width of filter.
1782 Specify the band-width of a filter in width_type units.
1783 Applies only to double-pole filter.
1784 The default is 0.707q and gives a Butterworth response.
1789 Mix channels with specific gain levels. The filter accepts the output
1790 channel layout followed by a set of channels definitions.
1792 This filter is also designed to efficiently remap the channels of an audio
1795 The filter accepts parameters of the form:
1796 "@var{l}|@var{outdef}|@var{outdef}|..."
1800 output channel layout or number of channels
1803 output channel specification, of the form:
1804 "@var{out_name}=[@var{gain}*]@var{in_name}[+[@var{gain}*]@var{in_name}...]"
1807 output channel to define, either a channel name (FL, FR, etc.) or a channel
1808 number (c0, c1, etc.)
1811 multiplicative coefficient for the channel, 1 leaving the volume unchanged
1814 input channel to use, see out_name for details; it is not possible to mix
1815 named and numbered input channels
1818 If the `=' in a channel specification is replaced by `<', then the gains for
1819 that specification will be renormalized so that the total is 1, thus
1820 avoiding clipping noise.
1822 @subsection Mixing examples
1824 For example, if you want to down-mix from stereo to mono, but with a bigger
1825 factor for the left channel:
1827 pan=1c|c0=0.9*c0+0.1*c1
1830 A customized down-mix to stereo that works automatically for 3-, 4-, 5- and
1831 7-channels surround:
1833 pan=stereo| FL < FL + 0.5*FC + 0.6*BL + 0.6*SL | FR < FR + 0.5*FC + 0.6*BR + 0.6*SR
1836 Note that @command{ffmpeg} integrates a default down-mix (and up-mix) system
1837 that should be preferred (see "-ac" option) unless you have very specific
1840 @subsection Remapping examples
1842 The channel remapping will be effective if, and only if:
1845 @item gain coefficients are zeroes or ones,
1846 @item only one input per channel output,
1849 If all these conditions are satisfied, the filter will notify the user ("Pure
1850 channel mapping detected"), and use an optimized and lossless method to do the
1853 For example, if you have a 5.1 source and want a stereo audio stream by
1854 dropping the extra channels:
1856 pan="stereo| c0=FL | c1=FR"
1859 Given the same source, you can also switch front left and front right channels
1860 and keep the input channel layout:
1862 pan="5.1| c0=c1 | c1=c0 | c2=c2 | c3=c3 | c4=c4 | c5=c5"
1865 If the input is a stereo audio stream, you can mute the front left channel (and
1866 still keep the stereo channel layout) with:
1871 Still with a stereo audio stream input, you can copy the right channel in both
1872 front left and right:
1874 pan="stereo| c0=FR | c1=FR"
1879 ReplayGain scanner filter. This filter takes an audio stream as an input and
1880 outputs it unchanged.
1881 At end of filtering it displays @code{track_gain} and @code{track_peak}.
1885 Convert the audio sample format, sample rate and channel layout. It is
1886 not meant to be used directly.
1888 @section silencedetect
1890 Detect silence in an audio stream.
1892 This filter logs a message when it detects that the input audio volume is less
1893 or equal to a noise tolerance value for a duration greater or equal to the
1894 minimum detected noise duration.
1896 The printed times and duration are expressed in seconds.
1898 The filter accepts the following options:
1902 Set silence duration until notification (default is 2 seconds).
1905 Set noise tolerance. Can be specified in dB (in case "dB" is appended to the
1906 specified value) or amplitude ratio. Default is -60dB, or 0.001.
1909 @subsection Examples
1913 Detect 5 seconds of silence with -50dB noise tolerance:
1915 silencedetect=n=-50dB:d=5
1919 Complete example with @command{ffmpeg} to detect silence with 0.0001 noise
1920 tolerance in @file{silence.mp3}:
1922 ffmpeg -i silence.mp3 -af silencedetect=noise=0.0001 -f null -
1926 @section silenceremove
1928 Remove silence from the beginning, middle or end of the audio.
1930 The filter accepts the following options:
1934 This value is used to indicate if audio should be trimmed at beginning of
1935 the audio. A value of zero indicates no silence should be trimmed from the
1936 beginning. When specifying a non-zero value, it trims audio up until it
1937 finds non-silence. Normally, when trimming silence from beginning of audio
1938 the @var{start_periods} will be @code{1} but it can be increased to higher
1939 values to trim all audio up to specific count of non-silence periods.
1940 Default value is @code{0}.
1942 @item start_duration
1943 Specify the amount of time that non-silence must be detected before it stops
1944 trimming audio. By increasing the duration, bursts of noises can be treated
1945 as silence and trimmed off. Default value is @code{0}.
1947 @item start_threshold
1948 This indicates what sample value should be treated as silence. For digital
1949 audio, a value of @code{0} may be fine but for audio recorded from analog,
1950 you may wish to increase the value to account for background noise.
1951 Can be specified in dB (in case "dB" is appended to the specified value)
1952 or amplitude ratio. Default value is @code{0}.
1955 Set the count for trimming silence from the end of audio.
1956 To remove silence from the middle of a file, specify a @var{stop_periods}
1957 that is negative. This value is then treated as a positive value and is
1958 used to indicate the effect should restart processing as specified by
1959 @var{start_periods}, making it suitable for removing periods of silence
1960 in the middle of the audio.
1961 Default value is @code{0}.
1964 Specify a duration of silence that must exist before audio is not copied any
1965 more. By specifying a higher duration, silence that is wanted can be left in
1967 Default value is @code{0}.
1969 @item stop_threshold
1970 This is the same as @option{start_threshold} but for trimming silence from
1972 Can be specified in dB (in case "dB" is appended to the specified value)
1973 or amplitude ratio. Default value is @code{0}.
1976 This indicate that @var{stop_duration} length of audio should be left intact
1977 at the beginning of each period of silence.
1978 For example, if you want to remove long pauses between words but do not want
1979 to remove the pauses completely. Default value is @code{0}.
1983 @subsection Examples
1987 The following example shows how this filter can be used to start a recording
1988 that does not contain the delay at the start which usually occurs between
1989 pressing the record button and the start of the performance:
1991 silenceremove=1:5:0.02
1997 Boost or cut treble (upper) frequencies of the audio using a two-pole
1998 shelving filter with a response similar to that of a standard
1999 hi-fi's tone-controls. This is also known as shelving equalisation (EQ).
2001 The filter accepts the following options:
2005 Give the gain at whichever is the lower of ~22 kHz and the
2006 Nyquist frequency. Its useful range is about -20 (for a large cut)
2007 to +20 (for a large boost). Beware of clipping when using a positive gain.
2010 Set the filter's central frequency and so can be used
2011 to extend or reduce the frequency range to be boosted or cut.
2012 The default value is @code{3000} Hz.
2015 Set method to specify band-width of filter.
2028 Determine how steep is the filter's shelf transition.
2033 Adjust the input audio volume.
2035 It accepts the following parameters:
2039 Set audio volume expression.
2041 Output values are clipped to the maximum value.
2043 The output audio volume is given by the relation:
2045 @var{output_volume} = @var{volume} * @var{input_volume}
2048 The default value for @var{volume} is "1.0".
2051 This parameter represents the mathematical precision.
2053 It determines which input sample formats will be allowed, which affects the
2054 precision of the volume scaling.
2058 8-bit fixed-point; this limits input sample format to U8, S16, and S32.
2060 32-bit floating-point; this limits input sample format to FLT. (default)
2062 64-bit floating-point; this limits input sample format to DBL.
2066 Choose the behaviour on encountering ReplayGain side data in input frames.
2070 Remove ReplayGain side data, ignoring its contents (the default).
2073 Ignore ReplayGain side data, but leave it in the frame.
2076 Prefer the track gain, if present.
2079 Prefer the album gain, if present.
2082 @item replaygain_preamp
2083 Pre-amplification gain in dB to apply to the selected replaygain gain.
2085 Default value for @var{replaygain_preamp} is 0.0.
2088 Set when the volume expression is evaluated.
2090 It accepts the following values:
2093 only evaluate expression once during the filter initialization, or
2094 when the @samp{volume} command is sent
2097 evaluate expression for each incoming frame
2100 Default value is @samp{once}.
2103 The volume expression can contain the following parameters.
2107 frame number (starting at zero)
2110 @item nb_consumed_samples
2111 number of samples consumed by the filter
2113 number of samples in the current frame
2115 original frame position in the file
2121 PTS at start of stream
2123 time at start of stream
2129 last set volume value
2132 Note that when @option{eval} is set to @samp{once} only the
2133 @var{sample_rate} and @var{tb} variables are available, all other
2134 variables will evaluate to NAN.
2136 @subsection Commands
2138 This filter supports the following commands:
2141 Modify the volume expression.
2142 The command accepts the same syntax of the corresponding option.
2144 If the specified expression is not valid, it is kept at its current
2146 @item replaygain_noclip
2147 Prevent clipping by limiting the gain applied.
2149 Default value for @var{replaygain_noclip} is 1.
2153 @subsection Examples
2157 Halve the input audio volume:
2161 volume=volume=-6.0206dB
2164 In all the above example the named key for @option{volume} can be
2165 omitted, for example like in:
2171 Increase input audio power by 6 decibels using fixed-point precision:
2173 volume=volume=6dB:precision=fixed
2177 Fade volume after time 10 with an annihilation period of 5 seconds:
2179 volume='if(lt(t,10),1,max(1-(t-10)/5,0))':eval=frame
2183 @section volumedetect
2185 Detect the volume of the input video.
2187 The filter has no parameters. The input is not modified. Statistics about
2188 the volume will be printed in the log when the input stream end is reached.
2190 In particular it will show the mean volume (root mean square), maximum
2191 volume (on a per-sample basis), and the beginning of a histogram of the
2192 registered volume values (from the maximum value to a cumulated 1/1000 of
2195 All volumes are in decibels relative to the maximum PCM value.
2197 @subsection Examples
2199 Here is an excerpt of the output:
2201 [Parsed_volumedetect_0 @ 0xa23120] mean_volume: -27 dB
2202 [Parsed_volumedetect_0 @ 0xa23120] max_volume: -4 dB
2203 [Parsed_volumedetect_0 @ 0xa23120] histogram_4db: 6
2204 [Parsed_volumedetect_0 @ 0xa23120] histogram_5db: 62
2205 [Parsed_volumedetect_0 @ 0xa23120] histogram_6db: 286
2206 [Parsed_volumedetect_0 @ 0xa23120] histogram_7db: 1042
2207 [Parsed_volumedetect_0 @ 0xa23120] histogram_8db: 2551
2208 [Parsed_volumedetect_0 @ 0xa23120] histogram_9db: 4609
2209 [Parsed_volumedetect_0 @ 0xa23120] histogram_10db: 8409
2215 The mean square energy is approximately -27 dB, or 10^-2.7.
2217 The largest sample is at -4 dB, or more precisely between -4 dB and -5 dB.
2219 There are 6 samples at -4 dB, 62 at -5 dB, 286 at -6 dB, etc.
2222 In other words, raising the volume by +4 dB does not cause any clipping,
2223 raising it by +5 dB causes clipping for 6 samples, etc.
2225 @c man end AUDIO FILTERS
2227 @chapter Audio Sources
2228 @c man begin AUDIO SOURCES
2230 Below is a description of the currently available audio sources.
2234 Buffer audio frames, and make them available to the filter chain.
2236 This source is mainly intended for a programmatic use, in particular
2237 through the interface defined in @file{libavfilter/asrc_abuffer.h}.
2239 It accepts the following parameters:
2243 The timebase which will be used for timestamps of submitted frames. It must be
2244 either a floating-point number or in @var{numerator}/@var{denominator} form.
2247 The sample rate of the incoming audio buffers.
2250 The sample format of the incoming audio buffers.
2251 Either a sample format name or its corresponding integer representation from
2252 the enum AVSampleFormat in @file{libavutil/samplefmt.h}
2254 @item channel_layout
2255 The channel layout of the incoming audio buffers.
2256 Either a channel layout name from channel_layout_map in
2257 @file{libavutil/channel_layout.c} or its corresponding integer representation
2258 from the AV_CH_LAYOUT_* macros in @file{libavutil/channel_layout.h}
2261 The number of channels of the incoming audio buffers.
2262 If both @var{channels} and @var{channel_layout} are specified, then they
2267 @subsection Examples
2270 abuffer=sample_rate=44100:sample_fmt=s16p:channel_layout=stereo
2273 will instruct the source to accept planar 16bit signed stereo at 44100Hz.
2274 Since the sample format with name "s16p" corresponds to the number
2275 6 and the "stereo" channel layout corresponds to the value 0x3, this is
2278 abuffer=sample_rate=44100:sample_fmt=6:channel_layout=0x3
2283 Generate an audio signal specified by an expression.
2285 This source accepts in input one or more expressions (one for each
2286 channel), which are evaluated and used to generate a corresponding
2289 This source accepts the following options:
2293 Set the '|'-separated expressions list for each separate channel. In case the
2294 @option{channel_layout} option is not specified, the selected channel layout
2295 depends on the number of provided expressions. Otherwise the last
2296 specified expression is applied to the remaining output channels.
2298 @item channel_layout, c
2299 Set the channel layout. The number of channels in the specified layout
2300 must be equal to the number of specified expressions.
2303 Set the minimum duration of the sourced audio. See
2304 @ref{time duration syntax,,the Time duration section in the ffmpeg-utils(1) manual,ffmpeg-utils}
2305 for the accepted syntax.
2306 Note that the resulting duration may be greater than the specified
2307 duration, as the generated audio is always cut at the end of a
2310 If not specified, or the expressed duration is negative, the audio is
2311 supposed to be generated forever.
2314 Set the number of samples per channel per each output frame,
2317 @item sample_rate, s
2318 Specify the sample rate, default to 44100.
2321 Each expression in @var{exprs} can contain the following constants:
2325 number of the evaluated sample, starting from 0
2328 time of the evaluated sample expressed in seconds, starting from 0
2335 @subsection Examples
2345 Generate a sin signal with frequency of 440 Hz, set sample rate to
2348 aevalsrc="sin(440*2*PI*t):s=8000"
2352 Generate a two channels signal, specify the channel layout (Front
2353 Center + Back Center) explicitly:
2355 aevalsrc="sin(420*2*PI*t)|cos(430*2*PI*t):c=FC|BC"
2359 Generate white noise:
2361 aevalsrc="-2+random(0)"
2365 Generate an amplitude modulated signal:
2367 aevalsrc="sin(10*2*PI*t)*sin(880*2*PI*t)"
2371 Generate 2.5 Hz binaural beats on a 360 Hz carrier:
2373 aevalsrc="0.1*sin(2*PI*(360-2.5/2)*t) | 0.1*sin(2*PI*(360+2.5/2)*t)"
2380 The null audio source, return unprocessed audio frames. It is mainly useful
2381 as a template and to be employed in analysis / debugging tools, or as
2382 the source for filters which ignore the input data (for example the sox
2385 This source accepts the following options:
2389 @item channel_layout, cl
2391 Specifies the channel layout, and can be either an integer or a string
2392 representing a channel layout. The default value of @var{channel_layout}
2395 Check the channel_layout_map definition in
2396 @file{libavutil/channel_layout.c} for the mapping between strings and
2397 channel layout values.
2399 @item sample_rate, r
2400 Specifies the sample rate, and defaults to 44100.
2403 Set the number of samples per requested frames.
2407 @subsection Examples
2411 Set the sample rate to 48000 Hz and the channel layout to AV_CH_LAYOUT_MONO.
2413 anullsrc=r=48000:cl=4
2417 Do the same operation with a more obvious syntax:
2419 anullsrc=r=48000:cl=mono
2423 All the parameters need to be explicitly defined.
2427 Synthesize a voice utterance using the libflite library.
2429 To enable compilation of this filter you need to configure FFmpeg with
2430 @code{--enable-libflite}.
2432 Note that the flite library is not thread-safe.
2434 The filter accepts the following options:
2439 If set to 1, list the names of the available voices and exit
2440 immediately. Default value is 0.
2443 Set the maximum number of samples per frame. Default value is 512.
2446 Set the filename containing the text to speak.
2449 Set the text to speak.
2452 Set the voice to use for the speech synthesis. Default value is
2453 @code{kal}. See also the @var{list_voices} option.
2456 @subsection Examples
2460 Read from file @file{speech.txt}, and synthesize the text using the
2461 standard flite voice:
2463 flite=textfile=speech.txt
2467 Read the specified text selecting the @code{slt} voice:
2469 flite=text='So fare thee well, poor devil of a Sub-Sub, whose commentator I am':voice=slt
2473 Input text to ffmpeg:
2475 ffmpeg -f lavfi -i flite=text='So fare thee well, poor devil of a Sub-Sub, whose commentator I am':voice=slt
2479 Make @file{ffplay} speak the specified text, using @code{flite} and
2480 the @code{lavfi} device:
2482 ffplay -f lavfi flite=text='No more be grieved for which that thou hast done.'
2486 For more information about libflite, check:
2487 @url{http://www.speech.cs.cmu.edu/flite/}
2491 Generate an audio signal made of a sine wave with amplitude 1/8.
2493 The audio signal is bit-exact.
2495 The filter accepts the following options:
2500 Set the carrier frequency. Default is 440 Hz.
2502 @item beep_factor, b
2503 Enable a periodic beep every second with frequency @var{beep_factor} times
2504 the carrier frequency. Default is 0, meaning the beep is disabled.
2506 @item sample_rate, r
2507 Specify the sample rate, default is 44100.
2510 Specify the duration of the generated audio stream.
2512 @item samples_per_frame
2513 Set the number of samples per output frame, default is 1024.
2516 @subsection Examples
2521 Generate a simple 440 Hz sine wave:
2527 Generate a 220 Hz sine wave with a 880 Hz beep each second, for 5 seconds:
2531 sine=frequency=220:beep_factor=4:duration=5
2536 @c man end AUDIO SOURCES
2538 @chapter Audio Sinks
2539 @c man begin AUDIO SINKS
2541 Below is a description of the currently available audio sinks.
2543 @section abuffersink
2545 Buffer audio frames, and make them available to the end of filter chain.
2547 This sink is mainly intended for programmatic use, in particular
2548 through the interface defined in @file{libavfilter/buffersink.h}
2549 or the options system.
2551 It accepts a pointer to an AVABufferSinkContext structure, which
2552 defines the incoming buffers' formats, to be passed as the opaque
2553 parameter to @code{avfilter_init_filter} for initialization.
2556 Null audio sink; do absolutely nothing with the input audio. It is
2557 mainly useful as a template and for use in analysis / debugging
2560 @c man end AUDIO SINKS
2562 @chapter Video Filters
2563 @c man begin VIDEO FILTERS
2565 When you configure your FFmpeg build, you can disable any of the
2566 existing filters using @code{--disable-filters}.
2567 The configure output will show the video filters included in your
2570 Below is a description of the currently available video filters.
2572 @section alphaextract
2574 Extract the alpha component from the input as a grayscale video. This
2575 is especially useful with the @var{alphamerge} filter.
2579 Add or replace the alpha component of the primary input with the
2580 grayscale value of a second input. This is intended for use with
2581 @var{alphaextract} to allow the transmission or storage of frame
2582 sequences that have alpha in a format that doesn't support an alpha
2585 For example, to reconstruct full frames from a normal YUV-encoded video
2586 and a separate video created with @var{alphaextract}, you might use:
2588 movie=in_alpha.mkv [alpha]; [in][alpha] alphamerge [out]
2591 Since this filter is designed for reconstruction, it operates on frame
2592 sequences without considering timestamps, and terminates when either
2593 input reaches end of stream. This will cause problems if your encoding
2594 pipeline drops frames. If you're trying to apply an image as an
2595 overlay to a video stream, consider the @var{overlay} filter instead.
2599 Same as the @ref{subtitles} filter, except that it doesn't require libavcodec
2600 and libavformat to work. On the other hand, it is limited to ASS (Advanced
2601 Substation Alpha) subtitles files.
2603 This filter accepts the following option in addition to the common options from
2604 the @ref{subtitles} filter:
2608 Set the shaping engine
2610 Available values are:
2613 The default libass shaping engine, which is the best available.
2615 Fast, font-agnostic shaper that can do only substitutions
2617 Slower shaper using OpenType for substitutions and positioning
2620 The default is @code{auto}.
2625 Compute the bounding box for the non-black pixels in the input frame
2628 This filter computes the bounding box containing all the pixels with a
2629 luminance value greater than the minimum allowed value.
2630 The parameters describing the bounding box are printed on the filter
2633 The filter accepts the following option:
2637 Set the minimal luminance value. Default is @code{16}.
2640 @section blackdetect
2642 Detect video intervals that are (almost) completely black. Can be
2643 useful to detect chapter transitions, commercials, or invalid
2644 recordings. Output lines contains the time for the start, end and
2645 duration of the detected black interval expressed in seconds.
2647 In order to display the output lines, you need to set the loglevel at
2648 least to the AV_LOG_INFO value.
2650 The filter accepts the following options:
2653 @item black_min_duration, d
2654 Set the minimum detected black duration expressed in seconds. It must
2655 be a non-negative floating point number.
2657 Default value is 2.0.
2659 @item picture_black_ratio_th, pic_th
2660 Set the threshold for considering a picture "black".
2661 Express the minimum value for the ratio:
2663 @var{nb_black_pixels} / @var{nb_pixels}
2666 for which a picture is considered black.
2667 Default value is 0.98.
2669 @item pixel_black_th, pix_th
2670 Set the threshold for considering a pixel "black".
2672 The threshold expresses the maximum pixel luminance value for which a
2673 pixel is considered "black". The provided value is scaled according to
2674 the following equation:
2676 @var{absolute_threshold} = @var{luminance_minimum_value} + @var{pixel_black_th} * @var{luminance_range_size}
2679 @var{luminance_range_size} and @var{luminance_minimum_value} depend on
2680 the input video format, the range is [0-255] for YUV full-range
2681 formats and [16-235] for YUV non full-range formats.
2683 Default value is 0.10.
2686 The following example sets the maximum pixel threshold to the minimum
2687 value, and detects only black intervals of 2 or more seconds:
2689 blackdetect=d=2:pix_th=0.00
2694 Detect frames that are (almost) completely black. Can be useful to
2695 detect chapter transitions or commercials. Output lines consist of
2696 the frame number of the detected frame, the percentage of blackness,
2697 the position in the file if known or -1 and the timestamp in seconds.
2699 In order to display the output lines, you need to set the loglevel at
2700 least to the AV_LOG_INFO value.
2702 It accepts the following parameters:
2707 The percentage of the pixels that have to be below the threshold; it defaults to
2710 @item threshold, thresh
2711 The threshold below which a pixel value is considered black; it defaults to
2716 @section blend, tblend
2718 Blend two video frames into each other.
2720 The @code{blend} filter takes two input streams and outputs one
2721 stream, the first input is the "top" layer and second input is
2722 "bottom" layer. Output terminates when shortest input terminates.
2724 The @code{tblend} (time blend) filter takes two consecutive frames
2725 from one single stream, and outputs the result obtained by blending
2726 the new frame on top of the old frame.
2728 A description of the accepted options follows.
2736 Set blend mode for specific pixel component or all pixel components in case
2737 of @var{all_mode}. Default value is @code{normal}.
2739 Available values for component modes are:
2773 Set blend opacity for specific pixel component or all pixel components in case
2774 of @var{all_opacity}. Only used in combination with pixel component blend modes.
2781 Set blend expression for specific pixel component or all pixel components in case
2782 of @var{all_expr}. Note that related mode options will be ignored if those are set.
2784 The expressions can use the following variables:
2788 The sequential number of the filtered frame, starting from @code{0}.
2792 the coordinates of the current sample
2796 the width and height of currently filtered plane
2800 Width and height scale depending on the currently filtered plane. It is the
2801 ratio between the corresponding luma plane number of pixels and the current
2802 plane ones. E.g. for YUV4:2:0 the values are @code{1,1} for the luma plane, and
2803 @code{0.5,0.5} for chroma planes.
2806 Time of the current frame, expressed in seconds.
2809 Value of pixel component at current location for first video frame (top layer).
2812 Value of pixel component at current location for second video frame (bottom layer).
2816 Force termination when the shortest input terminates. Default is
2817 @code{0}. This option is only defined for the @code{blend} filter.
2820 Continue applying the last bottom frame after the end of the stream. A value of
2821 @code{0} disable the filter after the last frame of the bottom layer is reached.
2822 Default is @code{1}. This option is only defined for the @code{blend} filter.
2825 @subsection Examples
2829 Apply transition from bottom layer to top layer in first 10 seconds:
2831 blend=all_expr='A*(if(gte(T,10),1,T/10))+B*(1-(if(gte(T,10),1,T/10)))'
2835 Apply 1x1 checkerboard effect:
2837 blend=all_expr='if(eq(mod(X,2),mod(Y,2)),A,B)'
2841 Apply uncover left effect:
2843 blend=all_expr='if(gte(N*SW+X,W),A,B)'
2847 Apply uncover down effect:
2849 blend=all_expr='if(gte(Y-N*SH,0),A,B)'
2853 Apply uncover up-left effect:
2855 blend=all_expr='if(gte(T*SH*40+Y,H)*gte((T*40*SW+X)*W/H,W),A,B)'
2859 Display differences between the current and the previous frame:
2861 tblend=all_mode=difference128
2867 Apply a boxblur algorithm to the input video.
2869 It accepts the following parameters:
2873 @item luma_radius, lr
2874 @item luma_power, lp
2875 @item chroma_radius, cr
2876 @item chroma_power, cp
2877 @item alpha_radius, ar
2878 @item alpha_power, ap
2882 A description of the accepted options follows.
2885 @item luma_radius, lr
2886 @item chroma_radius, cr
2887 @item alpha_radius, ar
2888 Set an expression for the box radius in pixels used for blurring the
2889 corresponding input plane.
2891 The radius value must be a non-negative number, and must not be
2892 greater than the value of the expression @code{min(w,h)/2} for the
2893 luma and alpha planes, and of @code{min(cw,ch)/2} for the chroma
2896 Default value for @option{luma_radius} is "2". If not specified,
2897 @option{chroma_radius} and @option{alpha_radius} default to the
2898 corresponding value set for @option{luma_radius}.
2900 The expressions can contain the following constants:
2904 The input width and height in pixels.
2908 The input chroma image width and height in pixels.
2912 The horizontal and vertical chroma subsample values. For example, for the
2913 pixel format "yuv422p", @var{hsub} is 2 and @var{vsub} is 1.
2916 @item luma_power, lp
2917 @item chroma_power, cp
2918 @item alpha_power, ap
2919 Specify how many times the boxblur filter is applied to the
2920 corresponding plane.
2922 Default value for @option{luma_power} is 2. If not specified,
2923 @option{chroma_power} and @option{alpha_power} default to the
2924 corresponding value set for @option{luma_power}.
2926 A value of 0 will disable the effect.
2929 @subsection Examples
2933 Apply a boxblur filter with the luma, chroma, and alpha radii
2936 boxblur=luma_radius=2:luma_power=1
2941 Set the luma radius to 2, and alpha and chroma radius to 0:
2943 boxblur=2:1:cr=0:ar=0
2947 Set the luma and chroma radii to a fraction of the video dimension:
2949 boxblur=luma_radius=min(h\,w)/10:luma_power=1:chroma_radius=min(cw\,ch)/10:chroma_power=1
2955 Visualize information exported by some codecs.
2957 Some codecs can export information through frames using side-data or other
2958 means. For example, some MPEG based codecs export motion vectors through the
2959 @var{export_mvs} flag in the codec @option{flags2} option.
2961 The filter accepts the following option:
2965 Set motion vectors to visualize.
2967 Available flags for @var{mv} are:
2971 forward predicted MVs of P-frames
2973 forward predicted MVs of B-frames
2975 backward predicted MVs of B-frames
2979 @subsection Examples
2983 Visualizes multi-directionals MVs from P and B-Frames using @command{ffplay}:
2985 ffplay -flags2 +export_mvs input.mpg -vf codecview=mv=pf+bf+bb
2989 @section colorbalance
2990 Modify intensity of primary colors (red, green and blue) of input frames.
2992 The filter allows an input frame to be adjusted in the shadows, midtones or highlights
2993 regions for the red-cyan, green-magenta or blue-yellow balance.
2995 A positive adjustment value shifts the balance towards the primary color, a negative
2996 value towards the complementary color.
2998 The filter accepts the following options:
3004 Adjust red, green and blue shadows (darkest pixels).
3009 Adjust red, green and blue midtones (medium pixels).
3014 Adjust red, green and blue highlights (brightest pixels).
3016 Allowed ranges for options are @code{[-1.0, 1.0]}. Defaults are @code{0}.
3019 @subsection Examples
3023 Add red color cast to shadows:
3029 @section colorlevels
3031 Adjust video input frames using levels.
3033 The filter accepts the following options:
3040 Adjust red, green, blue and alpha input black point.
3041 Allowed ranges for options are @code{[-1.0, 1.0]}. Defaults are @code{0}.
3047 Adjust red, green, blue and alpha input white point.
3048 Allowed ranges for options are @code{[-1.0, 1.0]}. Defaults are @code{1}.
3050 Input levels are used to lighten highlights (bright tones), darken shadows
3051 (dark tones), change the balance of bright and dark tones.
