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
540 select inverted quarter of sine wave
542 select inverted half of sine wave
544 select double-exponential seat
546 select double-exponential sigmoid
554 Fade in first 15 seconds of audio:
560 Fade out last 25 seconds of a 900 seconds audio:
562 afade=t=out:st=875:d=25
569 Set output format constraints for the input audio. The framework will
570 negotiate the most appropriate format to minimize conversions.
572 It accepts the following parameters:
576 A '|'-separated list of requested sample formats.
579 A '|'-separated list of requested sample rates.
581 @item channel_layouts
582 A '|'-separated list of requested channel layouts.
584 See @ref{channel layout syntax,,the Channel Layout section in the ffmpeg-utils(1) manual,ffmpeg-utils}
585 for the required syntax.
588 If a parameter is omitted, all values are allowed.
590 Force the output to either unsigned 8-bit or signed 16-bit stereo
592 aformat=sample_fmts=u8|s16:channel_layouts=stereo
597 Apply a two-pole all-pass filter with central frequency (in Hz)
598 @var{frequency}, and filter-width @var{width}.
599 An all-pass filter changes the audio's frequency to phase relationship
600 without changing its frequency to amplitude relationship.
602 The filter accepts the following options:
609 Set method to specify band-width of filter.
622 Specify the band-width of a filter in width_type units.
627 Merge two or more audio streams into a single multi-channel stream.
629 The filter accepts the following options:
634 Set the number of inputs. Default is 2.
638 If the channel layouts of the inputs are disjoint, and therefore compatible,
639 the channel layout of the output will be set accordingly and the channels
640 will be reordered as necessary. If the channel layouts of the inputs are not
641 disjoint, the output will have all the channels of the first input then all
642 the channels of the second input, in that order, and the channel layout of
643 the output will be the default value corresponding to the total number of
646 For example, if the first input is in 2.1 (FL+FR+LF) and the second input
647 is FC+BL+BR, then the output will be in 5.1, with the channels in the
648 following order: a1, a2, b1, a3, b2, b3 (a1 is the first channel of the
649 first input, b1 is the first channel of the second input).
651 On the other hand, if both input are in stereo, the output channels will be
652 in the default order: a1, a2, b1, b2, and the channel layout will be
653 arbitrarily set to 4.0, which may or may not be the expected value.
655 All inputs must have the same sample rate, and format.
657 If inputs do not have the same duration, the output will stop with the
664 Merge two mono files into a stereo stream:
666 amovie=left.wav [l] ; amovie=right.mp3 [r] ; [l] [r] amerge
670 Multiple merges assuming 1 video stream and 6 audio streams in @file{input.mkv}:
672 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
678 Mixes multiple audio inputs into a single output.
680 Note that this filter only supports float samples (the @var{amerge}
681 and @var{pan} audio filters support many formats). If the @var{amix}
682 input has integer samples then @ref{aresample} will be automatically
683 inserted to perform the conversion to float samples.
687 ffmpeg -i INPUT1 -i INPUT2 -i INPUT3 -filter_complex amix=inputs=3:duration=first:dropout_transition=3 OUTPUT
689 will mix 3 input audio streams to a single output with the same duration as the
690 first input and a dropout transition time of 3 seconds.
692 It accepts the following parameters:
696 The number of inputs. If unspecified, it defaults to 2.
699 How to determine the end-of-stream.
703 The duration of the longest input. (default)
706 The duration of the shortest input.
709 The duration of the first input.
713 @item dropout_transition
714 The transition time, in seconds, for volume renormalization when an input
715 stream ends. The default value is 2 seconds.
721 Pass the audio source unchanged to the output.
725 Pad the end of an audio stream with silence.
727 This can be used together with @command{ffmpeg} @option{-shortest} to
728 extend audio streams to the same length as the video stream.
730 A description of the accepted options follows.
734 Set silence packet size. Default value is 4096.
737 Set the number of samples of silence to add to the end. After the
738 value is reached, the stream is terminated. This option is mutually
739 exclusive with @option{whole_len}.
742 Set the minimum total number of samples in the output audio stream. If
743 the value is longer than the input audio length, silence is added to
744 the end, until the value is reached. This option is mutually exclusive
745 with @option{pad_len}.
748 If neither the @option{pad_len} nor the @option{whole_len} option is
749 set, the filter will add silence to the end of the input stream
756 Add 1024 samples of silence to the end of the input:
762 Make sure the audio output will contain at least 10000 samples, pad
763 the input with silence if required:
769 Use @command{ffmpeg} to pad the audio input with silence, so that the
770 video stream will always result the shortest and will be converted
771 until the end in the output file when using the @option{shortest}
774 ffmpeg -i VIDEO -i AUDIO -filter_complex "[1:0]apad" -shortest OUTPUT
779 Add a phasing effect to the input audio.
781 A phaser filter creates series of peaks and troughs in the frequency spectrum.
782 The position of the peaks and troughs are modulated so that they vary over time, creating a sweeping effect.
784 A description of the accepted parameters follows.
788 Set input gain. Default is 0.4.
791 Set output gain. Default is 0.74
794 Set delay in milliseconds. Default is 3.0.
797 Set decay. Default is 0.4.
800 Set modulation speed in Hz. Default is 0.5.
803 Set modulation type. Default is triangular.
805 It accepts the following values:
815 Resample the input audio to the specified parameters, using the
816 libswresample library. If none are specified then the filter will
817 automatically convert between its input and output.
819 This filter is also able to stretch/squeeze the audio data to make it match
820 the timestamps or to inject silence / cut out audio to make it match the
821 timestamps, do a combination of both or do neither.
823 The filter accepts the syntax
824 [@var{sample_rate}:]@var{resampler_options}, where @var{sample_rate}
825 expresses a sample rate and @var{resampler_options} is a list of
826 @var{key}=@var{value} pairs, separated by ":". See the
827 ffmpeg-resampler manual for the complete list of supported options.
833 Resample the input audio to 44100Hz:
839 Stretch/squeeze samples to the given timestamps, with a maximum of 1000
840 samples per second compensation:
846 @section asetnsamples
848 Set the number of samples per each output audio frame.
850 The last output packet may contain a different number of samples, as
851 the filter will flush all the remaining samples when the input audio
854 The filter accepts the following options:
858 @item nb_out_samples, n
859 Set the number of frames per each output audio frame. The number is
860 intended as the number of samples @emph{per each channel}.
861 Default value is 1024.
864 If set to 1, the filter will pad the last audio frame with zeroes, so
865 that the last frame will contain the same number of samples as the
866 previous ones. Default value is 1.
869 For example, to set the number of per-frame samples to 1234 and
870 disable padding for the last frame, use:
872 asetnsamples=n=1234:p=0
877 Set the sample rate without altering the PCM data.
878 This will result in a change of speed and pitch.
880 The filter accepts the following options:
884 Set the output sample rate. Default is 44100 Hz.
889 Show a line containing various information for each input audio frame.
890 The input audio is not modified.
892 The shown line contains a sequence of key/value pairs of the form
893 @var{key}:@var{value}.
895 The following values are shown in the output:
899 The (sequential) number of the input frame, starting from 0.
902 The presentation timestamp of the input frame, in time base units; the time base
903 depends on the filter input pad, and is usually 1/@var{sample_rate}.
906 The presentation timestamp of the input frame in seconds.
909 position of the frame in the input stream, -1 if this information in
910 unavailable and/or meaningless (for example in case of synthetic audio)
919 The sample rate for the audio frame.
922 The number of samples (per channel) in the frame.
925 The Adler-32 checksum (printed in hexadecimal) of the audio data. For planar
926 audio, the data is treated as if all the planes were concatenated.
928 @item plane_checksums
929 A list of Adler-32 checksums for each data plane.
935 Display time domain statistical information about the audio channels.
936 Statistics are calculated and displayed for each audio channel and,
937 where applicable, an overall figure is also given.
939 It accepts the following option:
942 Short window length in seconds, used for peak and trough RMS measurement.
943 Default is @code{0.05} (50 milliseconds). Allowed range is @code{[0.1 - 10]}.
947 Set metadata injection. All the metadata keys are prefixed with @code{lavfi.astats.X},
948 where @code{X} is channel number starting from 1 or string @code{Overall}. Default is
951 Available keys for each channel are:
974 For example full key look like this @code{lavfi.astats.1.DC_offset} or
975 this @code{lavfi.astats.Overall.Peak_count}.
977 For description what each key means read bellow.
981 A description of each shown parameter follows:
985 Mean amplitude displacement from zero.
988 Minimal sample level.
991 Maximal sample level.
995 Standard peak and RMS level measured in dBFS.
999 Peak and trough values for RMS level measured over a short window.
1002 Standard ratio of peak to RMS level (note: not in dB).
1005 Flatness (i.e. consecutive samples with the same value) of the signal at its peak levels
1006 (i.e. either @var{Min level} or @var{Max level}).
1009 Number of occasions (not the number of samples) that the signal attained either
1010 @var{Min level} or @var{Max level}.
1013 @section astreamsync
1015 Forward two audio streams and control the order the buffers are forwarded.
1017 The filter accepts the following options:
1021 Set the expression deciding which stream should be
1022 forwarded next: if the result is negative, the first stream is forwarded; if
1023 the result is positive or zero, the second stream is forwarded. It can use
1024 the following variables:
1028 number of buffers forwarded so far on each stream
1030 number of samples forwarded so far on each stream
1032 current timestamp of each stream
1035 The default value is @code{t1-t2}, which means to always forward the stream
1036 that has a smaller timestamp.
1039 @subsection Examples
1041 Stress-test @code{amerge} by randomly sending buffers on the wrong
1042 input, while avoiding too much of a desynchronization:
1044 amovie=file.ogg [a] ; amovie=file.mp3 [b] ;
1045 [a] [b] astreamsync=(2*random(1))-1+tanh(5*(t1-t2)) [a2] [b2] ;
1051 Synchronize audio data with timestamps by squeezing/stretching it and/or
1052 dropping samples/adding silence when needed.
1054 This filter is not built by default, please use @ref{aresample} to do squeezing/stretching.
1056 It accepts the following parameters:
1060 Enable stretching/squeezing the data to make it match the timestamps. Disabled
1061 by default. When disabled, time gaps are covered with silence.
1064 The minimum difference between timestamps and audio data (in seconds) to trigger
1065 adding/dropping samples. The default value is 0.1. If you get an imperfect
1066 sync with this filter, try setting this parameter to 0.
1069 The maximum compensation in samples per second. Only relevant with compensate=1.
1070 The default value is 500.
1073 Assume that the first PTS should be this value. The time base is 1 / sample
1074 rate. This allows for padding/trimming at the start of the stream. By default,
1075 no assumption is made about the first frame's expected PTS, so no padding or
1076 trimming is done. For example, this could be set to 0 to pad the beginning with
1077 silence if an audio stream starts after the video stream or to trim any samples
1078 with a negative PTS due to encoder delay.
1086 The filter accepts exactly one parameter, the audio tempo. If not
1087 specified then the filter will assume nominal 1.0 tempo. Tempo must
1088 be in the [0.5, 2.0] range.
1090 @subsection Examples
1094 Slow down audio to 80% tempo:
1100 To speed up audio to 125% tempo:
1108 Trim the input so that the output contains one continuous subpart of the input.
1110 It accepts the following parameters:
1113 Timestamp (in seconds) of the start of the section to keep. I.e. the audio
1114 sample with the timestamp @var{start} will be the first sample in the output.
1117 Specify time of the first audio sample that will be dropped, i.e. the
1118 audio sample immediately preceding the one with the timestamp @var{end} will be
1119 the last sample in the output.
1122 Same as @var{start}, except this option sets the start timestamp in samples
1126 Same as @var{end}, except this option sets the end timestamp in samples instead
1130 The maximum duration of the output in seconds.
1133 The number of the first sample that should be output.
1136 The number of the first sample that should be dropped.
1139 @option{start}, @option{end}, and @option{duration} are expressed as time
1140 duration specifications; see
1141 @ref{time duration syntax,,the Time duration section in the ffmpeg-utils(1) manual,ffmpeg-utils}.
1143 Note that the first two sets of the start/end options and the @option{duration}
1144 option look at the frame timestamp, while the _sample options simply count the
1145 samples that pass through the filter. So start/end_pts and start/end_sample will
1146 give different results when the timestamps are wrong, inexact or do not start at
1147 zero. Also note that this filter does not modify the timestamps. If you wish
1148 to have the output timestamps start at zero, insert the asetpts filter after the
1151 If multiple start or end options are set, this filter tries to be greedy and
1152 keep all samples that match at least one of the specified constraints. To keep
1153 only the part that matches all the constraints at once, chain multiple atrim
1156 The defaults are such that all the input is kept. So it is possible to set e.g.
1157 just the end values to keep everything before the specified time.
1162 Drop everything except the second minute of input:
1164 ffmpeg -i INPUT -af atrim=60:120
1168 Keep only the first 1000 samples:
1170 ffmpeg -i INPUT -af atrim=end_sample=1000
1177 Apply a two-pole Butterworth band-pass filter with central
1178 frequency @var{frequency}, and (3dB-point) band-width width.
1179 The @var{csg} option selects a constant skirt gain (peak gain = Q)
1180 instead of the default: constant 0dB peak gain.
1181 The filter roll off at 6dB per octave (20dB per decade).
1183 The filter accepts the following options:
1187 Set the filter's central frequency. Default is @code{3000}.
1190 Constant skirt gain if set to 1. Defaults to 0.
1193 Set method to specify band-width of filter.
1206 Specify the band-width of a filter in width_type units.
1211 Apply a two-pole Butterworth band-reject filter with central
1212 frequency @var{frequency}, and (3dB-point) band-width @var{width}.
1213 The filter roll off at 6dB per octave (20dB per decade).
1215 The filter accepts the following options:
1219 Set the filter's central frequency. Default is @code{3000}.
1222 Set method to specify band-width of filter.
1235 Specify the band-width of a filter in width_type units.
1240 Boost or cut the bass (lower) frequencies of the audio using a two-pole
1241 shelving filter with a response similar to that of a standard
1242 hi-fi's tone-controls. This is also known as shelving equalisation (EQ).
1244 The filter accepts the following options:
1248 Give the gain at 0 Hz. Its useful range is about -20
1249 (for a large cut) to +20 (for a large boost).
1250 Beware of clipping when using a positive gain.
1253 Set the filter's central frequency and so can be used
1254 to extend or reduce the frequency range to be boosted or cut.
1255 The default value is @code{100} Hz.
1258 Set method to specify band-width of filter.
1271 Determine how steep is the filter's shelf transition.
1276 Apply a biquad IIR filter with the given coefficients.
1277 Where @var{b0}, @var{b1}, @var{b2} and @var{a0}, @var{a1}, @var{a2}
1278 are the numerator and denominator coefficients respectively.
1281 Bauer stereo to binaural transformation, which improves headphone listening of
1282 stereo audio records.
1284 It accepts the following parameters:
1288 Pre-defined crossfeed level.
1292 Default level (fcut=700, feed=50).
1295 Chu Moy circuit (fcut=700, feed=60).
1298 Jan Meier circuit (fcut=650, feed=95).
1303 Cut frequency (in Hz).
1312 Remap input channels to new locations.
1314 It accepts the following parameters:
1316 @item channel_layout
1317 The channel layout of the output stream.
1320 Map channels from input to output. The argument is a '|'-separated list of
1321 mappings, each in the @code{@var{in_channel}-@var{out_channel}} or
1322 @var{in_channel} form. @var{in_channel} can be either the name of the input
1323 channel (e.g. FL for front left) or its index in the input channel layout.
1324 @var{out_channel} is the name of the output channel or its index in the output
1325 channel layout. If @var{out_channel} is not given then it is implicitly an
1326 index, starting with zero and increasing by one for each mapping.
1329 If no mapping is present, the filter will implicitly map input channels to
1330 output channels, preserving indices.
1332 For example, assuming a 5.1+downmix input MOV file,
1334 ffmpeg -i in.mov -filter 'channelmap=map=DL-FL|DR-FR' out.wav
1336 will create an output WAV file tagged as stereo from the downmix channels of
1339 To fix a 5.1 WAV improperly encoded in AAC's native channel order
1341 ffmpeg -i in.wav -filter 'channelmap=1|2|0|5|3|4:5.1' out.wav
1344 @section channelsplit
1346 Split each channel from an input audio stream into a separate output stream.
1348 It accepts the following parameters:
1350 @item channel_layout
1351 The channel layout of the input stream. The default is "stereo".
1354 For example, assuming a stereo input MP3 file,
1356 ffmpeg -i in.mp3 -filter_complex channelsplit out.mkv
1358 will create an output Matroska file with two audio streams, one containing only
1359 the left channel and the other the right channel.