3057 Adjust red, green, blue and alpha output black point.
3058 Allowed ranges for options are @code{[0, 1.0]}. Defaults are @code{0}.
3064 Adjust red, green, blue and alpha output white point.
3065 Allowed ranges for options are @code{[0, 1.0]}. Defaults are @code{1}.
3067 Output levels allows manual selection of a constrained output level range.
3070 @subsection Examples
3074 Make video output darker:
3076 colorlevels=rimin=0.058:gimin=0.058:bimin=0.058
3082 colorlevels=rimin=0.039:gimin=0.039:bimin=0.039:rimax=0.96:gimax=0.96:bimax=0.96
3086 Make video output lighter:
3088 colorlevels=rimax=0.902:gimax=0.902:bimax=0.902
3092 Increase brightness:
3094 colorlevels=romin=0.5:gomin=0.5:bomin=0.5
3098 @section colorchannelmixer
3100 Adjust video input frames by re-mixing color channels.
3102 This filter modifies a color channel by adding the values associated to
3103 the other channels of the same pixels. For example if the value to
3104 modify is red, the output value will be:
3106 @var{red}=@var{red}*@var{rr} + @var{blue}*@var{rb} + @var{green}*@var{rg} + @var{alpha}*@var{ra}
3109 The filter accepts the following options:
3116 Adjust contribution of input red, green, blue and alpha channels for output red channel.
3117 Default is @code{1} for @var{rr}, and @code{0} for @var{rg}, @var{rb} and @var{ra}.
3123 Adjust contribution of input red, green, blue and alpha channels for output green channel.
3124 Default is @code{1} for @var{gg}, and @code{0} for @var{gr}, @var{gb} and @var{ga}.
3130 Adjust contribution of input red, green, blue and alpha channels for output blue channel.
3131 Default is @code{1} for @var{bb}, and @code{0} for @var{br}, @var{bg} and @var{ba}.
3137 Adjust contribution of input red, green, blue and alpha channels for output alpha channel.
3138 Default is @code{1} for @var{aa}, and @code{0} for @var{ar}, @var{ag} and @var{ab}.
3140 Allowed ranges for options are @code{[-2.0, 2.0]}.
3143 @subsection Examples
3147 Convert source to grayscale:
3149 colorchannelmixer=.3:.4:.3:0:.3:.4:.3:0:.3:.4:.3
3152 Simulate sepia tones:
3154 colorchannelmixer=.393:.769:.189:0:.349:.686:.168:0:.272:.534:.131
3158 @section colormatrix
3160 Convert color matrix.
3162 The filter accepts the following options:
3167 Specify the source and destination color matrix. Both values must be
3170 The accepted values are:
3186 For example to convert from BT.601 to SMPTE-240M, use the command:
3188 colormatrix=bt601:smpte240m
3193 Copy the input source unchanged to the output. This is mainly useful for
3198 Crop the input video to given dimensions.
3200 It accepts the following parameters:
3204 The width of the output video. It defaults to @code{iw}.
3205 This expression is evaluated only once during the filter
3209 The height of the output video. It defaults to @code{ih}.
3210 This expression is evaluated only once during the filter
3214 The horizontal position, in the input video, of the left edge of the output
3215 video. It defaults to @code{(in_w-out_w)/2}.
3216 This expression is evaluated per-frame.
3219 The vertical position, in the input video, of the top edge of the output video.
3220 It defaults to @code{(in_h-out_h)/2}.
3221 This expression is evaluated per-frame.
3224 If set to 1 will force the output display aspect ratio
3225 to be the same of the input, by changing the output sample aspect
3226 ratio. It defaults to 0.
3229 The @var{out_w}, @var{out_h}, @var{x}, @var{y} parameters are
3230 expressions containing the following constants:
3235 The computed values for @var{x} and @var{y}. They are evaluated for
3240 The input width and height.
3244 These are the same as @var{in_w} and @var{in_h}.
3248 The output (cropped) width and height.
3252 These are the same as @var{out_w} and @var{out_h}.
3255 same as @var{iw} / @var{ih}
3258 input sample aspect ratio
3261 input display aspect ratio, it is the same as (@var{iw} / @var{ih}) * @var{sar}
3265 horizontal and vertical chroma subsample values. For example for the
3266 pixel format "yuv422p" @var{hsub} is 2 and @var{vsub} is 1.
3269 The number of the input frame, starting from 0.
3272 the position in the file of the input frame, NAN if unknown
3275 The timestamp expressed in seconds. It's NAN if the input timestamp is unknown.
3279 The expression for @var{out_w} may depend on the value of @var{out_h},
3280 and the expression for @var{out_h} may depend on @var{out_w}, but they
3281 cannot depend on @var{x} and @var{y}, as @var{x} and @var{y} are
3282 evaluated after @var{out_w} and @var{out_h}.
3284 The @var{x} and @var{y} parameters specify the expressions for the
3285 position of the top-left corner of the output (non-cropped) area. They
3286 are evaluated for each frame. If the evaluated value is not valid, it
3287 is approximated to the nearest valid value.
3289 The expression for @var{x} may depend on @var{y}, and the expression
3290 for @var{y} may depend on @var{x}.
3292 @subsection Examples
3296 Crop area with size 100x100 at position (12,34).
3301 Using named options, the example above becomes:
3303 crop=w=100:h=100:x=12:y=34
3307 Crop the central input area with size 100x100:
3313 Crop the central input area with size 2/3 of the input video:
3315 crop=2/3*in_w:2/3*in_h
3319 Crop the input video central square:
3326 Delimit the rectangle with the top-left corner placed at position
3327 100:100 and the right-bottom corner corresponding to the right-bottom
3328 corner of the input image.
3330 crop=in_w-100:in_h-100:100:100
3334 Crop 10 pixels from the left and right borders, and 20 pixels from
3335 the top and bottom borders
3337 crop=in_w-2*10:in_h-2*20
3341 Keep only the bottom right quarter of the input image:
3343 crop=in_w/2:in_h/2:in_w/2:in_h/2
3347 Crop height for getting Greek harmony:
3349 crop=in_w:1/PHI*in_w
3353 Apply trembling effect:
3355 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)
3359 Apply erratic camera effect depending on timestamp:
3361 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)"
3365 Set x depending on the value of y:
3367 crop=in_w/2:in_h/2:y:10+10*sin(n/10)
3373 Auto-detect the crop size.
3375 It calculates the necessary cropping parameters and prints the
3376 recommended parameters via the logging system. The detected dimensions
3377 correspond to the non-black area of the input video.
3379 It accepts the following parameters:
3384 Set higher black value threshold, which can be optionally specified
3385 from nothing (0) to everything (255 for 8bit based formats). An intensity
3386 value greater to the set value is considered non-black. It defaults to 24.
3387 You can also specify a value between 0.0 and 1.0 which will be scaled depending
3388 on the bitdepth of the pixel format.
3391 The value which the width/height should be divisible by. It defaults to
3392 16. The offset is automatically adjusted to center the video. Use 2 to
3393 get only even dimensions (needed for 4:2:2 video). 16 is best when
3394 encoding to most video codecs.
3396 @item reset_count, reset
3397 Set the counter that determines after how many frames cropdetect will
3398 reset the previously detected largest video area and start over to
3399 detect the current optimal crop area. Default value is 0.
3401 This can be useful when channel logos distort the video area. 0
3402 indicates 'never reset', and returns the largest area encountered during
3409 Apply color adjustments using curves.
3411 This filter is similar to the Adobe Photoshop and GIMP curves tools. Each
3412 component (red, green and blue) has its values defined by @var{N} key points
3413 tied from each other using a smooth curve. The x-axis represents the pixel
3414 values from the input frame, and the y-axis the new pixel values to be set for
3417 By default, a component curve is defined by the two points @var{(0;0)} and
3418 @var{(1;1)}. This creates a straight line where each original pixel value is
3419 "adjusted" to its own value, which means no change to the image.
3421 The filter allows you to redefine these two points and add some more. A new
3422 curve (using a natural cubic spline interpolation) will be define to pass
3423 smoothly through all these new coordinates. The new defined points needs to be
3424 strictly increasing over the x-axis, and their @var{x} and @var{y} values must
3425 be in the @var{[0;1]} interval. If the computed curves happened to go outside
3426 the vector spaces, the values will be clipped accordingly.
3428 If there is no key point defined in @code{x=0}, the filter will automatically
3429 insert a @var{(0;0)} point. In the same way, if there is no key point defined
3430 in @code{x=1}, the filter will automatically insert a @var{(1;1)} point.
3432 The filter accepts the following options:
3436 Select one of the available color presets. This option can be used in addition
3437 to the @option{r}, @option{g}, @option{b} parameters; in this case, the later
3438 options takes priority on the preset values.
3439 Available presets are:
3442 @item color_negative
3445 @item increase_contrast
3447 @item linear_contrast
3448 @item medium_contrast
3450 @item strong_contrast
3453 Default is @code{none}.
3455 Set the master key points. These points will define a second pass mapping. It
3456 is sometimes called a "luminance" or "value" mapping. It can be used with
3457 @option{r}, @option{g}, @option{b} or @option{all} since it acts like a
3458 post-processing LUT.
3460 Set the key points for the red component.
3462 Set the key points for the green component.
3464 Set the key points for the blue component.
3466 Set the key points for all components (not including master).
3467 Can be used in addition to the other key points component
3468 options. In this case, the unset component(s) will fallback on this
3469 @option{all} setting.
3471 Specify a Photoshop curves file (@code{.asv}) to import the settings from.
3474 To avoid some filtergraph syntax conflicts, each key points list need to be
3475 defined using the following syntax: @code{x0/y0 x1/y1 x2/y2 ...}.
3477 @subsection Examples
3481 Increase slightly the middle level of blue:
3483 curves=blue='0.5/0.58'
3489 curves=r='0/0.11 .42/.51 1/0.95':g='0.50/0.48':b='0/0.22 .49/.44 1/0.8'
3491 Here we obtain the following coordinates for each components:
3494 @code{(0;0.11) (0.42;0.51) (1;0.95)}
3496 @code{(0;0) (0.50;0.48) (1;1)}
3498 @code{(0;0.22) (0.49;0.44) (1;0.80)}
3502 The previous example can also be achieved with the associated built-in preset:
3504 curves=preset=vintage
3514 Use a Photoshop preset and redefine the points of the green component:
3516 curves=psfile='MyCurvesPresets/purple.asv':green='0.45/0.53'
3522 Denoise frames using 2D DCT (frequency domain filtering).
3524 This filter is not designed for real time.
3526 The filter accepts the following options:
3530 Set the noise sigma constant.
3532 This @var{sigma} defines a hard threshold of @code{3 * sigma}; every DCT
3533 coefficient (absolute value) below this threshold with be dropped.
3535 If you need a more advanced filtering, see @option{expr}.
3537 Default is @code{0}.
3540 Set number overlapping pixels for each block. Since the filter can be slow, you
3541 may want to reduce this value, at the cost of a less effective filter and the
3542 risk of various artefacts.
3544 If the overlapping value doesn't permit processing the whole input width or
3545 height, a warning will be displayed and according borders won't be denoised.
3547 Default value is @var{blocksize}-1, which is the best possible setting.
3550 Set the coefficient factor expression.
3552 For each coefficient of a DCT block, this expression will be evaluated as a
3553 multiplier value for the coefficient.
3555 If this is option is set, the @option{sigma} option will be ignored.
3557 The absolute value of the coefficient can be accessed through the @var{c}
3561 Set the @var{blocksize} using the number of bits. @code{1<<@var{n}} defines the
3562 @var{blocksize}, which is the width and height of the processed blocks.
3564 The default value is @var{3} (8x8) and can be raised to @var{4} for a
3565 @var{blocksize} of 16x16. Note that changing this setting has huge consequences
3566 on the speed processing. Also, a larger block size does not necessarily means a
3570 @subsection Examples
3572 Apply a denoise with a @option{sigma} of @code{4.5}:
3577 The same operation can be achieved using the expression system:
3579 dctdnoiz=e='gte(c, 4.5*3)'
3582 Violent denoise using a block size of @code{16x16}:
3590 Drop duplicated frames at regular intervals.
3592 The filter accepts the following options:
3596 Set the number of frames from which one will be dropped. Setting this to
3597 @var{N} means one frame in every batch of @var{N} frames will be dropped.
3598 Default is @code{5}.
3601 Set the threshold for duplicate detection. If the difference metric for a frame
3602 is less than or equal to this value, then it is declared as duplicate. Default
3606 Set scene change threshold. Default is @code{15}.
3610 Set the size of the x and y-axis blocks used during metric calculations.
3611 Larger blocks give better noise suppression, but also give worse detection of
3612 small movements. Must be a power of two. Default is @code{32}.
3615 Mark main input as a pre-processed input and activate clean source input
3616 stream. This allows the input to be pre-processed with various filters to help
3617 the metrics calculation while keeping the frame selection lossless. When set to
3618 @code{1}, the first stream is for the pre-processed input, and the second
3619 stream is the clean source from where the kept frames are chosen. Default is
3623 Set whether or not chroma is considered in the metric calculations. Default is
3629 Remove judder produced by partially interlaced telecined content.
3631 Judder can be introduced, for instance, by @ref{pullup} filter. If the original
3632 source was partially telecined content then the output of @code{pullup,dejudder}
3633 will have a variable frame rate. May change the recorded frame rate of the
3634 container. Aside from that change, this filter will not affect constant frame
3637 The option available in this filter is:
3641 Specify the length of the window over which the judder repeats.
3643 Accepts any integer greater than 1. Useful values are:
3647 If the original was telecined from 24 to 30 fps (Film to NTSC).
3650 If the original was telecined from 25 to 30 fps (PAL to NTSC).
3653 If a mixture of the two.
3656 The default is @samp{4}.
3661 Suppress a TV station logo by a simple interpolation of the surrounding
3662 pixels. Just set a rectangle covering the logo and watch it disappear
3663 (and sometimes something even uglier appear - your mileage may vary).
3665 It accepts the following parameters:
3670 Specify the top left corner coordinates of the logo. They must be
3675 Specify the width and height of the logo to clear. They must be
3679 Specify the thickness of the fuzzy edge of the rectangle (added to
3680 @var{w} and @var{h}). The default value is 4.
3683 When set to 1, a green rectangle is drawn on the screen to simplify
3684 finding the right @var{x}, @var{y}, @var{w}, and @var{h} parameters.
3685 The default value is 0.
3687 The rectangle is drawn on the outermost pixels which will be (partly)
3688 replaced with interpolated values. The values of the next pixels
3689 immediately outside this rectangle in each direction will be used to
3690 compute the interpolated pixel values inside the rectangle.
3694 @subsection Examples
3698 Set a rectangle covering the area with top left corner coordinates 0,0
3699 and size 100x77, and a band of size 10:
3701 delogo=x=0:y=0:w=100:h=77:band=10
3708 Attempt to fix small changes in horizontal and/or vertical shift. This
3709 filter helps remove camera shake from hand-holding a camera, bumping a
3710 tripod, moving on a vehicle, etc.
3712 The filter accepts the following options:
3720 Specify a rectangular area where to limit the search for motion
3722 If desired the search for motion vectors can be limited to a
3723 rectangular area of the frame defined by its top left corner, width
3724 and height. These parameters have the same meaning as the drawbox
3725 filter which can be used to visualise the position of the bounding
3728 This is useful when simultaneous movement of subjects within the frame
3729 might be confused for camera motion by the motion vector search.
3731 If any or all of @var{x}, @var{y}, @var{w} and @var{h} are set to -1
3732 then the full frame is used. This allows later options to be set
3733 without specifying the bounding box for the motion vector search.
3735 Default - search the whole frame.
3739 Specify the maximum extent of movement in x and y directions in the
3740 range 0-64 pixels. Default 16.
3743 Specify how to generate pixels to fill blanks at the edge of the
3744 frame. Available values are:
3747 Fill zeroes at blank locations
3749 Original image at blank locations
3751 Extruded edge value at blank locations
3753 Mirrored edge at blank locations
3755 Default value is @samp{mirror}.
3758 Specify the blocksize to use for motion search. Range 4-128 pixels,
3762 Specify the contrast threshold for blocks. Only blocks with more than
3763 the specified contrast (difference between darkest and lightest
3764 pixels) will be considered. Range 1-255, default 125.
3767 Specify the search strategy. Available values are:
3770 Set exhaustive search
3772 Set less exhaustive search.
3774 Default value is @samp{exhaustive}.
3777 If set then a detailed log of the motion search is written to the
3781 If set to 1, specify using OpenCL capabilities, only available if
3782 FFmpeg was configured with @code{--enable-opencl}. Default value is 0.
3788 Apply an exact inverse of the telecine operation. It requires a predefined
3789 pattern specified using the pattern option which must be the same as that passed
3790 to the telecine filter.
3792 This filter accepts the following options:
3801 The default value is @code{top}.
3805 A string of numbers representing the pulldown pattern you wish to apply.
3806 The default value is @code{23}.
3809 A number representing position of the first frame with respect to the telecine
3810 pattern. This is to be used if the stream is cut. The default value is @code{0}.
3815 Draw a colored box on the input image.
3817 It accepts the following parameters:
3822 The expressions which specify the top left corner coordinates of the box. It defaults to 0.
3826 The expressions which specify the width and height of the box; if 0 they are interpreted as
3827 the input width and height. It defaults to 0.
3830 Specify the color of the box to write. For the general syntax of this option,
3831 check the "Color" section in the ffmpeg-utils manual. If the special
3832 value @code{invert} is used, the box edge color is the same as the
3833 video with inverted luma.
3836 The expression which sets the thickness of the box edge. Default value is @code{3}.
3838 See below for the list of accepted constants.
3841 The parameters for @var{x}, @var{y}, @var{w} and @var{h} and @var{t} are expressions containing the
3842 following constants:
3846 The input display aspect ratio, it is the same as (@var{w} / @var{h}) * @var{sar}.
3850 horizontal and vertical chroma subsample values. For example for the
3851 pixel format "yuv422p" @var{hsub} is 2 and @var{vsub} is 1.
3855 The input width and height.
3858 The input sample aspect ratio.
3862 The x and y offset coordinates where the box is drawn.
3866 The width and height of the drawn box.
3869 The thickness of the drawn box.
3871 These constants allow the @var{x}, @var{y}, @var{w}, @var{h} and @var{t} expressions to refer to
3872 each other, so you may for example specify @code{y=x/dar} or @code{h=w/dar}.
3876 @subsection Examples
3880 Draw a black box around the edge of the input image:
3886 Draw a box with color red and an opacity of 50%:
3888 drawbox=10:20:200:60:red@@0.5
3891 The previous example can be specified as:
3893 drawbox=x=10:y=20:w=200:h=60:color=red@@0.5
3897 Fill the box with pink color:
3899 drawbox=x=10:y=10:w=100:h=100:color=pink@@0.5:t=max
3903 Draw a 2-pixel red 2.40:1 mask:
3905 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
3911 Draw a grid on the input image.
3913 It accepts the following parameters:
3918 The expressions which specify the coordinates of some point of grid intersection (meant to configure offset). Both default to 0.
3922 The expressions which specify the width and height of the grid cell, if 0 they are interpreted as the
3923 input width and height, respectively, minus @code{thickness}, so image gets
3924 framed. Default to 0.
3927 Specify the color of the grid. For the general syntax of this option,
3928 check the "Color" section in the ffmpeg-utils manual. If the special
3929 value @code{invert} is used, the grid color is the same as the
3930 video with inverted luma.
3933 The expression which sets the thickness of the grid line. Default value is @code{1}.
3935 See below for the list of accepted constants.
3938 The parameters for @var{x}, @var{y}, @var{w} and @var{h} and @var{t} are expressions containing the
3939 following constants:
3943 The input display aspect ratio, it is the same as (@var{w} / @var{h}) * @var{sar}.
3947 horizontal and vertical chroma subsample values. For example for the
3948 pixel format "yuv422p" @var{hsub} is 2 and @var{vsub} is 1.
3952 The input grid cell width and height.
3955 The input sample aspect ratio.
3959 The x and y coordinates of some point of grid intersection (meant to configure offset).
3963 The width and height of the drawn cell.
3966 The thickness of the drawn cell.
3968 These constants allow the @var{x}, @var{y}, @var{w}, @var{h} and @var{t} expressions to refer to
3969 each other, so you may for example specify @code{y=x/dar} or @code{h=w/dar}.
3973 @subsection Examples
3977 Draw a grid with cell 100x100 pixels, thickness 2 pixels, with color red and an opacity of 50%:
3979 drawgrid=width=100:height=100:thickness=2:color=red@@0.5
3983 Draw a white 3x3 grid with an opacity of 50%:
3985 drawgrid=w=iw/3:h=ih/3:t=2:c=white@@0.5
3992 Draw a text string or text from a specified file on top of a video, using the
3993 libfreetype library.
3995 To enable compilation of this filter, you need to configure FFmpeg with
3996 @code{--enable-libfreetype}.
3997 To enable default font fallback and the @var{font} option you need to
3998 configure FFmpeg with @code{--enable-libfontconfig}.
3999 To enable the @var{text_shaping} option, you need to configure FFmpeg with
4000 @code{--enable-libfribidi}.
4004 It accepts the following parameters:
4009 Used to draw a box around text using the background color.
4010 The value must be either 1 (enable) or 0 (disable).
4011 The default value of @var{box} is 0.
4014 Set the width of the border to be drawn around the box using @var{boxcolor}.
4015 The default value of @var{boxborderw} is 0.
4018 The color to be used for drawing box around text. For the syntax of this
4019 option, check the "Color" section in the ffmpeg-utils manual.
4021 The default value of @var{boxcolor} is "white".
4024 Set the width of the border to be drawn around the text using @var{bordercolor}.
4025 The default value of @var{borderw} is 0.
4028 Set the color to be used for drawing border around text. For the syntax of this
4029 option, check the "Color" section in the ffmpeg-utils manual.
4031 The default value of @var{bordercolor} is "black".
4034 Select how the @var{text} is expanded. Can be either @code{none},
4035 @code{strftime} (deprecated) or
4036 @code{normal} (default). See the @ref{drawtext_expansion, Text expansion} section
4040 If true, check and fix text coords to avoid clipping.
4043 The color to be used for drawing fonts. For the syntax of this option, check
4044 the "Color" section in the ffmpeg-utils manual.
4046 The default value of @var{fontcolor} is "black".
4048 @item fontcolor_expr
4049 String which is expanded the same way as @var{text} to obtain dynamic
4050 @var{fontcolor} value. By default this option has empty value and is not
4051 processed. When this option is set, it overrides @var{fontcolor} option.
4054 The font family to be used for drawing text. By default Sans.
4057 The font file to be used for drawing text. The path must be included.
4058 This parameter is mandatory if the fontconfig support is disabled.
4061 This option does not exist, please see the timeline system
4064 Draw the text applying alpha blending. The value can
4065 be either a number between 0.0 and 1.0
4066 The expression accepts the same variables @var{x, y} do.
4067 The default value is 1.
4068 Please see fontcolor_expr
4071 The font size to be used for drawing text.
4072 The default value of @var{fontsize} is 16.
4075 If set to 1, attempt to shape the text (for example, reverse the order of
4076 right-to-left text and join Arabic characters) before drawing it.
4077 Otherwise, just draw the text exactly as given.
4078 By default 1 (if supported).
4081 The flags to be used for loading the fonts.
4083 The flags map the corresponding flags supported by libfreetype, and are
4084 a combination of the following values:
4091 @item vertical_layout
4092 @item force_autohint
4095 @item ignore_global_advance_width
4097 @item ignore_transform
4103 Default value is "default".
4105 For more information consult the documentation for the FT_LOAD_*
4109 The color to be used for drawing a shadow behind the drawn text. For the
4110 syntax of this option, check the "Color" section in the ffmpeg-utils manual.
4112 The default value of @var{shadowcolor} is "black".
4116 The x and y offsets for the text shadow position with respect to the
4117 position of the text. They can be either positive or negative
4118 values. The default value for both is "0".
4121 The starting frame number for the n/frame_num variable. The default value
4125 The size in number of spaces to use for rendering the tab.
4129 Set the initial timecode representation in "hh:mm:ss[:;.]ff"
4130 format. It can be used with or without text parameter. @var{timecode_rate}
4131 option must be specified.
4133 @item timecode_rate, rate, r
4134 Set the timecode frame rate (timecode only).
4137 The text string to be drawn. The text must be a sequence of UTF-8
4139 This parameter is mandatory if no file is specified with the parameter
4143 A text file containing text to be drawn. The text must be a sequence
4144 of UTF-8 encoded characters.
4146 This parameter is mandatory if no text string is specified with the
4147 parameter @var{text}.
4149 If both @var{text} and @var{textfile} are specified, an error is thrown.
4152 If set to 1, the @var{textfile} will be reloaded before each frame.
4153 Be sure to update it atomically, or it may be read partially, or even fail.
4157 The expressions which specify the offsets where text will be drawn
4158 within the video frame. They are relative to the top/left border of the
4161 The default value of @var{x} and @var{y} is "0".
4163 See below for the list of accepted constants and functions.
4166 The parameters for @var{x} and @var{y} are expressions containing the
4167 following constants and functions:
4171 input display aspect ratio, it is the same as (@var{w} / @var{h}) * @var{sar}
4175 horizontal and vertical chroma subsample values. For example for the
4176 pixel format "yuv422p" @var{hsub} is 2 and @var{vsub} is 1.
4179 the height of each text line
4187 @item max_glyph_a, ascent
4188 the maximum distance from the baseline to the highest/upper grid
4189 coordinate used to place a glyph outline point, for all the rendered
4191 It is a positive value, due to the grid's orientation with the Y axis
4194 @item max_glyph_d, descent
4195 the maximum distance from the baseline to the lowest grid coordinate
4196 used to place a glyph outline point, for all the rendered glyphs.
4197 This is a negative value, due to the grid's orientation, with the Y axis
4201 maximum glyph height, that is the maximum height for all the glyphs
4202 contained in the rendered text, it is equivalent to @var{ascent} -
4206 maximum glyph width, that is the maximum width for all the glyphs
4207 contained in the rendered text
4210 the number of input frame, starting from 0
4212 @item rand(min, max)
4213 return a random number included between @var{min} and @var{max}
4216 The input sample aspect ratio.
4219 timestamp expressed in seconds, NAN if the input timestamp is unknown
4222 the height of the rendered text
4225 the width of the rendered text
4229 the x and y offset coordinates where the text is drawn.
4231 These parameters allow the @var{x} and @var{y} expressions to refer
4232 each other, so you can for example specify @code{y=x/dar}.
4235 @anchor{drawtext_expansion}
4236 @subsection Text expansion
4238 If @option{expansion} is set to @code{strftime},
4239 the filter recognizes strftime() sequences in the provided text and
4240 expands them accordingly. Check the documentation of strftime(). This
4241 feature is deprecated.
4243 If @option{expansion} is set to @code{none}, the text is printed verbatim.
4245 If @option{expansion} is set to @code{normal} (which is the default),
4246 the following expansion mechanism is used.
4248 The backslash character @samp{\}, followed by any character, always expands to
4249 the second character.
4251 Sequence of the form @code{%@{...@}} are expanded. The text between the
4252 braces is a function name, possibly followed by arguments separated by ':'.
4253 If the arguments contain special characters or delimiters (':' or '@}'),
4254 they should be escaped.
4256 Note that they probably must also be escaped as the value for the
4257 @option{text} option in the filter argument string and as the filter
4258 argument in the filtergraph description, and possibly also for the shell,
4259 that makes up to four levels of escaping; using a text file avoids these
4262 The following functions are available:
4267 The expression evaluation result.
4269 It must take one argument specifying the expression to be evaluated,
4270 which accepts the same constants and functions as the @var{x} and
4271 @var{y} values. Note that not all constants should be used, for
4272 example the text size is not known when evaluating the expression, so
4273 the constants @var{text_w} and @var{text_h} will have an undefined
4276 @item expr_int_format, eif
4277 Evaluate the expression's value and output as formatted integer.
4279 The first argument is the expression to be evaluated, just as for the @var{expr} function.
4280 The second argument specifies the output format. Allowed values are @samp{x},
4281 @samp{X}, @samp{d} and @samp{u}. They are treated exactly as in the
4282 @code{printf} function.
4283 The third parameter is optional and sets the number of positions taken by the output.
4284 It can be used to add padding with zeros from the left.
4287 The time at which the filter is running, expressed in UTC.
4288 It can accept an argument: a strftime() format string.
4291 The time at which the filter is running, expressed in the local time zone.
4292 It can accept an argument: a strftime() format string.
4295 Frame metadata. It must take one argument specifying metadata key.
4298 The frame number, starting from 0.
4301 A 1 character description of the current picture type.
4304 The timestamp of the current frame.
4305 It can take up to two arguments.
4307 The first argument is the format of the timestamp; it defaults to @code{flt}
4308 for seconds as a decimal number with microsecond accuracy; @code{hms} stands
4309 for a formatted @var{[-]HH:MM:SS.mmm} timestamp with millisecond accuracy.
4311 The second argument is an offset added to the timestamp.
4315 @subsection Examples
4319 Draw "Test Text" with font FreeSerif, using the default values for the
4320 optional parameters.
4323 drawtext="fontfile=/usr/share/fonts/truetype/freefont/FreeSerif.ttf: text='Test Text'"
4327 Draw 'Test Text' with font FreeSerif of size 24 at position x=100
4328 and y=50 (counting from the top-left corner of the screen), text is
4329 yellow with a red box around it. Both the text and the box have an
4333 drawtext="fontfile=/usr/share/fonts/truetype/freefont/FreeSerif.ttf: text='Test Text':\
4334 x=100: y=50: fontsize=24: fontcolor=yellow@@0.2: box=1: boxcolor=red@@0.2"
4337 Note that the double quotes are not necessary if spaces are not used
4338 within the parameter list.
4341 Show the text at the center of the video frame:
4343 drawtext="fontsize=30:fontfile=FreeSerif.ttf:text='hello world':x=(w-text_w)/2:y=(h-text_h-line_h)/2"
4347 Show a text line sliding from right to left in the last row of the video
4348 frame. The file @file{LONG_LINE} is assumed to contain a single line
4351 drawtext="fontsize=15:fontfile=FreeSerif.ttf:text=LONG_LINE:y=h-line_h:x=-50*t"
4355 Show the content of file @file{CREDITS} off the bottom of the frame and scroll up.
4357 drawtext="fontsize=20:fontfile=FreeSerif.ttf:textfile=CREDITS:y=h-20*t"
4361 Draw a single green letter "g", at the center of the input video.
4362 The glyph baseline is placed at half screen height.
4364 drawtext="fontsize=60:fontfile=FreeSerif.ttf:fontcolor=green:text=g:x=(w-max_glyph_w)/2:y=h/2-ascent"
4368 Show text for 1 second every 3 seconds:
4370 drawtext="fontfile=FreeSerif.ttf:fontcolor=white:x=100:y=x/dar:enable=lt(mod(t\,3)\,1):text='blink'"
4374 Use fontconfig to set the font. Note that the colons need to be escaped.
4376 drawtext='fontfile=Linux Libertine O-40\:style=Semibold:text=FFmpeg'
4380 Print the date of a real-time encoding (see strftime(3)):
4382 drawtext='fontfile=FreeSans.ttf:text=%@{localtime\:%a %b %d %Y@}'
4386 Show text fading in and out (appearing/disappearing):
4389 DS=1.0 # display start
4390 DE=10.0 # display end
4391 FID=1.5 # fade in duration
4392 FOD=5 # fade out duration
4393 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 @}"
4398 For more information about libfreetype, check:
4399 @url{http://www.freetype.org/}.
4401 For more information about fontconfig, check:
4402 @url{http://freedesktop.org/software/fontconfig/fontconfig-user.html}.
4404 For more information about libfribidi, check:
4405 @url{http://fribidi.org/}.
4409 Detect and draw edges. The filter uses the Canny Edge Detection algorithm.
4411 The filter accepts the following options:
4416 Set low and high threshold values used by the Canny thresholding
4419 The high threshold selects the "strong" edge pixels, which are then
4420 connected through 8-connectivity with the "weak" edge pixels selected
4421 by the low threshold.
4423 @var{low} and @var{high} threshold values must be chosen in the range
4424 [0,1], and @var{low} should be lesser or equal to @var{high}.
4426 Default value for @var{low} is @code{20/255}, and default value for @var{high}
4430 Define the drawing mode.
4434 Draw white/gray wires on black background.
4437 Mix the colors to create a paint/cartoon effect.
4440 Default value is @var{wires}.
4443 @subsection Examples
4447 Standard edge detection with custom values for the hysteresis thresholding:
4449 edgedetect=low=0.1:high=0.4
4453 Painting effect without thresholding:
4455 edgedetect=mode=colormix:high=0
4460 Set brightness, contrast, saturation and approximate gamma adjustment.
4462 The filter accepts the following options:
4466 Set the contrast expression. The value must be a float value in range
4467 @code{-2.0} to @code{2.0}. The default value is "0".
4470 Set the brightness expression. The value must be a float value in
4471 range @code{-1.0} to @code{1.0}. The default value is "0".
4474 Set the saturation expression. The value must be a float in
4475 range @code{0.0} to @code{3.0}. The default value is "1".
4478 Set the gamma expression. The value must be a float in range
4479 @code{0.1} to @code{10.0}. The default value is "1".
4482 Set the gamma expression for red. The value must be a float in
4483 range @code{0.1} to @code{10.0}. The default value is "1".
4486 Set the gamma expression for green. The value must be a float in range
4487 @code{0.1} to @code{10.0}. The default value is "1".
4490 Set the gamma expression for blue. The value must be a float in range
4491 @code{0.1} to @code{10.0}. The default value is "1".
4494 Set the gamma weight expression. It can be used to reduce the effect
4495 of a high gamma value on bright image areas, e.g. keep them from
4496 getting overamplified and just plain white. The value must be a float
4497 in range @code{0.0} to @code{1.0}. A value of @code{0.0} turns the
4498 gamma correction all the way down while @code{1.0} leaves it at its
4499 full strength. Default is "1".
4502 Set when the expressions for brightness, contrast, saturation and
4503 gamma expressions are evaluated.
4505 It accepts the following values:
4508 only evaluate expressions once during the filter initialization or
4509 when a command is processed
4512 evaluate expressions for each incoming frame
4515 Default value is @samp{init}.
4518 The expressions accept the following parameters:
4521 frame count of the input frame starting from 0
4524 byte position of the corresponding packet in the input file, NAN if
4528 frame rate of the input video, NAN if the input frame rate is unknown
4531 timestamp expressed in seconds, NAN if the input timestamp is unknown
4534 @subsection Commands
4535 The filter supports the following commands:
4539 Set the contrast expression.
4542 Set the brightness expression.
4545 Set the saturation expression.
4548 Set the gamma expression.
4551 Set the gamma_r expression.
4554 Set gamma_g expression.
4557 Set gamma_b expression.
4560 Set gamma_weight expression.
4562 The command accepts the same syntax of the corresponding option.
4564 If the specified expression is not valid, it is kept at its current
4569 @section extractplanes
4571 Extract color channel components from input video stream into
4572 separate grayscale video streams.
4574 The filter accepts the following option:
4578 Set plane(s) to extract.
4580 Available values for planes are:
4591 Choosing planes not available in the input will result in an error.
4592 That means you cannot select @code{r}, @code{g}, @code{b} planes
4593 with @code{y}, @code{u}, @code{v} planes at same time.
4596 @subsection Examples
4600 Extract luma, u and v color channel component from input video frame
4601 into 3 grayscale outputs:
4603 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
4609 Apply a posterize effect using the ELBG (Enhanced LBG) algorithm.
4611 For each input image, the filter will compute the optimal mapping from
4612 the input to the output given the codebook length, that is the number
4613 of distinct output colors.
4615 This filter accepts the following options.
4618 @item codebook_length, l
4619 Set codebook length. The value must be a positive integer, and
4620 represents the number of distinct output colors. Default value is 256.
4623 Set the maximum number of iterations to apply for computing the optimal
4624 mapping. The higher the value the better the result and the higher the
4625 computation time. Default value is 1.
4628 Set a random seed, must be an integer included between 0 and
4629 UINT32_MAX. If not specified, or if explicitly set to -1, the filter
4630 will try to use a good random seed on a best effort basis.
4635 Apply a fade-in/out effect to the input video.
4637 It accepts the following parameters:
4641 The effect type can be either "in" for a fade-in, or "out" for a fade-out
4643 Default is @code{in}.
4645 @item start_frame, s
4646 Specify the number of the frame to start applying the fade
4647 effect at. Default is 0.
4650 The number of frames that the fade effect lasts. At the end of the
4651 fade-in effect, the output video will have the same intensity as the input video.
4652 At the end of the fade-out transition, the output video will be filled with the
4653 selected @option{color}.
4657 If set to 1, fade only alpha channel, if one exists on the input.
4660 @item start_time, st
4661 Specify the timestamp (in seconds) of the frame to start to apply the fade
4662 effect. If both start_frame and start_time are specified, the fade will start at
4663 whichever comes last. Default is 0.
4666 The number of seconds for which the fade effect has to last. At the end of the
4667 fade-in effect the output video will have the same intensity as the input video,
4668 at the end of the fade-out transition the output video will be filled with the
4669 selected @option{color}.
4670 If both duration and nb_frames are specified, duration is used. Default is 0
4671 (nb_frames is used by default).
4674 Specify the color of the fade. Default is "black".
4677 @subsection Examples
4681 Fade in the first 30 frames of video:
4686 The command above is equivalent to:
4692 Fade out the last 45 frames of a 200-frame video:
4695 fade=type=out:start_frame=155:nb_frames=45
4699 Fade in the first 25 frames and fade out the last 25 frames of a 1000-frame video:
4701 fade=in:0:25, fade=out:975:25
4705 Make the first 5 frames yellow, then fade in from frame 5-24:
4707 fade=in:5:20:color=yellow
4711 Fade in alpha over first 25 frames of video:
4713 fade=in:0:25:alpha=1
4717 Make the first 5.5 seconds black, then fade in for 0.5 seconds:
4719 fade=t=in:st=5.5:d=0.5
4725 Apply arbitrary expressions to samples in frequency domain
4729 Adjust the dc value (gain) of the luma plane of the image. The filter
4730 accepts an integer value in range @code{0} to @code{1000}. The default
4731 value is set to @code{0}.
4734 Adjust the dc value (gain) of the 1st chroma plane of the image. The
4735 filter accepts an integer value in range @code{0} to @code{1000}. The
4736 default value is set to @code{0}.
4739 Adjust the dc value (gain) of the 2nd chroma plane of the image. The
4740 filter accepts an integer value in range @code{0} to @code{1000}. The
4741 default value is set to @code{0}.
4744 Set the frequency domain weight expression for the luma plane.
4747 Set the frequency domain weight expression for the 1st chroma plane.
4750 Set the frequency domain weight expression for the 2nd chroma plane.
4752 The filter accepts the following variables:
4755 The coordinates of the current sample.
4759 The width and height of the image.
4762 @subsection Examples
4768 fftfilt=dc_Y=128:weight_Y='squish(1-(Y+X)/100)'
4774 fftfilt=dc_Y=0:weight_Y='squish((Y+X)/100-1)'
4780 fftfilt=dc_Y=0:weight_Y='1+squish(1-(Y+X)/100)'
4787 Extract a single field from an interlaced image using stride
4788 arithmetic to avoid wasting CPU time. The output frames are marked as
4791 The filter accepts the following options:
4795 Specify whether to extract the top (if the value is @code{0} or
4796 @code{top}) or the bottom field (if the value is @code{1} or
4802 Field matching filter for inverse telecine. It is meant to reconstruct the
4803 progressive frames from a telecined stream. The filter does not drop duplicated
4804 frames, so to achieve a complete inverse telecine @code{fieldmatch} needs to be
4805 followed by a decimation filter such as @ref{decimate} in the filtergraph.
4807 The separation of the field matching and the decimation is notably motivated by
4808 the possibility of inserting a de-interlacing filter fallback between the two.
4809 If the source has mixed telecined and real interlaced content,
4810 @code{fieldmatch} will not be able to match fields for the interlaced parts.
4811 But these remaining combed frames will be marked as interlaced, and thus can be
4812 de-interlaced by a later filter such as @ref{yadif} before decimation.
4814 In addition to the various configuration options, @code{fieldmatch} can take an
4815 optional second stream, activated through the @option{ppsrc} option. If
4816 enabled, the frames reconstruction will be based on the fields and frames from
4817 this second stream. This allows the first input to be pre-processed in order to
4818 help the various algorithms of the filter, while keeping the output lossless
4819 (assuming the fields are matched properly). Typically, a field-aware denoiser,
4820 or brightness/contrast adjustments can help.
4822 Note that this filter uses the same algorithms as TIVTC/TFM (AviSynth project)
4823 and VIVTC/VFM (VapourSynth project). The later is a light clone of TFM from
4824 which @code{fieldmatch} is based on. While the semantic and usage are very
4825 close, some behaviour and options names can differ.
4827 The @ref{decimate} filter currently only works for constant frame rate input.
4828 Do not use @code{fieldmatch} and @ref{decimate} if your input has mixed
4829 telecined and progressive content with changing framerate.
4831 The filter accepts the following options:
4835 Specify the assumed field order of the input stream. Available values are:
4839 Auto detect parity (use FFmpeg's internal parity value).
4841 Assume bottom field first.
4843 Assume top field first.
4846 Note that it is sometimes recommended not to trust the parity announced by the
4849 Default value is @var{auto}.
4852 Set the matching mode or strategy to use. @option{pc} mode is the safest in the
4853 sense that it won't risk creating jerkiness due to duplicate frames when
4854 possible, but if there are bad edits or blended fields it will end up
4855 outputting combed frames when a good match might actually exist. On the other
4856 hand, @option{pcn_ub} mode is the most risky in terms of creating jerkiness,
4857 but will almost always find a good frame if there is one. The other values are
4858 all somewhere in between @option{pc} and @option{pcn_ub} in terms of risking
4859 jerkiness and creating duplicate frames versus finding good matches in sections
4860 with bad edits, orphaned fields, blended fields, etc.
4862 More details about p/c/n/u/b are available in @ref{p/c/n/u/b meaning} section.
4864 Available values are:
4868 2-way matching (p/c)
4870 2-way matching, and trying 3rd match if still combed (p/c + n)
4872 2-way matching, and trying 3rd match (same order) if still combed (p/c + u)
4874 2-way matching, trying 3rd match if still combed, and trying 4th/5th matches if
4875 still combed (p/c + n + u/b)
4877 3-way matching (p/c/n)
4879 3-way matching, and trying 4th/5th matches if all 3 of the original matches are
4880 detected as combed (p/c/n + u/b)
4883 The parenthesis at the end indicate the matches that would be used for that
4884 mode assuming @option{order}=@var{tff} (and @option{field} on @var{auto} or
4887 In terms of speed @option{pc} mode is by far the fastest and @option{pcn_ub} is
4890 Default value is @var{pc_n}.
4893 Mark the main input stream as a pre-processed input, and enable the secondary
4894 input stream as the clean source to pick the fields from. See the filter
4895 introduction for more details. It is similar to the @option{clip2} feature from
4898 Default value is @code{0} (disabled).
4901 Set the field to match from. It is recommended to set this to the same value as
4902 @option{order} unless you experience matching failures with that setting. In
4903 certain circumstances changing the field that is used to match from can have a
4904 large impact on matching performance. Available values are:
4908 Automatic (same value as @option{order}).
4910 Match from the bottom field.
4912 Match from the top field.
4915 Default value is @var{auto}.
4918 Set whether or not chroma is included during the match comparisons. In most
4919 cases it is recommended to leave this enabled. You should set this to @code{0}
4920 only if your clip has bad chroma problems such as heavy rainbowing or other
4921 artifacts. Setting this to @code{0} could also be used to speed things up at
4922 the cost of some accuracy.
4924 Default value is @code{1}.
4928 These define an exclusion band which excludes the lines between @option{y0} and
4929 @option{y1} from being included in the field matching decision. An exclusion
4930 band can be used to ignore subtitles, a logo, or other things that may
4931 interfere with the matching. @option{y0} sets the starting scan line and
4932 @option{y1} sets the ending line; all lines in between @option{y0} and
4933 @option{y1} (including @option{y0} and @option{y1}) will be ignored. Setting
4934 @option{y0} and @option{y1} to the same value will disable the feature.
4935 @option{y0} and @option{y1} defaults to @code{0}.
4938 Set the scene change detection threshold as a percentage of maximum change on
4939 the luma plane. Good values are in the @code{[8.0, 14.0]} range. Scene change
4940 detection is only relevant in case @option{combmatch}=@var{sc}. The range for
4941 @option{scthresh} is @code{[0.0, 100.0]}.
4943 Default value is @code{12.0}.
4946 When @option{combatch} is not @var{none}, @code{fieldmatch} will take into
4947 account the combed scores of matches when deciding what match to use as the
4948 final match. Available values are:
4952 No final matching based on combed scores.
4954 Combed scores are only used when a scene change is detected.
4956 Use combed scores all the time.
4959 Default is @var{sc}.
4962 Force @code{fieldmatch} to calculate the combed metrics for certain matches and
4963 print them. This setting is known as @option{micout} in TFM/VFM vocabulary.
4964 Available values are:
4968 No forced calculation.
4970 Force p/c/n calculations.
4972 Force p/c/n/u/b calculations.
4975 Default value is @var{none}.
4978 This is the area combing threshold used for combed frame detection. This
4979 essentially controls how "strong" or "visible" combing must be to be detected.
4980 Larger values mean combing must be more visible and smaller values mean combing
4981 can be less visible or strong and still be detected. Valid settings are from
4982 @code{-1} (every pixel will be detected as combed) to @code{255} (no pixel will
4983 be detected as combed). This is basically a pixel difference value. A good
4984 range is @code{[8, 12]}.
4986 Default value is @code{9}.
4989 Sets whether or not chroma is considered in the combed frame decision. Only
4990 disable this if your source has chroma problems (rainbowing, etc.) that are
4991 causing problems for the combed frame detection with chroma enabled. Actually,
4992 using @option{chroma}=@var{0} is usually more reliable, except for the case
4993 where there is chroma only combing in the source.
4995 Default value is @code{0}.
4999 Respectively set the x-axis and y-axis size of the window used during combed
5000 frame detection. This has to do with the size of the area in which
5001 @option{combpel} pixels are required to be detected as combed for a frame to be
5002 declared combed. See the @option{combpel} parameter description for more info.
5003 Possible values are any number that is a power of 2 starting at 4 and going up
5006 Default value is @code{16}.
5009 The number of combed pixels inside any of the @option{blocky} by
5010 @option{blockx} size blocks on the frame for the frame to be detected as
5011 combed. While @option{cthresh} controls how "visible" the combing must be, this
5012 setting controls "how much" combing there must be in any localized area (a
5013 window defined by the @option{blockx} and @option{blocky} settings) on the
5014 frame. Minimum value is @code{0} and maximum is @code{blocky x blockx} (at
5015 which point no frames will ever be detected as combed). This setting is known
5016 as @option{MI} in TFM/VFM vocabulary.
5018 Default value is @code{80}.
5021 @anchor{p/c/n/u/b meaning}
5022 @subsection p/c/n/u/b meaning
5024 @subsubsection p/c/n
5026 We assume the following telecined stream:
5029 Top fields: 1 2 2 3 4
5030 Bottom fields: 1 2 3 4 4
5033 The numbers correspond to the progressive frame the fields relate to. Here, the
5034 first two frames are progressive, the 3rd and 4th are combed, and so on.
5036 When @code{fieldmatch} is configured to run a matching from bottom
5037 (@option{field}=@var{bottom}) this is how this input stream get transformed:
5042 B 1 2 3 4 4 <-- matching reference
5051 As a result of the field matching, we can see that some frames get duplicated.
5052 To perform a complete inverse telecine, you need to rely on a decimation filter
5053 after this operation. See for instance the @ref{decimate} filter.
5055 The same operation now matching from top fields (@option{field}=@var{top})
5060 T 1 2 2 3 4 <-- matching reference
5070 In these examples, we can see what @var{p}, @var{c} and @var{n} mean;
5071 basically, they refer to the frame and field of the opposite parity:
5074 @item @var{p} matches the field of the opposite parity in the previous frame
5075 @item @var{c} matches the field of the opposite parity in the current frame
5076 @item @var{n} matches the field of the opposite parity in the next frame
5081 The @var{u} and @var{b} matching are a bit special in the sense that they match
5082 from the opposite parity flag. In the following examples, we assume that we are
5083 currently matching the 2nd frame (Top:2, bottom:2). According to the match, a
5084 'x' is placed above and below each matched fields.
5086 With bottom matching (@option{field}=@var{bottom}):
5091 Top 1 2 2 1 2 2 1 2 2 1 2 2 1 2 2
5092 Bottom 1 2 3 1 2 3 1 2 3 1 2 3 1 2 3
5100 With top matching (@option{field}=@var{top}):
5105 Top 1 2 2 1 2 2 1 2 2 1 2 2 1 2 2
5106 Bottom 1 2 3 1 2 3 1 2 3 1 2 3 1 2 3
5114 @subsection Examples
5116 Simple IVTC of a top field first telecined stream:
5118 fieldmatch=order=tff:combmatch=none, decimate
5121 Advanced IVTC, with fallback on @ref{yadif} for still combed frames:
5123 fieldmatch=order=tff:combmatch=full, yadif=deint=interlaced, decimate
5128 Transform the field order of the input video.
5130 It accepts the following parameters:
5135 The output field order. Valid values are @var{tff} for top field first or @var{bff}
5136 for bottom field first.
5139 The default value is @samp{tff}.
5141 The transformation is done by shifting the picture content up or down
5142 by one line, and filling the remaining line with appropriate picture content.
5143 This method is consistent with most broadcast field order converters.
5145 If the input video is not flagged as being interlaced, or it is already
5146 flagged as being of the required output field order, then this filter does
5147 not alter the incoming video.
5149 It is very useful when converting to or from PAL DV material,
5150 which is bottom field first.
5154 ffmpeg -i in.vob -vf "fieldorder=bff" out.dv
5159 Buffer input images and send them when they are requested.
5161 It is mainly useful when auto-inserted by the libavfilter
5164 It does not take parameters.
5168 Find a rectangular object
5170 It accepts the following options:
5174 Filepath of the object image, needs to be in gray8.
5177 Detection threshold, default is 0.5.
5180 Number of mipmaps, default is 3.
5182 @item xmin, ymin, xmax, ymax
5183 Specifies the rectangle in which to search.
5186 @subsection Examples
5190 Generate a representative palette of a given video using @command{ffmpeg}:
5192 ffmpeg -i file.ts -vf find_rect=newref.pgm,cover_rect=cover.jpg:mode=cover new.mkv
5198 Cover a rectangular object
5200 It accepts the following options:
5204 Filepath of the optional cover image, needs to be in yuv420.
5209 It accepts the following values:
5212 cover it by the supplied image
5214 cover it by interpolating the surrounding pixels
5217 Default value is @var{blur}.
5220 @subsection Examples
5224 Generate a representative palette of a given video using @command{ffmpeg}:
5226 ffmpeg -i file.ts -vf find_rect=newref.pgm,cover_rect=cover.jpg:mode=cover new.mkv
5233 Convert the input video to one of the specified pixel formats.
5234 Libavfilter will try to pick one that is suitable as input to
5237 It accepts the following parameters:
5241 A '|'-separated list of pixel format names, such as
5242 "pix_fmts=yuv420p|monow|rgb24".
5246 @subsection Examples
5250 Convert the input video to the @var{yuv420p} format
5252 format=pix_fmts=yuv420p
5255 Convert the input video to any of the formats in the list
5257 format=pix_fmts=yuv420p|yuv444p|yuv410p
5264 Convert the video to specified constant frame rate by duplicating or dropping
5265 frames as necessary.
5267 It accepts the following parameters:
5271 The desired output frame rate. The default is @code{25}.
5276 Possible values are:
5279 zero round towards 0
5283 round towards -infinity
5285 round towards +infinity
5289 The default is @code{near}.
5292 Assume the first PTS should be the given value, in seconds. This allows for
5293 padding/trimming at the start of stream. By default, no assumption is made
5294 about the first frame's expected PTS, so no padding or trimming is done.
5295 For example, this could be set to 0 to pad the beginning with duplicates of
5296 the first frame if a video stream starts after the audio stream or to trim any
5297 frames with a negative PTS.
5301 Alternatively, the options can be specified as a flat string:
5302 @var{fps}[:@var{round}].
5304 See also the @ref{setpts} filter.
5306 @subsection Examples
5310 A typical usage in order to set the fps to 25:
5316 Sets the fps to 24, using abbreviation and rounding method to round to nearest:
5318 fps=fps=film:round=near
5324 Pack two different video streams into a stereoscopic video, setting proper
5325 metadata on supported codecs. The two views should have the same size and
5326 framerate and processing will stop when the shorter video ends. Please note
5327 that you may conveniently adjust view properties with the @ref{scale} and
5330 It accepts the following parameters:
5334 The desired packing format. Supported values are:
5339 The views are next to each other (default).
5342 The views are on top of each other.
5345 The views are packed by line.
5348 The views are packed by column.
5351 The views are temporally interleaved.
5360 # Convert left and right views into a frame-sequential video
5361 ffmpeg -i LEFT -i RIGHT -filter_complex framepack=frameseq OUTPUT
5363 # Convert views into a side-by-side video with the same output resolution as the input
5364 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
5369 Select one frame every N-th frame.
5371 This filter accepts the following option:
5374 Select frame after every @code{step} frames.
5375 Allowed values are positive integers higher than 0. Default value is @code{1}.
5381 Apply a frei0r effect to the input video.
5383 To enable the compilation of this filter, you need to install the frei0r
5384 header and configure FFmpeg with @code{--enable-frei0r}.
5386 It accepts the following parameters:
5391 The name of the frei0r effect to load. If the environment variable
5392 @env{FREI0R_PATH} is defined, the frei0r effect is searched for in each of the
5393 directories specified by the colon-separated list in @env{FREIOR_PATH}.
5394 Otherwise, the standard frei0r paths are searched, in this order:
5395 @file{HOME/.frei0r-1/lib/}, @file{/usr/local/lib/frei0r-1/},
5396 @file{/usr/lib/frei0r-1/}.
5399 A '|'-separated list of parameters to pass to the frei0r effect.
5403 A frei0r effect parameter can be a boolean (its value is either
5404 "y" or "n"), a double, a color (specified as
5405 @var{R}/@var{G}/@var{B}, where @var{R}, @var{G}, and @var{B} are floating point
5406 numbers between 0.0 and 1.0, inclusive) or by a color description specified in the "Color"
5407 section in the ffmpeg-utils manual), a position (specified as @var{X}/@var{Y}, where
5408 @var{X} and @var{Y} are floating point numbers) and/or a string.
5410 The number and types of parameters depend on the loaded effect. If an
5411 effect parameter is not specified, the default value is set.
5413 @subsection Examples
5417 Apply the distort0r effect, setting the first two double parameters:
5419 frei0r=filter_name=distort0r:filter_params=0.5|0.01
5423 Apply the colordistance effect, taking a color as the first parameter:
5425 frei0r=colordistance:0.2/0.3/0.4
5426 frei0r=colordistance:violet
5427 frei0r=colordistance:0x112233
5431 Apply the perspective effect, specifying the top left and top right image
5434 frei0r=perspective:0.2/0.2|0.8/0.2
5438 For more information, see
5439 @url{http://frei0r.dyne.org}
5443 Apply fast and simple postprocessing. It is a faster version of @ref{spp}.
5445 It splits (I)DCT into horizontal/vertical passes. Unlike the simple post-
5446 processing filter, one of them is performed once per block, not per pixel.
5447 This allows for much higher speed.
5449 The filter accepts the following options:
5453 Set quality. This option defines the number of levels for averaging. It accepts
5454 an integer in the range 4-5. Default value is @code{4}.
5457 Force a constant quantization parameter. It accepts an integer in range 0-63.
5458 If not set, the filter will use the QP from the video stream (if available).
5461 Set filter strength. It accepts an integer in range -15 to 32. Lower values mean
5462 more details but also more artifacts, while higher values make the image smoother
5463 but also blurrier. Default value is @code{0} − PSNR optimal.
5466 Enable the use of the QP from the B-Frames if set to @code{1}. Using this
5467 option may cause flicker since the B-Frames have often larger QP. Default is
5468 @code{0} (not enabled).
5474 The filter accepts the following options:
5478 Set the luminance expression.
5480 Set the chrominance blue expression.
5482 Set the chrominance red expression.
5484 Set the alpha expression.
5486 Set the red expression.
5488 Set the green expression.
5490 Set the blue expression.
5493 The colorspace is selected according to the specified options. If one
5494 of the @option{lum_expr}, @option{cb_expr}, or @option{cr_expr}
5495 options is specified, the filter will automatically select a YCbCr
5496 colorspace. If one of the @option{red_expr}, @option{green_expr}, or
5497 @option{blue_expr} options is specified, it will select an RGB
5500 If one of the chrominance expression is not defined, it falls back on the other
5501 one. If no alpha expression is specified it will evaluate to opaque value.
5502 If none of chrominance expressions are specified, they will evaluate
5503 to the luminance expression.
5505 The expressions can use the following variables and functions:
5509 The sequential number of the filtered frame, starting from @code{0}.
5513 The coordinates of the current sample.
5517 The width and height of the image.
5521 Width and height scale depending on the currently filtered plane. It is the
5522 ratio between the corresponding luma plane number of pixels and the current
5523 plane ones. E.g. for YUV4:2:0 the values are @code{1,1} for the luma plane, and
5524 @code{0.5,0.5} for chroma planes.
5527 Time of the current frame, expressed in seconds.
5530 Return the value of the pixel at location (@var{x},@var{y}) of the current
5534 Return the value of the pixel at location (@var{x},@var{y}) of the luminance
5538 Return the value of the pixel at location (@var{x},@var{y}) of the
5539 blue-difference chroma plane. Return 0 if there is no such plane.
5542 Return the value of the pixel at location (@var{x},@var{y}) of the
5543 red-difference chroma plane. Return 0 if there is no such plane.
5548 Return the value of the pixel at location (@var{x},@var{y}) of the
5549 red/green/blue component. Return 0 if there is no such component.
5552 Return the value of the pixel at location (@var{x},@var{y}) of the alpha
5553 plane. Return 0 if there is no such plane.
5556 For functions, if @var{x} and @var{y} are outside the area, the value will be
5557 automatically clipped to the closer edge.
5559 @subsection Examples
5563 Flip the image horizontally:
5569 Generate a bidimensional sine wave, with angle @code{PI/3} and a
5570 wavelength of 100 pixels:
5572 geq=128 + 100*sin(2*(PI/100)*(cos(PI/3)*(X-50*T) + sin(PI/3)*Y)):128:128
5576 Generate a fancy enigmatic moving light:
5578 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
5582 Generate a quick emboss effect:
5584 format=gray,geq=lum_expr='(p(X,Y)+(256-p(X-4,Y-4)))/2'
5588 Modify RGB components depending on pixel position:
5590 geq=r='X/W*r(X,Y)':g='(1-X/W)*g(X,Y)':b='(H-Y)/H*b(X,Y)'
5594 Create a radial gradient that is the same size as the input (also see
5595 the @ref{vignette} filter):
5597 geq=lum=255*gauss((X/W-0.5)*3)*gauss((Y/H-0.5)*3)/gauss(0)/gauss(0),format=gray
5601 Create a linear gradient to use as a mask for another filter, then
5602 compose with @ref{overlay}. In this example the video will gradually
5603 become more blurry from the top to the bottom of the y-axis as defined
5604 by the linear gradient:
5606 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
5612 Fix the banding artifacts that are sometimes introduced into nearly flat
5613 regions by truncation to 8bit color depth.
5614 Interpolate the gradients that should go where the bands are, and
5617 It is designed for playback only. Do not use it prior to
5618 lossy compression, because compression tends to lose the dither and
5619 bring back the bands.
5621 It accepts the following parameters:
5626 The maximum amount by which the filter will change any one pixel. This is also
5627 the threshold for detecting nearly flat regions. Acceptable values range from
5628 .51 to 64; the default value is 1.2. Out-of-range values will be clipped to the
5632 The neighborhood to fit the gradient to. A larger radius makes for smoother
5633 gradients, but also prevents the filter from modifying the pixels near detailed
5634 regions. Acceptable values are 8-32; the default value is 16. Out-of-range
5635 values will be clipped to the valid range.
5639 Alternatively, the options can be specified as a flat string:
5640 @var{strength}[:@var{radius}]
5642 @subsection Examples
5646 Apply the filter with a @code{3.5} strength and radius of @code{8}:
5652 Specify radius, omitting the strength (which will fall-back to the default
5663 Apply a Hald CLUT to a video stream.
5665 First input is the video stream to process, and second one is the Hald CLUT.
5666 The Hald CLUT input can be a simple picture or a complete video stream.
5668 The filter accepts the following options:
5672 Force termination when the shortest input terminates. Default is @code{0}.
5674 Continue applying the last CLUT after the end of the stream. A value of
5675 @code{0} disable the filter after the last frame of the CLUT is reached.
5676 Default is @code{1}.
5679 @code{haldclut} also has the same interpolation options as @ref{lut3d} (both
5680 filters share the same internals).
5682 More information about the Hald CLUT can be found on Eskil Steenberg's website
5683 (Hald CLUT author) at @url{http://www.quelsolaar.com/technology/clut.html}.