1361 Split a 5.1 WAV file into per-channel files:
1363 ffmpeg -i in.wav -filter_complex
1364 'channelsplit=channel_layout=5.1[FL][FR][FC][LFE][SL][SR]'
1365 -map '[FL]' front_left.wav -map '[FR]' front_right.wav -map '[FC]'
1366 front_center.wav -map '[LFE]' lfe.wav -map '[SL]' side_left.wav -map '[SR]'
1371 Add a chorus effect to the audio.
1373 Can make a single vocal sound like a chorus, but can also be applied to instrumentation.
1375 Chorus resembles an echo effect with a short delay, but whereas with echo the delay is
1376 constant, with chorus, it is varied using using sinusoidal or triangular modulation.
1377 The modulation depth defines the range the modulated delay is played before or after
1378 the delay. Hence the delayed sound will sound slower or faster, that is the delayed
1379 sound tuned around the original one, like in a chorus where some vocals are slightly
1382 It accepts the following parameters:
1385 Set input gain. Default is 0.4.
1388 Set output gain. Default is 0.4.
1391 Set delays. A typical delay is around 40ms to 60ms.
1403 @subsection Examples
1409 chorus=0.7:0.9:55:0.4:0.25:2
1415 chorus=0.6:0.9:50|60:0.4|0.32:0.25|0.4:2|1.3
1419 Fuller sounding chorus with three delays:
1421 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
1426 Compress or expand the audio's dynamic range.
1428 It accepts the following parameters:
1434 A list of times in seconds for each channel over which the instantaneous level
1435 of the input signal is averaged to determine its volume. @var{attacks} refers to
1436 increase of volume and @var{decays} refers to decrease of volume. For most
1437 situations, the attack time (response to the audio getting louder) should be
1438 shorter than the decay time, because the human ear is more sensitive to sudden
1439 loud audio than sudden soft audio. A typical value for attack is 0.3 seconds and
1440 a typical value for decay is 0.8 seconds.
1443 A list of points for the transfer function, specified in dB relative to the
1444 maximum possible signal amplitude. Each key points list must be defined using
1445 the following syntax: @code{x0/y0|x1/y1|x2/y2|....} or
1446 @code{x0/y0 x1/y1 x2/y2 ....}
1448 The input values must be in strictly increasing order but the transfer function
1449 does not have to be monotonically rising. The point @code{0/0} is assumed but
1450 may be overridden (by @code{0/out-dBn}). Typical values for the transfer
1451 function are @code{-70/-70|-60/-20}.
1454 Set the curve radius in dB for all joints. It defaults to 0.01.
1457 Set the additional gain in dB to be applied at all points on the transfer
1458 function. This allows for easy adjustment of the overall gain.
1462 Set an initial volume, in dB, to be assumed for each channel when filtering
1463 starts. This permits the user to supply a nominal level initially, so that, for
1464 example, a very large gain is not applied to initial signal levels before the
1465 companding has begun to operate. A typical value for audio which is initially
1466 quiet is -90 dB. It defaults to 0.
1469 Set a delay, in seconds. The input audio is analyzed immediately, but audio is
1470 delayed before being fed to the volume adjuster. Specifying a delay
1471 approximately equal to the attack/decay times allows the filter to effectively
1472 operate in predictive rather than reactive mode. It defaults to 0.
1476 @subsection Examples
1480 Make music with both quiet and loud passages suitable for listening to in a
1483 compand=.3|.3:1|1:-90/-60|-60/-40|-40/-30|-20/-20:6:0:-90:0.2
1486 Another example for audio with whisper and explosion parts:
1488 compand=0|0:1|1:-90/-900|-70/-70|-30/-9|0/-3:6:0:0:0
1492 A noise gate for when the noise is at a lower level than the signal:
1494 compand=.1|.1:.2|.2:-900/-900|-50.1/-900|-50/-50:.01:0:-90:.1
1498 Here is another noise gate, this time for when the noise is at a higher level
1499 than the signal (making it, in some ways, similar to squelch):
1501 compand=.1|.1:.1|.1:-45.1/-45.1|-45/-900|0/-900:.01:45:-90:.1
1506 Apply a DC shift to the audio.
1508 This can be useful to remove a DC offset (caused perhaps by a hardware problem
1509 in the recording chain) from the audio. The effect of a DC offset is reduced
1510 headroom and hence volume. The @ref{astats} filter can be used to determine if
1511 a signal has a DC offset.
1515 Set the DC shift, allowed range is [-1, 1]. It indicates the amount to shift
1519 Optional. It should have a value much less than 1 (e.g. 0.05 or 0.02) and is
1520 used to prevent clipping.
1525 Make audio easier to listen to on headphones.
1527 This filter adds `cues' to 44.1kHz stereo (i.e. audio CD format) audio
1528 so that when listened to on headphones the stereo image is moved from
1529 inside your head (standard for headphones) to outside and in front of
1530 the listener (standard for speakers).
1536 Apply a two-pole peaking equalisation (EQ) filter. With this
1537 filter, the signal-level at and around a selected frequency can
1538 be increased or decreased, whilst (unlike bandpass and bandreject
1539 filters) that at all other frequencies is unchanged.
1541 In order to produce complex equalisation curves, this filter can
1542 be given several times, each with a different central frequency.
1544 The filter accepts the following options:
1548 Set the filter's central frequency in Hz.
1551 Set method to specify band-width of filter.
1564 Specify the band-width of a filter in width_type units.
1567 Set the required gain or attenuation in dB.
1568 Beware of clipping when using a positive gain.
1571 @subsection Examples
1574 Attenuate 10 dB at 1000 Hz, with a bandwidth of 200 Hz:
1576 equalizer=f=1000:width_type=h:width=200:g=-10
1580 Apply 2 dB gain at 1000 Hz with Q 1 and attenuate 5 dB at 100 Hz with Q 2:
1582 equalizer=f=1000:width_type=q:width=1:g=2,equalizer=f=100:width_type=q:width=2:g=-5
1587 Apply a flanging effect to the audio.
1589 The filter accepts the following options:
1593 Set base delay in milliseconds. Range from 0 to 30. Default value is 0.
1596 Set added swep delay in milliseconds. Range from 0 to 10. Default value is 2.
1599 Set percentage regeneration (delayed signal feedback). Range from -95 to 95.
1603 Set percentage of delayed signal mixed with original. Range from 0 to 100.
1604 Default value is 71.
1607 Set sweeps per second (Hz). Range from 0.1 to 10. Default value is 0.5.
1610 Set swept wave shape, can be @var{triangular} or @var{sinusoidal}.
1611 Default value is @var{sinusoidal}.
1614 Set swept wave percentage-shift for multi channel. Range from 0 to 100.
1615 Default value is 25.
1618 Set delay-line interpolation, @var{linear} or @var{quadratic}.
1619 Default is @var{linear}.
1624 Apply a high-pass filter with 3dB point frequency.
1625 The filter can be either single-pole, or double-pole (the default).
1626 The filter roll off at 6dB per pole per octave (20dB per pole per decade).
1628 The filter accepts the following options:
1632 Set frequency in Hz. Default is 3000.
1635 Set number of poles. Default is 2.
1638 Set method to specify band-width of filter.
1651 Specify the band-width of a filter in width_type units.
1652 Applies only to double-pole filter.
1653 The default is 0.707q and gives a Butterworth response.
1658 Join multiple input streams into one multi-channel stream.
1660 It accepts the following parameters:
1664 The number of input streams. It defaults to 2.
1666 @item channel_layout
1667 The desired output channel layout. It defaults to stereo.
1670 Map channels from inputs to output. The argument is a '|'-separated list of
1671 mappings, each in the @code{@var{input_idx}.@var{in_channel}-@var{out_channel}}
1672 form. @var{input_idx} is the 0-based index of the input stream. @var{in_channel}
1673 can be either the name of the input channel (e.g. FL for front left) or its
1674 index in the specified input stream. @var{out_channel} is the name of the output
1678 The filter will attempt to guess the mappings when they are not specified
1679 explicitly. It does so by first trying to find an unused matching input channel
1680 and if that fails it picks the first unused input channel.
1682 Join 3 inputs (with properly set channel layouts):
1684 ffmpeg -i INPUT1 -i INPUT2 -i INPUT3 -filter_complex join=inputs=3 OUTPUT
1687 Build a 5.1 output from 6 single-channel streams:
1689 ffmpeg -i fl -i fr -i fc -i sl -i sr -i lfe -filter_complex
1690 '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'
1696 Load a LADSPA (Linux Audio Developer's Simple Plugin API) plugin.
1698 To enable compilation of this filter you need to configure FFmpeg with
1699 @code{--enable-ladspa}.
1703 Specifies the name of LADSPA plugin library to load. If the environment
1704 variable @env{LADSPA_PATH} is defined, the LADSPA plugin is searched in
1705 each one of the directories specified by the colon separated list in
1706 @env{LADSPA_PATH}, otherwise in the standard LADSPA paths, which are in
1707 this order: @file{HOME/.ladspa/lib/}, @file{/usr/local/lib/ladspa/},
1708 @file{/usr/lib/ladspa/}.
1711 Specifies the plugin within the library. Some libraries contain only
1712 one plugin, but others contain many of them. If this is not set filter
1713 will list all available plugins within the specified library.
1716 Set the '|' separated list of controls which are zero or more floating point
1717 values that determine the behavior of the loaded plugin (for example delay,
1719 Controls need to be defined using the following syntax:
1720 c0=@var{value0}|c1=@var{value1}|c2=@var{value2}|..., where
1721 @var{valuei} is the value set on the @var{i}-th control.
1722 If @option{controls} is set to @code{help}, all available controls and
1723 their valid ranges are printed.
1725 @item sample_rate, s
1726 Specify the sample rate, default to 44100. Only used if plugin have
1730 Set the number of samples per channel per each output frame, default
1731 is 1024. Only used if plugin have zero inputs.
1734 Set the minimum duration of the sourced audio. See
1735 @ref{time duration syntax,,the Time duration section in the ffmpeg-utils(1) manual,ffmpeg-utils}
1736 for the accepted syntax.
1737 Note that the resulting duration may be greater than the specified duration,
1738 as the generated audio is always cut at the end of a complete frame.
1739 If not specified, or the expressed duration is negative, the audio is
1740 supposed to be generated forever.
1741 Only used if plugin have zero inputs.
1745 @subsection Examples
1749 List all available plugins within amp (LADSPA example plugin) library:
1755 List all available controls and their valid ranges for @code{vcf_notch}
1756 plugin from @code{VCF} library:
1758 ladspa=f=vcf:p=vcf_notch:c=help
1762 Simulate low quality audio equipment using @code{Computer Music Toolkit} (CMT)
1765 ladspa=file=cmt:plugin=lofi:controls=c0=22|c1=12|c2=12
1769 Add reverberation to the audio using TAP-plugins
1770 (Tom's Audio Processing plugins):
1772 ladspa=file=tap_reverb:tap_reverb
1776 Generate white noise, with 0.2 amplitude:
1778 ladspa=file=cmt:noise_source_white:c=c0=.2
1782 Generate 20 bpm clicks using plugin @code{C* Click - Metronome} from the
1783 @code{C* Audio Plugin Suite} (CAPS) library:
1785 ladspa=file=caps:Click:c=c1=20'
1789 Apply @code{C* Eq10X2 - Stereo 10-band equaliser} effect:
1791 ladspa=caps:Eq10X2:c=c0=-48|c9=-24|c3=12|c4=2
1795 @subsection Commands
1797 This filter supports the following commands:
1800 Modify the @var{N}-th control value.
1802 If the specified value is not valid, it is ignored and prior one is kept.
1807 Apply a low-pass filter with 3dB point frequency.
1808 The filter can be either single-pole or double-pole (the default).
1809 The filter roll off at 6dB per pole per octave (20dB per pole per decade).
1811 The filter accepts the following options:
1815 Set frequency in Hz. Default is 500.
1818 Set number of poles. Default is 2.
1821 Set method to specify band-width of filter.
1834 Specify the band-width of a filter in width_type units.
1835 Applies only to double-pole filter.
1836 The default is 0.707q and gives a Butterworth response.
1841 Mix channels with specific gain levels. The filter accepts the output
1842 channel layout followed by a set of channels definitions.
1844 This filter is also designed to efficiently remap the channels of an audio
1847 The filter accepts parameters of the form:
1848 "@var{l}|@var{outdef}|@var{outdef}|..."
1852 output channel layout or number of channels
1855 output channel specification, of the form:
1856 "@var{out_name}=[@var{gain}*]@var{in_name}[+[@var{gain}*]@var{in_name}...]"
1859 output channel to define, either a channel name (FL, FR, etc.) or a channel
1860 number (c0, c1, etc.)
1863 multiplicative coefficient for the channel, 1 leaving the volume unchanged
1866 input channel to use, see out_name for details; it is not possible to mix
1867 named and numbered input channels
1870 If the `=' in a channel specification is replaced by `<', then the gains for
1871 that specification will be renormalized so that the total is 1, thus
1872 avoiding clipping noise.
1874 @subsection Mixing examples
1876 For example, if you want to down-mix from stereo to mono, but with a bigger
1877 factor for the left channel:
1879 pan=1c|c0=0.9*c0+0.1*c1
1882 A customized down-mix to stereo that works automatically for 3-, 4-, 5- and
1883 7-channels surround:
1885 pan=stereo| FL < FL + 0.5*FC + 0.6*BL + 0.6*SL | FR < FR + 0.5*FC + 0.6*BR + 0.6*SR
1888 Note that @command{ffmpeg} integrates a default down-mix (and up-mix) system
1889 that should be preferred (see "-ac" option) unless you have very specific
1892 @subsection Remapping examples
1894 The channel remapping will be effective if, and only if:
1897 @item gain coefficients are zeroes or ones,
1898 @item only one input per channel output,
1901 If all these conditions are satisfied, the filter will notify the user ("Pure
1902 channel mapping detected"), and use an optimized and lossless method to do the
1905 For example, if you have a 5.1 source and want a stereo audio stream by
1906 dropping the extra channels:
1908 pan="stereo| c0=FL | c1=FR"
1911 Given the same source, you can also switch front left and front right channels
1912 and keep the input channel layout:
1914 pan="5.1| c0=c1 | c1=c0 | c2=c2 | c3=c3 | c4=c4 | c5=c5"
1917 If the input is a stereo audio stream, you can mute the front left channel (and
1918 still keep the stereo channel layout) with:
1923 Still with a stereo audio stream input, you can copy the right channel in both
1924 front left and right:
1926 pan="stereo| c0=FR | c1=FR"
1931 ReplayGain scanner filter. This filter takes an audio stream as an input and
1932 outputs it unchanged.
1933 At end of filtering it displays @code{track_gain} and @code{track_peak}.
1937 Convert the audio sample format, sample rate and channel layout. It is
1938 not meant to be used directly.
1940 @section silencedetect
1942 Detect silence in an audio stream.
1944 This filter logs a message when it detects that the input audio volume is less
1945 or equal to a noise tolerance value for a duration greater or equal to the
1946 minimum detected noise duration.
1948 The printed times and duration are expressed in seconds.
1950 The filter accepts the following options:
1954 Set silence duration until notification (default is 2 seconds).
1957 Set noise tolerance. Can be specified in dB (in case "dB" is appended to the
1958 specified value) or amplitude ratio. Default is -60dB, or 0.001.
1961 @subsection Examples
1965 Detect 5 seconds of silence with -50dB noise tolerance:
1967 silencedetect=n=-50dB:d=5
1971 Complete example with @command{ffmpeg} to detect silence with 0.0001 noise
1972 tolerance in @file{silence.mp3}:
1974 ffmpeg -i silence.mp3 -af silencedetect=noise=0.0001 -f null -
1978 @section silenceremove
1980 Remove silence from the beginning, middle or end of the audio.
1982 The filter accepts the following options:
1986 This value is used to indicate if audio should be trimmed at beginning of
1987 the audio. A value of zero indicates no silence should be trimmed from the
1988 beginning. When specifying a non-zero value, it trims audio up until it
1989 finds non-silence. Normally, when trimming silence from beginning of audio
1990 the @var{start_periods} will be @code{1} but it can be increased to higher
1991 values to trim all audio up to specific count of non-silence periods.
1992 Default value is @code{0}.
1994 @item start_duration
1995 Specify the amount of time that non-silence must be detected before it stops
1996 trimming audio. By increasing the duration, bursts of noises can be treated
1997 as silence and trimmed off. Default value is @code{0}.
1999 @item start_threshold
2000 This indicates what sample value should be treated as silence. For digital
2001 audio, a value of @code{0} may be fine but for audio recorded from analog,
2002 you may wish to increase the value to account for background noise.
2003 Can be specified in dB (in case "dB" is appended to the specified value)
2004 or amplitude ratio. Default value is @code{0}.
2007 Set the count for trimming silence from the end of audio.
2008 To remove silence from the middle of a file, specify a @var{stop_periods}
2009 that is negative. This value is then treated as a positive value and is
2010 used to indicate the effect should restart processing as specified by
2011 @var{start_periods}, making it suitable for removing periods of silence
2012 in the middle of the audio.
2013 Default value is @code{0}.
2016 Specify a duration of silence that must exist before audio is not copied any
2017 more. By specifying a higher duration, silence that is wanted can be left in
2019 Default value is @code{0}.
2021 @item stop_threshold
2022 This is the same as @option{start_threshold} but for trimming silence from
2024 Can be specified in dB (in case "dB" is appended to the specified value)
2025 or amplitude ratio. Default value is @code{0}.
2028 This indicate that @var{stop_duration} length of audio should be left intact
2029 at the beginning of each period of silence.
2030 For example, if you want to remove long pauses between words but do not want
2031 to remove the pauses completely. Default value is @code{0}.
2035 @subsection Examples
2039 The following example shows how this filter can be used to start a recording
2040 that does not contain the delay at the start which usually occurs between
2041 pressing the record button and the start of the performance:
2043 silenceremove=1:5:0.02
2049 Boost or cut treble (upper) frequencies of the audio using a two-pole
2050 shelving filter with a response similar to that of a standard
2051 hi-fi's tone-controls. This is also known as shelving equalisation (EQ).
2053 The filter accepts the following options:
2057 Give the gain at whichever is the lower of ~22 kHz and the
2058 Nyquist frequency. Its useful range is about -20 (for a large cut)
2059 to +20 (for a large boost). Beware of clipping when using a positive gain.
2062 Set the filter's central frequency and so can be used
2063 to extend or reduce the frequency range to be boosted or cut.
2064 The default value is @code{3000} Hz.
2067 Set method to specify band-width of filter.
2080 Determine how steep is the filter's shelf transition.
2085 Adjust the input audio volume.
2087 It accepts the following parameters:
2091 Set audio volume expression.
2093 Output values are clipped to the maximum value.
2095 The output audio volume is given by the relation:
2097 @var{output_volume} = @var{volume} * @var{input_volume}
2100 The default value for @var{volume} is "1.0".
2103 This parameter represents the mathematical precision.
2105 It determines which input sample formats will be allowed, which affects the
2106 precision of the volume scaling.
2110 8-bit fixed-point; this limits input sample format to U8, S16, and S32.
2112 32-bit floating-point; this limits input sample format to FLT. (default)
2114 64-bit floating-point; this limits input sample format to DBL.
2118 Choose the behaviour on encountering ReplayGain side data in input frames.
2122 Remove ReplayGain side data, ignoring its contents (the default).
2125 Ignore ReplayGain side data, but leave it in the frame.
2128 Prefer the track gain, if present.
2131 Prefer the album gain, if present.
2134 @item replaygain_preamp
2135 Pre-amplification gain in dB to apply to the selected replaygain gain.
2137 Default value for @var{replaygain_preamp} is 0.0.
2140 Set when the volume expression is evaluated.
2142 It accepts the following values:
2145 only evaluate expression once during the filter initialization, or
2146 when the @samp{volume} command is sent
2149 evaluate expression for each incoming frame
2152 Default value is @samp{once}.
2155 The volume expression can contain the following parameters.
2159 frame number (starting at zero)
2162 @item nb_consumed_samples
2163 number of samples consumed by the filter
2165 number of samples in the current frame
2167 original frame position in the file
2173 PTS at start of stream
2175 time at start of stream
2181 last set volume value
2184 Note that when @option{eval} is set to @samp{once} only the
2185 @var{sample_rate} and @var{tb} variables are available, all other
2186 variables will evaluate to NAN.
2188 @subsection Commands
2190 This filter supports the following commands:
2193 Modify the volume expression.
2194 The command accepts the same syntax of the corresponding option.
2196 If the specified expression is not valid, it is kept at its current
2198 @item replaygain_noclip
2199 Prevent clipping by limiting the gain applied.
2201 Default value for @var{replaygain_noclip} is 1.
2205 @subsection Examples
2209 Halve the input audio volume:
2213 volume=volume=-6.0206dB
2216 In all the above example the named key for @option{volume} can be
2217 omitted, for example like in:
2223 Increase input audio power by 6 decibels using fixed-point precision:
2225 volume=volume=6dB:precision=fixed
2229 Fade volume after time 10 with an annihilation period of 5 seconds:
2231 volume='if(lt(t,10),1,max(1-(t-10)/5,0))':eval=frame
2235 @section volumedetect
2237 Detect the volume of the input video.
2239 The filter has no parameters. The input is not modified. Statistics about
2240 the volume will be printed in the log when the input stream end is reached.
2242 In particular it will show the mean volume (root mean square), maximum
2243 volume (on a per-sample basis), and the beginning of a histogram of the
2244 registered volume values (from the maximum value to a cumulated 1/1000 of
2247 All volumes are in decibels relative to the maximum PCM value.
2249 @subsection Examples
2251 Here is an excerpt of the output:
2253 [Parsed_volumedetect_0 @ 0xa23120] mean_volume: -27 dB
2254 [Parsed_volumedetect_0 @ 0xa23120] max_volume: -4 dB
2255 [Parsed_volumedetect_0 @ 0xa23120] histogram_4db: 6
2256 [Parsed_volumedetect_0 @ 0xa23120] histogram_5db: 62
2257 [Parsed_volumedetect_0 @ 0xa23120] histogram_6db: 286
2258 [Parsed_volumedetect_0 @ 0xa23120] histogram_7db: 1042
2259 [Parsed_volumedetect_0 @ 0xa23120] histogram_8db: 2551
2260 [Parsed_volumedetect_0 @ 0xa23120] histogram_9db: 4609
2261 [Parsed_volumedetect_0 @ 0xa23120] histogram_10db: 8409
2267 The mean square energy is approximately -27 dB, or 10^-2.7.
2269 The largest sample is at -4 dB, or more precisely between -4 dB and -5 dB.
2271 There are 6 samples at -4 dB, 62 at -5 dB, 286 at -6 dB, etc.
2274 In other words, raising the volume by +4 dB does not cause any clipping,
2275 raising it by +5 dB causes clipping for 6 samples, etc.
2277 @c man end AUDIO FILTERS
2279 @chapter Audio Sources
2280 @c man begin AUDIO SOURCES
2282 Below is a description of the currently available audio sources.
2286 Buffer audio frames, and make them available to the filter chain.
2288 This source is mainly intended for a programmatic use, in particular
2289 through the interface defined in @file{libavfilter/asrc_abuffer.h}.
2291 It accepts the following parameters:
2295 The timebase which will be used for timestamps of submitted frames. It must be
2296 either a floating-point number or in @var{numerator}/@var{denominator} form.
2299 The sample rate of the incoming audio buffers.
2302 The sample format of the incoming audio buffers.
2303 Either a sample format name or its corresponding integer representation from
2304 the enum AVSampleFormat in @file{libavutil/samplefmt.h}
2306 @item channel_layout
2307 The channel layout of the incoming audio buffers.
2308 Either a channel layout name from channel_layout_map in
2309 @file{libavutil/channel_layout.c} or its corresponding integer representation
2310 from the AV_CH_LAYOUT_* macros in @file{libavutil/channel_layout.h}
2313 The number of channels of the incoming audio buffers.
2314 If both @var{channels} and @var{channel_layout} are specified, then they
2319 @subsection Examples
2322 abuffer=sample_rate=44100:sample_fmt=s16p:channel_layout=stereo
2325 will instruct the source to accept planar 16bit signed stereo at 44100Hz.
2326 Since the sample format with name "s16p" corresponds to the number
2327 6 and the "stereo" channel layout corresponds to the value 0x3, this is
2330 abuffer=sample_rate=44100:sample_fmt=6:channel_layout=0x3
2335 Generate an audio signal specified by an expression.
2337 This source accepts in input one or more expressions (one for each
2338 channel), which are evaluated and used to generate a corresponding
2341 This source accepts the following options:
2345 Set the '|'-separated expressions list for each separate channel. In case the
2346 @option{channel_layout} option is not specified, the selected channel layout
2347 depends on the number of provided expressions. Otherwise the last
2348 specified expression is applied to the remaining output channels.
2350 @item channel_layout, c
2351 Set the channel layout. The number of channels in the specified layout
2352 must be equal to the number of specified expressions.
2355 Set the minimum duration of the sourced audio. See
2356 @ref{time duration syntax,,the Time duration section in the ffmpeg-utils(1) manual,ffmpeg-utils}
2357 for the accepted syntax.
2358 Note that the resulting duration may be greater than the specified
2359 duration, as the generated audio is always cut at the end of a
2362 If not specified, or the expressed duration is negative, the audio is
2363 supposed to be generated forever.
2366 Set the number of samples per channel per each output frame,
2369 @item sample_rate, s
2370 Specify the sample rate, default to 44100.
2373 Each expression in @var{exprs} can contain the following constants:
2377 number of the evaluated sample, starting from 0
2380 time of the evaluated sample expressed in seconds, starting from 0
2387 @subsection Examples
2397 Generate a sin signal with frequency of 440 Hz, set sample rate to
2400 aevalsrc="sin(440*2*PI*t):s=8000"
2404 Generate a two channels signal, specify the channel layout (Front
2405 Center + Back Center) explicitly:
2407 aevalsrc="sin(420*2*PI*t)|cos(430*2*PI*t):c=FC|BC"
2411 Generate white noise:
2413 aevalsrc="-2+random(0)"
2417 Generate an amplitude modulated signal:
2419 aevalsrc="sin(10*2*PI*t)*sin(880*2*PI*t)"
2423 Generate 2.5 Hz binaural beats on a 360 Hz carrier:
2425 aevalsrc="0.1*sin(2*PI*(360-2.5/2)*t) | 0.1*sin(2*PI*(360+2.5/2)*t)"
2432 The null audio source, return unprocessed audio frames. It is mainly useful
2433 as a template and to be employed in analysis / debugging tools, or as
2434 the source for filters which ignore the input data (for example the sox
2437 This source accepts the following options:
2441 @item channel_layout, cl
2443 Specifies the channel layout, and can be either an integer or a string
2444 representing a channel layout. The default value of @var{channel_layout}
2447 Check the channel_layout_map definition in
2448 @file{libavutil/channel_layout.c} for the mapping between strings and
2449 channel layout values.
2451 @item sample_rate, r
2452 Specifies the sample rate, and defaults to 44100.
2455 Set the number of samples per requested frames.
2459 @subsection Examples
2463 Set the sample rate to 48000 Hz and the channel layout to AV_CH_LAYOUT_MONO.
2465 anullsrc=r=48000:cl=4
2469 Do the same operation with a more obvious syntax:
2471 anullsrc=r=48000:cl=mono
2475 All the parameters need to be explicitly defined.
2479 Synthesize a voice utterance using the libflite library.
2481 To enable compilation of this filter you need to configure FFmpeg with
2482 @code{--enable-libflite}.
2484 Note that the flite library is not thread-safe.
2486 The filter accepts the following options:
2491 If set to 1, list the names of the available voices and exit
2492 immediately. Default value is 0.
2495 Set the maximum number of samples per frame. Default value is 512.
2498 Set the filename containing the text to speak.
2501 Set the text to speak.
2504 Set the voice to use for the speech synthesis. Default value is
2505 @code{kal}. See also the @var{list_voices} option.
2508 @subsection Examples
2512 Read from file @file{speech.txt}, and synthesize the text using the
2513 standard flite voice:
2515 flite=textfile=speech.txt
2519 Read the specified text selecting the @code{slt} voice:
2521 flite=text='So fare thee well, poor devil of a Sub-Sub, whose commentator I am':voice=slt
2525 Input text to ffmpeg:
2527 ffmpeg -f lavfi -i flite=text='So fare thee well, poor devil of a Sub-Sub, whose commentator I am':voice=slt
2531 Make @file{ffplay} speak the specified text, using @code{flite} and
2532 the @code{lavfi} device:
2534 ffplay -f lavfi flite=text='No more be grieved for which that thou hast done.'
2538 For more information about libflite, check:
2539 @url{http://www.speech.cs.cmu.edu/flite/}
2543 Generate an audio signal made of a sine wave with amplitude 1/8.
2545 The audio signal is bit-exact.
2547 The filter accepts the following options:
2552 Set the carrier frequency. Default is 440 Hz.
2554 @item beep_factor, b
2555 Enable a periodic beep every second with frequency @var{beep_factor} times
2556 the carrier frequency. Default is 0, meaning the beep is disabled.
2558 @item sample_rate, r
2559 Specify the sample rate, default is 44100.
2562 Specify the duration of the generated audio stream.
2564 @item samples_per_frame
2565 Set the number of samples per output frame, default is 1024.
2568 @subsection Examples
2573 Generate a simple 440 Hz sine wave:
2579 Generate a 220 Hz sine wave with a 880 Hz beep each second, for 5 seconds:
2583 sine=frequency=220:beep_factor=4:duration=5
2588 @c man end AUDIO SOURCES
2590 @chapter Audio Sinks
2591 @c man begin AUDIO SINKS
2593 Below is a description of the currently available audio sinks.
2595 @section abuffersink
2597 Buffer audio frames, and make them available to the end of filter chain.
2599 This sink is mainly intended for programmatic use, in particular
2600 through the interface defined in @file{libavfilter/buffersink.h}
2601 or the options system.
2603 It accepts a pointer to an AVABufferSinkContext structure, which
2604 defines the incoming buffers' formats, to be passed as the opaque
2605 parameter to @code{avfilter_init_filter} for initialization.
2608 Null audio sink; do absolutely nothing with the input audio. It is
2609 mainly useful as a template and for use in analysis / debugging
2612 @c man end AUDIO SINKS
2614 @chapter Video Filters
2615 @c man begin VIDEO FILTERS
2617 When you configure your FFmpeg build, you can disable any of the
2618 existing filters using @code{--disable-filters}.
2619 The configure output will show the video filters included in your
2622 Below is a description of the currently available video filters.
2624 @section alphaextract
2626 Extract the alpha component from the input as a grayscale video. This
2627 is especially useful with the @var{alphamerge} filter.
2631 Add or replace the alpha component of the primary input with the
2632 grayscale value of a second input. This is intended for use with
2633 @var{alphaextract} to allow the transmission or storage of frame
2634 sequences that have alpha in a format that doesn't support an alpha
2637 For example, to reconstruct full frames from a normal YUV-encoded video
2638 and a separate video created with @var{alphaextract}, you might use:
2640 movie=in_alpha.mkv [alpha]; [in][alpha] alphamerge [out]
2643 Since this filter is designed for reconstruction, it operates on frame
2644 sequences without considering timestamps, and terminates when either
2645 input reaches end of stream. This will cause problems if your encoding
2646 pipeline drops frames. If you're trying to apply an image as an
2647 overlay to a video stream, consider the @var{overlay} filter instead.
2651 Same as the @ref{subtitles} filter, except that it doesn't require libavcodec
2652 and libavformat to work. On the other hand, it is limited to ASS (Advanced
2653 Substation Alpha) subtitles files.
2655 This filter accepts the following option in addition to the common options from
2656 the @ref{subtitles} filter:
2660 Set the shaping engine
2662 Available values are:
2665 The default libass shaping engine, which is the best available.
2667 Fast, font-agnostic shaper that can do only substitutions
2669 Slower shaper using OpenType for substitutions and positioning
2672 The default is @code{auto}.
2677 Compute the bounding box for the non-black pixels in the input frame
2680 This filter computes the bounding box containing all the pixels with a
2681 luminance value greater than the minimum allowed value.