5685 @subsection Workflow examples
5687 @subsubsection Hald CLUT video stream
5689 Generate an identity Hald CLUT stream altered with various effects:
5691 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
5694 Note: make sure you use a lossless codec.
5696 Then use it with @code{haldclut} to apply it on some random stream:
5698 ffmpeg -f lavfi -i mandelbrot -i clut.nut -filter_complex '[0][1] haldclut' -t 20 mandelclut.mkv
5701 The Hald CLUT will be applied to the 10 first seconds (duration of
5702 @file{clut.nut}), then the latest picture of that CLUT stream will be applied
5703 to the remaining frames of the @code{mandelbrot} stream.
5705 @subsubsection Hald CLUT with preview
5707 A Hald CLUT is supposed to be a squared image of @code{Level*Level*Level} by
5708 @code{Level*Level*Level} pixels. For a given Hald CLUT, FFmpeg will select the
5709 biggest possible square starting at the top left of the picture. The remaining
5710 padding pixels (bottom or right) will be ignored. This area can be used to add
5711 a preview of the Hald CLUT.
5713 Typically, the following generated Hald CLUT will be supported by the
5714 @code{haldclut} filter:
5717 ffmpeg -f lavfi -i @ref{haldclutsrc}=8 -vf "
5718 pad=iw+320 [padded_clut];
5719 smptebars=s=320x256, split [a][b];
5720 [padded_clut][a] overlay=W-320:h, curves=color_negative [main];
5721 [main][b] overlay=W-320" -frames:v 1 clut.png
5724 It contains the original and a preview of the effect of the CLUT: SMPTE color
5725 bars are displayed on the right-top, and below the same color bars processed by
5728 Then, the effect of this Hald CLUT can be visualized with:
5730 ffplay input.mkv -vf "movie=clut.png, [in] haldclut"
5735 Flip the input video horizontally.
5737 For example, to horizontally flip the input video with @command{ffmpeg}:
5739 ffmpeg -i in.avi -vf "hflip" out.avi
5743 This filter applies a global color histogram equalization on a
5746 It can be used to correct video that has a compressed range of pixel
5747 intensities. The filter redistributes the pixel intensities to
5748 equalize their distribution across the intensity range. It may be
5749 viewed as an "automatically adjusting contrast filter". This filter is
5750 useful only for correcting degraded or poorly captured source
5753 The filter accepts the following options:
5757 Determine the amount of equalization to be applied. As the strength
5758 is reduced, the distribution of pixel intensities more-and-more
5759 approaches that of the input frame. The value must be a float number
5760 in the range [0,1] and defaults to 0.200.
5763 Set the maximum intensity that can generated and scale the output
5764 values appropriately. The strength should be set as desired and then
5765 the intensity can be limited if needed to avoid washing-out. The value
5766 must be a float number in the range [0,1] and defaults to 0.210.
5769 Set the antibanding level. If enabled the filter will randomly vary
5770 the luminance of output pixels by a small amount to avoid banding of
5771 the histogram. Possible values are @code{none}, @code{weak} or
5772 @code{strong}. It defaults to @code{none}.
5777 Compute and draw a color distribution histogram for the input video.
5779 The computed histogram is a representation of the color component
5780 distribution in an image.
5782 The filter accepts the following options:
5788 It accepts the following values:
5791 Standard histogram that displays the color components distribution in an
5792 image. Displays color graph for each color component. Shows distribution of
5793 the Y, U, V, A or R, G, B components, depending on input format, in the
5794 current frame. Below each graph a color component scale meter is shown.
5797 Displays chroma values (U/V color placement) in a two dimensional
5798 graph (which is called a vectorscope). The brighter a pixel in the
5799 vectorscope, the more pixels of the input frame correspond to that pixel
5800 (i.e., more pixels have this chroma value). The V component is displayed on
5801 the horizontal (X) axis, with the leftmost side being V = 0 and the rightmost
5802 side being V = 255. The U component is displayed on the vertical (Y) axis,
5803 with the top representing U = 0 and the bottom representing U = 255.
5805 The position of a white pixel in the graph corresponds to the chroma value of
5806 a pixel of the input clip. The graph can therefore be used to read the hue
5807 (color flavor) and the saturation (the dominance of the hue in the color). As
5808 the hue of a color changes, it moves around the square. At the center of the
5809 square the saturation is zero, which means that the corresponding pixel has no
5810 color. If the amount of a specific color is increased (while leaving the other
5811 colors unchanged) the saturation increases, and the indicator moves towards
5812 the edge of the square.
5815 Chroma values in vectorscope, similar as @code{color} but actual chroma values
5819 Per row/column color component graph. In row mode, the graph on the left side
5820 represents color component value 0 and the right side represents value = 255.
5821 In column mode, the top side represents color component value = 0 and bottom
5822 side represents value = 255.
5824 Default value is @code{levels}.
5827 Set height of level in @code{levels}. Default value is @code{200}.
5828 Allowed range is [50, 2048].
5831 Set height of color scale in @code{levels}. Default value is @code{12}.
5832 Allowed range is [0, 40].
5835 Set step for @code{waveform} mode. Smaller values are useful to find out how
5836 many values of the same luminance are distributed across input rows/columns.
5837 Default value is @code{10}. Allowed range is [1, 255].
5840 Set mode for @code{waveform}. Can be either @code{row}, or @code{column}.
5841 Default is @code{row}.
5843 @item waveform_mirror
5844 Set mirroring mode for @code{waveform}. @code{0} means unmirrored, @code{1}
5845 means mirrored. In mirrored mode, higher values will be represented on the left
5846 side for @code{row} mode and at the top for @code{column} mode. Default is
5847 @code{0} (unmirrored).
5850 Set display mode for @code{waveform} and @code{levels}.
5851 It accepts the following values:
5854 Display separate graph for the color components side by side in
5855 @code{row} waveform mode or one below the other in @code{column} waveform mode
5856 for @code{waveform} histogram mode. For @code{levels} histogram mode,
5857 per color component graphs are placed below each other.
5859 Using this display mode in @code{waveform} histogram mode makes it easy to
5860 spot color casts in the highlights and shadows of an image, by comparing the
5861 contours of the top and the bottom graphs of each waveform. Since whites,
5862 grays, and blacks are characterized by exactly equal amounts of red, green,
5863 and blue, neutral areas of the picture should display three waveforms of
5864 roughly equal width/height. If not, the correction is easy to perform by
5865 making level adjustments the three waveforms.
5868 Presents information identical to that in the @code{parade}, except
5869 that the graphs representing color components are superimposed directly
5872 This display mode in @code{waveform} histogram mode makes it easier to spot
5873 relative differences or similarities in overlapping areas of the color
5874 components that are supposed to be identical, such as neutral whites, grays,
5877 Default is @code{parade}.
5880 Set mode for @code{levels}. Can be either @code{linear}, or @code{logarithmic}.
5881 Default is @code{linear}.
5884 @subsection Examples
5889 Calculate and draw histogram:
5891 ffplay -i input -vf histogram
5899 This is a high precision/quality 3d denoise filter. It aims to reduce
5900 image noise, producing smooth images and making still images really
5901 still. It should enhance compressibility.
5903 It accepts the following optional parameters:
5907 A non-negative floating point number which specifies spatial luma strength.
5910 @item chroma_spatial
5911 A non-negative floating point number which specifies spatial chroma strength.
5912 It defaults to 3.0*@var{luma_spatial}/4.0.
5915 A floating point number which specifies luma temporal strength. It defaults to
5916 6.0*@var{luma_spatial}/4.0.
5919 A floating point number which specifies chroma temporal strength. It defaults to
5920 @var{luma_tmp}*@var{chroma_spatial}/@var{luma_spatial}.
5925 Apply a high-quality magnification filter designed for pixel art. This filter
5926 was originally created by Maxim Stepin.
5928 It accepts the following option:
5932 Set the scaling dimension: @code{2} for @code{hq2x}, @code{3} for
5933 @code{hq3x} and @code{4} for @code{hq4x}.
5934 Default is @code{3}.
5939 Modify the hue and/or the saturation of the input.
5941 It accepts the following parameters:
5945 Specify the hue angle as a number of degrees. It accepts an expression,
5946 and defaults to "0".
5949 Specify the saturation in the [-10,10] range. It accepts an expression and
5953 Specify the hue angle as a number of radians. It accepts an
5954 expression, and defaults to "0".
5957 Specify the brightness in the [-10,10] range. It accepts an expression and
5961 @option{h} and @option{H} are mutually exclusive, and can't be
5962 specified at the same time.
5964 The @option{b}, @option{h}, @option{H} and @option{s} option values are
5965 expressions containing the following constants:
5969 frame count of the input frame starting from 0
5972 presentation timestamp of the input frame expressed in time base units
5975 frame rate of the input video, NAN if the input frame rate is unknown
5978 timestamp expressed in seconds, NAN if the input timestamp is unknown
5981 time base of the input video
5984 @subsection Examples
5988 Set the hue to 90 degrees and the saturation to 1.0:
5994 Same command but expressing the hue in radians:
6000 Rotate hue and make the saturation swing between 0
6001 and 2 over a period of 1 second:
6003 hue="H=2*PI*t: s=sin(2*PI*t)+1"
6007 Apply a 3 seconds saturation fade-in effect starting at 0:
6012 The general fade-in expression can be written as:
6014 hue="s=min(0\, max((t-START)/DURATION\, 1))"
6018 Apply a 3 seconds saturation fade-out effect starting at 5 seconds:
6020 hue="s=max(0\, min(1\, (8-t)/3))"
6023 The general fade-out expression can be written as:
6025 hue="s=max(0\, min(1\, (START+DURATION-t)/DURATION))"
6030 @subsection Commands
6032 This filter supports the following commands:
6038 Modify the hue and/or the saturation and/or brightness of the input video.
6039 The command accepts the same syntax of the corresponding option.
6041 If the specified expression is not valid, it is kept at its current
6047 Detect video interlacing type.
6049 This filter tries to detect if the input frames as interlaced, progressive,
6050 top or bottom field first. It will also try and detect fields that are
6051 repeated between adjacent frames (a sign of telecine).
6053 Single frame detection considers only immediately adjacent frames when classifying each frame.
6054 Multiple frame detection incorporates the classification history of previous frames.
6056 The filter will log these metadata values:
6059 @item single.current_frame
6060 Detected type of current frame using single-frame detection. One of:
6061 ``tff'' (top field first), ``bff'' (bottom field first),
6062 ``progressive'', or ``undetermined''
6065 Cumulative number of frames detected as top field first using single-frame detection.
6068 Cumulative number of frames detected as top field first using multiple-frame detection.
6071 Cumulative number of frames detected as bottom field first using single-frame detection.
6073 @item multiple.current_frame
6074 Detected type of current frame using multiple-frame detection. One of:
6075 ``tff'' (top field first), ``bff'' (bottom field first),
6076 ``progressive'', or ``undetermined''
6079 Cumulative number of frames detected as bottom field first using multiple-frame detection.
6081 @item single.progressive
6082 Cumulative number of frames detected as progressive using single-frame detection.
6084 @item multiple.progressive
6085 Cumulative number of frames detected as progressive using multiple-frame detection.
6087 @item single.undetermined
6088 Cumulative number of frames that could not be classified using single-frame detection.
6090 @item multiple.undetermined
6091 Cumulative number of frames that could not be classified using multiple-frame detection.
6093 @item repeated.current_frame
6094 Which field in the current frame is repeated from the last. One of ``neither'', ``top'', or ``bottom''.
6096 @item repeated.neither
6097 Cumulative number of frames with no repeated field.
6100 Cumulative number of frames with the top field repeated from the previous frame's top field.
6102 @item repeated.bottom
6103 Cumulative number of frames with the bottom field repeated from the previous frame's bottom field.
6106 The filter accepts the following options:
6110 Set interlacing threshold.
6112 Set progressive threshold.
6114 Threshold for repeated field detection.
6116 Number of frames after which a given frame's contribution to the
6117 statistics is halved (i.e., it contributes only 0.5 to it's
6118 classification). The default of 0 means that all frames seen are given
6119 full weight of 1.0 forever.
6120 @item analyze_interlaced_flag
6121 When this is not 0 then idet will use the specified number of frames to determine
6122 if the interlaced flag is accurate, it will not count undetermined frames.
6123 If the flag is found to be accurate it will be used without any further
6124 computations, if it is found to be inaccuarte it will be cleared without any
6125 further computations. This allows inserting the idet filter as a low computational
6126 method to clean up the interlaced flag
6131 Deinterleave or interleave fields.
6133 This filter allows one to process interlaced images fields without
6134 deinterlacing them. Deinterleaving splits the input frame into 2
6135 fields (so called half pictures). Odd lines are moved to the top
6136 half of the output image, even lines to the bottom half.
6137 You can process (filter) them independently and then re-interleave them.
6139 The filter accepts the following options:
6143 @item chroma_mode, c
6145 Available values for @var{luma_mode}, @var{chroma_mode} and
6146 @var{alpha_mode} are:
6152 @item deinterleave, d
6153 Deinterleave fields, placing one above the other.
6156 Interleave fields. Reverse the effect of deinterleaving.
6158 Default value is @code{none}.
6161 @item chroma_swap, cs
6162 @item alpha_swap, as
6163 Swap luma/chroma/alpha fields. Exchange even & odd lines. Default value is @code{0}.
6168 Simple interlacing filter from progressive contents. This interleaves upper (or
6169 lower) lines from odd frames with lower (or upper) lines from even frames,
6170 halving the frame rate and preserving image height.
6173 Original Original New Frame
6174 Frame 'j' Frame 'j+1' (tff)
6175 ========== =========== ==================
6176 Line 0 --------------------> Frame 'j' Line 0
6177 Line 1 Line 1 ----> Frame 'j+1' Line 1
6178 Line 2 ---------------------> Frame 'j' Line 2
6179 Line 3 Line 3 ----> Frame 'j+1' Line 3
6181 New Frame + 1 will be generated by Frame 'j+2' and Frame 'j+3' and so on
6184 It accepts the following optional parameters:
6188 This determines whether the interlaced frame is taken from the even
6189 (tff - default) or odd (bff) lines of the progressive frame.
6192 Enable (default) or disable the vertical lowpass filter to avoid twitter
6193 interlacing and reduce moire patterns.
6198 Deinterlace input video by applying Donald Graft's adaptive kernel
6199 deinterling. Work on interlaced parts of a video to produce
6202 The description of the accepted parameters follows.
6206 Set the threshold which affects the filter's tolerance when
6207 determining if a pixel line must be processed. It must be an integer
6208 in the range [0,255] and defaults to 10. A value of 0 will result in
6209 applying the process on every pixels.
6212 Paint pixels exceeding the threshold value to white if set to 1.
6216 Set the fields order. Swap fields if set to 1, leave fields alone if
6220 Enable additional sharpening if set to 1. Default is 0.
6223 Enable twoway sharpening if set to 1. Default is 0.
6226 @subsection Examples
6230 Apply default values:
6232 kerndeint=thresh=10:map=0:order=0:sharp=0:twoway=0
6236 Enable additional sharpening:
6242 Paint processed pixels in white:
6248 @section lenscorrection
6250 Correct radial lens distortion
6252 This filter can be used to correct for radial distortion as can result from the use
6253 of wide angle lenses, and thereby re-rectify the image. To find the right parameters
6254 one can use tools available for example as part of opencv or simply trial-and-error.
6255 To use opencv use the calibration sample (under samples/cpp) from the opencv sources
6256 and extract the k1 and k2 coefficients from the resulting matrix.
6258 Note that effectively the same filter is available in the open-source tools Krita and
6259 Digikam from the KDE project.
6261 In contrast to the @ref{vignette} filter, which can also be used to compensate lens errors,
6262 this filter corrects the distortion of the image, whereas @ref{vignette} corrects the
6263 brightness distribution, so you may want to use both filters together in certain
6264 cases, though you will have to take care of ordering, i.e. whether vignetting should
6265 be applied before or after lens correction.
6269 The filter accepts the following options:
6273 Relative x-coordinate of the focal point of the image, and thereby the center of the
6274 distortion. This value has a range [0,1] and is expressed as fractions of the image
6277 Relative y-coordinate of the focal point of the image, and thereby the center of the
6278 distortion. This value has a range [0,1] and is expressed as fractions of the image
6281 Coefficient of the quadratic correction term. 0.5 means no correction.
6283 Coefficient of the double quadratic correction term. 0.5 means no correction.
6286 The formula that generates the correction is:
6288 @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)
6290 where @var{r_0} is halve of the image diagonal and @var{r_src} and @var{r_tgt} are the
6291 distances from the focal point in the source and target images, respectively.
6296 Apply a 3D LUT to an input video.
6298 The filter accepts the following options:
6302 Set the 3D LUT file name.
6304 Currently supported formats:
6316 Select interpolation mode.
6318 Available values are:
6322 Use values from the nearest defined point.
6324 Interpolate values using the 8 points defining a cube.
6326 Interpolate values using a tetrahedron.
6330 @section lut, lutrgb, lutyuv
6332 Compute a look-up table for binding each pixel component input value
6333 to an output value, and apply it to the input video.
6335 @var{lutyuv} applies a lookup table to a YUV input video, @var{lutrgb}
6336 to an RGB input video.
6338 These filters accept the following parameters:
6341 set first pixel component expression
6343 set second pixel component expression
6345 set third pixel component expression
6347 set fourth pixel component expression, corresponds to the alpha component
6350 set red component expression
6352 set green component expression
6354 set blue component expression
6356 alpha component expression
6359 set Y/luminance component expression
6361 set U/Cb component expression
6363 set V/Cr component expression
6366 Each of them specifies the expression to use for computing the lookup table for
6367 the corresponding pixel component values.
6369 The exact component associated to each of the @var{c*} options depends on the
6372 The @var{lut} filter requires either YUV or RGB pixel formats in input,
6373 @var{lutrgb} requires RGB pixel formats in input, and @var{lutyuv} requires YUV.
6375 The expressions can contain the following constants and functions:
6380 The input width and height.
6383 The input value for the pixel component.
6386 The input value, clipped to the @var{minval}-@var{maxval} range.
6389 The maximum value for the pixel component.
6392 The minimum value for the pixel component.
6395 The negated value for the pixel component value, clipped to the
6396 @var{minval}-@var{maxval} range; it corresponds to the expression
6397 "maxval-clipval+minval".
6400 The computed value in @var{val}, clipped to the
6401 @var{minval}-@var{maxval} range.
6403 @item gammaval(gamma)
6404 The computed gamma correction value of the pixel component value,
6405 clipped to the @var{minval}-@var{maxval} range. It corresponds to the
6407 "pow((clipval-minval)/(maxval-minval)\,@var{gamma})*(maxval-minval)+minval"
6411 All expressions default to "val".
6413 @subsection Examples
6419 lutrgb="r=maxval+minval-val:g=maxval+minval-val:b=maxval+minval-val"
6420 lutyuv="y=maxval+minval-val:u=maxval+minval-val:v=maxval+minval-val"
6423 The above is the same as:
6425 lutrgb="r=negval:g=negval:b=negval"
6426 lutyuv="y=negval:u=negval:v=negval"
6436 Remove chroma components, turning the video into a graytone image:
6438 lutyuv="u=128:v=128"
6442 Apply a luma burning effect:
6448 Remove green and blue components:
6454 Set a constant alpha channel value on input:
6456 format=rgba,lutrgb=a="maxval-minval/2"
6460 Correct luminance gamma by a factor of 0.5:
6462 lutyuv=y=gammaval(0.5)
6466 Discard least significant bits of luma:
6468 lutyuv=y='bitand(val, 128+64+32)'
6472 @section mergeplanes
6474 Merge color channel components from several video streams.
6476 The filter accepts up to 4 input streams, and merge selected input
6477 planes to the output video.
6479 This filter accepts the following options:
6482 Set input to output plane mapping. Default is @code{0}.
6484 The mappings is specified as a bitmap. It should be specified as a
6485 hexadecimal number in the form 0xAa[Bb[Cc[Dd]]]. 'Aa' describes the
6486 mapping for the first plane of the output stream. 'A' sets the number of
6487 the input stream to use (from 0 to 3), and 'a' the plane number of the
6488 corresponding input to use (from 0 to 3). The rest of the mappings is
6489 similar, 'Bb' describes the mapping for the output stream second
6490 plane, 'Cc' describes the mapping for the output stream third plane and
6491 'Dd' describes the mapping for the output stream fourth plane.
6494 Set output pixel format. Default is @code{yuva444p}.
6497 @subsection Examples
6501 Merge three gray video streams of same width and height into single video stream:
6503 [a0][a1][a2]mergeplanes=0x001020:yuv444p
6507 Merge 1st yuv444p stream and 2nd gray video stream into yuva444p video stream:
6509 [a0][a1]mergeplanes=0x00010210:yuva444p
6513 Swap Y and A plane in yuva444p stream:
6515 format=yuva444p,mergeplanes=0x03010200:yuva444p
6519 Swap U and V plane in yuv420p stream:
6521 format=yuv420p,mergeplanes=0x000201:yuv420p
6525 Cast a rgb24 clip to yuv444p:
6527 format=rgb24,mergeplanes=0x000102:yuv444p
6533 Apply motion-compensation deinterlacing.
6535 It needs one field per frame as input and must thus be used together
6536 with yadif=1/3 or equivalent.
6538 This filter accepts the following options:
6541 Set the deinterlacing mode.
6543 It accepts one of the following values:
6548 use iterative motion estimation
6550 like @samp{slow}, but use multiple reference frames.
6552 Default value is @samp{fast}.
6555 Set the picture field parity assumed for the input video. It must be
6556 one of the following values:
6560 assume top field first
6562 assume bottom field first
6565 Default value is @samp{bff}.
6568 Set per-block quantization parameter (QP) used by the internal
6571 Higher values should result in a smoother motion vector field but less
6572 optimal individual vectors. Default value is 1.
6577 Drop frames that do not differ greatly from the previous frame in
6578 order to reduce frame rate.
6580 The main use of this filter is for very-low-bitrate encoding
6581 (e.g. streaming over dialup modem), but it could in theory be used for
6582 fixing movies that were inverse-telecined incorrectly.
6584 A description of the accepted options follows.
6588 Set the maximum number of consecutive frames which can be dropped (if
6589 positive), or the minimum interval between dropped frames (if
6590 negative). If the value is 0, the frame is dropped unregarding the
6591 number of previous sequentially dropped frames.
6598 Set the dropping threshold values.
6600 Values for @option{hi} and @option{lo} are for 8x8 pixel blocks and
6601 represent actual pixel value differences, so a threshold of 64
6602 corresponds to 1 unit of difference for each pixel, or the same spread
6603 out differently over the block.
6605 A frame is a candidate for dropping if no 8x8 blocks differ by more
6606 than a threshold of @option{hi}, and if no more than @option{frac} blocks (1
6607 meaning the whole image) differ by more than a threshold of @option{lo}.
6609 Default value for @option{hi} is 64*12, default value for @option{lo} is
6610 64*5, and default value for @option{frac} is 0.33.
6618 It accepts an integer in input; if non-zero it negates the
6619 alpha component (if available). The default value in input is 0.
6623 Force libavfilter not to use any of the specified pixel formats for the
6624 input to the next filter.
6626 It accepts the following parameters:
6630 A '|'-separated list of pixel format names, such as
6631 apix_fmts=yuv420p|monow|rgb24".
6635 @subsection Examples
6639 Force libavfilter to use a format different from @var{yuv420p} for the
6640 input to the vflip filter:
6642 noformat=pix_fmts=yuv420p,vflip
6646 Convert the input video to any of the formats not contained in the list:
6648 noformat=yuv420p|yuv444p|yuv410p
6654 Add noise on video input frame.
6656 The filter accepts the following options:
6664 Set noise seed for specific pixel component or all pixel components in case
6665 of @var{all_seed}. Default value is @code{123457}.
6667 @item all_strength, alls
6668 @item c0_strength, c0s
6669 @item c1_strength, c1s
6670 @item c2_strength, c2s
6671 @item c3_strength, c3s
6672 Set noise strength for specific pixel component or all pixel components in case
6673 @var{all_strength}. Default value is @code{0}. Allowed range is [0, 100].
6675 @item all_flags, allf
6680 Set pixel component flags or set flags for all components if @var{all_flags}.
6681 Available values for component flags are:
6684 averaged temporal noise (smoother)
6686 mix random noise with a (semi)regular pattern
6688 temporal noise (noise pattern changes between frames)
6690 uniform noise (gaussian otherwise)
6694 @subsection Examples
6696 Add temporal and uniform noise to input video:
6698 noise=alls=20:allf=t+u
6703 Pass the video source unchanged to the output.
6707 Apply a video transform using libopencv.
6709 To enable this filter, install the libopencv library and headers and
6710 configure FFmpeg with @code{--enable-libopencv}.
6712 It accepts the following parameters:
6717 The name of the libopencv filter to apply.
6720 The parameters to pass to the libopencv filter. If not specified, the default
6725 Refer to the official libopencv documentation for more precise
6727 @url{http://docs.opencv.org/master/modules/imgproc/doc/filtering.html}
6729 Several libopencv filters are supported; see the following subsections.
6734 Dilate an image by using a specific structuring element.
6735 It corresponds to the libopencv function @code{cvDilate}.
6737 It accepts the parameters: @var{struct_el}|@var{nb_iterations}.
6739 @var{struct_el} represents a structuring element, and has the syntax:
6740 @var{cols}x@var{rows}+@var{anchor_x}x@var{anchor_y}/@var{shape}
6742 @var{cols} and @var{rows} represent the number of columns and rows of
6743 the structuring element, @var{anchor_x} and @var{anchor_y} the anchor
6744 point, and @var{shape} the shape for the structuring element. @var{shape}
6745 must be "rect", "cross", "ellipse", or "custom".
6747 If the value for @var{shape} is "custom", it must be followed by a
6748 string of the form "=@var{filename}". The file with name
6749 @var{filename} is assumed to represent a binary image, with each
6750 printable character corresponding to a bright pixel. When a custom
6751 @var{shape} is used, @var{cols} and @var{rows} are ignored, the number
6752 or columns and rows of the read file are assumed instead.
6754 The default value for @var{struct_el} is "3x3+0x0/rect".
6756 @var{nb_iterations} specifies the number of times the transform is
6757 applied to the image, and defaults to 1.
6761 # Use the default values
6764 # Dilate using a structuring element with a 5x5 cross, iterating two times
6765 ocv=filter_name=dilate:filter_params=5x5+2x2/cross|2
6767 # Read the shape from the file diamond.shape, iterating two times.
6768 # The file diamond.shape may contain a pattern of characters like this
6774 # The specified columns and rows are ignored
6775 # but the anchor point coordinates are not
6776 ocv=dilate:0x0+2x2/custom=diamond.shape|2
6781 Erode an image by using a specific structuring element.
6782 It corresponds to the libopencv function @code{cvErode}.
6784 It accepts the parameters: @var{struct_el}:@var{nb_iterations},
6785 with the same syntax and semantics as the @ref{dilate} filter.
6789 Smooth the input video.
6791 The filter takes the following parameters:
6792 @var{type}|@var{param1}|@var{param2}|@var{param3}|@var{param4}.
6794 @var{type} is the type of smooth filter to apply, and must be one of
6795 the following values: "blur", "blur_no_scale", "median", "gaussian",
6796 or "bilateral". The default value is "gaussian".
6798 The meaning of @var{param1}, @var{param2}, @var{param3}, and @var{param4}
6799 depend on the smooth type. @var{param1} and
6800 @var{param2} accept integer positive values or 0. @var{param3} and
6801 @var{param4} accept floating point values.
6803 The default value for @var{param1} is 3. The default value for the
6804 other parameters is 0.
6806 These parameters correspond to the parameters assigned to the
6807 libopencv function @code{cvSmooth}.
6812 Overlay one video on top of another.
6814 It takes two inputs and has one output. The first input is the "main"
6815 video on which the second input is overlaid.
6817 It accepts the following parameters:
6819 A description of the accepted options follows.
6824 Set the expression for the x and y coordinates of the overlaid video
6825 on the main video. Default value is "0" for both expressions. In case
6826 the expression is invalid, it is set to a huge value (meaning that the
6827 overlay will not be displayed within the output visible area).
6830 The action to take when EOF is encountered on the secondary input; it accepts
6831 one of the following values:
6835 Repeat the last frame (the default).
6839 Pass the main input through.
6843 Set when the expressions for @option{x}, and @option{y} are evaluated.
6845 It accepts the following values:
6848 only evaluate expressions once during the filter initialization or
6849 when a command is processed
6852 evaluate expressions for each incoming frame
6855 Default value is @samp{frame}.
6858 If set to 1, force the output to terminate when the shortest input
6859 terminates. Default value is 0.
6862 Set the format for the output video.
6864 It accepts the following values:
6879 Default value is @samp{yuv420}.
6881 @item rgb @emph{(deprecated)}
6882 If set to 1, force the filter to accept inputs in the RGB
6883 color space. Default value is 0. This option is deprecated, use
6884 @option{format} instead.
6887 If set to 1, force the filter to draw the last overlay frame over the
6888 main input until the end of the stream. A value of 0 disables this
6889 behavior. Default value is 1.
6892 The @option{x}, and @option{y} expressions can contain the following
6898 The main input width and height.
6902 The overlay input width and height.
6906 The computed values for @var{x} and @var{y}. They are evaluated for
6911 horizontal and vertical chroma subsample values of the output
6912 format. For example for the pixel format "yuv422p" @var{hsub} is 2 and
6916 the number of input frame, starting from 0
6919 the position in the file of the input frame, NAN if unknown
6922 The timestamp, expressed in seconds. It's NAN if the input timestamp is unknown.
6926 Note that the @var{n}, @var{pos}, @var{t} variables are available only
6927 when evaluation is done @emph{per frame}, and will evaluate to NAN
6928 when @option{eval} is set to @samp{init}.
6930 Be aware that frames are taken from each input video in timestamp
6931 order, hence, if their initial timestamps differ, it is a good idea
6932 to pass the two inputs through a @var{setpts=PTS-STARTPTS} filter to
6933 have them begin in the same zero timestamp, as the example for
6934 the @var{movie} filter does.
6936 You can chain together more overlays but you should test the
6937 efficiency of such approach.
6939 @subsection Commands
6941 This filter supports the following commands:
6945 Modify the x and y of the overlay input.
6946 The command accepts the same syntax of the corresponding option.
6948 If the specified expression is not valid, it is kept at its current
6952 @subsection Examples
6956 Draw the overlay at 10 pixels from the bottom right corner of the main
6959 overlay=main_w-overlay_w-10:main_h-overlay_h-10
6962 Using named options the example above becomes:
6964 overlay=x=main_w-overlay_w-10:y=main_h-overlay_h-10
6968 Insert a transparent PNG logo in the bottom left corner of the input,
6969 using the @command{ffmpeg} tool with the @code{-filter_complex} option:
6971 ffmpeg -i input -i logo -filter_complex 'overlay=10:main_h-overlay_h-10' output
6975 Insert 2 different transparent PNG logos (second logo on bottom
6976 right corner) using the @command{ffmpeg} tool:
6978 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
6982 Add a transparent color layer on top of the main video; @code{WxH}
6983 must specify the size of the main input to the overlay filter:
6985 color=color=red@@.3:size=WxH [over]; [in][over] overlay [out]
6989 Play an original video and a filtered version (here with the deshake
6990 filter) side by side using the @command{ffplay} tool:
6992 ffplay input.avi -vf 'split[a][b]; [a]pad=iw*2:ih[src]; [b]deshake[filt]; [src][filt]overlay=w'
6995 The above command is the same as:
6997 ffplay input.avi -vf 'split[b], pad=iw*2[src], [b]deshake, [src]overlay=w'
7001 Make a sliding overlay appearing from the left to the right top part of the
7002 screen starting since time 2:
7004 overlay=x='if(gte(t,2), -w+(t-2)*20, NAN)':y=0
7008 Compose output by putting two input videos side to side:
7010 ffmpeg -i left.avi -i right.avi -filter_complex "
7011 nullsrc=size=200x100 [background];
7012 [0:v] setpts=PTS-STARTPTS, scale=100x100 [left];
7013 [1:v] setpts=PTS-STARTPTS, scale=100x100 [right];
7014 [background][left] overlay=shortest=1 [background+left];
7015 [background+left][right] overlay=shortest=1:x=100 [left+right]
7020 Mask 10-20 seconds of a video by applying the delogo filter to a section
7022 ffmpeg -i test.avi -codec:v:0 wmv2 -ar 11025 -b:v 9000k
7023 -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]'
7028 Chain several overlays in cascade:
7030 nullsrc=s=200x200 [bg];
7031 testsrc=s=100x100, split=4 [in0][in1][in2][in3];
7032 [in0] lutrgb=r=0, [bg] overlay=0:0 [mid0];
7033 [in1] lutrgb=g=0, [mid0] overlay=100:0 [mid1];
7034 [in2] lutrgb=b=0, [mid1] overlay=0:100 [mid2];
7035 [in3] null, [mid2] overlay=100:100 [out0]
7042 Apply Overcomplete Wavelet denoiser.
7044 The filter accepts the following options:
7050 Larger depth values will denoise lower frequency components more, but
7051 slow down filtering.
7053 Must be an int in the range 8-16, default is @code{8}.
7055 @item luma_strength, ls
7058 Must be a double value in the range 0-1000, default is @code{1.0}.
7060 @item chroma_strength, cs
7061 Set chroma strength.
7063 Must be a double value in the range 0-1000, default is @code{1.0}.
7068 Add paddings to the input image, and place the original input at the
7069 provided @var{x}, @var{y} coordinates.
7071 It accepts the following parameters:
7076 Specify an expression for the size of the output image with the
7077 paddings added. If the value for @var{width} or @var{height} is 0, the
7078 corresponding input size is used for the output.
7080 The @var{width} expression can reference the value set by the
7081 @var{height} expression, and vice versa.
7083 The default value of @var{width} and @var{height} is 0.
7087 Specify the offsets to place the input image at within the padded area,
7088 with respect to the top/left border of the output image.
7090 The @var{x} expression can reference the value set by the @var{y}
7091 expression, and vice versa.
7093 The default value of @var{x} and @var{y} is 0.
7096 Specify the color of the padded area. For the syntax of this option,
7097 check the "Color" section in the ffmpeg-utils manual.
7099 The default value of @var{color} is "black".
7102 The value for the @var{width}, @var{height}, @var{x}, and @var{y}
7103 options are expressions containing the following constants:
7108 The input video width and height.
7112 These are the same as @var{in_w} and @var{in_h}.
7116 The output width and height (the size of the padded area), as
7117 specified by the @var{width} and @var{height} expressions.
7121 These are the same as @var{out_w} and @var{out_h}.
7125 The x and y offsets as specified by the @var{x} and @var{y}
7126 expressions, or NAN if not yet specified.
7129 same as @var{iw} / @var{ih}
7132 input sample aspect ratio
7135 input display aspect ratio, it is the same as (@var{iw} / @var{ih}) * @var{sar}
7139 The horizontal and vertical chroma subsample values. For example for the
7140 pixel format "yuv422p" @var{hsub} is 2 and @var{vsub} is 1.
7143 @subsection Examples
7147 Add paddings with the color "violet" to the input video. The output video
7148 size is 640x480, and the top-left corner of the input video is placed at
7151 pad=640:480:0:40:violet
7154 The example above is equivalent to the following command:
7156 pad=width=640:height=480:x=0:y=40:color=violet
7160 Pad the input to get an output with dimensions increased by 3/2,
7161 and put the input video at the center of the padded area:
7163 pad="3/2*iw:3/2*ih:(ow-iw)/2:(oh-ih)/2"
7167 Pad the input to get a squared output with size equal to the maximum
7168 value between the input width and height, and put the input video at
7169 the center of the padded area:
7171 pad="max(iw\,ih):ow:(ow-iw)/2:(oh-ih)/2"
7175 Pad the input to get a final w/h ratio of 16:9:
7177 pad="ih*16/9:ih:(ow-iw)/2:(oh-ih)/2"
7181 In case of anamorphic video, in order to set the output display aspect
7182 correctly, it is necessary to use @var{sar} in the expression,
7183 according to the relation:
7185 (ih * X / ih) * sar = output_dar
7186 X = output_dar / sar
7189 Thus the previous example needs to be modified to:
7191 pad="ih*16/9/sar:ih:(ow-iw)/2:(oh-ih)/2"
7195 Double the output size and put the input video in the bottom-right
7196 corner of the output padded area:
7198 pad="2*iw:2*ih:ow-iw:oh-ih"
7205 Generate one palette for a whole video stream.
7207 It accepts the following options:
7211 Set the maximum number of colors to quantize in the palette.
7212 Note: the palette will still contain 256 colors; the unused palette entries
7215 @item reserve_transparent
7216 Create a palette of 255 colors maximum and reserve the last one for
7217 transparency. Reserving the transparency color is useful for GIF optimization.
7218 If not set, the maximum of colors in the palette will be 256. You probably want
7219 to disable this option for a standalone image.
7223 Set statistics mode.
7225 It accepts the following values:
7228 Compute full frame histograms.
7230 Compute histograms only for the part that differs from previous frame. This
7231 might be relevant to give more importance to the moving part of your input if
7232 the background is static.
7235 Default value is @var{full}.
7238 The filter also exports the frame metadata @code{lavfi.color_quant_ratio}
7239 (@code{nb_color_in / nb_color_out}) which you can use to evaluate the degree of
7240 color quantization of the palette. This information is also visible at
7241 @var{info} logging level.
7243 @subsection Examples
7247 Generate a representative palette of a given video using @command{ffmpeg}:
7249 ffmpeg -i input.mkv -vf palettegen palette.png
7255 Use a palette to downsample an input video stream.
7257 The filter takes two inputs: one video stream and a palette. The palette must
7258 be a 256 pixels image.
7260 It accepts the following options:
7264 Select dithering mode. Available algorithms are:
7267 Ordered 8x8 bayer dithering (deterministic)
7269 Dithering as defined by Paul Heckbert in 1982 (simple error diffusion).
7270 Note: this dithering is sometimes considered "wrong" and is included as a
7272 @item floyd_steinberg
7273 Floyd and Steingberg dithering (error diffusion)
7275 Frankie Sierra dithering v2 (error diffusion)
7277 Frankie Sierra dithering v2 "Lite" (error diffusion)
7280 Default is @var{sierra2_4a}.
7283 When @var{bayer} dithering is selected, this option defines the scale of the
7284 pattern (how much the crosshatch pattern is visible). A low value means more
7285 visible pattern for less banding, and higher value means less visible pattern
7286 at the cost of more banding.
7288 The option must be an integer value in the range [0,5]. Default is @var{2}.
7291 If set, define the zone to process
7295 Only the changing rectangle will be reprocessed. This is similar to GIF
7296 cropping/offsetting compression mechanism. This option can be useful for speed
7297 if only a part of the image is changing, and has use cases such as limiting the
7298 scope of the error diffusal @option{dither} to the rectangle that bounds the
7299 moving scene (it leads to more deterministic output if the scene doesn't change
7300 much, and as a result less moving noise and better GIF compression).
7303 Default is @var{none}.
7306 @subsection Examples
7310 Use a palette (generated for example with @ref{palettegen}) to encode a GIF
7311 using @command{ffmpeg}:
7313 ffmpeg -i input.mkv -i palette.png -lavfi paletteuse output.gif
7317 @section perspective
7319 Correct perspective of video not recorded perpendicular to the screen.
7321 A description of the accepted parameters follows.
7332 Set coordinates expression for top left, top right, bottom left and bottom right corners.
7333 Default values are @code{0:0:W:0:0:H:W:H} with which perspective will remain unchanged.
7334 If the @code{sense} option is set to @code{source}, then the specified points will be sent
7335 to the corners of the destination. If the @code{sense} option is set to @code{destination},
7336 then the corners of the source will be sent to the specified coordinates.
7338 The expressions can use the following variables:
7343 the width and height of video frame.
7347 Set interpolation for perspective correction.
7349 It accepts the following values:
7355 Default value is @samp{linear}.
7358 Set interpretation of coordinate options.
7360 It accepts the following values:
7364 Send point in the source specified by the given coordinates to
7365 the corners of the destination.
7367 @item 1, destination
7369 Send the corners of the source to the point in the destination specified
7370 by the given coordinates.
7372 Default value is @samp{source}.
7378 Delay interlaced video by one field time so that the field order changes.
7380 The intended use is to fix PAL movies that have been captured with the
7381 opposite field order to the film-to-video transfer.
7383 A description of the accepted parameters follows.
7389 It accepts the following values:
7392 Capture field order top-first, transfer bottom-first.
7393 Filter will delay the bottom field.
7396 Capture field order bottom-first, transfer top-first.
7397 Filter will delay the top field.
7400 Capture and transfer with the same field order. This mode only exists
7401 for the documentation of the other options to refer to, but if you
7402 actually select it, the filter will faithfully do nothing.
7405 Capture field order determined automatically by field flags, transfer
7407 Filter selects among @samp{t} and @samp{b} modes on a frame by frame
7408 basis using field flags. If no field information is available,
7409 then this works just like @samp{u}.
7412 Capture unknown or varying, transfer opposite.
7413 Filter selects among @samp{t} and @samp{b} on a frame by frame basis by
7414 analyzing the images and selecting the alternative that produces best
7415 match between the fields.
7418 Capture top-first, transfer unknown or varying.
7419 Filter selects among @samp{t} and @samp{p} using image analysis.
7422 Capture bottom-first, transfer unknown or varying.
7423 Filter selects among @samp{b} and @samp{p} using image analysis.
7426 Capture determined by field flags, transfer unknown or varying.
7427 Filter selects among @samp{t}, @samp{b} and @samp{p} using field flags and
7428 image analysis. If no field information is available, then this works just
7429 like @samp{U}. This is the default mode.
7432 Both capture and transfer unknown or varying.
7433 Filter selects among @samp{t}, @samp{b} and @samp{p} using image analysis only.
7437 @section pixdesctest
7439 Pixel format descriptor test filter, mainly useful for internal
7440 testing. The output video should be equal to the input video.
7444 format=monow, pixdesctest
7447 can be used to test the monowhite pixel format descriptor definition.
7451 Enable the specified chain of postprocessing subfilters using libpostproc. This
7452 library should be automatically selected with a GPL build (@code{--enable-gpl}).
7453 Subfilters must be separated by '/' and can be disabled by prepending a '-'.
7454 Each subfilter and some options have a short and a long name that can be used
7455 interchangeably, i.e. dr/dering are the same.
7457 The filters accept the following options:
7461 Set postprocessing subfilters string.
7464 All subfilters share common options to determine their scope:
7468 Honor the quality commands for this subfilter.
7471 Do chrominance filtering, too (default).
7474 Do luminance filtering only (no chrominance).
7477 Do chrominance filtering only (no luminance).
7480 These options can be appended after the subfilter name, separated by a '|'.
7482 Available subfilters are:
7485 @item hb/hdeblock[|difference[|flatness]]
7486 Horizontal deblocking filter
7489 Difference factor where higher values mean more deblocking (default: @code{32}).
7491 Flatness threshold where lower values mean more deblocking (default: @code{39}).
7494 @item vb/vdeblock[|difference[|flatness]]
7495 Vertical deblocking filter
7498 Difference factor where higher values mean more deblocking (default: @code{32}).
7500 Flatness threshold where lower values mean more deblocking (default: @code{39}).
7503 @item ha/hadeblock[|difference[|flatness]]
7504 Accurate horizontal deblocking filter
7507 Difference factor where higher values mean more deblocking (default: @code{32}).
7509 Flatness threshold where lower values mean more deblocking (default: @code{39}).
7512 @item va/vadeblock[|difference[|flatness]]
7513 Accurate vertical deblocking filter
7516 Difference factor where higher values mean more deblocking (default: @code{32}).
7518 Flatness threshold where lower values mean more deblocking (default: @code{39}).
7522 The horizontal and vertical deblocking filters share the difference and
7523 flatness values so you cannot set different horizontal and vertical
7528 Experimental horizontal deblocking filter
7531 Experimental vertical deblocking filter
7536 @item tn/tmpnoise[|threshold1[|threshold2[|threshold3]]], temporal noise reducer
7539 larger -> stronger filtering
7541 larger -> stronger filtering
7543 larger -> stronger filtering
7546 @item al/autolevels[:f/fullyrange], automatic brightness / contrast correction
7549 Stretch luminance to @code{0-255}.
7552 @item lb/linblenddeint
7553 Linear blend deinterlacing filter that deinterlaces the given block by
7554 filtering all lines with a @code{(1 2 1)} filter.
7556 @item li/linipoldeint
7557 Linear interpolating deinterlacing filter that deinterlaces the given block by
7558 linearly interpolating every second line.
7560 @item ci/cubicipoldeint
7561 Cubic interpolating deinterlacing filter deinterlaces the given block by
7562 cubically interpolating every second line.
7564 @item md/mediandeint
7565 Median deinterlacing filter that deinterlaces the given block by applying a
7566 median filter to every second line.
7568 @item fd/ffmpegdeint
7569 FFmpeg deinterlacing filter that deinterlaces the given block by filtering every
7570 second line with a @code{(-1 4 2 4 -1)} filter.
7573 Vertically applied FIR lowpass deinterlacing filter that deinterlaces the given
7574 block by filtering all lines with a @code{(-1 2 6 2 -1)} filter.
7576 @item fq/forceQuant[|quantizer]
7577 Overrides the quantizer table from the input with the constant quantizer you
7585 Default pp filter combination (@code{hb|a,vb|a,dr|a})
7588 Fast pp filter combination (@code{h1|a,v1|a,dr|a})
7591 High quality pp filter combination (@code{ha|a|128|7,va|a,dr|a})
7594 @subsection Examples
7598 Apply horizontal and vertical deblocking, deringing and automatic
7599 brightness/contrast:
7605 Apply default filters without brightness/contrast correction:
7611 Apply default filters and temporal denoiser:
7613 pp=default/tmpnoise|1|2|3
7617 Apply deblocking on luminance only, and switch vertical deblocking on or off
7618 automatically depending on available CPU time:
7625 Apply Postprocessing filter 7. It is variant of the @ref{spp} filter,
7626 similar to spp = 6 with 7 point DCT, where only the center sample is
7629 The filter accepts the following options:
7633 Force a constant quantization parameter. It accepts an integer in range
7634 0 to 63. If not set, the filter will use the QP from the video stream
7638 Set thresholding mode. Available modes are:
7642 Set hard thresholding.
7644 Set soft thresholding (better de-ringing effect, but likely blurrier).
7646 Set medium thresholding (good results, default).
7652 Obtain the average, maximum and minimum PSNR (Peak Signal to Noise
7653 Ratio) between two input videos.
7655 This filter takes in input two input videos, the first input is
7656 considered the "main" source and is passed unchanged to the
7657 output. The second input is used as a "reference" video for computing
7660 Both video inputs must have the same resolution and pixel format for
7661 this filter to work correctly. Also it assumes that both inputs
7662 have the same number of frames, which are compared one by one.
7664 The obtained average PSNR is printed through the logging system.
7666 The filter stores the accumulated MSE (mean squared error) of each
7667 frame, and at the end of the processing it is averaged across all frames
7668 equally, and the following formula is applied to obtain the PSNR:
7671 PSNR = 10*log10(MAX^2/MSE)
7674 Where MAX is the average of the maximum values of each component of the
7677 The description of the accepted parameters follows.
7681 If specified the filter will use the named file to save the PSNR of
7682 each individual frame.
7685 The file printed if @var{stats_file} is selected, contains a sequence of
7686 key/value pairs of the form @var{key}:@var{value} for each compared
7689 A description of each shown parameter follows:
7693 sequential number of the input frame, starting from 1
7696 Mean Square Error pixel-by-pixel average difference of the compared
7697 frames, averaged over all the image components.
7699 @item mse_y, mse_u, mse_v, mse_r, mse_g, mse_g, mse_a
7700 Mean Square Error pixel-by-pixel average difference of the compared
7701 frames for the component specified by the suffix.
7703 @item psnr_y, psnr_u, psnr_v, psnr_r, psnr_g, psnr_b, psnr_a
7704 Peak Signal to Noise ratio of the compared frames for the component
7705 specified by the suffix.
7710 movie=ref_movie.mpg, setpts=PTS-STARTPTS [main];
7711 [main][ref] psnr="stats_file=stats.log" [out]
7714 On this example the input file being processed is compared with the
7715 reference file @file{ref_movie.mpg}. The PSNR of each individual frame
7716 is stored in @file{stats.log}.
7721 Pulldown reversal (inverse telecine) filter, capable of handling mixed
7722 hard-telecine, 24000/1001 fps progressive, and 30000/1001 fps progressive
7725 The pullup filter is designed to take advantage of future context in making
7726 its decisions. This filter is stateless in the sense that it does not lock
7727 onto a pattern to follow, but it instead looks forward to the following
7728 fields in order to identify matches and rebuild progressive frames.
7730 To produce content with an even framerate, insert the fps filter after
7731 pullup, use @code{fps=24000/1001} if the input frame rate is 29.97fps,
7732 @code{fps=24} for 30fps and the (rare) telecined 25fps input.
7734 The filter accepts the following options:
7741 These options set the amount of "junk" to ignore at the left, right, top, and
7742 bottom of the image, respectively. Left and right are in units of 8 pixels,
7743 while top and bottom are in units of 2 lines.
7744 The default is 8 pixels on each side.
7747 Set the strict breaks. Setting this option to 1 will reduce the chances of
7748 filter generating an occasional mismatched frame, but it may also cause an
7749 excessive number of frames to be dropped during high motion sequences.
7750 Conversely, setting it to -1 will make filter match fields more easily.
7751 This may help processing of video where there is slight blurring between
7752 the fields, but may also cause there to be interlaced frames in the output.
7753 Default value is @code{0}.
7756 Set the metric plane to use. It accepts the following values:
7762 Use chroma blue plane.
7765 Use chroma red plane.
7768 This option may be set to use chroma plane instead of the default luma plane
7769 for doing filter's computations. This may improve accuracy on very clean
7770 source material, but more likely will decrease accuracy, especially if there
7771 is chroma noise (rainbow effect) or any grayscale video.
7772 The main purpose of setting @option{mp} to a chroma plane is to reduce CPU
7773 load and make pullup usable in realtime on slow machines.
7776 For best results (without duplicated frames in the output file) it is
7777 necessary to change the output frame rate. For example, to inverse
7778 telecine NTSC input:
7780 ffmpeg -i input -vf pullup -r 24000/1001 ...
7785 Change video quantization parameters (QP).
7787 The filter accepts the following option:
7791 Set expression for quantization parameter.
7794 The expression is evaluated through the eval API and can contain, among others,
7795 the following constants:
7799 1 if index is not 129, 0 otherwise.
7802 Sequentional index starting from -129 to 128.
7805 @subsection Examples
7817 Suppress a TV station logo, using an image file to determine which
7818 pixels comprise the logo. It works by filling in the pixels that
7819 comprise the logo with neighboring pixels.
7821 The filter accepts the following options:
7825 Set the filter bitmap file, which can be any image format supported by
7826 libavformat. The width and height of the image file must match those of the
7827 video stream being processed.
7830 Pixels in the provided bitmap image with a value of zero are not
7831 considered part of the logo, non-zero pixels are considered part of
7832 the logo. If you use white (255) for the logo and black (0) for the
7833 rest, you will be safe. For making the filter bitmap, it is
7834 recommended to take a screen capture of a black frame with the logo
7835 visible, and then using a threshold filter followed by the erode
7836 filter once or twice.
7838 If needed, little splotches can be fixed manually. Remember that if
7839 logo pixels are not covered, the filter quality will be much
7840 reduced. Marking too many pixels as part of the logo does not hurt as
7841 much, but it will increase the amount of blurring needed to cover over
7842 the image and will destroy more information than necessary, and extra
7843 pixels will slow things down on a large logo.
7845 @section repeatfields
7847 This filter uses the repeat_field flag from the Video ES headers and hard repeats
7848 fields based on its value.
7852 Rotate video by an arbitrary angle expressed in radians.
7854 The filter accepts the following options:
7856 A description of the optional parameters follows.
7859 Set an expression for the angle by which to rotate the input video
7860 clockwise, expressed as a number of radians. A negative value will
7861 result in a counter-clockwise rotation. By default it is set to "0".
7863 This expression is evaluated for each frame.
7866 Set the output width expression, default value is "iw".
7867 This expression is evaluated just once during configuration.
7870 Set the output height expression, default value is "ih".
7871 This expression is evaluated just once during configuration.
7874 Enable bilinear interpolation if set to 1, a value of 0 disables
7875 it. Default value is 1.
7878 Set the color used to fill the output area not covered by the rotated
7879 image. For the general syntax of this option, check the "Color" section in the
7880 ffmpeg-utils manual. If the special value "none" is selected then no
7881 background is printed (useful for example if the background is never shown).
7883 Default value is "black".
7886 The expressions for the angle and the output size can contain the
7887 following constants and functions:
7891 sequential number of the input frame, starting from 0. It is always NAN
7892 before the first frame is filtered.
7895 time in seconds of the input frame, it is set to 0 when the filter is
7896 configured. It is always NAN before the first frame is filtered.
7900 horizontal and vertical chroma subsample values. For example for the
7901 pixel format "yuv422p" @var{hsub} is 2 and @var{vsub} is 1.
7905 the input video width and height
7909 the output width and height, that is the size of the padded area as
7910 specified by the @var{width} and @var{height} expressions
7914 the minimal width/height required for completely containing the input
7915 video rotated by @var{a} radians.
7917 These are only available when computing the @option{out_w} and
7918 @option{out_h} expressions.
7921 @subsection Examples
7925 Rotate the input by PI/6 radians clockwise:
7931 Rotate the input by PI/6 radians counter-clockwise:
7937 Rotate the input by 45 degrees clockwise:
7943 Apply a constant rotation with period T, starting from an angle of PI/3:
7945 rotate=PI/3+2*PI*t/T
7949 Make the input video rotation oscillating with a period of T
7950 seconds and an amplitude of A radians:
7952 rotate=A*sin(2*PI/T*t)
7956 Rotate the video, output size is chosen so that the whole rotating
7957 input video is always completely contained in the output:
7959 rotate='2*PI*t:ow=hypot(iw,ih):oh=ow'
7963 Rotate the video, reduce the output size so that no background is ever
7966 rotate=2*PI*t:ow='min(iw,ih)/sqrt(2)':oh=ow:c=none
7970 @subsection Commands
7972 The filter supports the following commands:
7976 Set the angle expression.
7977 The command accepts the same syntax of the corresponding option.
7979 If the specified expression is not valid, it is kept at its current
7985 Apply Shape Adaptive Blur.
7987 The filter accepts the following options:
7990 @item luma_radius, lr
7991 Set luma blur filter strength, must be a value in range 0.1-4.0, default
7992 value is 1.0. A greater value will result in a more blurred image, and
7993 in slower processing.
7995 @item luma_pre_filter_radius, lpfr
7996 Set luma pre-filter radius, must be a value in the 0.1-2.0 range, default
7999 @item luma_strength, ls
8000 Set luma maximum difference between pixels to still be considered, must
8001 be a value in the 0.1-100.0 range, default value is 1.0.
8003 @item chroma_radius, cr
8004 Set chroma blur filter strength, must be a value in range 0.1-4.0. A
8005 greater value will result in a more blurred image, and in slower
8008 @item chroma_pre_filter_radius, cpfr
8009 Set chroma pre-filter radius, must be a value in the 0.1-2.0 range.
8011 @item chroma_strength, cs
8012 Set chroma maximum difference between pixels to still be considered,
8013 must be a value in the 0.1-100.0 range.
8016 Each chroma option value, if not explicitly specified, is set to the
8017 corresponding luma option value.
8022 Scale (resize) the input video, using the libswscale library.
8024 The scale filter forces the output display aspect ratio to be the same
8025 of the input, by changing the output sample aspect ratio.
8027 If the input image format is different from the format requested by
8028 the next filter, the scale filter will convert the input to the
8032 The filter accepts the following options, or any of the options
8033 supported by the libswscale scaler.
8035 See @ref{scaler_options,,the ffmpeg-scaler manual,ffmpeg-scaler} for
8036 the complete list of scaler options.
8041 Set the output video dimension expression. Default value is the input
8044 If the value is 0, the input width is used for the output.
8046 If one of the values is -1, the scale filter will use a value that
8047 maintains the aspect ratio of the input image, calculated from the
8048 other specified dimension. If both of them are -1, the input size is
8051 If one of the values is -n with n > 1, the scale filter will also use a value
8052 that maintains the aspect ratio of the input image, calculated from the other
8053 specified dimension. After that it will, however, make sure that the calculated
8054 dimension is divisible by n and adjust the value if necessary.
8056 See below for the list of accepted constants for use in the dimension
8060 Set the interlacing mode. It accepts the following values:
8064 Force interlaced aware scaling.
8067 Do not apply interlaced scaling.
8070 Select interlaced aware scaling depending on whether the source frames
8071 are flagged as interlaced or not.
8074 Default value is @samp{0}.
8077 Set libswscale scaling flags. See
8078 @ref{sws_flags,,the ffmpeg-scaler manual,ffmpeg-scaler} for the
8079 complete list of values. If not explicitly specified the filter applies
8083 Set the video size. For the syntax of this option, check the
8084 @ref{video size syntax,,"Video size" section in the ffmpeg-utils manual,ffmpeg-utils}.
8086 @item in_color_matrix
8087 @item out_color_matrix
8088 Set in/output YCbCr color space type.
8090 This allows the autodetected value to be overridden as well as allows forcing
8091 a specific value used for the output and encoder.
8093 If not specified, the color space type depends on the pixel format.
8099 Choose automatically.
8102 Format conforming to International Telecommunication Union (ITU)
8103 Recommendation BT.709.
8106 Set color space conforming to the United States Federal Communications
8107 Commission (FCC) Code of Federal Regulations (CFR) Title 47 (2003) 73.682 (a).
8110 Set color space conforming to:
8114 ITU Radiocommunication Sector (ITU-R) Recommendation BT.601
8117 ITU-R Rec. BT.470-6 (1998) Systems B, B1, and G
8120 Society of Motion Picture and Television Engineers (SMPTE) ST 170:2004
8125 Set color space conforming to SMPTE ST 240:1999.
8130 Set in/output YCbCr sample range.
8132 This allows the autodetected value to be overridden as well as allows forcing
8133 a specific value used for the output and encoder. If not specified, the
8134 range depends on the pixel format. Possible values:
8138 Choose automatically.
8141 Set full range (0-255 in case of 8-bit luma).
8144 Set "MPEG" range (16-235 in case of 8-bit luma).
8147 @item force_original_aspect_ratio
8148 Enable decreasing or increasing output video width or height if necessary to
8149 keep the original aspect ratio. Possible values:
8153 Scale the video as specified and disable this feature.
8156 The output video dimensions will automatically be decreased if needed.
8159 The output video dimensions will automatically be increased if needed.
8163 One useful instance of this option is that when you know a specific device's
8164 maximum allowed resolution, you can use this to limit the output video to
8165 that, while retaining the aspect ratio. For example, device A allows
8166 1280x720 playback, and your video is 1920x800. Using this option (set it to
8167 decrease) and specifying 1280x720 to the command line makes the output
8170 Please note that this is a different thing than specifying -1 for @option{w}
8171 or @option{h}, you still need to specify the output resolution for this option
8176 The values of the @option{w} and @option{h} options are expressions
8177 containing the following constants:
8182 The input width and height
8186 These are the same as @var{in_w} and @var{in_h}.
8190 The output (scaled) width and height
8194 These are the same as @var{out_w} and @var{out_h}
8197 The same as @var{iw} / @var{ih}
8200 input sample aspect ratio
8203 The input display aspect ratio. Calculated from @code{(iw / ih) * sar}.
8207 horizontal and vertical input chroma subsample values. For example for the
8208 pixel format "yuv422p" @var{hsub} is 2 and @var{vsub} is 1.
8212 horizontal and vertical output chroma subsample values. For example for the
8213 pixel format "yuv422p" @var{hsub} is 2 and @var{vsub} is 1.
8216 @subsection Examples
8220 Scale the input video to a size of 200x100
8225 This is equivalent to:
8236 Specify a size abbreviation for the output size:
8241 which can also be written as:
8247 Scale the input to 2x:
8253 The above is the same as:
8259 Scale the input to 2x with forced interlaced scaling:
8261 scale=2*iw:2*ih:interl=1
8265 Scale the input to half size:
8271 Increase the width, and set the height to the same size:
8284 Increase the height, and set the width to 3/2 of the height:
8286 scale=w=3/2*oh:h=3/5*ih
8290 Increase the size, making the size a multiple of the chroma
8293 scale="trunc(3/2*iw/hsub)*hsub:trunc(3/2*ih/vsub)*vsub"
8297 Increase the width to a maximum of 500 pixels,
8298 keeping the same aspect ratio as the input:
8300 scale=w='min(500\, iw*3/2):h=-1'
8304 @section separatefields
8306 The @code{separatefields} takes a frame-based video input and splits
8307 each frame into its components fields, producing a new half height clip
8308 with twice the frame rate and twice the frame count.