2682 The parameters describing the bounding box are printed on the filter
2685 The filter accepts the following option:
2689 Set the minimal luminance value. Default is @code{16}.
2692 @section blackdetect
2694 Detect video intervals that are (almost) completely black. Can be
2695 useful to detect chapter transitions, commercials, or invalid
2696 recordings. Output lines contains the time for the start, end and
2697 duration of the detected black interval expressed 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 The filter accepts the following options:
2705 @item black_min_duration, d
2706 Set the minimum detected black duration expressed in seconds. It must
2707 be a non-negative floating point number.
2709 Default value is 2.0.
2711 @item picture_black_ratio_th, pic_th
2712 Set the threshold for considering a picture "black".
2713 Express the minimum value for the ratio:
2715 @var{nb_black_pixels} / @var{nb_pixels}
2718 for which a picture is considered black.
2719 Default value is 0.98.
2721 @item pixel_black_th, pix_th
2722 Set the threshold for considering a pixel "black".
2724 The threshold expresses the maximum pixel luminance value for which a
2725 pixel is considered "black". The provided value is scaled according to
2726 the following equation:
2728 @var{absolute_threshold} = @var{luminance_minimum_value} + @var{pixel_black_th} * @var{luminance_range_size}
2731 @var{luminance_range_size} and @var{luminance_minimum_value} depend on
2732 the input video format, the range is [0-255] for YUV full-range
2733 formats and [16-235] for YUV non full-range formats.
2735 Default value is 0.10.
2738 The following example sets the maximum pixel threshold to the minimum
2739 value, and detects only black intervals of 2 or more seconds:
2741 blackdetect=d=2:pix_th=0.00
2746 Detect frames that are (almost) completely black. Can be useful to
2747 detect chapter transitions or commercials. Output lines consist of
2748 the frame number of the detected frame, the percentage of blackness,
2749 the position in the file if known or -1 and the timestamp in seconds.
2751 In order to display the output lines, you need to set the loglevel at
2752 least to the AV_LOG_INFO value.
2754 It accepts the following parameters:
2759 The percentage of the pixels that have to be below the threshold; it defaults to
2762 @item threshold, thresh
2763 The threshold below which a pixel value is considered black; it defaults to
2768 @section blend, tblend
2770 Blend two video frames into each other.
2772 The @code{blend} filter takes two input streams and outputs one
2773 stream, the first input is the "top" layer and second input is
2774 "bottom" layer. Output terminates when shortest input terminates.
2776 The @code{tblend} (time blend) filter takes two consecutive frames
2777 from one single stream, and outputs the result obtained by blending
2778 the new frame on top of the old frame.
2780 A description of the accepted options follows.
2788 Set blend mode for specific pixel component or all pixel components in case
2789 of @var{all_mode}. Default value is @code{normal}.
2791 Available values for component modes are:
2828 Set blend opacity for specific pixel component or all pixel components in case
2829 of @var{all_opacity}. Only used in combination with pixel component blend modes.
2836 Set blend expression for specific pixel component or all pixel components in case
2837 of @var{all_expr}. Note that related mode options will be ignored if those are set.
2839 The expressions can use the following variables:
2843 The sequential number of the filtered frame, starting from @code{0}.
2847 the coordinates of the current sample
2851 the width and height of currently filtered plane
2855 Width and height scale depending on the currently filtered plane. It is the
2856 ratio between the corresponding luma plane number of pixels and the current
2857 plane ones. E.g. for YUV4:2:0 the values are @code{1,1} for the luma plane, and
2858 @code{0.5,0.5} for chroma planes.
2861 Time of the current frame, expressed in seconds.
2864 Value of pixel component at current location for first video frame (top layer).
2867 Value of pixel component at current location for second video frame (bottom layer).
2871 Force termination when the shortest input terminates. Default is
2872 @code{0}. This option is only defined for the @code{blend} filter.
2875 Continue applying the last bottom frame after the end of the stream. A value of
2876 @code{0} disable the filter after the last frame of the bottom layer is reached.
2877 Default is @code{1}. This option is only defined for the @code{blend} filter.
2880 @subsection Examples
2884 Apply transition from bottom layer to top layer in first 10 seconds:
2886 blend=all_expr='A*(if(gte(T,10),1,T/10))+B*(1-(if(gte(T,10),1,T/10)))'
2890 Apply 1x1 checkerboard effect:
2892 blend=all_expr='if(eq(mod(X,2),mod(Y,2)),A,B)'
2896 Apply uncover left effect:
2898 blend=all_expr='if(gte(N*SW+X,W),A,B)'
2902 Apply uncover down effect:
2904 blend=all_expr='if(gte(Y-N*SH,0),A,B)'
2908 Apply uncover up-left effect:
2910 blend=all_expr='if(gte(T*SH*40+Y,H)*gte((T*40*SW+X)*W/H,W),A,B)'
2914 Display differences between the current and the previous frame:
2916 tblend=all_mode=difference128
2922 Apply a boxblur algorithm to the input video.
2924 It accepts the following parameters:
2928 @item luma_radius, lr
2929 @item luma_power, lp
2930 @item chroma_radius, cr
2931 @item chroma_power, cp
2932 @item alpha_radius, ar
2933 @item alpha_power, ap
2937 A description of the accepted options follows.
2940 @item luma_radius, lr
2941 @item chroma_radius, cr
2942 @item alpha_radius, ar
2943 Set an expression for the box radius in pixels used for blurring the
2944 corresponding input plane.
2946 The radius value must be a non-negative number, and must not be
2947 greater than the value of the expression @code{min(w,h)/2} for the
2948 luma and alpha planes, and of @code{min(cw,ch)/2} for the chroma
2951 Default value for @option{luma_radius} is "2". If not specified,
2952 @option{chroma_radius} and @option{alpha_radius} default to the
2953 corresponding value set for @option{luma_radius}.
2955 The expressions can contain the following constants:
2959 The input width and height in pixels.
2963 The input chroma image width and height in pixels.
2967 The horizontal and vertical chroma subsample values. For example, for the
2968 pixel format "yuv422p", @var{hsub} is 2 and @var{vsub} is 1.
2971 @item luma_power, lp
2972 @item chroma_power, cp
2973 @item alpha_power, ap
2974 Specify how many times the boxblur filter is applied to the
2975 corresponding plane.
2977 Default value for @option{luma_power} is 2. If not specified,
2978 @option{chroma_power} and @option{alpha_power} default to the
2979 corresponding value set for @option{luma_power}.
2981 A value of 0 will disable the effect.
2984 @subsection Examples
2988 Apply a boxblur filter with the luma, chroma, and alpha radii
2991 boxblur=luma_radius=2:luma_power=1
2996 Set the luma radius to 2, and alpha and chroma radius to 0:
2998 boxblur=2:1:cr=0:ar=0
3002 Set the luma and chroma radii to a fraction of the video dimension:
3004 boxblur=luma_radius=min(h\,w)/10:luma_power=1:chroma_radius=min(cw\,ch)/10:chroma_power=1
3010 Visualize information exported by some codecs.
3012 Some codecs can export information through frames using side-data or other
3013 means. For example, some MPEG based codecs export motion vectors through the
3014 @var{export_mvs} flag in the codec @option{flags2} option.
3016 The filter accepts the following option:
3020 Set motion vectors to visualize.
3022 Available flags for @var{mv} are:
3026 forward predicted MVs of P-frames
3028 forward predicted MVs of B-frames
3030 backward predicted MVs of B-frames
3034 @subsection Examples
3038 Visualizes multi-directionals MVs from P and B-Frames using @command{ffplay}:
3040 ffplay -flags2 +export_mvs input.mpg -vf codecview=mv=pf+bf+bb
3044 @section colorbalance
3045 Modify intensity of primary colors (red, green and blue) of input frames.
3047 The filter allows an input frame to be adjusted in the shadows, midtones or highlights
3048 regions for the red-cyan, green-magenta or blue-yellow balance.
3050 A positive adjustment value shifts the balance towards the primary color, a negative
3051 value towards the complementary color.
3053 The filter accepts the following options:
3059 Adjust red, green and blue shadows (darkest pixels).
3064 Adjust red, green and blue midtones (medium pixels).
3069 Adjust red, green and blue highlights (brightest pixels).
3071 Allowed ranges for options are @code{[-1.0, 1.0]}. Defaults are @code{0}.
3074 @subsection Examples
3078 Add red color cast to shadows:
3085 RGB colorspace color keying.
3087 The filter accepts the following options:
3091 The color which will be replaced with transparency.
3094 Similarity percentage with the key color.
3096 0.01 matches only the exact key color, while 1.0 matches everything.
3101 0.0 makes pixels either fully transparent, or not transparent at all.
3103 Higher values result in semi-transparent pixels, with a higher transparency
3104 the more similar the pixels color is to the key color.
3107 @subsection Examples
3111 Make every green pixel in the input image transparent:
3113 ffmpeg -i input.png -vf colorkey=green out.png
3117 Overlay a greenscreen-video on top of a static background image.
3119 ffmpeg -i background.png -i video.mp4 -filter_complex "[1:v]colorkey=0x3BBD1E:0.3:0.2[ckout];[0:v][ckout]overlay[out]" -map "[out]" output.flv
3123 @section colorlevels
3125 Adjust video input frames using levels.
3127 The filter accepts the following options:
3134 Adjust red, green, blue and alpha input black point.
3135 Allowed ranges for options are @code{[-1.0, 1.0]}. Defaults are @code{0}.
3141 Adjust red, green, blue and alpha input white point.
3142 Allowed ranges for options are @code{[-1.0, 1.0]}. Defaults are @code{1}.
3144 Input levels are used to lighten highlights (bright tones), darken shadows
3145 (dark tones), change the balance of bright and dark tones.
3151 Adjust red, green, blue and alpha output black point.
3152 Allowed ranges for options are @code{[0, 1.0]}. Defaults are @code{0}.
3158 Adjust red, green, blue and alpha output white point.
3159 Allowed ranges for options are @code{[0, 1.0]}. Defaults are @code{1}.
3161 Output levels allows manual selection of a constrained output level range.
3164 @subsection Examples
3168 Make video output darker:
3170 colorlevels=rimin=0.058:gimin=0.058:bimin=0.058
3176 colorlevels=rimin=0.039:gimin=0.039:bimin=0.039:rimax=0.96:gimax=0.96:bimax=0.96
3180 Make video output lighter:
3182 colorlevels=rimax=0.902:gimax=0.902:bimax=0.902
3186 Increase brightness:
3188 colorlevels=romin=0.5:gomin=0.5:bomin=0.5
3192 @section colorchannelmixer
3194 Adjust video input frames by re-mixing color channels.
3196 This filter modifies a color channel by adding the values associated to
3197 the other channels of the same pixels. For example if the value to
3198 modify is red, the output value will be:
3200 @var{red}=@var{red}*@var{rr} + @var{blue}*@var{rb} + @var{green}*@var{rg} + @var{alpha}*@var{ra}
3203 The filter accepts the following options:
3210 Adjust contribution of input red, green, blue and alpha channels for output red channel.
3211 Default is @code{1} for @var{rr}, and @code{0} for @var{rg}, @var{rb} and @var{ra}.
3217 Adjust contribution of input red, green, blue and alpha channels for output green channel.
3218 Default is @code{1} for @var{gg}, and @code{0} for @var{gr}, @var{gb} and @var{ga}.
3224 Adjust contribution of input red, green, blue and alpha channels for output blue channel.
3225 Default is @code{1} for @var{bb}, and @code{0} for @var{br}, @var{bg} and @var{ba}.
3231 Adjust contribution of input red, green, blue and alpha channels for output alpha channel.
3232 Default is @code{1} for @var{aa}, and @code{0} for @var{ar}, @var{ag} and @var{ab}.
3234 Allowed ranges for options are @code{[-2.0, 2.0]}.
3237 @subsection Examples
3241 Convert source to grayscale:
3243 colorchannelmixer=.3:.4:.3:0:.3:.4:.3:0:.3:.4:.3
3246 Simulate sepia tones:
3248 colorchannelmixer=.393:.769:.189:0:.349:.686:.168:0:.272:.534:.131
3252 @section colormatrix
3254 Convert color matrix.
3256 The filter accepts the following options:
3261 Specify the source and destination color matrix. Both values must be
3264 The accepted values are:
3280 For example to convert from BT.601 to SMPTE-240M, use the command:
3282 colormatrix=bt601:smpte240m
3287 Copy the input source unchanged to the output. This is mainly useful for
3292 Crop the input video to given dimensions.
3294 It accepts the following parameters:
3298 The width of the output video. It defaults to @code{iw}.
3299 This expression is evaluated only once during the filter
3303 The height of the output video. It defaults to @code{ih}.
3304 This expression is evaluated only once during the filter
3308 The horizontal position, in the input video, of the left edge of the output
3309 video. It defaults to @code{(in_w-out_w)/2}.
3310 This expression is evaluated per-frame.
3313 The vertical position, in the input video, of the top edge of the output video.
3314 It defaults to @code{(in_h-out_h)/2}.
3315 This expression is evaluated per-frame.
3318 If set to 1 will force the output display aspect ratio
3319 to be the same of the input, by changing the output sample aspect
3320 ratio. It defaults to 0.
3323 The @var{out_w}, @var{out_h}, @var{x}, @var{y} parameters are
3324 expressions containing the following constants:
3329 The computed values for @var{x} and @var{y}. They are evaluated for
3334 The input width and height.
3338 These are the same as @var{in_w} and @var{in_h}.
3342 The output (cropped) width and height.
3346 These are the same as @var{out_w} and @var{out_h}.
3349 same as @var{iw} / @var{ih}
3352 input sample aspect ratio
3355 input display aspect ratio, it is the same as (@var{iw} / @var{ih}) * @var{sar}
3359 horizontal and vertical chroma subsample values. For example for the
3360 pixel format "yuv422p" @var{hsub} is 2 and @var{vsub} is 1.
3363 The number of the input frame, starting from 0.
3366 the position in the file of the input frame, NAN if unknown
3369 The timestamp expressed in seconds. It's NAN if the input timestamp is unknown.
3373 The expression for @var{out_w} may depend on the value of @var{out_h},
3374 and the expression for @var{out_h} may depend on @var{out_w}, but they
3375 cannot depend on @var{x} and @var{y}, as @var{x} and @var{y} are
3376 evaluated after @var{out_w} and @var{out_h}.
3378 The @var{x} and @var{y} parameters specify the expressions for the
3379 position of the top-left corner of the output (non-cropped) area. They
3380 are evaluated for each frame. If the evaluated value is not valid, it
3381 is approximated to the nearest valid value.
3383 The expression for @var{x} may depend on @var{y}, and the expression
3384 for @var{y} may depend on @var{x}.
3386 @subsection Examples
3390 Crop area with size 100x100 at position (12,34).
3395 Using named options, the example above becomes:
3397 crop=w=100:h=100:x=12:y=34
3401 Crop the central input area with size 100x100:
3407 Crop the central input area with size 2/3 of the input video:
3409 crop=2/3*in_w:2/3*in_h
3413 Crop the input video central square:
3420 Delimit the rectangle with the top-left corner placed at position
3421 100:100 and the right-bottom corner corresponding to the right-bottom
3422 corner of the input image.
3424 crop=in_w-100:in_h-100:100:100
3428 Crop 10 pixels from the left and right borders, and 20 pixels from
3429 the top and bottom borders
3431 crop=in_w-2*10:in_h-2*20
3435 Keep only the bottom right quarter of the input image:
3437 crop=in_w/2:in_h/2:in_w/2:in_h/2
3441 Crop height for getting Greek harmony:
3443 crop=in_w:1/PHI*in_w
3447 Apply trembling effect:
3449 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)
3453 Apply erratic camera effect depending on timestamp:
3455 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)"
3459 Set x depending on the value of y:
3461 crop=in_w/2:in_h/2:y:10+10*sin(n/10)
3467 Auto-detect the crop size.
3469 It calculates the necessary cropping parameters and prints the
3470 recommended parameters via the logging system. The detected dimensions
3471 correspond to the non-black area of the input video.
3473 It accepts the following parameters:
3478 Set higher black value threshold, which can be optionally specified
3479 from nothing (0) to everything (255 for 8bit based formats). An intensity
3480 value greater to the set value is considered non-black. It defaults to 24.
3481 You can also specify a value between 0.0 and 1.0 which will be scaled depending
3482 on the bitdepth of the pixel format.
3485 The value which the width/height should be divisible by. It defaults to
3486 16. The offset is automatically adjusted to center the video. Use 2 to
3487 get only even dimensions (needed for 4:2:2 video). 16 is best when
3488 encoding to most video codecs.
3490 @item reset_count, reset
3491 Set the counter that determines after how many frames cropdetect will
3492 reset the previously detected largest video area and start over to
3493 detect the current optimal crop area. Default value is 0.
3495 This can be useful when channel logos distort the video area. 0
3496 indicates 'never reset', and returns the largest area encountered during
3503 Apply color adjustments using curves.
3505 This filter is similar to the Adobe Photoshop and GIMP curves tools. Each
3506 component (red, green and blue) has its values defined by @var{N} key points
3507 tied from each other using a smooth curve. The x-axis represents the pixel
3508 values from the input frame, and the y-axis the new pixel values to be set for
3511 By default, a component curve is defined by the two points @var{(0;0)} and
3512 @var{(1;1)}. This creates a straight line where each original pixel value is
3513 "adjusted" to its own value, which means no change to the image.
3515 The filter allows you to redefine these two points and add some more. A new
3516 curve (using a natural cubic spline interpolation) will be define to pass
3517 smoothly through all these new coordinates. The new defined points needs to be
3518 strictly increasing over the x-axis, and their @var{x} and @var{y} values must
3519 be in the @var{[0;1]} interval. If the computed curves happened to go outside
3520 the vector spaces, the values will be clipped accordingly.
3522 If there is no key point defined in @code{x=0}, the filter will automatically
3523 insert a @var{(0;0)} point. In the same way, if there is no key point defined
3524 in @code{x=1}, the filter will automatically insert a @var{(1;1)} point.
3526 The filter accepts the following options:
3530 Select one of the available color presets. This option can be used in addition
3531 to the @option{r}, @option{g}, @option{b} parameters; in this case, the later
3532 options takes priority on the preset values.
3533 Available presets are:
3536 @item color_negative
3539 @item increase_contrast
3541 @item linear_contrast
3542 @item medium_contrast
3544 @item strong_contrast
3547 Default is @code{none}.
3549 Set the master key points. These points will define a second pass mapping. It
3550 is sometimes called a "luminance" or "value" mapping. It can be used with
3551 @option{r}, @option{g}, @option{b} or @option{all} since it acts like a
3552 post-processing LUT.
3554 Set the key points for the red component.
3556 Set the key points for the green component.
3558 Set the key points for the blue component.
3560 Set the key points for all components (not including master).
3561 Can be used in addition to the other key points component
3562 options. In this case, the unset component(s) will fallback on this
3563 @option{all} setting.
3565 Specify a Photoshop curves file (@code{.asv}) to import the settings from.
3568 To avoid some filtergraph syntax conflicts, each key points list need to be
3569 defined using the following syntax: @code{x0/y0 x1/y1 x2/y2 ...}.
3571 @subsection Examples
3575 Increase slightly the middle level of blue:
3577 curves=blue='0.5/0.58'
3583 curves=r='0/0.11 .42/.51 1/0.95':g='0.50/0.48':b='0/0.22 .49/.44 1/0.8'
3585 Here we obtain the following coordinates for each components:
3588 @code{(0;0.11) (0.42;0.51) (1;0.95)}
3590 @code{(0;0) (0.50;0.48) (1;1)}
3592 @code{(0;0.22) (0.49;0.44) (1;0.80)}
3596 The previous example can also be achieved with the associated built-in preset:
3598 curves=preset=vintage
3608 Use a Photoshop preset and redefine the points of the green component:
3610 curves=psfile='MyCurvesPresets/purple.asv':green='0.45/0.53'
3616 Denoise frames using 2D DCT (frequency domain filtering).
3618 This filter is not designed for real time.
3620 The filter accepts the following options:
3624 Set the noise sigma constant.
3626 This @var{sigma} defines a hard threshold of @code{3 * sigma}; every DCT
3627 coefficient (absolute value) below this threshold with be dropped.
3629 If you need a more advanced filtering, see @option{expr}.
3631 Default is @code{0}.
3634 Set number overlapping pixels for each block. Since the filter can be slow, you
3635 may want to reduce this value, at the cost of a less effective filter and the
3636 risk of various artefacts.
3638 If the overlapping value doesn't permit processing the whole input width or
3639 height, a warning will be displayed and according borders won't be denoised.
3641 Default value is @var{blocksize}-1, which is the best possible setting.
3644 Set the coefficient factor expression.
3646 For each coefficient of a DCT block, this expression will be evaluated as a
3647 multiplier value for the coefficient.
3649 If this is option is set, the @option{sigma} option will be ignored.
3651 The absolute value of the coefficient can be accessed through the @var{c}
3655 Set the @var{blocksize} using the number of bits. @code{1<<@var{n}} defines the
3656 @var{blocksize}, which is the width and height of the processed blocks.
3658 The default value is @var{3} (8x8) and can be raised to @var{4} for a
3659 @var{blocksize} of 16x16. Note that changing this setting has huge consequences
3660 on the speed processing. Also, a larger block size does not necessarily means a
3664 @subsection Examples
3666 Apply a denoise with a @option{sigma} of @code{4.5}:
3671 The same operation can be achieved using the expression system:
3673 dctdnoiz=e='gte(c, 4.5*3)'
3676 Violent denoise using a block size of @code{16x16}:
3684 Drop duplicated frames at regular intervals.
3686 The filter accepts the following options:
3690 Set the number of frames from which one will be dropped. Setting this to
3691 @var{N} means one frame in every batch of @var{N} frames will be dropped.
3692 Default is @code{5}.
3695 Set the threshold for duplicate detection. If the difference metric for a frame
3696 is less than or equal to this value, then it is declared as duplicate. Default
3700 Set scene change threshold. Default is @code{15}.
3704 Set the size of the x and y-axis blocks used during metric calculations.
3705 Larger blocks give better noise suppression, but also give worse detection of
3706 small movements. Must be a power of two. Default is @code{32}.
3709 Mark main input as a pre-processed input and activate clean source input
3710 stream. This allows the input to be pre-processed with various filters to help
3711 the metrics calculation while keeping the frame selection lossless. When set to
3712 @code{1}, the first stream is for the pre-processed input, and the second
3713 stream is the clean source from where the kept frames are chosen. Default is
3717 Set whether or not chroma is considered in the metric calculations. Default is
3723 Remove judder produced by partially interlaced telecined content.
3725 Judder can be introduced, for instance, by @ref{pullup} filter. If the original
3726 source was partially telecined content then the output of @code{pullup,dejudder}
3727 will have a variable frame rate. May change the recorded frame rate of the
3728 container. Aside from that change, this filter will not affect constant frame
3731 The option available in this filter is:
3735 Specify the length of the window over which the judder repeats.
3737 Accepts any integer greater than 1. Useful values are:
3741 If the original was telecined from 24 to 30 fps (Film to NTSC).
3744 If the original was telecined from 25 to 30 fps (PAL to NTSC).
3747 If a mixture of the two.
3750 The default is @samp{4}.
3755 Suppress a TV station logo by a simple interpolation of the surrounding
3756 pixels. Just set a rectangle covering the logo and watch it disappear
3757 (and sometimes something even uglier appear - your mileage may vary).
3759 It accepts the following parameters:
3764 Specify the top left corner coordinates of the logo. They must be
3769 Specify the width and height of the logo to clear. They must be
3773 Specify the thickness of the fuzzy edge of the rectangle (added to
3774 @var{w} and @var{h}). The default value is 4.
3777 When set to 1, a green rectangle is drawn on the screen to simplify
3778 finding the right @var{x}, @var{y}, @var{w}, and @var{h} parameters.
3779 The default value is 0.
3781 The rectangle is drawn on the outermost pixels which will be (partly)
3782 replaced with interpolated values. The values of the next pixels
3783 immediately outside this rectangle in each direction will be used to
3784 compute the interpolated pixel values inside the rectangle.
3788 @subsection Examples
3792 Set a rectangle covering the area with top left corner coordinates 0,0
3793 and size 100x77, and a band of size 10:
3795 delogo=x=0:y=0:w=100:h=77:band=10
3802 Attempt to fix small changes in horizontal and/or vertical shift. This
3803 filter helps remove camera shake from hand-holding a camera, bumping a
3804 tripod, moving on a vehicle, etc.
3806 The filter accepts the following options:
3814 Specify a rectangular area where to limit the search for motion
3816 If desired the search for motion vectors can be limited to a
3817 rectangular area of the frame defined by its top left corner, width
3818 and height. These parameters have the same meaning as the drawbox
3819 filter which can be used to visualise the position of the bounding
3822 This is useful when simultaneous movement of subjects within the frame
3823 might be confused for camera motion by the motion vector search.
3825 If any or all of @var{x}, @var{y}, @var{w} and @var{h} are set to -1
3826 then the full frame is used. This allows later options to be set
3827 without specifying the bounding box for the motion vector search.
3829 Default - search the whole frame.
3833 Specify the maximum extent of movement in x and y directions in the
3834 range 0-64 pixels. Default 16.
3837 Specify how to generate pixels to fill blanks at the edge of the
3838 frame. Available values are:
3841 Fill zeroes at blank locations
3843 Original image at blank locations
3845 Extruded edge value at blank locations
3847 Mirrored edge at blank locations
3849 Default value is @samp{mirror}.
3852 Specify the blocksize to use for motion search. Range 4-128 pixels,
3856 Specify the contrast threshold for blocks. Only blocks with more than
3857 the specified contrast (difference between darkest and lightest
3858 pixels) will be considered. Range 1-255, default 125.
3861 Specify the search strategy. Available values are:
3864 Set exhaustive search
3866 Set less exhaustive search.
3868 Default value is @samp{exhaustive}.
3871 If set then a detailed log of the motion search is written to the
3875 If set to 1, specify using OpenCL capabilities, only available if
3876 FFmpeg was configured with @code{--enable-opencl}. Default value is 0.
3882 Apply an exact inverse of the telecine operation. It requires a predefined
3883 pattern specified using the pattern option which must be the same as that passed
3884 to the telecine filter.
3886 This filter accepts the following options:
3895 The default value is @code{top}.
3899 A string of numbers representing the pulldown pattern you wish to apply.
3900 The default value is @code{23}.
3903 A number representing position of the first frame with respect to the telecine
3904 pattern. This is to be used if the stream is cut. The default value is @code{0}.
3909 Draw a colored box on the input image.
3911 It accepts the following parameters:
3916 The expressions which specify the top left corner coordinates of the box. It defaults to 0.
3920 The expressions which specify the width and height of the box; if 0 they are interpreted as
3921 the input width and height. It defaults to 0.
3924 Specify the color of the box to write. For the general syntax of this option,
3925 check the "Color" section in the ffmpeg-utils manual. If the special
3926 value @code{invert} is used, the box edge color is the same as the
3927 video with inverted luma.
3930 The expression which sets the thickness of the box edge. Default value is @code{3}.
3932 See below for the list of accepted constants.
3935 The parameters for @var{x}, @var{y}, @var{w} and @var{h} and @var{t} are expressions containing the
3936 following constants:
3940 The input display aspect ratio, it is the same as (@var{w} / @var{h}) * @var{sar}.
3944 horizontal and vertical chroma subsample values. For example for the
3945 pixel format "yuv422p" @var{hsub} is 2 and @var{vsub} is 1.
3949 The input width and height.
3952 The input sample aspect ratio.
3956 The x and y offset coordinates where the box is drawn.
3960 The width and height of the drawn box.
3963 The thickness of the drawn box.
3965 These constants allow the @var{x}, @var{y}, @var{w}, @var{h} and @var{t} expressions to refer to
3966 each other, so you may for example specify @code{y=x/dar} or @code{h=w/dar}.
3970 @subsection Examples
3974 Draw a black box around the edge of the input image:
3980 Draw a box with color red and an opacity of 50%:
3982 drawbox=10:20:200:60:red@@0.5
3985 The previous example can be specified as:
3987 drawbox=x=10:y=20:w=200:h=60:color=red@@0.5
3991 Fill the box with pink color:
3993 drawbox=x=10:y=10:w=100:h=100:color=pink@@0.5:t=max
3997 Draw a 2-pixel red 2.40:1 mask:
3999 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
4003 @section drawgraph, adrawgraph
4005 Draw a graph using input video or audio metadata.
4007 It accepts the following parameters:
4011 Set 1st frame metadata key from which metadata values will be used to draw a graph.
4014 Set 1st foreground color expression.
4017 Set 2nd frame metadata key from which metadata values will be used to draw a graph.
4020 Set 2nd foreground color expression.
4023 Set 3rd frame metadata key from which metadata values will be used to draw a graph.
4026 Set 3rd foreground color expression.
4029 Set 4th frame metadata key from which metadata values will be used to draw a graph.
4032 Set 4th foreground color expression.
4035 Set minimal value of metadata value.
4038 Set maximal value of metadata value.
4041 Set graph background color. Default is white.
4046 Available values for mode is:
4053 Default is @code{line}.
4058 Available values for slide is:
4061 Draw new frame when right border is reached.
4064 Replace old columns with new ones.
4067 Scroll from right to left.
4070 Default is @code{frame}.
4073 Set size of graph video. For the syntax of this option, check the
4074 @ref{video size syntax,,"Video size" section in the ffmpeg-utils manual,ffmpeg-utils}.
4075 The default value is @code{900x256}.
4077 The foreground color expressions can use the following variables:
4080 Minimal value of metadata value.
4083 Maximal value of metadata value.
4086 Current metadata key value.
4089 The color is defined as 0xAABBGGRR.
4092 Example using metadata from @ref{signalstats} filter:
4094 signalstats,drawgraph=lavfi.signalstats.YAVG:min=0:max=255
4097 Example using metadata from @ref{ebur128} filter:
4099 ebur128=metadata=1,adrawgraph=lavfi.r128.M:min=-120:max=5
4104 Draw a grid on the input image.
4106 It accepts the following parameters:
4111 The expressions which specify the coordinates of some point of grid intersection (meant to configure offset). Both default to 0.
4115 The expressions which specify the width and height of the grid cell, if 0 they are interpreted as the
4116 input width and height, respectively, minus @code{thickness}, so image gets
4117 framed. Default to 0.
4120 Specify the color of the grid. For the general syntax of this option,
4121 check the "Color" section in the ffmpeg-utils manual. If the special
4122 value @code{invert} is used, the grid color is the same as the
4123 video with inverted luma.
4126 The expression which sets the thickness of the grid line. Default value is @code{1}.
4128 See below for the list of accepted constants.
4131 The parameters for @var{x}, @var{y}, @var{w} and @var{h} and @var{t} are expressions containing the
4132 following constants:
4136 The input display aspect ratio, it is the same as (@var{w} / @var{h}) * @var{sar}.
4140 horizontal and vertical chroma subsample values. For example for the
4141 pixel format "yuv422p" @var{hsub} is 2 and @var{vsub} is 1.
4145 The input grid cell width and height.
4148 The input sample aspect ratio.
4152 The x and y coordinates of some point of grid intersection (meant to configure offset).
4156 The width and height of the drawn cell.
4159 The thickness of the drawn cell.
4161 These constants allow the @var{x}, @var{y}, @var{w}, @var{h} and @var{t} expressions to refer to
4162 each other, so you may for example specify @code{y=x/dar} or @code{h=w/dar}.
4166 @subsection Examples
4170 Draw a grid with cell 100x100 pixels, thickness 2 pixels, with color red and an opacity of 50%:
4172 drawgrid=width=100:height=100:thickness=2:color=red@@0.5
4176 Draw a white 3x3 grid with an opacity of 50%:
4178 drawgrid=w=iw/3:h=ih/3:t=2:c=white@@0.5
4185 Draw a text string or text from a specified file on top of a video, using the
4186 libfreetype library.
4188 To enable compilation of this filter, you need to configure FFmpeg with
4189 @code{--enable-libfreetype}.
4190 To enable default font fallback and the @var{font} option you need to
4191 configure FFmpeg with @code{--enable-libfontconfig}.
4192 To enable the @var{text_shaping} option, you need to configure FFmpeg with
4193 @code{--enable-libfribidi}.
4197 It accepts the following parameters:
4202 Used to draw a box around text using the background color.
4203 The value must be either 1 (enable) or 0 (disable).
4204 The default value of @var{box} is 0.
4207 Set the width of the border to be drawn around the box using @var{boxcolor}.
4208 The default value of @var{boxborderw} is 0.
4211 The color to be used for drawing box around text. For the syntax of this
4212 option, check the "Color" section in the ffmpeg-utils manual.