8310 This filter use field-dominance information in frame to decide which
8311 of each pair of fields to place first in the output.
8312 If it gets it wrong use @ref{setfield} filter before @code{separatefields} filter.
8314 @section setdar, setsar
8316 The @code{setdar} filter sets the Display Aspect Ratio for the filter
8319 This is done by changing the specified Sample (aka Pixel) Aspect
8320 Ratio, according to the following equation:
8322 @var{DAR} = @var{HORIZONTAL_RESOLUTION} / @var{VERTICAL_RESOLUTION} * @var{SAR}
8325 Keep in mind that the @code{setdar} filter does not modify the pixel
8326 dimensions of the video frame. Also, the display aspect ratio set by
8327 this filter may be changed by later filters in the filterchain,
8328 e.g. in case of scaling or if another "setdar" or a "setsar" filter is
8331 The @code{setsar} filter sets the Sample (aka Pixel) Aspect Ratio for
8332 the filter output video.
8334 Note that as a consequence of the application of this filter, the
8335 output display aspect ratio will change according to the equation
8338 Keep in mind that the sample aspect ratio set by the @code{setsar}
8339 filter may be changed by later filters in the filterchain, e.g. if
8340 another "setsar" or a "setdar" filter is applied.
8342 It accepts the following parameters:
8345 @item r, ratio, dar (@code{setdar} only), sar (@code{setsar} only)
8346 Set the aspect ratio used by the filter.
8348 The parameter can be a floating point number string, an expression, or
8349 a string of the form @var{num}:@var{den}, where @var{num} and
8350 @var{den} are the numerator and denominator of the aspect ratio. If
8351 the parameter is not specified, it is assumed the value "0".
8352 In case the form "@var{num}:@var{den}" is used, the @code{:} character
8356 Set the maximum integer value to use for expressing numerator and
8357 denominator when reducing the expressed aspect ratio to a rational.
8358 Default value is @code{100}.
8362 The parameter @var{sar} is an expression containing
8363 the following constants:
8367 These are approximated values for the mathematical constants e
8368 (Euler's number), pi (Greek pi), and phi (the golden ratio).
8371 The input width and height.
8374 These are the same as @var{w} / @var{h}.
8377 The input sample aspect ratio.
8380 The input display aspect ratio. It is the same as
8381 (@var{w} / @var{h}) * @var{sar}.
8384 Horizontal and vertical chroma subsample values. For example, for the
8385 pixel format "yuv422p" @var{hsub} is 2 and @var{vsub} is 1.
8388 @subsection Examples
8393 To change the display aspect ratio to 16:9, specify one of the following:
8401 To change the sample aspect ratio to 10:11, specify:
8407 To set a display aspect ratio of 16:9, and specify a maximum integer value of
8408 1000 in the aspect ratio reduction, use the command:
8410 setdar=ratio=16/9:max=1000
8418 Force field for the output video frame.
8420 The @code{setfield} filter marks the interlace type field for the
8421 output frames. It does not change the input frame, but only sets the
8422 corresponding property, which affects how the frame is treated by
8423 following filters (e.g. @code{fieldorder} or @code{yadif}).
8425 The filter accepts the following options:
8430 Available values are:
8434 Keep the same field property.
8437 Mark the frame as bottom-field-first.
8440 Mark the frame as top-field-first.
8443 Mark the frame as progressive.
8449 Show a line containing various information for each input video frame.
8450 The input video is not modified.
8452 The shown line contains a sequence of key/value pairs of the form
8453 @var{key}:@var{value}.
8455 The following values are shown in the output:
8459 The (sequential) number of the input frame, starting from 0.
8462 The Presentation TimeStamp of the input frame, expressed as a number of
8463 time base units. The time base unit depends on the filter input pad.
8466 The Presentation TimeStamp of the input frame, expressed as a number of
8470 The position of the frame in the input stream, or -1 if this information is
8471 unavailable and/or meaningless (for example in case of synthetic video).
8474 The pixel format name.
8477 The sample aspect ratio of the input frame, expressed in the form
8478 @var{num}/@var{den}.
8481 The size of the input frame. For the syntax of this option, check the
8482 @ref{video size syntax,,"Video size" section in the ffmpeg-utils manual,ffmpeg-utils}.
8485 The type of interlaced mode ("P" for "progressive", "T" for top field first, "B"
8486 for bottom field first).
8489 This is 1 if the frame is a key frame, 0 otherwise.
8492 The picture type of the input frame ("I" for an I-frame, "P" for a
8493 P-frame, "B" for a B-frame, or "?" for an unknown type).
8494 Also refer to the documentation of the @code{AVPictureType} enum and of
8495 the @code{av_get_picture_type_char} function defined in
8496 @file{libavutil/avutil.h}.
8499 The Adler-32 checksum (printed in hexadecimal) of all the planes of the input frame.
8501 @item plane_checksum
8502 The Adler-32 checksum (printed in hexadecimal) of each plane of the input frame,
8503 expressed in the form "[@var{c0} @var{c1} @var{c2} @var{c3}]".
8506 @section showpalette
8508 Displays the 256 colors palette of each frame. This filter is only relevant for
8509 @var{pal8} pixel format frames.
8511 It accepts the following option:
8515 Set the size of the box used to represent one palette color entry. Default is
8516 @code{30} (for a @code{30x30} pixel box).
8519 @section shuffleplanes
8521 Reorder and/or duplicate video planes.
8523 It accepts the following parameters:
8528 The index of the input plane to be used as the first output plane.
8531 The index of the input plane to be used as the second output plane.
8534 The index of the input plane to be used as the third output plane.
8537 The index of the input plane to be used as the fourth output plane.
8541 The first plane has the index 0. The default is to keep the input unchanged.
8543 Swap the second and third planes of the input:
8545 ffmpeg -i INPUT -vf shuffleplanes=0:2:1:3 OUTPUT
8548 @section signalstats
8549 Evaluate various visual metrics that assist in determining issues associated
8550 with the digitization of analog video media.
8552 By default the filter will log these metadata values:
8556 Display the minimal Y value contained within the input frame. Expressed in
8560 Display the Y value at the 10% percentile within the input frame. Expressed in
8564 Display the average Y value within the input frame. Expressed in range of
8568 Display the Y value at the 90% percentile within the input frame. Expressed in
8572 Display the maximum Y value contained within the input frame. Expressed in
8576 Display the minimal U value contained within the input frame. Expressed in
8580 Display the U value at the 10% percentile within the input frame. Expressed in
8584 Display the average U value within the input frame. Expressed in range of
8588 Display the U value at the 90% percentile within the input frame. Expressed in
8592 Display the maximum U value contained within the input frame. Expressed in
8596 Display the minimal V value contained within the input frame. Expressed in
8600 Display the V value at the 10% percentile within the input frame. Expressed in
8604 Display the average V value within the input frame. Expressed in range of
8608 Display the V value at the 90% percentile within the input frame. Expressed in
8612 Display the maximum V value contained within the input frame. Expressed in
8616 Display the minimal saturation value contained within the input frame.
8617 Expressed in range of [0-~181.02].
8620 Display the saturation value at the 10% percentile within the input frame.
8621 Expressed in range of [0-~181.02].
8624 Display the average saturation value within the input frame. Expressed in range
8628 Display the saturation value at the 90% percentile within the input frame.
8629 Expressed in range of [0-~181.02].
8632 Display the maximum saturation value contained within the input frame.
8633 Expressed in range of [0-~181.02].
8636 Display the median value for hue within the input frame. Expressed in range of
8640 Display the average value for hue within the input frame. Expressed in range of
8644 Display the average of sample value difference between all values of the Y
8645 plane in the current frame and corresponding values of the previous input frame.
8646 Expressed in range of [0-255].
8649 Display the average of sample value difference between all values of the U
8650 plane in the current frame and corresponding values of the previous input frame.
8651 Expressed in range of [0-255].
8654 Display the average of sample value difference between all values of the V
8655 plane in the current frame and corresponding values of the previous input frame.
8656 Expressed in range of [0-255].
8659 The filter accepts the following options:
8665 @option{stat} specify an additional form of image analysis.
8666 @option{out} output video with the specified type of pixel highlighted.
8668 Both options accept the following values:
8672 Identify @var{temporal outliers} pixels. A @var{temporal outlier} is a pixel
8673 unlike the neighboring pixels of the same field. Examples of temporal outliers
8674 include the results of video dropouts, head clogs, or tape tracking issues.
8677 Identify @var{vertical line repetition}. Vertical line repetition includes
8678 similar rows of pixels within a frame. In born-digital video vertical line
8679 repetition is common, but this pattern is uncommon in video digitized from an
8680 analog source. When it occurs in video that results from the digitization of an
8681 analog source it can indicate concealment from a dropout compensator.
8684 Identify pixels that fall outside of legal broadcast range.
8688 Set the highlight color for the @option{out} option. The default color is
8692 @subsection Examples
8696 Output data of various video metrics:
8698 ffprobe -f lavfi movie=example.mov,signalstats="stat=tout+vrep+brng" -show_frames
8702 Output specific data about the minimum and maximum values of the Y plane per frame:
8704 ffprobe -f lavfi movie=example.mov,signalstats -show_entries frame_tags=lavfi.signalstats.YMAX,lavfi.signalstats.YMIN
8708 Playback video while highlighting pixels that are outside of broadcast range in red.
8710 ffplay example.mov -vf signalstats="out=brng:color=red"
8714 Playback video with signalstats metadata drawn over the frame.
8716 ffplay example.mov -vf signalstats=stat=brng+vrep+tout,drawtext=fontfile=FreeSerif.ttf:textfile=signalstat_drawtext.txt
8719 The contents of signalstat_drawtext.txt used in the command are:
8722 Y (%@{metadata:lavfi.signalstats.YMIN@}-%@{metadata:lavfi.signalstats.YMAX@})
8723 U (%@{metadata:lavfi.signalstats.UMIN@}-%@{metadata:lavfi.signalstats.UMAX@})
8724 V (%@{metadata:lavfi.signalstats.VMIN@}-%@{metadata:lavfi.signalstats.VMAX@})
8725 saturation maximum: %@{metadata:lavfi.signalstats.SATMAX@}
8733 Blur the input video without impacting the outlines.
8735 It accepts the following options:
8738 @item luma_radius, lr
8739 Set the luma radius. The option value must be a float number in
8740 the range [0.1,5.0] that specifies the variance of the gaussian filter
8741 used to blur the image (slower if larger). Default value is 1.0.
8743 @item luma_strength, ls
8744 Set the luma strength. The option value must be a float number
8745 in the range [-1.0,1.0] that configures the blurring. A value included
8746 in [0.0,1.0] will blur the image whereas a value included in
8747 [-1.0,0.0] will sharpen the image. Default value is 1.0.
8749 @item luma_threshold, lt
8750 Set the luma threshold used as a coefficient to determine
8751 whether a pixel should be blurred or not. The option value must be an
8752 integer in the range [-30,30]. A value of 0 will filter all the image,
8753 a value included in [0,30] will filter flat areas and a value included
8754 in [-30,0] will filter edges. Default value is 0.
8756 @item chroma_radius, cr
8757 Set the chroma radius. The option value must be a float number in
8758 the range [0.1,5.0] that specifies the variance of the gaussian filter
8759 used to blur the image (slower if larger). Default value is 1.0.
8761 @item chroma_strength, cs
8762 Set the chroma strength. The option value must be a float number
8763 in the range [-1.0,1.0] that configures the blurring. A value included
8764 in [0.0,1.0] will blur the image whereas a value included in
8765 [-1.0,0.0] will sharpen the image. Default value is 1.0.
8767 @item chroma_threshold, ct
8768 Set the chroma threshold used as a coefficient to determine
8769 whether a pixel should be blurred or not. The option value must be an
8770 integer in the range [-30,30]. A value of 0 will filter all the image,
8771 a value included in [0,30] will filter flat areas and a value included
8772 in [-30,0] will filter edges. Default value is 0.
8775 If a chroma option is not explicitly set, the corresponding luma value
8780 Convert between different stereoscopic image formats.
8782 The filters accept the following options:
8786 Set stereoscopic image format of input.
8788 Available values for input image formats are:
8791 side by side parallel (left eye left, right eye right)
8794 side by side crosseye (right eye left, left eye right)
8797 side by side parallel with half width resolution
8798 (left eye left, right eye right)
8801 side by side crosseye with half width resolution
8802 (right eye left, left eye right)
8805 above-below (left eye above, right eye below)
8808 above-below (right eye above, left eye below)
8811 above-below with half height resolution
8812 (left eye above, right eye below)
8815 above-below with half height resolution
8816 (right eye above, left eye below)
8819 alternating frames (left eye first, right eye second)
8822 alternating frames (right eye first, left eye second)
8824 Default value is @samp{sbsl}.
8828 Set stereoscopic image format of output.
8830 Available values for output image formats are all the input formats as well as:
8833 anaglyph red/blue gray
8834 (red filter on left eye, blue filter on right eye)
8837 anaglyph red/green gray
8838 (red filter on left eye, green filter on right eye)
8841 anaglyph red/cyan gray
8842 (red filter on left eye, cyan filter on right eye)
8845 anaglyph red/cyan half colored
8846 (red filter on left eye, cyan filter on right eye)
8849 anaglyph red/cyan color
8850 (red filter on left eye, cyan filter on right eye)
8853 anaglyph red/cyan color optimized with the least squares projection of dubois
8854 (red filter on left eye, cyan filter on right eye)
8857 anaglyph green/magenta gray
8858 (green filter on left eye, magenta filter on right eye)
8861 anaglyph green/magenta half colored
8862 (green filter on left eye, magenta filter on right eye)
8865 anaglyph green/magenta colored
8866 (green filter on left eye, magenta filter on right eye)
8869 anaglyph green/magenta color optimized with the least squares projection of dubois
8870 (green filter on left eye, magenta filter on right eye)
8873 anaglyph yellow/blue gray
8874 (yellow filter on left eye, blue filter on right eye)
8877 anaglyph yellow/blue half colored
8878 (yellow filter on left eye, blue filter on right eye)
8881 anaglyph yellow/blue colored
8882 (yellow filter on left eye, blue filter on right eye)
8885 anaglyph yellow/blue color optimized with the least squares projection of dubois
8886 (yellow filter on left eye, blue filter on right eye)
8889 interleaved rows (left eye has top row, right eye starts on next row)
8892 interleaved rows (right eye has top row, left eye starts on next row)
8895 mono output (left eye only)
8898 mono output (right eye only)
8901 Default value is @samp{arcd}.
8904 @subsection Examples
8908 Convert input video from side by side parallel to anaglyph yellow/blue dubois:
8914 Convert input video from above bellow (left eye above, right eye below) to side by side crosseye.
8923 Apply a simple postprocessing filter that compresses and decompresses the image
8924 at several (or - in the case of @option{quality} level @code{6} - all) shifts
8925 and average the results.
8927 The filter accepts the following options:
8931 Set quality. This option defines the number of levels for averaging. It accepts
8932 an integer in the range 0-6. If set to @code{0}, the filter will have no
8933 effect. A value of @code{6} means the higher quality. For each increment of
8934 that value the speed drops by a factor of approximately 2. Default value is
8938 Force a constant quantization parameter. If not set, the filter will use the QP
8939 from the video stream (if available).
8942 Set thresholding mode. Available modes are:
8946 Set hard thresholding (default).
8948 Set soft thresholding (better de-ringing effect, but likely blurrier).
8952 Enable the use of the QP from the B-Frames if set to @code{1}. Using this
8953 option may cause flicker since the B-Frames have often larger QP. Default is
8954 @code{0} (not enabled).
8960 Draw subtitles on top of input video using the libass library.
8962 To enable compilation of this filter you need to configure FFmpeg with
8963 @code{--enable-libass}. This filter also requires a build with libavcodec and
8964 libavformat to convert the passed subtitles file to ASS (Advanced Substation
8965 Alpha) subtitles format.
8967 The filter accepts the following options:
8971 Set the filename of the subtitle file to read. It must be specified.
8974 Specify the size of the original video, the video for which the ASS file
8975 was composed. For the syntax of this option, check the
8976 @ref{video size syntax,,"Video size" section in the ffmpeg-utils manual,ffmpeg-utils}.
8977 Due to a misdesign in ASS aspect ratio arithmetic, this is necessary to
8978 correctly scale the fonts if the aspect ratio has been changed.
8981 Set subtitles input character encoding. @code{subtitles} filter only. Only
8982 useful if not UTF-8.
8984 @item stream_index, si
8985 Set subtitles stream index. @code{subtitles} filter only.
8988 Override default style or script info parameters of the subtitles. It accepts a
8989 string containing ASS style format @code{KEY=VALUE} couples separated by ",".
8992 If the first key is not specified, it is assumed that the first value
8993 specifies the @option{filename}.
8995 For example, to render the file @file{sub.srt} on top of the input
8996 video, use the command:
9001 which is equivalent to:
9003 subtitles=filename=sub.srt
9006 To render the default subtitles stream from file @file{video.mkv}, use:
9011 To render the second subtitles stream from that file, use:
9013 subtitles=video.mkv:si=1
9016 To make the subtitles stream from @file{sub.srt} appear in transparent green
9017 @code{DejaVu Serif}, use:
9019 subtitles=sub.srt:force_style='FontName=DejaVu Serif,PrimaryColour=&HAA00FF00'
9024 Scale the input by 2x and smooth using the Super2xSaI (Scale and
9025 Interpolate) pixel art scaling algorithm.
9027 Useful for enlarging pixel art images without reducing sharpness.
9034 Apply telecine process to the video.
9036 This filter accepts the following options:
9045 The default value is @code{top}.
9049 A string of numbers representing the pulldown pattern you wish to apply.
9050 The default value is @code{23}.
9054 Some typical patterns:
9059 24p: 2332 (preferred)
9066 24p: 222222222223 ("Euro pulldown")
9072 Select the most representative frame in a given sequence of consecutive frames.
9074 The filter accepts the following options:
9078 Set the frames batch size to analyze; in a set of @var{n} frames, the filter
9079 will pick one of them, and then handle the next batch of @var{n} frames until
9080 the end. Default is @code{100}.
9083 Since the filter keeps track of the whole frames sequence, a bigger @var{n}
9084 value will result in a higher memory usage, so a high value is not recommended.
9086 @subsection Examples
9090 Extract one picture each 50 frames:
9096 Complete example of a thumbnail creation with @command{ffmpeg}:
9098 ffmpeg -i in.avi -vf thumbnail,scale=300:200 -frames:v 1 out.png
9104 Tile several successive frames together.
9106 The filter accepts the following options:
9111 Set the grid size (i.e. the number of lines and columns). For the syntax of
9112 this option, check the
9113 @ref{video size syntax,,"Video size" section in the ffmpeg-utils manual,ffmpeg-utils}.
9116 Set the maximum number of frames to render in the given area. It must be less
9117 than or equal to @var{w}x@var{h}. The default value is @code{0}, meaning all
9118 the area will be used.
9121 Set the outer border margin in pixels.
9124 Set the inner border thickness (i.e. the number of pixels between frames). For
9125 more advanced padding options (such as having different values for the edges),
9126 refer to the pad video filter.
9129 Specify the color of the unused area. For the syntax of this option, check the
9130 "Color" section in the ffmpeg-utils manual. The default value of @var{color}
9134 @subsection Examples
9138 Produce 8x8 PNG tiles of all keyframes (@option{-skip_frame nokey}) in a movie:
9140 ffmpeg -skip_frame nokey -i file.avi -vf 'scale=128:72,tile=8x8' -an -vsync 0 keyframes%03d.png
9142 The @option{-vsync 0} is necessary to prevent @command{ffmpeg} from
9143 duplicating each output frame to accommodate the originally detected frame
9147 Display @code{5} pictures in an area of @code{3x2} frames,
9148 with @code{7} pixels between them, and @code{2} pixels of initial margin, using
9149 mixed flat and named options:
9151 tile=3x2:nb_frames=5:padding=7:margin=2
9157 Perform various types of temporal field interlacing.
9159 Frames are counted starting from 1, so the first input frame is
9162 The filter accepts the following options:
9167 Specify the mode of the interlacing. This option can also be specified
9168 as a value alone. See below for a list of values for this option.
9170 Available values are:
9174 Move odd frames into the upper field, even into the lower field,
9175 generating a double height frame at half frame rate.
9179 Frame 1 Frame 2 Frame 3 Frame 4
9181 11111 22222 33333 44444
9182 11111 22222 33333 44444
9183 11111 22222 33333 44444
9184 11111 22222 33333 44444
9198 Only output even frames, odd frames are dropped, generating a frame with
9199 unchanged height at half frame rate.
9204 Frame 1 Frame 2 Frame 3 Frame 4
9206 11111 22222 33333 44444
9207 11111 22222 33333 44444
9208 11111 22222 33333 44444
9209 11111 22222 33333 44444
9219 Only output odd frames, even frames are dropped, generating a frame with
9220 unchanged height at half frame rate.
9225 Frame 1 Frame 2 Frame 3 Frame 4
9227 11111 22222 33333 44444
9228 11111 22222 33333 44444
9229 11111 22222 33333 44444
9230 11111 22222 33333 44444
9240 Expand each frame to full height, but pad alternate lines with black,
9241 generating a frame with double height at the same input frame rate.
9246 Frame 1 Frame 2 Frame 3 Frame 4
9248 11111 22222 33333 44444
9249 11111 22222 33333 44444
9250 11111 22222 33333 44444
9251 11111 22222 33333 44444
9254 11111 ..... 33333 .....
9255 ..... 22222 ..... 44444
9256 11111 ..... 33333 .....
9257 ..... 22222 ..... 44444
9258 11111 ..... 33333 .....
9259 ..... 22222 ..... 44444
9260 11111 ..... 33333 .....
9261 ..... 22222 ..... 44444
9265 @item interleave_top, 4
9266 Interleave the upper field from odd frames with the lower field from
9267 even frames, generating a frame with unchanged height at half frame rate.
9272 Frame 1 Frame 2 Frame 3 Frame 4
9274 11111<- 22222 33333<- 44444
9275 11111 22222<- 33333 44444<-
9276 11111<- 22222 33333<- 44444
9277 11111 22222<- 33333 44444<-
9287 @item interleave_bottom, 5
9288 Interleave the lower field from odd frames with the upper field from
9289 even frames, generating a frame with unchanged height at half frame rate.
9294 Frame 1 Frame 2 Frame 3 Frame 4
9296 11111 22222<- 33333 44444<-
9297 11111<- 22222 33333<- 44444
9298 11111 22222<- 33333 44444<-
9299 11111<- 22222 33333<- 44444
9309 @item interlacex2, 6
9310 Double frame rate with unchanged height. Frames are inserted each
9311 containing the second temporal field from the previous input frame and
9312 the first temporal field from the next input frame. This mode relies on
9313 the top_field_first flag. Useful for interlaced video displays with no
9314 field synchronisation.
9319 Frame 1 Frame 2 Frame 3 Frame 4
9321 11111 22222 33333 44444
9322 11111 22222 33333 44444
9323 11111 22222 33333 44444
9324 11111 22222 33333 44444
9327 11111 22222 22222 33333 33333 44444 44444
9328 11111 11111 22222 22222 33333 33333 44444
9329 11111 22222 22222 33333 33333 44444 44444
9330 11111 11111 22222 22222 33333 33333 44444
9336 Numeric values are deprecated but are accepted for backward
9337 compatibility reasons.
9339 Default mode is @code{merge}.
9342 Specify flags influencing the filter process.
9344 Available value for @var{flags} is:
9347 @item low_pass_filter, vlfp
9348 Enable vertical low-pass filtering in the filter.
9349 Vertical low-pass filtering is required when creating an interlaced
9350 destination from a progressive source which contains high-frequency
9351 vertical detail. Filtering will reduce interlace 'twitter' and Moire
9354 Vertical low-pass filtering can only be enabled for @option{mode}
9355 @var{interleave_top} and @var{interleave_bottom}.
9362 Transpose rows with columns in the input video and optionally flip it.
9364 It accepts the following parameters:
9369 Specify the transposition direction.
9371 Can assume the following values:
9373 @item 0, 4, cclock_flip
9374 Rotate by 90 degrees counterclockwise and vertically flip (default), that is:
9382 Rotate by 90 degrees clockwise, that is:
9390 Rotate by 90 degrees counterclockwise, that is:
9397 @item 3, 7, clock_flip
9398 Rotate by 90 degrees clockwise and vertically flip, that is:
9406 For values between 4-7, the transposition is only done if the input
9407 video geometry is portrait and not landscape. These values are
9408 deprecated, the @code{passthrough} option should be used instead.
9410 Numerical values are deprecated, and should be dropped in favor of
9414 Do not apply the transposition if the input geometry matches the one
9415 specified by the specified value. It accepts the following values:
9418 Always apply transposition.
9420 Preserve portrait geometry (when @var{height} >= @var{width}).
9422 Preserve landscape geometry (when @var{width} >= @var{height}).
9425 Default value is @code{none}.
9428 For example to rotate by 90 degrees clockwise and preserve portrait
9431 transpose=dir=1:passthrough=portrait
9434 The command above can also be specified as:
9436 transpose=1:portrait
9440 Trim the input so that the output contains one continuous subpart of the input.
9442 It accepts the following parameters:
9445 Specify the time of the start of the kept section, i.e. the frame with the
9446 timestamp @var{start} will be the first frame in the output.
9449 Specify the time of the first frame that will be dropped, i.e. the frame
9450 immediately preceding the one with the timestamp @var{end} will be the last
9451 frame in the output.
9454 This is the same as @var{start}, except this option sets the start timestamp
9455 in timebase units instead of seconds.
9458 This is the same as @var{end}, except this option sets the end timestamp
9459 in timebase units instead of seconds.
9462 The maximum duration of the output in seconds.
9465 The number of the first frame that should be passed to the output.
9468 The number of the first frame that should be dropped.
9471 @option{start}, @option{end}, and @option{duration} are expressed as time
9472 duration specifications; see
9473 @ref{time duration syntax,,the Time duration section in the ffmpeg-utils(1) manual,ffmpeg-utils}
9474 for the accepted syntax.
9476 Note that the first two sets of the start/end options and the @option{duration}
9477 option look at the frame timestamp, while the _frame variants simply count the
9478 frames that pass through the filter. Also note that this filter does not modify
9479 the timestamps. If you wish for the output timestamps to start at zero, insert a
9480 setpts filter after the trim filter.
9482 If multiple start or end options are set, this filter tries to be greedy and
9483 keep all the frames that match at least one of the specified constraints. To keep
9484 only the part that matches all the constraints at once, chain multiple trim
9487 The defaults are such that all the input is kept. So it is possible to set e.g.
9488 just the end values to keep everything before the specified time.
9493 Drop everything except the second minute of input:
9495 ffmpeg -i INPUT -vf trim=60:120
9499 Keep only the first second:
9501 ffmpeg -i INPUT -vf trim=duration=1
9510 Sharpen or blur the input video.
9512 It accepts the following parameters:
9515 @item luma_msize_x, lx
9516 Set the luma matrix horizontal size. It must be an odd integer between
9517 3 and 63. The default value is 5.
9519 @item luma_msize_y, ly
9520 Set the luma matrix vertical size. It must be an odd integer between 3
9521 and 63. The default value is 5.
9523 @item luma_amount, la
9524 Set the luma effect strength. It must be a floating point number, reasonable
9525 values lay between -1.5 and 1.5.
9527 Negative values will blur the input video, while positive values will
9528 sharpen it, a value of zero will disable the effect.
9530 Default value is 1.0.
9532 @item chroma_msize_x, cx
9533 Set the chroma matrix horizontal size. It must be an odd integer
9534 between 3 and 63. The default value is 5.
9536 @item chroma_msize_y, cy
9537 Set the chroma matrix vertical size. It must be an odd integer
9538 between 3 and 63. The default value is 5.
9540 @item chroma_amount, ca
9541 Set the chroma effect strength. It must be a floating point number, reasonable
9542 values lay between -1.5 and 1.5.
9544 Negative values will blur the input video, while positive values will
9545 sharpen it, a value of zero will disable the effect.
9547 Default value is 0.0.
9550 If set to 1, specify using OpenCL capabilities, only available if
9551 FFmpeg was configured with @code{--enable-opencl}. Default value is 0.
9555 All parameters are optional and default to the equivalent of the
9556 string '5:5:1.0:5:5:0.0'.
9558 @subsection Examples
9562 Apply strong luma sharpen effect:
9564 unsharp=luma_msize_x=7:luma_msize_y=7:luma_amount=2.5
9568 Apply a strong blur of both luma and chroma parameters:
9570 unsharp=7:7:-2:7:7:-2
9576 Apply ultra slow/simple postprocessing filter that compresses and decompresses
9577 the image at several (or - in the case of @option{quality} level @code{8} - all)
9578 shifts and average the results.
9580 The way this differs from the behavior of spp is that uspp actually encodes &
9581 decodes each case with libavcodec Snow, whereas spp uses a simplified intra only 8x8
9582 DCT similar to MJPEG.
9584 The filter accepts the following options:
9588 Set quality. This option defines the number of levels for averaging. It accepts
9589 an integer in the range 0-8. If set to @code{0}, the filter will have no
9590 effect. A value of @code{8} means the higher quality. For each increment of
9591 that value the speed drops by a factor of approximately 2. Default value is
9595 Force a constant quantization parameter. If not set, the filter will use the QP
9596 from the video stream (if available).
9599 @anchor{vidstabdetect}
9600 @section vidstabdetect
9602 Analyze video stabilization/deshaking. Perform pass 1 of 2, see
9603 @ref{vidstabtransform} for pass 2.
9605 This filter generates a file with relative translation and rotation
9606 transform information about subsequent frames, which is then used by
9607 the @ref{vidstabtransform} filter.
9609 To enable compilation of this filter you need to configure FFmpeg with
9610 @code{--enable-libvidstab}.
9612 This filter accepts the following options:
9616 Set the path to the file used to write the transforms information.
9617 Default value is @file{transforms.trf}.
9620 Set how shaky the video is and how quick the camera is. It accepts an
9621 integer in the range 1-10, a value of 1 means little shakiness, a
9622 value of 10 means strong shakiness. Default value is 5.
9625 Set the accuracy of the detection process. It must be a value in the
9626 range 1-15. A value of 1 means low accuracy, a value of 15 means high
9627 accuracy. Default value is 15.
9630 Set stepsize of the search process. The region around minimum is
9631 scanned with 1 pixel resolution. Default value is 6.
9634 Set minimum contrast. Below this value a local measurement field is
9635 discarded. Must be a floating point value in the range 0-1. Default
9639 Set reference frame number for tripod mode.
9641 If enabled, the motion of the frames is compared to a reference frame
9642 in the filtered stream, identified by the specified number. The idea
9643 is to compensate all movements in a more-or-less static scene and keep
9644 the camera view absolutely still.
9646 If set to 0, it is disabled. The frames are counted starting from 1.
9649 Show fields and transforms in the resulting frames. It accepts an
9650 integer in the range 0-2. Default value is 0, which disables any
9654 @subsection Examples
9664 Analyze strongly shaky movie and put the results in file
9665 @file{mytransforms.trf}:
9667 vidstabdetect=shakiness=10:accuracy=15:result="mytransforms.trf"
9671 Visualize the result of internal transformations in the resulting
9674 vidstabdetect=show=1
9678 Analyze a video with medium shakiness using @command{ffmpeg}:
9680 ffmpeg -i input -vf vidstabdetect=shakiness=5:show=1 dummy.avi
9684 @anchor{vidstabtransform}
9685 @section vidstabtransform
9687 Video stabilization/deshaking: pass 2 of 2,
9688 see @ref{vidstabdetect} for pass 1.
9690 Read a file with transform information for each frame and
9691 apply/compensate them. Together with the @ref{vidstabdetect}
9692 filter this can be used to deshake videos. See also
9693 @url{http://public.hronopik.de/vid.stab}. It is important to also use
9694 the @ref{unsharp} filter, see below.
9696 To enable compilation of this filter you need to configure FFmpeg with
9697 @code{--enable-libvidstab}.
9703 Set path to the file used to read the transforms. Default value is
9704 @file{transforms.trf}.
9707 Set the number of frames (value*2 + 1) used for lowpass filtering the
9708 camera movements. Default value is 10.
9710 For example a number of 10 means that 21 frames are used (10 in the
9711 past and 10 in the future) to smoothen the motion in the video. A
9712 larger value leads to a smoother video, but limits the acceleration of
9713 the camera (pan/tilt movements). 0 is a special case where a static
9714 camera is simulated.
9717 Set the camera path optimization algorithm.
9719 Accepted values are:
9722 gaussian kernel low-pass filter on camera motion (default)
9724 averaging on transformations
9728 Set maximal number of pixels to translate frames. Default value is -1,
9732 Set maximal angle in radians (degree*PI/180) to rotate frames. Default
9733 value is -1, meaning no limit.
9736 Specify how to deal with borders that may be visible due to movement
9739 Available values are:
9742 keep image information from previous frame (default)
9744 fill the border black
9748 Invert transforms if set to 1. Default value is 0.
9751 Consider transforms as relative to previous frame if set to 1,
9752 absolute if set to 0. Default value is 0.
9755 Set percentage to zoom. A positive value will result in a zoom-in
9756 effect, a negative value in a zoom-out effect. Default value is 0 (no
9760 Set optimal zooming to avoid borders.
9762 Accepted values are:
9767 optimal static zoom value is determined (only very strong movements
9768 will lead to visible borders) (default)
9770 optimal adaptive zoom value is determined (no borders will be
9771 visible), see @option{zoomspeed}
9774 Note that the value given at zoom is added to the one calculated here.
9777 Set percent to zoom maximally each frame (enabled when
9778 @option{optzoom} is set to 2). Range is from 0 to 5, default value is
9782 Specify type of interpolation.
9784 Available values are:
9789 linear only horizontal
9791 linear in both directions (default)
9793 cubic in both directions (slow)
9797 Enable virtual tripod mode if set to 1, which is equivalent to
9798 @code{relative=0:smoothing=0}. Default value is 0.
9800 Use also @code{tripod} option of @ref{vidstabdetect}.
9803 Increase log verbosity if set to 1. Also the detected global motions
9804 are written to the temporary file @file{global_motions.trf}. Default
9808 @subsection Examples
9812 Use @command{ffmpeg} for a typical stabilization with default values:
9814 ffmpeg -i inp.mpeg -vf vidstabtransform,unsharp=5:5:0.8:3:3:0.4 inp_stabilized.mpeg
9817 Note the use of the @ref{unsharp} filter which is always recommended.
9820 Zoom in a bit more and load transform data from a given file:
9822 vidstabtransform=zoom=5:input="mytransforms.trf"
9826 Smoothen the video even more:
9828 vidstabtransform=smoothing=30
9834 Flip the input video vertically.
9836 For example, to vertically flip a video with @command{ffmpeg}:
9838 ffmpeg -i in.avi -vf "vflip" out.avi
9844 Make or reverse a natural vignetting effect.
9846 The filter accepts the following options:
9850 Set lens angle expression as a number of radians.
9852 The value is clipped in the @code{[0,PI/2]} range.
9854 Default value: @code{"PI/5"}
9858 Set center coordinates expressions. Respectively @code{"w/2"} and @code{"h/2"}
9862 Set forward/backward mode.
9864 Available modes are:
9867 The larger the distance from the central point, the darker the image becomes.
9870 The larger the distance from the central point, the brighter the image becomes.
9871 This can be used to reverse a vignette effect, though there is no automatic
9872 detection to extract the lens @option{angle} and other settings (yet). It can
9873 also be used to create a burning effect.
9876 Default value is @samp{forward}.
9879 Set evaluation mode for the expressions (@option{angle}, @option{x0}, @option{y0}).
9881 It accepts the following values:
9884 Evaluate expressions only once during the filter initialization.
9887 Evaluate expressions for each incoming frame. This is way slower than the
9888 @samp{init} mode since it requires all the scalers to be re-computed, but it
9889 allows advanced dynamic expressions.
9892 Default value is @samp{init}.
9895 Set dithering to reduce the circular banding effects. Default is @code{1}
9899 Set vignette aspect. This setting allows one to adjust the shape of the vignette.
9900 Setting this value to the SAR of the input will make a rectangular vignetting
9901 following the dimensions of the video.
9903 Default is @code{1/1}.
9906 @subsection Expressions
9908 The @option{alpha}, @option{x0} and @option{y0} expressions can contain the
9909 following parameters.
9914 input width and height
9917 the number of input frame, starting from 0
9920 the PTS (Presentation TimeStamp) time of the filtered video frame, expressed in
9921 @var{TB} units, NAN if undefined
9924 frame rate of the input video, NAN if the input frame rate is unknown
9927 the PTS (Presentation TimeStamp) of the filtered video frame,
9928 expressed in seconds, NAN if undefined
9931 time base of the input video
9935 @subsection Examples
9939 Apply simple strong vignetting effect:
9945 Make a flickering vignetting:
9947 vignette='PI/4+random(1)*PI/50':eval=frame
9954 Deinterlace the input video ("w3fdif" stands for "Weston 3 Field
9955 Deinterlacing Filter").
9957 Based on the process described by Martin Weston for BBC R&D, and
9958 implemented based on the de-interlace algorithm written by Jim
9959 Easterbrook for BBC R&D, the Weston 3 field deinterlacing filter
9960 uses filter coefficients calculated by BBC R&D.
9962 There are two sets of filter coefficients, so called "simple":
9963 and "complex". Which set of filter coefficients is used can
9964 be set by passing an optional parameter:
9968 Set the interlacing filter coefficients. Accepts one of the following values:
9972 Simple filter coefficient set.
9974 More-complex filter coefficient set.
9976 Default value is @samp{complex}.
9979 Specify which frames to deinterlace. Accept one of the following values:
9983 Deinterlace all frames,
9985 Only deinterlace frames marked as interlaced.
9988 Default value is @samp{all}.
9992 Apply the xBR high-quality magnification filter which is designed for pixel
9993 art. It follows a set of edge-detection rules, see
9994 @url{http://www.libretro.com/forums/viewtopic.php?f=6&t=134}.
9996 It accepts the following option:
10000 Set the scaling dimension: @code{2} for @code{2xBR}, @code{3} for
10001 @code{3xBR} and @code{4} for @code{4xBR}.
10002 Default is @code{3}.
10008 Deinterlace the input video ("yadif" means "yet another deinterlacing
10011 It accepts the following parameters:
10017 The interlacing mode to adopt. It accepts one of the following values:
10020 @item 0, send_frame
10021 Output one frame for each frame.
10022 @item 1, send_field
10023 Output one frame for each field.
10024 @item 2, send_frame_nospatial
10025 Like @code{send_frame}, but it skips the spatial interlacing check.
10026 @item 3, send_field_nospatial
10027 Like @code{send_field}, but it skips the spatial interlacing check.
10030 The default value is @code{send_frame}.
10033 The picture field parity assumed for the input interlaced video. It accepts one
10034 of the following values:
10038 Assume the top field is first.
10040 Assume the bottom field is first.
10042 Enable automatic detection of field parity.
10045 The default value is @code{auto}.
10046 If the interlacing is unknown or the decoder does not export this information,
10047 top field first will be assumed.
10050 Specify which frames to deinterlace. Accept one of the following
10055 Deinterlace all frames.
10056 @item 1, interlaced
10057 Only deinterlace frames marked as interlaced.
10060 The default value is @code{all}.
10065 Apply Zoom & Pan effect.
10067 This filter accepts the following options:
10071 Set the zoom expression. Default is 1.
10075 Set the x and y expression. Default is 0.
10078 Set the duration expression in number of frames.
10079 This sets for how many number of frames effect will last for
10080 single input image.
10083 Set the output image size, default is 'hd720'.
10086 Each expression can contain the following constants:
10105 Output frame count.
10109 Last calculated 'x' and 'y' position from 'x' and 'y' expression
10110 for current input frame.
10114 'x' and 'y' of last output frame of previous input frame or 0 when there was
10115 not yet such frame (first input frame).
10118 Last calculated zoom from 'z' expression for current input frame.
10121 Last calculated zoom of last output frame of previous input frame.
10124 Number of output frames for current input frame. Calculated from 'd' expression
10125 for each input frame.
10128 number of output frames created for previous input frame
10131 Rational number: input width / input height
10134 sample aspect ratio
10137 display aspect ratio
10141 @subsection Examples
10145 Zoom-in up to 1.5 and pan at same time to some spot near center of picture:
10147 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
10151 @c man end VIDEO FILTERS
10153 @chapter Video Sources
10154 @c man begin VIDEO SOURCES
10156 Below is a description of the currently available video sources.
10160 Buffer video frames, and make them available to the filter chain.
10162 This source is mainly intended for a programmatic use, in particular
10163 through the interface defined in @file{libavfilter/vsrc_buffer.h}.
10165 It accepts the following parameters:
10170 Specify the size (width and height) of the buffered video frames. For the
10171 syntax of this option, check the
10172 @ref{video size syntax,,"Video size" section in the ffmpeg-utils manual,ffmpeg-utils}.
10175 The input video width.
10178 The input video height.
10181 A string representing the pixel format of the buffered video frames.
10182 It may be a number corresponding to a pixel format, or a pixel format
10186 Specify the timebase assumed by the timestamps of the buffered frames.
10189 Specify the frame rate expected for the video stream.
10191 @item pixel_aspect, sar
10192 The sample (pixel) aspect ratio of the input video.
10195 Specify the optional parameters to be used for the scale filter which
10196 is automatically inserted when an input change is detected in the
10197 input size or format.
10202 buffer=width=320:height=240:pix_fmt=yuv410p:time_base=1/24:sar=1
10205 will instruct the source to accept video frames with size 320x240 and
10206 with format "yuv410p", assuming 1/24 as the timestamps timebase and
10207 square pixels (1:1 sample aspect ratio).
10208 Since the pixel format with name "yuv410p" corresponds to the number 6
10209 (check the enum AVPixelFormat definition in @file{libavutil/pixfmt.h}),
10210 this example corresponds to:
10212 buffer=size=320x240:pixfmt=6:time_base=1/24:pixel_aspect=1/1
10215 Alternatively, the options can be specified as a flat string, but this
10216 syntax is deprecated:
10218 @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}]
10222 Create a pattern generated by an elementary cellular automaton.
10224 The initial state of the cellular automaton can be defined through the
10225 @option{filename}, and @option{pattern} options. If such options are
10226 not specified an initial state is created randomly.
10228 At each new frame a new row in the video is filled with the result of
10229 the cellular automaton next generation. The behavior when the whole
10230 frame is filled is defined by the @option{scroll} option.
10232 This source accepts the following options:
10236 Read the initial cellular automaton state, i.e. the starting row, from
10237 the specified file.
10238 In the file, each non-whitespace character is considered an alive
10239 cell, a newline will terminate the row, and further characters in the
10240 file will be ignored.
10243 Read the initial cellular automaton state, i.e. the starting row, from
10244 the specified string.
10246 Each non-whitespace character in the string is considered an alive
10247 cell, a newline will terminate the row, and further characters in the
10248 string will be ignored.
10251 Set the video rate, that is the number of frames generated per second.
10254 @item random_fill_ratio, ratio
10255 Set the random fill ratio for the initial cellular automaton row. It
10256 is a floating point number value ranging from 0 to 1, defaults to
10259 This option is ignored when a file or a pattern is specified.
10261 @item random_seed, seed
10262 Set the seed for filling randomly the initial row, must be an integer
10263 included between 0 and UINT32_MAX. If not specified, or if explicitly
10264 set to -1, the filter will try to use a good random seed on a best
10268 Set the cellular automaton rule, it is a number ranging from 0 to 255.
10269 Default value is 110.
10272 Set the size of the output video. For the syntax of this option, check the
10273 @ref{video size syntax,,"Video size" section in the ffmpeg-utils manual,ffmpeg-utils}.
10275 If @option{filename} or @option{pattern} is specified, the size is set
10276 by default to the width of the specified initial state row, and the
10277 height is set to @var{width} * PHI.
10279 If @option{size} is set, it must contain the width of the specified
10280 pattern string, and the specified pattern will be centered in the
10283 If a filename or a pattern string is not specified, the size value
10284 defaults to "320x518" (used for a randomly generated initial state).
10287 If set to 1, scroll the output upward when all the rows in the output
10288 have been already filled. If set to 0, the new generated row will be
10289 written over the top row just after the bottom row is filled.
10292 @item start_full, full
10293 If set to 1, completely fill the output with generated rows before
10294 outputting the first frame.
10295 This is the default behavior, for disabling set the value to 0.
10298 If set to 1, stitch the left and right row edges together.
10299 This is the default behavior, for disabling set the value to 0.
10302 @subsection Examples
10306 Read the initial state from @file{pattern}, and specify an output of
10309 cellauto=f=pattern:s=200x400
10313 Generate a random initial row with a width of 200 cells, with a fill
10316 cellauto=ratio=2/3:s=200x200
10320 Create a pattern generated by rule 18 starting by a single alive cell
10321 centered on an initial row with width 100:
10323 cellauto=p=@@:s=100x400:full=0:rule=18
10327 Specify a more elaborated initial pattern:
10329 cellauto=p='@@@@ @@ @@@@':s=100x400:full=0:rule=18
10334 @section mandelbrot
10336 Generate a Mandelbrot set fractal, and progressively zoom towards the
10337 point specified with @var{start_x} and @var{start_y}.
10339 This source accepts the following options:
10344 Set the terminal pts value. Default value is 400.
10347 Set the terminal scale value.
10348 Must be a floating point value. Default value is 0.3.
10351 Set the inner coloring mode, that is the algorithm used to draw the
10352 Mandelbrot fractal internal region.
10354 It shall assume one of the following values:
10359 Show time until convergence.
10361 Set color based on point closest to the origin of the iterations.
10366 Default value is @var{mincol}.
10369 Set the bailout value. Default value is 10.0.
10372 Set the maximum of iterations performed by the rendering
10373 algorithm. Default value is 7189.
10376 Set outer coloring mode.
10377 It shall assume one of following values:
10379 @item iteration_count
10380 Set iteration cound mode.
10381 @item normalized_iteration_count
10382 set normalized iteration count mode.
10384 Default value is @var{normalized_iteration_count}.
10387 Set frame rate, expressed as number of frames per second. Default
10391 Set frame size. For the syntax of this option, check the "Video
10392 size" section in the ffmpeg-utils manual. Default value is "640x480".
10395 Set the initial scale value. Default value is 3.0.
10398 Set the initial x position. Must be a floating point value between
10399 -100 and 100. Default value is -0.743643887037158704752191506114774.
10402 Set the initial y position. Must be a floating point value between
10403 -100 and 100. Default value is -0.131825904205311970493132056385139.
10408 Generate various test patterns, as generated by the MPlayer test filter.
10410 The size of the generated video is fixed, and is 256x256.
10411 This source is useful in particular for testing encoding features.
10413 This source accepts the following options:
10418 Specify the frame rate of the sourced video, as the number of frames
10419 generated per second. It has to be a string in the format
10420 @var{frame_rate_num}/@var{frame_rate_den}, an integer number, a floating point
10421 number or a valid video frame rate abbreviation. The default value is
10425 Set the duration of the sourced video. See
10426 @ref{time duration syntax,,the Time duration section in the ffmpeg-utils(1) manual,ffmpeg-utils}
10427 for the accepted syntax.
10429 If not specified, or the expressed duration is negative, the video is
10430 supposed to be generated forever.
10434 Set the number or the name of the test to perform. Supported tests are:
10450 Default value is "all", which will cycle through the list of all tests.
10455 mptestsrc=t=dc_luma
10458 will generate a "dc_luma" test pattern.
10460 @section frei0r_src
10462 Provide a frei0r source.
10464 To enable compilation of this filter you need to install the frei0r
10465 header and configure FFmpeg with @code{--enable-frei0r}.
10467 This source accepts the following parameters:
10472 The size of the video to generate. For the syntax of this option, check the
10473 @ref{video size syntax,,"Video size" section in the ffmpeg-utils manual,ffmpeg-utils}.
10476 The framerate of the generated video. It may be a string of the form
10477 @var{num}/@var{den} or a frame rate abbreviation.
10480 The name to the frei0r source to load. For more information regarding frei0r and
10481 how to set the parameters, read the @ref{frei0r} section in the video filters
10484 @item filter_params
10485 A '|'-separated list of parameters to pass to the frei0r source.
10489 For example, to generate a frei0r partik0l source with size 200x200
10490 and frame rate 10 which is overlaid on the overlay filter main input:
10492 frei0r_src=size=200x200:framerate=10:filter_name=partik0l:filter_params=1234 [overlay]; [in][overlay] overlay
10497 Generate a life pattern.
10499 This source is based on a generalization of John Conway's life game.
10501 The sourced input represents a life grid, each pixel represents a cell
10502 which can be in one of two possible states, alive or dead. Every cell
10503 interacts with its eight neighbours, which are the cells that are
10504 horizontally, vertically, or diagonally adjacent.
10506 At each interaction the grid evolves according to the adopted rule,
10507 which specifies the number of neighbor alive cells which will make a
10508 cell stay alive or born. The @option{rule} option allows one to specify
10511 This source accepts the following options:
10515 Set the file from which to read the initial grid state. In the file,
10516 each non-whitespace character is considered an alive cell, and newline
10517 is used to delimit the end of each row.
10519 If this option is not specified, the initial grid is generated
10523 Set the video rate, that is the number of frames generated per second.
10526 @item random_fill_ratio, ratio
10527 Set the random fill ratio for the initial random grid. It is a
10528 floating point number value ranging from 0 to 1, defaults to 1/PHI.
10529 It is ignored when a file is specified.
10531 @item random_seed, seed
10532 Set the seed for filling the initial random grid, must be an integer
10533 included between 0 and UINT32_MAX. If not specified, or if explicitly
10534 set to -1, the filter will try to use a good random seed on a best
10540 A rule can be specified with a code of the kind "S@var{NS}/B@var{NB}",
10541 where @var{NS} and @var{NB} are sequences of numbers in the range 0-8,
10542 @var{NS} specifies the number of alive neighbor cells which make a
10543 live cell stay alive, and @var{NB} the number of alive neighbor cells
10544 which make a dead cell to become alive (i.e. to "born").
10545 "s" and "b" can be used in place of "S" and "B", respectively.
10547 Alternatively a rule can be specified by an 18-bits integer. The 9
10548 high order bits are used to encode the next cell state if it is alive
10549 for each number of neighbor alive cells, the low order bits specify
10550 the rule for "borning" new cells. Higher order bits encode for an
10551 higher number of neighbor cells.
10552 For example the number 6153 = @code{(12<<9)+9} specifies a stay alive
10553 rule of 12 and a born rule of 9, which corresponds to "S23/B03".
10555 Default value is "S23/B3", which is the original Conway's game of life
10556 rule, and will keep a cell alive if it has 2 or 3 neighbor alive
10557 cells, and will born a new cell if there are three alive cells around
10561 Set the size of the output video. For the syntax of this option, check the
10562 @ref{video size syntax,,"Video size" section in the ffmpeg-utils manual,ffmpeg-utils}.
10564 If @option{filename} is specified, the size is set by default to the
10565 same size of the input file. If @option{size} is set, it must contain
10566 the size specified in the input file, and the initial grid defined in
10567 that file is centered in the larger resulting area.
10569 If a filename is not specified, the size value defaults to "320x240"
10570 (used for a randomly generated initial grid).
10573 If set to 1, stitch the left and right grid edges together, and the
10574 top and bottom edges also. Defaults to 1.
10577 Set cell mold speed. If set, a dead cell will go from @option{death_color} to
10578 @option{mold_color} with a step of @option{mold}. @option{mold} can have a
10579 value from 0 to 255.
10582 Set the color of living (or new born) cells.
10585 Set the color of dead cells. If @option{mold} is set, this is the first color
10586 used to represent a dead cell.
10589 Set mold color, for definitely dead and moldy cells.
10591 For the syntax of these 3 color options, check the "Color" section in the
10592 ffmpeg-utils manual.
10595 @subsection Examples
10599 Read a grid from @file{pattern}, and center it on a grid of size
10602 life=f=pattern:s=300x300
10606 Generate a random grid of size 200x200, with a fill ratio of 2/3:
10608 life=ratio=2/3:s=200x200
10612 Specify a custom rule for evolving a randomly generated grid:
10618 Full example with slow death effect (mold) using @command{ffplay}:
10620 ffplay -f lavfi life=s=300x200:mold=10:r=60:ratio=0.1:death_color=#C83232:life_color=#00ff00,scale=1200:800:flags=16
10625 @anchor{haldclutsrc}
10627 @anchor{rgbtestsrc}
10629 @anchor{smptehdbars}
10631 @section color, haldclutsrc, nullsrc, rgbtestsrc, smptebars, smptehdbars, testsrc
10633 The @code{color} source provides an uniformly colored input.
10635 The @code{haldclutsrc} source provides an identity Hald CLUT. See also
10636 @ref{haldclut} filter.
10638 The @code{nullsrc} source returns unprocessed video frames. It is
10639 mainly useful to be employed in analysis / debugging tools, or as the
10640 source for filters which ignore the input data.
10642 The @code{rgbtestsrc} source generates an RGB test pattern useful for
10643 detecting RGB vs BGR issues. You should see a red, green and blue
10644 stripe from top to bottom.
10646 The @code{smptebars} source generates a color bars pattern, based on
10647 the SMPTE Engineering Guideline EG 1-1990.
10649 The @code{smptehdbars} source generates a color bars pattern, based on
10650 the SMPTE RP 219-2002.
10652 The @code{testsrc} source generates a test video pattern, showing a
10653 color pattern, a scrolling gradient and a timestamp. This is mainly
10654 intended for testing purposes.
10656 The sources accept the following parameters:
10661 Specify the color of the source, only available in the @code{color}
10662 source. For the syntax of this option, check the "Color" section in the
10663 ffmpeg-utils manual.
10666 Specify the level of the Hald CLUT, only available in the @code{haldclutsrc}
10667 source. A level of @code{N} generates a picture of @code{N*N*N} by @code{N*N*N}
10668 pixels to be used as identity matrix for 3D lookup tables. Each component is
10669 coded on a @code{1/(N*N)} scale.
10672 Specify the size of the sourced video. For the syntax of this option, check the
10673 @ref{video size syntax,,"Video size" section in the ffmpeg-utils manual,ffmpeg-utils}.
10674 The default value is @code{320x240}.
10676 This option is not available with the @code{haldclutsrc} filter.
10679 Specify the frame rate of the sourced video, as the number of frames
10680 generated per second. It has to be a string in the format
10681 @var{frame_rate_num}/@var{frame_rate_den}, an integer number, a floating point
10682 number or a valid video frame rate abbreviation. The default value is
10686 Set the sample aspect ratio of the sourced video.
10689 Set the duration of the sourced video. See
10690 @ref{time duration syntax,,the Time duration section in the ffmpeg-utils(1) manual,ffmpeg-utils}
10691 for the accepted syntax.
10693 If not specified, or the expressed duration is negative, the video is
10694 supposed to be generated forever.
10697 Set the number of decimals to show in the timestamp, only available in the
10698 @code{testsrc} source.
10700 The displayed timestamp value will correspond to the original
10701 timestamp value multiplied by the power of 10 of the specified
10702 value. Default value is 0.
10705 For example the following:
10707 testsrc=duration=5.3:size=qcif:rate=10
10710 will generate a video with a duration of 5.3 seconds, with size
10711 176x144 and a frame rate of 10 frames per second.
10713 The following graph description will generate a red source
10714 with an opacity of 0.2, with size "qcif" and a frame rate of 10
10717 color=c=red@@0.2:s=qcif:r=10
10720 If the input content is to be ignored, @code{nullsrc} can be used. The
10721 following command generates noise in the luminance plane by employing
10722 the @code{geq} filter:
10724 nullsrc=s=256x256, geq=random(1)*255:128:128
10727 @subsection Commands
10729 The @code{color} source supports the following commands:
10733 Set the color of the created image. Accepts the same syntax of the
10734 corresponding @option{color} option.
10737 @c man end VIDEO SOURCES
10739 @chapter Video Sinks
10740 @c man begin VIDEO SINKS
10742 Below is a description of the currently available video sinks.
10744 @section buffersink
10746 Buffer video frames, and make them available to the end of the filter
10749 This sink is mainly intended for programmatic use, in particular
10750 through the interface defined in @file{libavfilter/buffersink.h}
10751 or the options system.
10753 It accepts a pointer to an AVBufferSinkContext structure, which
10754 defines the incoming buffers' formats, to be passed as the opaque
10755 parameter to @code{avfilter_init_filter} for initialization.
10759 Null video sink: do absolutely nothing with the input video. It is
10760 mainly useful as a template and for use in analysis / debugging
10763 @c man end VIDEO SINKS
10765 @chapter Multimedia Filters
10766 @c man begin MULTIMEDIA FILTERS
10768 Below is a description of the currently available multimedia filters.
10770 @section avectorscope
10772 Convert input audio to a video output, representing the audio vector
10775 The filter is used to measure the difference between channels of stereo
10776 audio stream. A monoaural signal, consisting of identical left and right
10777 signal, results in straight vertical line. Any stereo separation is visible
10778 as a deviation from this line, creating a Lissajous figure.
10779 If the straight (or deviation from it) but horizontal line appears this
10780 indicates that the left and right channels are out of phase.
10782 The filter accepts the following options:
10786 Set the vectorscope mode.
10788 Available values are:
10791 Lissajous rotated by 45 degrees.
10794 Same as above but not rotated.
10797 Default value is @samp{lissajous}.
10800 Set the video size for the output. For the syntax of this option, check the
10801 @ref{video size syntax,,"Video size" section in the ffmpeg-utils manual,ffmpeg-utils}.
10802 Default value is @code{400x400}.
10805 Set the output frame rate. Default value is @code{25}.
10810 Specify the red, green and blue contrast. Default values are @code{40}, @code{160} and @code{80}.
10811 Allowed range is @code{[0, 255]}.
10816 Specify the red, green and blue fade. Default values are @code{15}, @code{10} and @code{5}.
10817 Allowed range is @code{[0, 255]}.
10820 Set the zoom factor. Default value is @code{1}. Allowed range is @code{[1, 10]}.
10823 @subsection Examples
10827 Complete example using @command{ffplay}:
10829 ffplay -f lavfi 'amovie=input.mp3, asplit [a][out1];
10830 [a] avectorscope=zoom=1.3:rc=2:gc=200:bc=10:rf=1:gf=8:bf=7 [out0]'
10836 Concatenate audio and video streams, joining them together one after the
10839 The filter works on segments of synchronized video and audio streams. All
10840 segments must have the same number of streams of each type, and that will
10841 also be the number of streams at output.
10843 The filter accepts the following options:
10848 Set the number of segments. Default is 2.
10851 Set the number of output video streams, that is also the number of video
10852 streams in each segment. Default is 1.
10855 Set the number of output audio streams, that is also the number of audio
10856 streams in each segment. Default is 0.
10859 Activate unsafe mode: do not fail if segments have a different format.
10863 The filter has @var{v}+@var{a} outputs: first @var{v} video outputs, then
10864 @var{a} audio outputs.
10866 There are @var{n}x(@var{v}+@var{a}) inputs: first the inputs for the first
10867 segment, in the same order as the outputs, then the inputs for the second
10870 Related streams do not always have exactly the same duration, for various
10871 reasons including codec frame size or sloppy authoring. For that reason,
10872 related synchronized streams (e.g. a video and its audio track) should be
10873 concatenated at once. The concat filter will use the duration of the longest
10874 stream in each segment (except the last one), and if necessary pad shorter
10875 audio streams with silence.
10877 For this filter to work correctly, all segments must start at timestamp 0.
10879 All corresponding streams must have the same parameters in all segments; the
10880 filtering system will automatically select a common pixel format for video
10881 streams, and a common sample format, sample rate and channel layout for
10882 audio streams, but other settings, such as resolution, must be converted
10883 explicitly by the user.
10885 Different frame rates are acceptable but will result in variable frame rate
10886 at output; be sure to configure the output file to handle it.