4214 The default value of @var{boxcolor} is "white".
4217 Set the width of the border to be drawn around the text using @var{bordercolor}.
4218 The default value of @var{borderw} is 0.
4221 Set the color to be used for drawing border around text. For the syntax of this
4222 option, check the "Color" section in the ffmpeg-utils manual.
4224 The default value of @var{bordercolor} is "black".
4227 Select how the @var{text} is expanded. Can be either @code{none},
4228 @code{strftime} (deprecated) or
4229 @code{normal} (default). See the @ref{drawtext_expansion, Text expansion} section
4233 If true, check and fix text coords to avoid clipping.
4236 The color to be used for drawing fonts. For the syntax of this option, check
4237 the "Color" section in the ffmpeg-utils manual.
4239 The default value of @var{fontcolor} is "black".
4241 @item fontcolor_expr
4242 String which is expanded the same way as @var{text} to obtain dynamic
4243 @var{fontcolor} value. By default this option has empty value and is not
4244 processed. When this option is set, it overrides @var{fontcolor} option.
4247 The font family to be used for drawing text. By default Sans.
4250 The font file to be used for drawing text. The path must be included.
4251 This parameter is mandatory if the fontconfig support is disabled.
4254 This option does not exist, please see the timeline system
4257 Draw the text applying alpha blending. The value can
4258 be either a number between 0.0 and 1.0
4259 The expression accepts the same variables @var{x, y} do.
4260 The default value is 1.
4261 Please see fontcolor_expr
4264 The font size to be used for drawing text.
4265 The default value of @var{fontsize} is 16.
4268 If set to 1, attempt to shape the text (for example, reverse the order of
4269 right-to-left text and join Arabic characters) before drawing it.
4270 Otherwise, just draw the text exactly as given.
4271 By default 1 (if supported).
4274 The flags to be used for loading the fonts.
4276 The flags map the corresponding flags supported by libfreetype, and are
4277 a combination of the following values:
4284 @item vertical_layout
4285 @item force_autohint
4288 @item ignore_global_advance_width
4290 @item ignore_transform
4296 Default value is "default".
4298 For more information consult the documentation for the FT_LOAD_*
4302 The color to be used for drawing a shadow behind the drawn text. For the
4303 syntax of this option, check the "Color" section in the ffmpeg-utils manual.
4305 The default value of @var{shadowcolor} is "black".
4309 The x and y offsets for the text shadow position with respect to the
4310 position of the text. They can be either positive or negative
4311 values. The default value for both is "0".
4314 The starting frame number for the n/frame_num variable. The default value
4318 The size in number of spaces to use for rendering the tab.
4322 Set the initial timecode representation in "hh:mm:ss[:;.]ff"
4323 format. It can be used with or without text parameter. @var{timecode_rate}
4324 option must be specified.
4326 @item timecode_rate, rate, r
4327 Set the timecode frame rate (timecode only).
4330 The text string to be drawn. The text must be a sequence of UTF-8
4332 This parameter is mandatory if no file is specified with the parameter
4336 A text file containing text to be drawn. The text must be a sequence
4337 of UTF-8 encoded characters.
4339 This parameter is mandatory if no text string is specified with the
4340 parameter @var{text}.
4342 If both @var{text} and @var{textfile} are specified, an error is thrown.
4345 If set to 1, the @var{textfile} will be reloaded before each frame.
4346 Be sure to update it atomically, or it may be read partially, or even fail.
4350 The expressions which specify the offsets where text will be drawn
4351 within the video frame. They are relative to the top/left border of the
4354 The default value of @var{x} and @var{y} is "0".
4356 See below for the list of accepted constants and functions.
4359 The parameters for @var{x} and @var{y} are expressions containing the
4360 following constants and functions:
4364 input display aspect ratio, it is the same as (@var{w} / @var{h}) * @var{sar}
4368 horizontal and vertical chroma subsample values. For example for the
4369 pixel format "yuv422p" @var{hsub} is 2 and @var{vsub} is 1.
4372 the height of each text line
4380 @item max_glyph_a, ascent
4381 the maximum distance from the baseline to the highest/upper grid
4382 coordinate used to place a glyph outline point, for all the rendered
4384 It is a positive value, due to the grid's orientation with the Y axis
4387 @item max_glyph_d, descent
4388 the maximum distance from the baseline to the lowest grid coordinate
4389 used to place a glyph outline point, for all the rendered glyphs.
4390 This is a negative value, due to the grid's orientation, with the Y axis
4394 maximum glyph height, that is the maximum height for all the glyphs
4395 contained in the rendered text, it is equivalent to @var{ascent} -
4399 maximum glyph width, that is the maximum width for all the glyphs
4400 contained in the rendered text
4403 the number of input frame, starting from 0
4405 @item rand(min, max)
4406 return a random number included between @var{min} and @var{max}
4409 The input sample aspect ratio.
4412 timestamp expressed in seconds, NAN if the input timestamp is unknown
4415 the height of the rendered text
4418 the width of the rendered text
4422 the x and y offset coordinates where the text is drawn.
4424 These parameters allow the @var{x} and @var{y} expressions to refer
4425 each other, so you can for example specify @code{y=x/dar}.
4428 @anchor{drawtext_expansion}
4429 @subsection Text expansion
4431 If @option{expansion} is set to @code{strftime},
4432 the filter recognizes strftime() sequences in the provided text and
4433 expands them accordingly. Check the documentation of strftime(). This
4434 feature is deprecated.
4436 If @option{expansion} is set to @code{none}, the text is printed verbatim.
4438 If @option{expansion} is set to @code{normal} (which is the default),
4439 the following expansion mechanism is used.
4441 The backslash character @samp{\}, followed by any character, always expands to
4442 the second character.
4444 Sequence of the form @code{%@{...@}} are expanded. The text between the
4445 braces is a function name, possibly followed by arguments separated by ':'.
4446 If the arguments contain special characters or delimiters (':' or '@}'),
4447 they should be escaped.
4449 Note that they probably must also be escaped as the value for the
4450 @option{text} option in the filter argument string and as the filter
4451 argument in the filtergraph description, and possibly also for the shell,
4452 that makes up to four levels of escaping; using a text file avoids these
4455 The following functions are available:
4460 The expression evaluation result.
4462 It must take one argument specifying the expression to be evaluated,
4463 which accepts the same constants and functions as the @var{x} and
4464 @var{y} values. Note that not all constants should be used, for
4465 example the text size is not known when evaluating the expression, so
4466 the constants @var{text_w} and @var{text_h} will have an undefined
4469 @item expr_int_format, eif
4470 Evaluate the expression's value and output as formatted integer.
4472 The first argument is the expression to be evaluated, just as for the @var{expr} function.
4473 The second argument specifies the output format. Allowed values are @samp{x},
4474 @samp{X}, @samp{d} and @samp{u}. They are treated exactly as in the
4475 @code{printf} function.
4476 The third parameter is optional and sets the number of positions taken by the output.
4477 It can be used to add padding with zeros from the left.
4480 The time at which the filter is running, expressed in UTC.
4481 It can accept an argument: a strftime() format string.
4484 The time at which the filter is running, expressed in the local time zone.
4485 It can accept an argument: a strftime() format string.
4488 Frame metadata. It must take one argument specifying metadata key.
4491 The frame number, starting from 0.
4494 A 1 character description of the current picture type.
4497 The timestamp of the current frame.
4498 It can take up to two arguments.
4500 The first argument is the format of the timestamp; it defaults to @code{flt}
4501 for seconds as a decimal number with microsecond accuracy; @code{hms} stands
4502 for a formatted @var{[-]HH:MM:SS.mmm} timestamp with millisecond accuracy.
4504 The second argument is an offset added to the timestamp.
4508 @subsection Examples
4512 Draw "Test Text" with font FreeSerif, using the default values for the
4513 optional parameters.
4516 drawtext="fontfile=/usr/share/fonts/truetype/freefont/FreeSerif.ttf: text='Test Text'"
4520 Draw 'Test Text' with font FreeSerif of size 24 at position x=100
4521 and y=50 (counting from the top-left corner of the screen), text is
4522 yellow with a red box around it. Both the text and the box have an
4526 drawtext="fontfile=/usr/share/fonts/truetype/freefont/FreeSerif.ttf: text='Test Text':\
4527 x=100: y=50: fontsize=24: fontcolor=yellow@@0.2: box=1: boxcolor=red@@0.2"
4530 Note that the double quotes are not necessary if spaces are not used
4531 within the parameter list.
4534 Show the text at the center of the video frame:
4536 drawtext="fontsize=30:fontfile=FreeSerif.ttf:text='hello world':x=(w-text_w)/2:y=(h-text_h-line_h)/2"
4540 Show a text line sliding from right to left in the last row of the video
4541 frame. The file @file{LONG_LINE} is assumed to contain a single line
4544 drawtext="fontsize=15:fontfile=FreeSerif.ttf:text=LONG_LINE:y=h-line_h:x=-50*t"
4548 Show the content of file @file{CREDITS} off the bottom of the frame and scroll up.
4550 drawtext="fontsize=20:fontfile=FreeSerif.ttf:textfile=CREDITS:y=h-20*t"
4554 Draw a single green letter "g", at the center of the input video.
4555 The glyph baseline is placed at half screen height.
4557 drawtext="fontsize=60:fontfile=FreeSerif.ttf:fontcolor=green:text=g:x=(w-max_glyph_w)/2:y=h/2-ascent"
4561 Show text for 1 second every 3 seconds:
4563 drawtext="fontfile=FreeSerif.ttf:fontcolor=white:x=100:y=x/dar:enable=lt(mod(t\,3)\,1):text='blink'"
4567 Use fontconfig to set the font. Note that the colons need to be escaped.
4569 drawtext='fontfile=Linux Libertine O-40\:style=Semibold:text=FFmpeg'
4573 Print the date of a real-time encoding (see strftime(3)):
4575 drawtext='fontfile=FreeSans.ttf:text=%@{localtime\:%a %b %d %Y@}'
4579 Show text fading in and out (appearing/disappearing):
4582 DS=1.0 # display start
4583 DE=10.0 # display end
4584 FID=1.5 # fade in duration
4585 FOD=5 # fade out duration
4586 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 @}"
4591 For more information about libfreetype, check:
4592 @url{http://www.freetype.org/}.
4594 For more information about fontconfig, check:
4595 @url{http://freedesktop.org/software/fontconfig/fontconfig-user.html}.
4597 For more information about libfribidi, check:
4598 @url{http://fribidi.org/}.
4602 Detect and draw edges. The filter uses the Canny Edge Detection algorithm.
4604 The filter accepts the following options:
4609 Set low and high threshold values used by the Canny thresholding
4612 The high threshold selects the "strong" edge pixels, which are then
4613 connected through 8-connectivity with the "weak" edge pixels selected
4614 by the low threshold.
4616 @var{low} and @var{high} threshold values must be chosen in the range
4617 [0,1], and @var{low} should be lesser or equal to @var{high}.
4619 Default value for @var{low} is @code{20/255}, and default value for @var{high}
4623 Define the drawing mode.
4627 Draw white/gray wires on black background.
4630 Mix the colors to create a paint/cartoon effect.
4633 Default value is @var{wires}.
4636 @subsection Examples
4640 Standard edge detection with custom values for the hysteresis thresholding:
4642 edgedetect=low=0.1:high=0.4
4646 Painting effect without thresholding:
4648 edgedetect=mode=colormix:high=0
4653 Set brightness, contrast, saturation and approximate gamma adjustment.
4655 The filter accepts the following options:
4659 Set the contrast expression. The value must be a float value in range
4660 @code{-2.0} to @code{2.0}. The default value is "0".
4663 Set the brightness expression. The value must be a float value in
4664 range @code{-1.0} to @code{1.0}. The default value is "0".
4667 Set the saturation expression. The value must be a float in
4668 range @code{0.0} to @code{3.0}. The default value is "1".
4671 Set the gamma expression. The value must be a float in range
4672 @code{0.1} to @code{10.0}. The default value is "1".
4675 Set the gamma expression for red. The value must be a float in
4676 range @code{0.1} to @code{10.0}. The default value is "1".
4679 Set the gamma expression for green. The value must be a float in range
4680 @code{0.1} to @code{10.0}. The default value is "1".
4683 Set the gamma expression for blue. The value must be a float in range
4684 @code{0.1} to @code{10.0}. The default value is "1".
4687 Set the gamma weight expression. It can be used to reduce the effect
4688 of a high gamma value on bright image areas, e.g. keep them from
4689 getting overamplified and just plain white. The value must be a float
4690 in range @code{0.0} to @code{1.0}. A value of @code{0.0} turns the
4691 gamma correction all the way down while @code{1.0} leaves it at its
4692 full strength. Default is "1".
4695 Set when the expressions for brightness, contrast, saturation and
4696 gamma expressions are evaluated.
4698 It accepts the following values:
4701 only evaluate expressions once during the filter initialization or
4702 when a command is processed
4705 evaluate expressions for each incoming frame
4708 Default value is @samp{init}.
4711 The expressions accept the following parameters:
4714 frame count of the input frame starting from 0
4717 byte position of the corresponding packet in the input file, NAN if
4721 frame rate of the input video, NAN if the input frame rate is unknown
4724 timestamp expressed in seconds, NAN if the input timestamp is unknown
4727 @subsection Commands
4728 The filter supports the following commands:
4732 Set the contrast expression.
4735 Set the brightness expression.
4738 Set the saturation expression.
4741 Set the gamma expression.
4744 Set the gamma_r expression.
4747 Set gamma_g expression.
4750 Set gamma_b expression.
4753 Set gamma_weight expression.
4755 The command accepts the same syntax of the corresponding option.
4757 If the specified expression is not valid, it is kept at its current
4762 @section extractplanes
4764 Extract color channel components from input video stream into
4765 separate grayscale video streams.
4767 The filter accepts the following option:
4771 Set plane(s) to extract.
4773 Available values for planes are:
4784 Choosing planes not available in the input will result in an error.
4785 That means you cannot select @code{r}, @code{g}, @code{b} planes
4786 with @code{y}, @code{u}, @code{v} planes at same time.
4789 @subsection Examples
4793 Extract luma, u and v color channel component from input video frame
4794 into 3 grayscale outputs:
4796 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
4802 Apply a posterize effect using the ELBG (Enhanced LBG) algorithm.
4804 For each input image, the filter will compute the optimal mapping from
4805 the input to the output given the codebook length, that is the number
4806 of distinct output colors.
4808 This filter accepts the following options.
4811 @item codebook_length, l
4812 Set codebook length. The value must be a positive integer, and
4813 represents the number of distinct output colors. Default value is 256.
4816 Set the maximum number of iterations to apply for computing the optimal
4817 mapping. The higher the value the better the result and the higher the
4818 computation time. Default value is 1.
4821 Set a random seed, must be an integer included between 0 and
4822 UINT32_MAX. If not specified, or if explicitly set to -1, the filter
4823 will try to use a good random seed on a best effort basis.
4828 Apply a fade-in/out effect to the input video.
4830 It accepts the following parameters:
4834 The effect type can be either "in" for a fade-in, or "out" for a fade-out
4836 Default is @code{in}.
4838 @item start_frame, s
4839 Specify the number of the frame to start applying the fade
4840 effect at. Default is 0.
4843 The number of frames that the fade effect lasts. At the end of the
4844 fade-in effect, the output video will have the same intensity as the input video.
4845 At the end of the fade-out transition, the output video will be filled with the
4846 selected @option{color}.
4850 If set to 1, fade only alpha channel, if one exists on the input.
4853 @item start_time, st
4854 Specify the timestamp (in seconds) of the frame to start to apply the fade
4855 effect. If both start_frame and start_time are specified, the fade will start at
4856 whichever comes last. Default is 0.
4859 The number of seconds for which the fade effect has to last. At the end of the
4860 fade-in effect the output video will have the same intensity as the input video,
4861 at the end of the fade-out transition the output video will be filled with the
4862 selected @option{color}.
4863 If both duration and nb_frames are specified, duration is used. Default is 0
4864 (nb_frames is used by default).
4867 Specify the color of the fade. Default is "black".
4870 @subsection Examples
4874 Fade in the first 30 frames of video:
4879 The command above is equivalent to:
4885 Fade out the last 45 frames of a 200-frame video:
4888 fade=type=out:start_frame=155:nb_frames=45
4892 Fade in the first 25 frames and fade out the last 25 frames of a 1000-frame video:
4894 fade=in:0:25, fade=out:975:25
4898 Make the first 5 frames yellow, then fade in from frame 5-24:
4900 fade=in:5:20:color=yellow
4904 Fade in alpha over first 25 frames of video:
4906 fade=in:0:25:alpha=1
4910 Make the first 5.5 seconds black, then fade in for 0.5 seconds:
4912 fade=t=in:st=5.5:d=0.5
4918 Apply arbitrary expressions to samples in frequency domain
4922 Adjust the dc value (gain) of the luma plane of the image. The filter
4923 accepts an integer value in range @code{0} to @code{1000}. The default
4924 value is set to @code{0}.
4927 Adjust the dc value (gain) of the 1st chroma plane of the image. The
4928 filter accepts an integer value in range @code{0} to @code{1000}. The
4929 default value is set to @code{0}.
4932 Adjust the dc value (gain) of the 2nd chroma plane of the image. The
4933 filter accepts an integer value in range @code{0} to @code{1000}. The
4934 default value is set to @code{0}.
4937 Set the frequency domain weight expression for the luma plane.
4940 Set the frequency domain weight expression for the 1st chroma plane.
4943 Set the frequency domain weight expression for the 2nd chroma plane.
4945 The filter accepts the following variables:
4948 The coordinates of the current sample.
4952 The width and height of the image.
4955 @subsection Examples
4961 fftfilt=dc_Y=128:weight_Y='squish(1-(Y+X)/100)'
4967 fftfilt=dc_Y=0:weight_Y='squish((Y+X)/100-1)'
4973 fftfilt=dc_Y=0:weight_Y='1+squish(1-(Y+X)/100)'
4980 Extract a single field from an interlaced image using stride
4981 arithmetic to avoid wasting CPU time. The output frames are marked as
4984 The filter accepts the following options:
4988 Specify whether to extract the top (if the value is @code{0} or
4989 @code{top}) or the bottom field (if the value is @code{1} or
4995 Field matching filter for inverse telecine. It is meant to reconstruct the
4996 progressive frames from a telecined stream. The filter does not drop duplicated
4997 frames, so to achieve a complete inverse telecine @code{fieldmatch} needs to be
4998 followed by a decimation filter such as @ref{decimate} in the filtergraph.
5000 The separation of the field matching and the decimation is notably motivated by
5001 the possibility of inserting a de-interlacing filter fallback between the two.
5002 If the source has mixed telecined and real interlaced content,
5003 @code{fieldmatch} will not be able to match fields for the interlaced parts.
5004 But these remaining combed frames will be marked as interlaced, and thus can be
5005 de-interlaced by a later filter such as @ref{yadif} before decimation.
5007 In addition to the various configuration options, @code{fieldmatch} can take an
5008 optional second stream, activated through the @option{ppsrc} option. If
5009 enabled, the frames reconstruction will be based on the fields and frames from
5010 this second stream. This allows the first input to be pre-processed in order to
5011 help the various algorithms of the filter, while keeping the output lossless
5012 (assuming the fields are matched properly). Typically, a field-aware denoiser,
5013 or brightness/contrast adjustments can help.
5015 Note that this filter uses the same algorithms as TIVTC/TFM (AviSynth project)
5016 and VIVTC/VFM (VapourSynth project). The later is a light clone of TFM from
5017 which @code{fieldmatch} is based on. While the semantic and usage are very
5018 close, some behaviour and options names can differ.
5020 The @ref{decimate} filter currently only works for constant frame rate input.
5021 Do not use @code{fieldmatch} and @ref{decimate} if your input has mixed
5022 telecined and progressive content with changing framerate.
5024 The filter accepts the following options:
5028 Specify the assumed field order of the input stream. Available values are:
5032 Auto detect parity (use FFmpeg's internal parity value).
5034 Assume bottom field first.
5036 Assume top field first.
5039 Note that it is sometimes recommended not to trust the parity announced by the
5042 Default value is @var{auto}.
5045 Set the matching mode or strategy to use. @option{pc} mode is the safest in the
5046 sense that it won't risk creating jerkiness due to duplicate frames when
5047 possible, but if there are bad edits or blended fields it will end up
5048 outputting combed frames when a good match might actually exist. On the other
5049 hand, @option{pcn_ub} mode is the most risky in terms of creating jerkiness,
5050 but will almost always find a good frame if there is one. The other values are
5051 all somewhere in between @option{pc} and @option{pcn_ub} in terms of risking
5052 jerkiness and creating duplicate frames versus finding good matches in sections
5053 with bad edits, orphaned fields, blended fields, etc.
5055 More details about p/c/n/u/b are available in @ref{p/c/n/u/b meaning} section.
5057 Available values are:
5061 2-way matching (p/c)
5063 2-way matching, and trying 3rd match if still combed (p/c + n)
5065 2-way matching, and trying 3rd match (same order) if still combed (p/c + u)
5067 2-way matching, trying 3rd match if still combed, and trying 4th/5th matches if
5068 still combed (p/c + n + u/b)
5070 3-way matching (p/c/n)
5072 3-way matching, and trying 4th/5th matches if all 3 of the original matches are
5073 detected as combed (p/c/n + u/b)
5076 The parenthesis at the end indicate the matches that would be used for that
5077 mode assuming @option{order}=@var{tff} (and @option{field} on @var{auto} or
5080 In terms of speed @option{pc} mode is by far the fastest and @option{pcn_ub} is
5083 Default value is @var{pc_n}.
5086 Mark the main input stream as a pre-processed input, and enable the secondary
5087 input stream as the clean source to pick the fields from. See the filter
5088 introduction for more details. It is similar to the @option{clip2} feature from
5091 Default value is @code{0} (disabled).
5094 Set the field to match from. It is recommended to set this to the same value as
5095 @option{order} unless you experience matching failures with that setting. In
5096 certain circumstances changing the field that is used to match from can have a
5097 large impact on matching performance. Available values are:
5101 Automatic (same value as @option{order}).
5103 Match from the bottom field.
5105 Match from the top field.
5108 Default value is @var{auto}.
5111 Set whether or not chroma is included during the match comparisons. In most
5112 cases it is recommended to leave this enabled. You should set this to @code{0}
5113 only if your clip has bad chroma problems such as heavy rainbowing or other
5114 artifacts. Setting this to @code{0} could also be used to speed things up at
5115 the cost of some accuracy.
5117 Default value is @code{1}.
5121 These define an exclusion band which excludes the lines between @option{y0} and
5122 @option{y1} from being included in the field matching decision. An exclusion
5123 band can be used to ignore subtitles, a logo, or other things that may
5124 interfere with the matching. @option{y0} sets the starting scan line and
5125 @option{y1} sets the ending line; all lines in between @option{y0} and
5126 @option{y1} (including @option{y0} and @option{y1}) will be ignored. Setting
5127 @option{y0} and @option{y1} to the same value will disable the feature.
5128 @option{y0} and @option{y1} defaults to @code{0}.
5131 Set the scene change detection threshold as a percentage of maximum change on
5132 the luma plane. Good values are in the @code{[8.0, 14.0]} range. Scene change
5133 detection is only relevant in case @option{combmatch}=@var{sc}. The range for
5134 @option{scthresh} is @code{[0.0, 100.0]}.
5136 Default value is @code{12.0}.
5139 When @option{combatch} is not @var{none}, @code{fieldmatch} will take into
5140 account the combed scores of matches when deciding what match to use as the
5141 final match. Available values are:
5145 No final matching based on combed scores.
5147 Combed scores are only used when a scene change is detected.
5149 Use combed scores all the time.
5152 Default is @var{sc}.
5155 Force @code{fieldmatch} to calculate the combed metrics for certain matches and
5156 print them. This setting is known as @option{micout} in TFM/VFM vocabulary.
5157 Available values are:
5161 No forced calculation.
5163 Force p/c/n calculations.
5165 Force p/c/n/u/b calculations.
5168 Default value is @var{none}.
5171 This is the area combing threshold used for combed frame detection. This
5172 essentially controls how "strong" or "visible" combing must be to be detected.
5173 Larger values mean combing must be more visible and smaller values mean combing
5174 can be less visible or strong and still be detected. Valid settings are from
5175 @code{-1} (every pixel will be detected as combed) to @code{255} (no pixel will
5176 be detected as combed). This is basically a pixel difference value. A good
5177 range is @code{[8, 12]}.
5179 Default value is @code{9}.
5182 Sets whether or not chroma is considered in the combed frame decision. Only
5183 disable this if your source has chroma problems (rainbowing, etc.) that are
5184 causing problems for the combed frame detection with chroma enabled. Actually,
5185 using @option{chroma}=@var{0} is usually more reliable, except for the case
5186 where there is chroma only combing in the source.
5188 Default value is @code{0}.
5192 Respectively set the x-axis and y-axis size of the window used during combed
5193 frame detection. This has to do with the size of the area in which
5194 @option{combpel} pixels are required to be detected as combed for a frame to be
5195 declared combed. See the @option{combpel} parameter description for more info.
5196 Possible values are any number that is a power of 2 starting at 4 and going up
5199 Default value is @code{16}.
5202 The number of combed pixels inside any of the @option{blocky} by
5203 @option{blockx} size blocks on the frame for the frame to be detected as
5204 combed. While @option{cthresh} controls how "visible" the combing must be, this
5205 setting controls "how much" combing there must be in any localized area (a
5206 window defined by the @option{blockx} and @option{blocky} settings) on the
5207 frame. Minimum value is @code{0} and maximum is @code{blocky x blockx} (at
5208 which point no frames will ever be detected as combed). This setting is known
5209 as @option{MI} in TFM/VFM vocabulary.
5211 Default value is @code{80}.
5214 @anchor{p/c/n/u/b meaning}
5215 @subsection p/c/n/u/b meaning
5217 @subsubsection p/c/n
5219 We assume the following telecined stream:
5222 Top fields: 1 2 2 3 4
5223 Bottom fields: 1 2 3 4 4
5226 The numbers correspond to the progressive frame the fields relate to. Here, the
5227 first two frames are progressive, the 3rd and 4th are combed, and so on.
5229 When @code{fieldmatch} is configured to run a matching from bottom
5230 (@option{field}=@var{bottom}) this is how this input stream get transformed:
5235 B 1 2 3 4 4 <-- matching reference
5244 As a result of the field matching, we can see that some frames get duplicated.
5245 To perform a complete inverse telecine, you need to rely on a decimation filter
5246 after this operation. See for instance the @ref{decimate} filter.
5248 The same operation now matching from top fields (@option{field}=@var{top})
5253 T 1 2 2 3 4 <-- matching reference
5263 In these examples, we can see what @var{p}, @var{c} and @var{n} mean;
5264 basically, they refer to the frame and field of the opposite parity:
5267 @item @var{p} matches the field of the opposite parity in the previous frame
5268 @item @var{c} matches the field of the opposite parity in the current frame
5269 @item @var{n} matches the field of the opposite parity in the next frame
5274 The @var{u} and @var{b} matching are a bit special in the sense that they match
5275 from the opposite parity flag. In the following examples, we assume that we are
5276 currently matching the 2nd frame (Top:2, bottom:2). According to the match, a
5277 'x' is placed above and below each matched fields.
5279 With bottom matching (@option{field}=@var{bottom}):
5284 Top 1 2 2 1 2 2 1 2 2 1 2 2 1 2 2
5285 Bottom 1 2 3 1 2 3 1 2 3 1 2 3 1 2 3
5293 With top matching (@option{field}=@var{top}):
5298 Top 1 2 2 1 2 2 1 2 2 1 2 2 1 2 2
5299 Bottom 1 2 3 1 2 3 1 2 3 1 2 3 1 2 3
5307 @subsection Examples
5309 Simple IVTC of a top field first telecined stream:
5311 fieldmatch=order=tff:combmatch=none, decimate
5314 Advanced IVTC, with fallback on @ref{yadif} for still combed frames:
5316 fieldmatch=order=tff:combmatch=full, yadif=deint=interlaced, decimate
5321 Transform the field order of the input video.
5323 It accepts the following parameters:
5328 The output field order. Valid values are @var{tff} for top field first or @var{bff}
5329 for bottom field first.
5332 The default value is @samp{tff}.
5334 The transformation is done by shifting the picture content up or down
5335 by one line, and filling the remaining line with appropriate picture content.
5336 This method is consistent with most broadcast field order converters.
5338 If the input video is not flagged as being interlaced, or it is already
5339 flagged as being of the required output field order, then this filter does
5340 not alter the incoming video.
5342 It is very useful when converting to or from PAL DV material,
5343 which is bottom field first.
5347 ffmpeg -i in.vob -vf "fieldorder=bff" out.dv
5352 Buffer input images and send them when they are requested.
5354 It is mainly useful when auto-inserted by the libavfilter
5357 It does not take parameters.
5361 Find a rectangular object
5363 It accepts the following options:
5367 Filepath of the object image, needs to be in gray8.
5370 Detection threshold, default is 0.5.
5373 Number of mipmaps, default is 3.
5375 @item xmin, ymin, xmax, ymax
5376 Specifies the rectangle in which to search.
5379 @subsection Examples
5383 Generate a representative palette of a given video using @command{ffmpeg}:
5385 ffmpeg -i file.ts -vf find_rect=newref.pgm,cover_rect=cover.jpg:mode=cover new.mkv
5391 Cover a rectangular object
5393 It accepts the following options:
5397 Filepath of the optional cover image, needs to be in yuv420.
5402 It accepts the following values:
5405 cover it by the supplied image
5407 cover it by interpolating the surrounding pixels
5410 Default value is @var{blur}.
5413 @subsection Examples
5417 Generate a representative palette of a given video using @command{ffmpeg}:
5419 ffmpeg -i file.ts -vf find_rect=newref.pgm,cover_rect=cover.jpg:mode=cover new.mkv
5426 Convert the input video to one of the specified pixel formats.
5427 Libavfilter will try to pick one that is suitable as input to
5430 It accepts the following parameters:
5434 A '|'-separated list of pixel format names, such as
5435 "pix_fmts=yuv420p|monow|rgb24".
5439 @subsection Examples
5443 Convert the input video to the @var{yuv420p} format
5445 format=pix_fmts=yuv420p
5448 Convert the input video to any of the formats in the list
5450 format=pix_fmts=yuv420p|yuv444p|yuv410p
5457 Convert the video to specified constant frame rate by duplicating or dropping
5458 frames as necessary.
5460 It accepts the following parameters:
5464 The desired output frame rate. The default is @code{25}.
5469 Possible values are:
5472 zero round towards 0
5476 round towards -infinity
5478 round towards +infinity
5482 The default is @code{near}.
5485 Assume the first PTS should be the given value, in seconds. This allows for
5486 padding/trimming at the start of stream. By default, no assumption is made
5487 about the first frame's expected PTS, so no padding or trimming is done.
5488 For example, this could be set to 0 to pad the beginning with duplicates of
5489 the first frame if a video stream starts after the audio stream or to trim any
5490 frames with a negative PTS.
5494 Alternatively, the options can be specified as a flat string:
5495 @var{fps}[:@var{round}].
5497 See also the @ref{setpts} filter.
5499 @subsection Examples
5503 A typical usage in order to set the fps to 25:
5509 Sets the fps to 24, using abbreviation and rounding method to round to nearest:
5511 fps=fps=film:round=near
5517 Pack two different video streams into a stereoscopic video, setting proper
5518 metadata on supported codecs. The two views should have the same size and
5519 framerate and processing will stop when the shorter video ends. Please note
5520 that you may conveniently adjust view properties with the @ref{scale} and
5523 It accepts the following parameters:
5527 The desired packing format. Supported values are:
5532 The views are next to each other (default).
5535 The views are on top of each other.
5538 The views are packed by line.
5541 The views are packed by column.
5544 The views are temporally interleaved.
5553 # Convert left and right views into a frame-sequential video
5554 ffmpeg -i LEFT -i RIGHT -filter_complex framepack=frameseq OUTPUT
5556 # Convert views into a side-by-side video with the same output resolution as the input
5557 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
5562 Select one frame every N-th frame.
5564 This filter accepts the following option:
5567 Select frame after every @code{step} frames.
5568 Allowed values are positive integers higher than 0. Default value is @code{1}.
5574 Apply a frei0r effect to the input video.
5576 To enable the compilation of this filter, you need to install the frei0r
5577 header and configure FFmpeg with @code{--enable-frei0r}.