10888 @subsection Examples
10892 Concatenate an opening, an episode and an ending, all in bilingual version
10893 (video in stream 0, audio in streams 1 and 2):
10895 ffmpeg -i opening.mkv -i episode.mkv -i ending.mkv -filter_complex \
10896 '[0:0] [0:1] [0:2] [1:0] [1:1] [1:2] [2:0] [2:1] [2:2]
10897 concat=n=3:v=1:a=2 [v] [a1] [a2]' \
10898 -map '[v]' -map '[a1]' -map '[a2]' output.mkv
10902 Concatenate two parts, handling audio and video separately, using the
10903 (a)movie sources, and adjusting the resolution:
10905 movie=part1.mp4, scale=512:288 [v1] ; amovie=part1.mp4 [a1] ;
10906 movie=part2.mp4, scale=512:288 [v2] ; amovie=part2.mp4 [a2] ;
10907 [v1] [v2] concat [outv] ; [a1] [a2] concat=v=0:a=1 [outa]
10909 Note that a desync will happen at the stitch if the audio and video streams
10910 do not have exactly the same duration in the first file.
10916 EBU R128 scanner filter. This filter takes an audio stream as input and outputs
10917 it unchanged. By default, it logs a message at a frequency of 10Hz with the
10918 Momentary loudness (identified by @code{M}), Short-term loudness (@code{S}),
10919 Integrated loudness (@code{I}) and Loudness Range (@code{LRA}).
10921 The filter also has a video output (see the @var{video} option) with a real
10922 time graph to observe the loudness evolution. The graphic contains the logged
10923 message mentioned above, so it is not printed anymore when this option is set,
10924 unless the verbose logging is set. The main graphing area contains the
10925 short-term loudness (3 seconds of analysis), and the gauge on the right is for
10926 the momentary loudness (400 milliseconds).
10928 More information about the Loudness Recommendation EBU R128 on
10929 @url{http://tech.ebu.ch/loudness}.
10931 The filter accepts the following options:
10936 Activate the video output. The audio stream is passed unchanged whether this
10937 option is set or no. The video stream will be the first output stream if
10938 activated. Default is @code{0}.
10941 Set the video size. This option is for video only. For the syntax of this
10943 @ref{video size syntax,,"Video size" section in the ffmpeg-utils manual,ffmpeg-utils}.
10944 Default and minimum resolution is @code{640x480}.
10947 Set the EBU scale meter. Default is @code{9}. Common values are @code{9} and
10948 @code{18}, respectively for EBU scale meter +9 and EBU scale meter +18. Any
10949 other integer value between this range is allowed.
10952 Set metadata injection. If set to @code{1}, the audio input will be segmented
10953 into 100ms output frames, each of them containing various loudness information
10954 in metadata. All the metadata keys are prefixed with @code{lavfi.r128.}.
10956 Default is @code{0}.
10959 Force the frame logging level.
10961 Available values are:
10964 information logging level
10966 verbose logging level
10969 By default, the logging level is set to @var{info}. If the @option{video} or
10970 the @option{metadata} options are set, it switches to @var{verbose}.
10975 Available modes can be cumulated (the option is a @code{flag} type). Possible
10979 Disable any peak mode (default).
10981 Enable sample-peak mode.
10983 Simple peak mode looking for the higher sample value. It logs a message
10984 for sample-peak (identified by @code{SPK}).
10986 Enable true-peak mode.
10988 If enabled, the peak lookup is done on an over-sampled version of the input
10989 stream for better peak accuracy. It logs a message for true-peak.
10990 (identified by @code{TPK}) and true-peak per frame (identified by @code{FTPK}).
10991 This mode requires a build with @code{libswresample}.
10996 @subsection Examples
11000 Real-time graph using @command{ffplay}, with a EBU scale meter +18:
11002 ffplay -f lavfi -i "amovie=input.mp3,ebur128=video=1:meter=18 [out0][out1]"
11006 Run an analysis with @command{ffmpeg}:
11008 ffmpeg -nostats -i input.mp3 -filter_complex ebur128 -f null -
11012 @section interleave, ainterleave
11014 Temporally interleave frames from several inputs.
11016 @code{interleave} works with video inputs, @code{ainterleave} with audio.
11018 These filters read frames from several inputs and send the oldest
11019 queued frame to the output.
11021 Input streams must have a well defined, monotonically increasing frame
11024 In order to submit one frame to output, these filters need to enqueue
11025 at least one frame for each input, so they cannot work in case one
11026 input is not yet terminated and will not receive incoming frames.
11028 For example consider the case when one input is a @code{select} filter
11029 which always drop input frames. The @code{interleave} filter will keep
11030 reading from that input, but it will never be able to send new frames
11031 to output until the input will send an end-of-stream signal.
11033 Also, depending on inputs synchronization, the filters will drop
11034 frames in case one input receives more frames than the other ones, and
11035 the queue is already filled.
11037 These filters accept the following options:
11041 Set the number of different inputs, it is 2 by default.
11044 @subsection Examples
11048 Interleave frames belonging to different streams using @command{ffmpeg}:
11050 ffmpeg -i bambi.avi -i pr0n.mkv -filter_complex "[0:v][1:v] interleave" out.avi
11054 Add flickering blur effect:
11056 select='if(gt(random(0), 0.2), 1, 2)':n=2 [tmp], boxblur=2:2, [tmp] interleave
11060 @section perms, aperms
11062 Set read/write permissions for the output frames.
11064 These filters are mainly aimed at developers to test direct path in the
11065 following filter in the filtergraph.
11067 The filters accept the following options:
11071 Select the permissions mode.
11073 It accepts the following values:
11076 Do nothing. This is the default.
11078 Set all the output frames read-only.
11080 Set all the output frames directly writable.
11082 Make the frame read-only if writable, and writable if read-only.
11084 Set each output frame read-only or writable randomly.
11088 Set the seed for the @var{random} mode, must be an integer included between
11089 @code{0} and @code{UINT32_MAX}. If not specified, or if explicitly set to
11090 @code{-1}, the filter will try to use a good random seed on a best effort
11094 Note: in case of auto-inserted filter between the permission filter and the
11095 following one, the permission might not be received as expected in that
11096 following filter. Inserting a @ref{format} or @ref{aformat} filter before the
11097 perms/aperms filter can avoid this problem.
11099 @section select, aselect
11101 Select frames to pass in output.
11103 This filter accepts the following options:
11108 Set expression, which is evaluated for each input frame.
11110 If the expression is evaluated to zero, the frame is discarded.
11112 If the evaluation result is negative or NaN, the frame is sent to the
11113 first output; otherwise it is sent to the output with index
11114 @code{ceil(val)-1}, assuming that the input index starts from 0.
11116 For example a value of @code{1.2} corresponds to the output with index
11117 @code{ceil(1.2)-1 = 2-1 = 1}, that is the second output.
11120 Set the number of outputs. The output to which to send the selected
11121 frame is based on the result of the evaluation. Default value is 1.
11124 The expression can contain the following constants:
11128 The (sequential) number of the filtered frame, starting from 0.
11131 The (sequential) number of the selected frame, starting from 0.
11133 @item prev_selected_n
11134 The sequential number of the last selected frame. It's NAN if undefined.
11137 The timebase of the input timestamps.
11140 The PTS (Presentation TimeStamp) of the filtered video frame,
11141 expressed in @var{TB} units. It's NAN if undefined.
11144 The PTS of the filtered video frame,
11145 expressed in seconds. It's NAN if undefined.
11148 The PTS of the previously filtered video frame. It's NAN if undefined.
11150 @item prev_selected_pts
11151 The PTS of the last previously filtered video frame. It's NAN if undefined.
11153 @item prev_selected_t
11154 The PTS of the last previously selected video frame. It's NAN if undefined.
11157 The PTS of the first video frame in the video. It's NAN if undefined.
11160 The time of the first video frame in the video. It's NAN if undefined.
11162 @item pict_type @emph{(video only)}
11163 The type of the filtered frame. It can assume one of the following
11175 @item interlace_type @emph{(video only)}
11176 The frame interlace type. It can assume one of the following values:
11179 The frame is progressive (not interlaced).
11181 The frame is top-field-first.
11183 The frame is bottom-field-first.
11186 @item consumed_sample_n @emph{(audio only)}
11187 the number of selected samples before the current frame
11189 @item samples_n @emph{(audio only)}
11190 the number of samples in the current frame
11192 @item sample_rate @emph{(audio only)}
11193 the input sample rate
11196 This is 1 if the filtered frame is a key-frame, 0 otherwise.
11199 the position in the file of the filtered frame, -1 if the information
11200 is not available (e.g. for synthetic video)
11202 @item scene @emph{(video only)}
11203 value between 0 and 1 to indicate a new scene; a low value reflects a low
11204 probability for the current frame to introduce a new scene, while a higher
11205 value means the current frame is more likely to be one (see the example below)
11209 The default value of the select expression is "1".
11211 @subsection Examples
11215 Select all frames in input:
11220 The example above is the same as:
11232 Select only I-frames:
11234 select='eq(pict_type\,I)'
11238 Select one frame every 100:
11240 select='not(mod(n\,100))'
11244 Select only frames contained in the 10-20 time interval:
11246 select=between(t\,10\,20)
11250 Select only I frames contained in the 10-20 time interval:
11252 select=between(t\,10\,20)*eq(pict_type\,I)
11256 Select frames with a minimum distance of 10 seconds:
11258 select='isnan(prev_selected_t)+gte(t-prev_selected_t\,10)'
11262 Use aselect to select only audio frames with samples number > 100:
11264 aselect='gt(samples_n\,100)'
11268 Create a mosaic of the first scenes:
11270 ffmpeg -i video.avi -vf select='gt(scene\,0.4)',scale=160:120,tile -frames:v 1 preview.png
11273 Comparing @var{scene} against a value between 0.3 and 0.5 is generally a sane
11277 Send even and odd frames to separate outputs, and compose them:
11279 select=n=2:e='mod(n, 2)+1' [odd][even]; [odd] pad=h=2*ih [tmp]; [tmp][even] overlay=y=h
11283 @section sendcmd, asendcmd
11285 Send commands to filters in the filtergraph.
11287 These filters read commands to be sent to other filters in the
11290 @code{sendcmd} must be inserted between two video filters,
11291 @code{asendcmd} must be inserted between two audio filters, but apart
11292 from that they act the same way.
11294 The specification of commands can be provided in the filter arguments
11295 with the @var{commands} option, or in a file specified by the
11296 @var{filename} option.
11298 These filters accept the following options:
11301 Set the commands to be read and sent to the other filters.
11303 Set the filename of the commands to be read and sent to the other
11307 @subsection Commands syntax
11309 A commands description consists of a sequence of interval
11310 specifications, comprising a list of commands to be executed when a
11311 particular event related to that interval occurs. The occurring event
11312 is typically the current frame time entering or leaving a given time
11315 An interval is specified by the following syntax:
11317 @var{START}[-@var{END}] @var{COMMANDS};
11320 The time interval is specified by the @var{START} and @var{END} times.
11321 @var{END} is optional and defaults to the maximum time.
11323 The current frame time is considered within the specified interval if
11324 it is included in the interval [@var{START}, @var{END}), that is when
11325 the time is greater or equal to @var{START} and is lesser than
11328 @var{COMMANDS} consists of a sequence of one or more command
11329 specifications, separated by ",", relating to that interval. The
11330 syntax of a command specification is given by:
11332 [@var{FLAGS}] @var{TARGET} @var{COMMAND} @var{ARG}
11335 @var{FLAGS} is optional and specifies the type of events relating to
11336 the time interval which enable sending the specified command, and must
11337 be a non-null sequence of identifier flags separated by "+" or "|" and
11338 enclosed between "[" and "]".
11340 The following flags are recognized:
11343 The command is sent when the current frame timestamp enters the
11344 specified interval. In other words, the command is sent when the
11345 previous frame timestamp was not in the given interval, and the
11349 The command is sent when the current frame timestamp leaves the
11350 specified interval. In other words, the command is sent when the
11351 previous frame timestamp was in the given interval, and the
11355 If @var{FLAGS} is not specified, a default value of @code{[enter]} is
11358 @var{TARGET} specifies the target of the command, usually the name of
11359 the filter class or a specific filter instance name.
11361 @var{COMMAND} specifies the name of the command for the target filter.
11363 @var{ARG} is optional and specifies the optional list of argument for
11364 the given @var{COMMAND}.
11366 Between one interval specification and another, whitespaces, or
11367 sequences of characters starting with @code{#} until the end of line,
11368 are ignored and can be used to annotate comments.
11370 A simplified BNF description of the commands specification syntax
11373 @var{COMMAND_FLAG} ::= "enter" | "leave"
11374 @var{COMMAND_FLAGS} ::= @var{COMMAND_FLAG} [(+|"|")@var{COMMAND_FLAG}]
11375 @var{COMMAND} ::= ["[" @var{COMMAND_FLAGS} "]"] @var{TARGET} @var{COMMAND} [@var{ARG}]
11376 @var{COMMANDS} ::= @var{COMMAND} [,@var{COMMANDS}]
11377 @var{INTERVAL} ::= @var{START}[-@var{END}] @var{COMMANDS}
11378 @var{INTERVALS} ::= @var{INTERVAL}[;@var{INTERVALS}]
11381 @subsection Examples
11385 Specify audio tempo change at second 4:
11387 asendcmd=c='4.0 atempo tempo 1.5',atempo
11391 Specify a list of drawtext and hue commands in a file.
11393 # show text in the interval 5-10
11394 5.0-10.0 [enter] drawtext reinit 'fontfile=FreeSerif.ttf:text=hello world',
11395 [leave] drawtext reinit 'fontfile=FreeSerif.ttf:text=';
11397 # desaturate the image in the interval 15-20
11398 15.0-20.0 [enter] hue s 0,
11399 [enter] drawtext reinit 'fontfile=FreeSerif.ttf:text=nocolor',
11401 [leave] drawtext reinit 'fontfile=FreeSerif.ttf:text=color';
11403 # apply an exponential saturation fade-out effect, starting from time 25
11404 25 [enter] hue s exp(25-t)
11407 A filtergraph allowing to read and process the above command list
11408 stored in a file @file{test.cmd}, can be specified with:
11410 sendcmd=f=test.cmd,drawtext=fontfile=FreeSerif.ttf:text='',hue
11415 @section setpts, asetpts
11417 Change the PTS (presentation timestamp) of the input frames.
11419 @code{setpts} works on video frames, @code{asetpts} on audio frames.
11421 This filter accepts the following options:
11426 The expression which is evaluated for each frame to construct its timestamp.
11430 The expression is evaluated through the eval API and can contain the following
11435 frame rate, only defined for constant frame-rate video
11438 The presentation timestamp in input
11441 The count of the input frame for video or the number of consumed samples,
11442 not including the current frame for audio, starting from 0.
11444 @item NB_CONSUMED_SAMPLES
11445 The number of consumed samples, not including the current frame (only
11448 @item NB_SAMPLES, S
11449 The number of samples in the current frame (only audio)
11451 @item SAMPLE_RATE, SR
11452 The audio sample rate.
11455 The PTS of the first frame.
11458 the time in seconds of the first frame
11461 State whether the current frame is interlaced.
11464 the time in seconds of the current frame
11467 original position in the file of the frame, or undefined if undefined
11468 for the current frame
11471 The previous input PTS.
11474 previous input time in seconds
11477 The previous output PTS.
11480 previous output time in seconds
11483 The wallclock (RTC) time in microseconds. This is deprecated, use time(0)
11487 The wallclock (RTC) time at the start of the movie in microseconds.
11490 The timebase of the input timestamps.
11494 @subsection Examples
11498 Start counting PTS from zero
11500 setpts=PTS-STARTPTS
11504 Apply fast motion effect:
11510 Apply slow motion effect:
11516 Set fixed rate of 25 frames per second:
11522 Set fixed rate 25 fps with some jitter:
11524 setpts='1/(25*TB) * (N + 0.05 * sin(N*2*PI/25))'
11528 Apply an offset of 10 seconds to the input PTS:
11534 Generate timestamps from a "live source" and rebase onto the current timebase:
11536 setpts='(RTCTIME - RTCSTART) / (TB * 1000000)'
11540 Generate timestamps by counting samples:
11547 @section settb, asettb
11549 Set the timebase to use for the output frames timestamps.
11550 It is mainly useful for testing timebase configuration.
11552 It accepts the following parameters:
11557 The expression which is evaluated into the output timebase.
11561 The value for @option{tb} is an arithmetic expression representing a
11562 rational. The expression can contain the constants "AVTB" (the default
11563 timebase), "intb" (the input timebase) and "sr" (the sample rate,
11564 audio only). Default value is "intb".
11566 @subsection Examples
11570 Set the timebase to 1/25:
11576 Set the timebase to 1/10:
11582 Set the timebase to 1001/1000:
11588 Set the timebase to 2*intb:
11594 Set the default timebase value:
11601 Convert input audio to a video output representing
11602 frequency spectrum logarithmically (using constant Q transform with
11603 Brown-Puckette algorithm), with musical tone scale, from E0 to D#10 (10 octaves).
11605 The filter accepts the following options:
11609 Specify transform volume (multiplier) expression. The expression can contain
11612 @item frequency, freq, f
11613 the frequency where transform is evaluated
11614 @item timeclamp, tc
11615 value of timeclamp option
11619 @item a_weighting(f)
11620 A-weighting of equal loudness
11621 @item b_weighting(f)
11622 B-weighting of equal loudness
11623 @item c_weighting(f)
11624 C-weighting of equal loudness
11626 Default value is @code{16}.
11629 Specify transform length expression. The expression can contain variables:
11631 @item frequency, freq, f
11632 the frequency where transform is evaluated
11633 @item timeclamp, tc
11634 value of timeclamp option
11636 Default value is @code{384/f*tc/(384/f+tc)}.
11639 Specify the transform timeclamp. At low frequency, there is trade-off between
11640 accuracy in time domain and frequency domain. If timeclamp is lower,
11641 event in time domain is represented more accurately (such as fast bass drum),
11642 otherwise event in frequency domain is represented more accurately
11643 (such as bass guitar). Acceptable value is [0.1, 1.0]. Default value is @code{0.17}.
11646 Specify the transform coeffclamp. If coeffclamp is lower, transform is
11647 more accurate, otherwise transform is faster. Acceptable value is [0.1, 10.0].
11648 Default value is @code{1.0}.
11651 Specify gamma. Lower gamma makes the spectrum more contrast, higher gamma
11652 makes the spectrum having more range. Acceptable value is [1.0, 7.0].
11653 Default value is @code{3.0}.
11656 Specify gamma of bargraph. Acceptable value is [1.0, 7.0].
11657 Default value is @code{1.0}.
11660 Specify font file for use with freetype. If not specified, use embedded font.
11663 Specify font color expression. This is arithmetic expression that should return
11664 integer value 0xRRGGBB. The expression can contain variables:
11666 @item frequency, freq, f
11667 the frequency where transform is evaluated
11668 @item timeclamp, tc
11669 value of timeclamp option
11674 midi number of frequency f, some midi numbers: E0(16), C1(24), C2(36), A4(69)
11675 @item r(x), g(x), b(x)
11676 red, green, and blue value of intensity x
11678 Default value is @code{st(0, (midi(f)-59.5)/12);
11679 st(1, if(between(ld(0),0,1), 0.5-0.5*cos(2*PI*ld(0)), 0));
11680 r(1-ld(1)) + b(ld(1))}
11683 If set to 1 (the default), the video size is 1920x1080 (full HD),
11684 if set to 0, the video size is 960x540. Use this option to make CPU usage lower.
11687 Specify video fps. Default value is @code{25}.
11690 Specify number of transform per frame, so there are fps*count transforms
11691 per second. Note that audio data rate must be divisible by fps*count.
11692 Default value is @code{6}.
11696 @subsection Examples
11700 Playing audio while showing the spectrum:
11702 ffplay -f lavfi 'amovie=a.mp3, asplit [a][out1]; [a] showcqt [out0]'
11706 Same as above, but with frame rate 30 fps:
11708 ffplay -f lavfi 'amovie=a.mp3, asplit [a][out1]; [a] showcqt=fps=30:count=5 [out0]'
11712 Playing at 960x540 and lower CPU usage:
11714 ffplay -f lavfi 'amovie=a.mp3, asplit [a][out1]; [a] showcqt=fullhd=0:count=3 [out0]'
11718 A1 and its harmonics: A1, A2, (near)E3, A3:
11720 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),
11721 asplit[a][out1]; [a] showcqt [out0]'
11725 Same as above, but with more accuracy in frequency domain (and slower):
11727 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),
11728 asplit[a][out1]; [a] showcqt=timeclamp=0.5 [out0]'
11732 B-weighting of equal loudness
11734 volume=16*b_weighting(f)
11740 tlength=100/f*tc/(100/f+tc)
11744 Custom fontcolor, C-note is colored green, others are colored blue
11746 fontcolor='if(mod(floor(midi(f)+0.5),12), 0x0000FF, g(1))'
11750 Custom gamma, now spectrum is linear to the amplitude.
11757 @section showspectrum
11759 Convert input audio to a video output, representing the audio frequency
11762 The filter accepts the following options:
11766 Specify the video size for the output. For the syntax of this option, check the
11767 @ref{video size syntax,,"Video size" section in the ffmpeg-utils manual,ffmpeg-utils}.
11768 Default value is @code{640x512}.
11771 Specify how the spectrum should slide along the window.
11773 It accepts the following values:
11776 the samples start again on the left when they reach the right
11778 the samples scroll from right to left
11780 frames are only produced when the samples reach the right
11783 Default value is @code{replace}.
11786 Specify display mode.
11788 It accepts the following values:
11791 all channels are displayed in the same row
11793 all channels are displayed in separate rows
11796 Default value is @samp{combined}.
11799 Specify display color mode.
11801 It accepts the following values:
11804 each channel is displayed in a separate color
11806 each channel is is displayed using the same color scheme
11809 Default value is @samp{channel}.
11812 Specify scale used for calculating intensity color values.
11814 It accepts the following values:
11819 square root, default
11826 Default value is @samp{sqrt}.
11829 Set saturation modifier for displayed colors. Negative values provide
11830 alternative color scheme. @code{0} is no saturation at all.
11831 Saturation must be in [-10.0, 10.0] range.
11832 Default value is @code{1}.
11835 Set window function.
11837 It accepts the following values:
11840 No samples pre-processing (do not expect this to be faster)
11849 Default value is @code{hann}.
11852 The usage is very similar to the showwaves filter; see the examples in that
11855 @subsection Examples
11859 Large window with logarithmic color scaling:
11861 showspectrum=s=1280x480:scale=log
11865 Complete example for a colored and sliding spectrum per channel using @command{ffplay}:
11867 ffplay -f lavfi 'amovie=input.mp3, asplit [a][out1];
11868 [a] showspectrum=mode=separate:color=intensity:slide=1:scale=cbrt [out0]'
11874 Convert input audio to a video output, representing the samples waves.
11876 The filter accepts the following options:
11880 Specify the video size for the output. For the syntax of this option, check the
11881 @ref{video size syntax,,"Video size" section in the ffmpeg-utils manual,ffmpeg-utils}.
11882 Default value is @code{600x240}.
11887 Available values are:
11890 Draw a point for each sample.
11893 Draw a vertical line for each sample.
11896 Draw a point for each sample and a line between them.
11899 Draw a centered vertical line for each sample.
11902 Default value is @code{point}.
11905 Set the number of samples which are printed on the same column. A
11906 larger value will decrease the frame rate. Must be a positive
11907 integer. This option can be set only if the value for @var{rate}
11908 is not explicitly specified.
11911 Set the (approximate) output frame rate. This is done by setting the
11912 option @var{n}. Default value is "25".
11914 @item split_channels
11915 Set if channels should be drawn separately or overlap. Default value is 0.
11919 @subsection Examples
11923 Output the input file audio and the corresponding video representation
11926 amovie=a.mp3,asplit[out0],showwaves[out1]
11930 Create a synthetic signal and show it with showwaves, forcing a
11931 frame rate of 30 frames per second:
11933 aevalsrc=sin(1*2*PI*t)*sin(880*2*PI*t):cos(2*PI*200*t),asplit[out0],showwaves=r=30[out1]
11937 @section showwavespic
11939 Convert input audio to a single video frame, representing the samples waves.
11941 The filter accepts the following options:
11945 Specify the video size for the output. For the syntax of this option, check the
11946 @ref{video size syntax,,"Video size" section in the ffmpeg-utils manual,ffmpeg-utils}.
11947 Default value is @code{600x240}.
11949 @item split_channels
11950 Set if channels should be drawn separately or overlap. Default value is 0.
11953 @subsection Examples
11957 Extract a channel split representation of the wave form of a whole audio track
11958 in a 1024x800 picture using @command{ffmpeg}:
11960 ffmpeg -i audio.flac -lavfi showwavespic=split_channels=1:s=1024x800 waveform.png
11964 @section split, asplit
11966 Split input into several identical outputs.
11968 @code{asplit} works with audio input, @code{split} with video.
11970 The filter accepts a single parameter which specifies the number of outputs. If
11971 unspecified, it defaults to 2.
11973 @subsection Examples
11977 Create two separate outputs from the same input:
11979 [in] split [out0][out1]
11983 To create 3 or more outputs, you need to specify the number of
11986 [in] asplit=3 [out0][out1][out2]
11990 Create two separate outputs from the same input, one cropped and
11993 [in] split [splitout1][splitout2];
11994 [splitout1] crop=100:100:0:0 [cropout];
11995 [splitout2] pad=200:200:100:100 [padout];
11999 Create 5 copies of the input audio with @command{ffmpeg}:
12001 ffmpeg -i INPUT -filter_complex asplit=5 OUTPUT
12007 Receive commands sent through a libzmq client, and forward them to
12008 filters in the filtergraph.
12010 @code{zmq} and @code{azmq} work as a pass-through filters. @code{zmq}
12011 must be inserted between two video filters, @code{azmq} between two
12014 To enable these filters you need to install the libzmq library and
12015 headers and configure FFmpeg with @code{--enable-libzmq}.
12017 For more information about libzmq see:
12018 @url{http://www.zeromq.org/}
12020 The @code{zmq} and @code{azmq} filters work as a libzmq server, which
12021 receives messages sent through a network interface defined by the
12022 @option{bind_address} option.
12024 The received message must be in the form:
12026 @var{TARGET} @var{COMMAND} [@var{ARG}]
12029 @var{TARGET} specifies the target of the command, usually the name of
12030 the filter class or a specific filter instance name.
12032 @var{COMMAND} specifies the name of the command for the target filter.
12034 @var{ARG} is optional and specifies the optional argument list for the
12035 given @var{COMMAND}.
12037 Upon reception, the message is processed and the corresponding command
12038 is injected into the filtergraph. Depending on the result, the filter
12039 will send a reply to the client, adopting the format:
12041 @var{ERROR_CODE} @var{ERROR_REASON}
12045 @var{MESSAGE} is optional.
12047 @subsection Examples
12049 Look at @file{tools/zmqsend} for an example of a zmq client which can
12050 be used to send commands processed by these filters.
12052 Consider the following filtergraph generated by @command{ffplay}
12054 ffplay -dumpgraph 1 -f lavfi "
12055 color=s=100x100:c=red [l];
12056 color=s=100x100:c=blue [r];
12057 nullsrc=s=200x100, zmq [bg];
12058 [bg][l] overlay [bg+l];
12059 [bg+l][r] overlay=x=100 "
12062 To change the color of the left side of the video, the following
12063 command can be used:
12065 echo Parsed_color_0 c yellow | tools/zmqsend
12068 To change the right side:
12070 echo Parsed_color_1 c pink | tools/zmqsend
12073 @c man end MULTIMEDIA FILTERS
12075 @chapter Multimedia Sources
12076 @c man begin MULTIMEDIA SOURCES
12078 Below is a description of the currently available multimedia sources.
12082 This is the same as @ref{movie} source, except it selects an audio
12088 Read audio and/or video stream(s) from a movie container.
12090 It accepts the following parameters:
12094 The name of the resource to read (not necessarily a file; it can also be a
12095 device or a stream accessed through some protocol).
12097 @item format_name, f
12098 Specifies the format assumed for the movie to read, and can be either
12099 the name of a container or an input device. If not specified, the
12100 format is guessed from @var{movie_name} or by probing.
12102 @item seek_point, sp
12103 Specifies the seek point in seconds. The frames will be output
12104 starting from this seek point. The parameter is evaluated with
12105 @code{av_strtod}, so the numerical value may be suffixed by an IS
12106 postfix. The default value is "0".
12109 Specifies the streams to read. Several streams can be specified,
12110 separated by "+". The source will then have as many outputs, in the
12111 same order. The syntax is explained in the ``Stream specifiers''
12112 section in the ffmpeg manual. Two special names, "dv" and "da" specify
12113 respectively the default (best suited) video and audio stream. Default
12114 is "dv", or "da" if the filter is called as "amovie".
12116 @item stream_index, si
12117 Specifies the index of the video stream to read. If the value is -1,
12118 the most suitable video stream will be automatically selected. The default
12119 value is "-1". Deprecated. If the filter is called "amovie", it will select
12120 audio instead of video.
12123 Specifies how many times to read the stream in sequence.
12124 If the value is less than 1, the stream will be read again and again.
12125 Default value is "1".
12127 Note that when the movie is looped the source timestamps are not
12128 changed, so it will generate non monotonically increasing timestamps.
12131 It allows overlaying a second video on top of the main input of
12132 a filtergraph, as shown in this graph:
12134 input -----------> deltapts0 --> overlay --> output
12137 movie --> scale--> deltapts1 -------+
12139 @subsection Examples
12143 Skip 3.2 seconds from the start of the AVI file in.avi, and overlay it
12144 on top of the input labelled "in":
12146 movie=in.avi:seek_point=3.2, scale=180:-1, setpts=PTS-STARTPTS [over];
12147 [in] setpts=PTS-STARTPTS [main];
12148 [main][over] overlay=16:16 [out]
12152 Read from a video4linux2 device, and overlay it on top of the input
12155 movie=/dev/video0:f=video4linux2, scale=180:-1, setpts=PTS-STARTPTS [over];
12156 [in] setpts=PTS-STARTPTS [main];
12157 [main][over] overlay=16:16 [out]
12161 Read the first video stream and the audio stream with id 0x81 from
12162 dvd.vob; the video is connected to the pad named "video" and the audio is
12163 connected to the pad named "audio":
12165 movie=dvd.vob:s=v:0+#0x81 [video] [audio]
12169 @c man end MULTIMEDIA SOURCES