5579 It accepts the following parameters:
5584 The name of the frei0r effect to load. If the environment variable
5585 @env{FREI0R_PATH} is defined, the frei0r effect is searched for in each of the
5586 directories specified by the colon-separated list in @env{FREIOR_PATH}.
5587 Otherwise, the standard frei0r paths are searched, in this order:
5588 @file{HOME/.frei0r-1/lib/}, @file{/usr/local/lib/frei0r-1/},
5589 @file{/usr/lib/frei0r-1/}.
5592 A '|'-separated list of parameters to pass to the frei0r effect.
5596 A frei0r effect parameter can be a boolean (its value is either
5597 "y" or "n"), a double, a color (specified as
5598 @var{R}/@var{G}/@var{B}, where @var{R}, @var{G}, and @var{B} are floating point
5599 numbers between 0.0 and 1.0, inclusive) or by a color description specified in the "Color"
5600 section in the ffmpeg-utils manual), a position (specified as @var{X}/@var{Y}, where
5601 @var{X} and @var{Y} are floating point numbers) and/or a string.
5603 The number and types of parameters depend on the loaded effect. If an
5604 effect parameter is not specified, the default value is set.
5606 @subsection Examples
5610 Apply the distort0r effect, setting the first two double parameters:
5612 frei0r=filter_name=distort0r:filter_params=0.5|0.01
5616 Apply the colordistance effect, taking a color as the first parameter:
5618 frei0r=colordistance:0.2/0.3/0.4
5619 frei0r=colordistance:violet
5620 frei0r=colordistance:0x112233
5624 Apply the perspective effect, specifying the top left and top right image
5627 frei0r=perspective:0.2/0.2|0.8/0.2
5631 For more information, see
5632 @url{http://frei0r.dyne.org}
5636 Apply fast and simple postprocessing. It is a faster version of @ref{spp}.
5638 It splits (I)DCT into horizontal/vertical passes. Unlike the simple post-
5639 processing filter, one of them is performed once per block, not per pixel.
5640 This allows for much higher speed.
5642 The filter accepts the following options:
5646 Set quality. This option defines the number of levels for averaging. It accepts
5647 an integer in the range 4-5. Default value is @code{4}.
5650 Force a constant quantization parameter. It accepts an integer in range 0-63.
5651 If not set, the filter will use the QP from the video stream (if available).
5654 Set filter strength. It accepts an integer in range -15 to 32. Lower values mean
5655 more details but also more artifacts, while higher values make the image smoother
5656 but also blurrier. Default value is @code{0} − PSNR optimal.
5659 Enable the use of the QP from the B-Frames if set to @code{1}. Using this
5660 option may cause flicker since the B-Frames have often larger QP. Default is
5661 @code{0} (not enabled).
5667 The filter accepts the following options:
5671 Set the luminance expression.
5673 Set the chrominance blue expression.
5675 Set the chrominance red expression.
5677 Set the alpha expression.
5679 Set the red expression.
5681 Set the green expression.
5683 Set the blue expression.
5686 The colorspace is selected according to the specified options. If one
5687 of the @option{lum_expr}, @option{cb_expr}, or @option{cr_expr}
5688 options is specified, the filter will automatically select a YCbCr
5689 colorspace. If one of the @option{red_expr}, @option{green_expr}, or
5690 @option{blue_expr} options is specified, it will select an RGB
5693 If one of the chrominance expression is not defined, it falls back on the other
5694 one. If no alpha expression is specified it will evaluate to opaque value.
5695 If none of chrominance expressions are specified, they will evaluate
5696 to the luminance expression.
5698 The expressions can use the following variables and functions:
5702 The sequential number of the filtered frame, starting from @code{0}.
5706 The coordinates of the current sample.
5710 The width and height of the image.
5714 Width and height scale depending on the currently filtered plane. It is the
5715 ratio between the corresponding luma plane number of pixels and the current
5716 plane ones. E.g. for YUV4:2:0 the values are @code{1,1} for the luma plane, and
5717 @code{0.5,0.5} for chroma planes.
5720 Time of the current frame, expressed in seconds.
5723 Return the value of the pixel at location (@var{x},@var{y}) of the current
5727 Return the value of the pixel at location (@var{x},@var{y}) of the luminance
5731 Return the value of the pixel at location (@var{x},@var{y}) of the
5732 blue-difference chroma plane. Return 0 if there is no such plane.
5735 Return the value of the pixel at location (@var{x},@var{y}) of the
5736 red-difference chroma plane. Return 0 if there is no such plane.
5741 Return the value of the pixel at location (@var{x},@var{y}) of the
5742 red/green/blue component. Return 0 if there is no such component.
5745 Return the value of the pixel at location (@var{x},@var{y}) of the alpha
5746 plane. Return 0 if there is no such plane.
5749 For functions, if @var{x} and @var{y} are outside the area, the value will be
5750 automatically clipped to the closer edge.
5752 @subsection Examples
5756 Flip the image horizontally:
5762 Generate a bidimensional sine wave, with angle @code{PI/3} and a
5763 wavelength of 100 pixels:
5765 geq=128 + 100*sin(2*(PI/100)*(cos(PI/3)*(X-50*T) + sin(PI/3)*Y)):128:128
5769 Generate a fancy enigmatic moving light:
5771 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
5775 Generate a quick emboss effect:
5777 format=gray,geq=lum_expr='(p(X,Y)+(256-p(X-4,Y-4)))/2'
5781 Modify RGB components depending on pixel position:
5783 geq=r='X/W*r(X,Y)':g='(1-X/W)*g(X,Y)':b='(H-Y)/H*b(X,Y)'
5787 Create a radial gradient that is the same size as the input (also see
5788 the @ref{vignette} filter):
5790 geq=lum=255*gauss((X/W-0.5)*3)*gauss((Y/H-0.5)*3)/gauss(0)/gauss(0),format=gray
5794 Create a linear gradient to use as a mask for another filter, then
5795 compose with @ref{overlay}. In this example the video will gradually
5796 become more blurry from the top to the bottom of the y-axis as defined
5797 by the linear gradient:
5799 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
5805 Fix the banding artifacts that are sometimes introduced into nearly flat
5806 regions by truncation to 8bit color depth.
5807 Interpolate the gradients that should go where the bands are, and
5810 It is designed for playback only. Do not use it prior to
5811 lossy compression, because compression tends to lose the dither and
5812 bring back the bands.
5814 It accepts the following parameters:
5819 The maximum amount by which the filter will change any one pixel. This is also
5820 the threshold for detecting nearly flat regions. Acceptable values range from
5821 .51 to 64; the default value is 1.2. Out-of-range values will be clipped to the
5825 The neighborhood to fit the gradient to. A larger radius makes for smoother
5826 gradients, but also prevents the filter from modifying the pixels near detailed
5827 regions. Acceptable values are 8-32; the default value is 16. Out-of-range
5828 values will be clipped to the valid range.
5832 Alternatively, the options can be specified as a flat string:
5833 @var{strength}[:@var{radius}]
5835 @subsection Examples
5839 Apply the filter with a @code{3.5} strength and radius of @code{8}:
5845 Specify radius, omitting the strength (which will fall-back to the default
5856 Apply a Hald CLUT to a video stream.
5858 First input is the video stream to process, and second one is the Hald CLUT.
5859 The Hald CLUT input can be a simple picture or a complete video stream.
5861 The filter accepts the following options:
5865 Force termination when the shortest input terminates. Default is @code{0}.
5867 Continue applying the last CLUT after the end of the stream. A value of
5868 @code{0} disable the filter after the last frame of the CLUT is reached.
5869 Default is @code{1}.
5872 @code{haldclut} also has the same interpolation options as @ref{lut3d} (both
5873 filters share the same internals).
5875 More information about the Hald CLUT can be found on Eskil Steenberg's website
5876 (Hald CLUT author) at @url{http://www.quelsolaar.com/technology/clut.html}.
5878 @subsection Workflow examples
5880 @subsubsection Hald CLUT video stream
5882 Generate an identity Hald CLUT stream altered with various effects:
5884 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
5887 Note: make sure you use a lossless codec.
5889 Then use it with @code{haldclut} to apply it on some random stream:
5891 ffmpeg -f lavfi -i mandelbrot -i clut.nut -filter_complex '[0][1] haldclut' -t 20 mandelclut.mkv
5894 The Hald CLUT will be applied to the 10 first seconds (duration of
5895 @file{clut.nut}), then the latest picture of that CLUT stream will be applied
5896 to the remaining frames of the @code{mandelbrot} stream.
5898 @subsubsection Hald CLUT with preview
5900 A Hald CLUT is supposed to be a squared image of @code{Level*Level*Level} by
5901 @code{Level*Level*Level} pixels. For a given Hald CLUT, FFmpeg will select the
5902 biggest possible square starting at the top left of the picture. The remaining
5903 padding pixels (bottom or right) will be ignored. This area can be used to add
5904 a preview of the Hald CLUT.
5906 Typically, the following generated Hald CLUT will be supported by the
5907 @code{haldclut} filter:
5910 ffmpeg -f lavfi -i @ref{haldclutsrc}=8 -vf "
5911 pad=iw+320 [padded_clut];
5912 smptebars=s=320x256, split [a][b];
5913 [padded_clut][a] overlay=W-320:h, curves=color_negative [main];
5914 [main][b] overlay=W-320" -frames:v 1 clut.png
5917 It contains the original and a preview of the effect of the CLUT: SMPTE color
5918 bars are displayed on the right-top, and below the same color bars processed by
5921 Then, the effect of this Hald CLUT can be visualized with:
5923 ffplay input.mkv -vf "movie=clut.png, [in] haldclut"
5928 Flip the input video horizontally.
5930 For example, to horizontally flip the input video with @command{ffmpeg}:
5932 ffmpeg -i in.avi -vf "hflip" out.avi
5936 This filter applies a global color histogram equalization on a
5939 It can be used to correct video that has a compressed range of pixel
5940 intensities. The filter redistributes the pixel intensities to
5941 equalize their distribution across the intensity range. It may be
5942 viewed as an "automatically adjusting contrast filter". This filter is
5943 useful only for correcting degraded or poorly captured source
5946 The filter accepts the following options:
5950 Determine the amount of equalization to be applied. As the strength
5951 is reduced, the distribution of pixel intensities more-and-more
5952 approaches that of the input frame. The value must be a float number
5953 in the range [0,1] and defaults to 0.200.
5956 Set the maximum intensity that can generated and scale the output
5957 values appropriately. The strength should be set as desired and then
5958 the intensity can be limited if needed to avoid washing-out. The value
5959 must be a float number in the range [0,1] and defaults to 0.210.
5962 Set the antibanding level. If enabled the filter will randomly vary
5963 the luminance of output pixels by a small amount to avoid banding of
5964 the histogram. Possible values are @code{none}, @code{weak} or
5965 @code{strong}. It defaults to @code{none}.
5970 Compute and draw a color distribution histogram for the input video.
5972 The computed histogram is a representation of the color component
5973 distribution in an image.
5975 The filter accepts the following options:
5981 It accepts the following values:
5984 Standard histogram that displays the color components distribution in an
5985 image. Displays color graph for each color component. Shows distribution of
5986 the Y, U, V, A or R, G, B components, depending on input format, in the
5987 current frame. Below each graph a color component scale meter is shown.
5990 Displays chroma values (U/V color placement) in a two dimensional
5991 graph (which is called a vectorscope). The brighter a pixel in the
5992 vectorscope, the more pixels of the input frame correspond to that pixel
5993 (i.e., more pixels have this chroma value). The V component is displayed on
5994 the horizontal (X) axis, with the leftmost side being V = 0 and the rightmost
5995 side being V = 255. The U component is displayed on the vertical (Y) axis,
5996 with the top representing U = 0 and the bottom representing U = 255.
5998 The position of a white pixel in the graph corresponds to the chroma value of
5999 a pixel of the input clip. The graph can therefore be used to read the hue
6000 (color flavor) and the saturation (the dominance of the hue in the color). As
6001 the hue of a color changes, it moves around the square. At the center of the
6002 square the saturation is zero, which means that the corresponding pixel has no
6003 color. If the amount of a specific color is increased (while leaving the other
6004 colors unchanged) the saturation increases, and the indicator moves towards
6005 the edge of the square.
6008 Chroma values in vectorscope, similar as @code{color} but actual chroma values
6012 Per row/column color component graph. In row mode, the graph on the left side
6013 represents color component value 0 and the right side represents value = 255.
6014 In column mode, the top side represents color component value = 0 and bottom
6015 side represents value = 255.
6017 Default value is @code{levels}.
6020 Set height of level in @code{levels}. Default value is @code{200}.
6021 Allowed range is [50, 2048].
6024 Set height of color scale in @code{levels}. Default value is @code{12}.
6025 Allowed range is [0, 40].
6028 Set step for @code{waveform} mode. Smaller values are useful to find out how
6029 many values of the same luminance are distributed across input rows/columns.
6030 Default value is @code{10}. Allowed range is [1, 255].
6033 Set mode for @code{waveform}. Can be either @code{row}, or @code{column}.
6034 Default is @code{row}.
6036 @item waveform_mirror
6037 Set mirroring mode for @code{waveform}. @code{0} means unmirrored, @code{1}
6038 means mirrored. In mirrored mode, higher values will be represented on the left
6039 side for @code{row} mode and at the top for @code{column} mode. Default is
6040 @code{0} (unmirrored).
6043 Set display mode for @code{waveform} and @code{levels}.
6044 It accepts the following values:
6047 Display separate graph for the color components side by side in
6048 @code{row} waveform mode or one below the other in @code{column} waveform mode
6049 for @code{waveform} histogram mode. For @code{levels} histogram mode,
6050 per color component graphs are placed below each other.
6052 Using this display mode in @code{waveform} histogram mode makes it easy to
6053 spot color casts in the highlights and shadows of an image, by comparing the
6054 contours of the top and the bottom graphs of each waveform. Since whites,
6055 grays, and blacks are characterized by exactly equal amounts of red, green,
6056 and blue, neutral areas of the picture should display three waveforms of
6057 roughly equal width/height. If not, the correction is easy to perform by
6058 making level adjustments the three waveforms.
6061 Presents information identical to that in the @code{parade}, except
6062 that the graphs representing color components are superimposed directly
6065 This display mode in @code{waveform} histogram mode makes it easier to spot
6066 relative differences or similarities in overlapping areas of the color
6067 components that are supposed to be identical, such as neutral whites, grays,
6070 Default is @code{parade}.
6073 Set mode for @code{levels}. Can be either @code{linear}, or @code{logarithmic}.
6074 Default is @code{linear}.
6077 @subsection Examples
6082 Calculate and draw histogram:
6084 ffplay -i input -vf histogram
6092 This is a high precision/quality 3d denoise filter. It aims to reduce
6093 image noise, producing smooth images and making still images really
6094 still. It should enhance compressibility.
6096 It accepts the following optional parameters:
6100 A non-negative floating point number which specifies spatial luma strength.
6103 @item chroma_spatial
6104 A non-negative floating point number which specifies spatial chroma strength.
6105 It defaults to 3.0*@var{luma_spatial}/4.0.
6108 A floating point number which specifies luma temporal strength. It defaults to
6109 6.0*@var{luma_spatial}/4.0.
6112 A floating point number which specifies chroma temporal strength. It defaults to
6113 @var{luma_tmp}*@var{chroma_spatial}/@var{luma_spatial}.
6118 Apply a high-quality magnification filter designed for pixel art. This filter
6119 was originally created by Maxim Stepin.
6121 It accepts the following option:
6125 Set the scaling dimension: @code{2} for @code{hq2x}, @code{3} for
6126 @code{hq3x} and @code{4} for @code{hq4x}.
6127 Default is @code{3}.
6132 Modify the hue and/or the saturation of the input.
6134 It accepts the following parameters:
6138 Specify the hue angle as a number of degrees. It accepts an expression,
6139 and defaults to "0".
6142 Specify the saturation in the [-10,10] range. It accepts an expression and
6146 Specify the hue angle as a number of radians. It accepts an
6147 expression, and defaults to "0".
6150 Specify the brightness in the [-10,10] range. It accepts an expression and
6154 @option{h} and @option{H} are mutually exclusive, and can't be
6155 specified at the same time.
6157 The @option{b}, @option{h}, @option{H} and @option{s} option values are
6158 expressions containing the following constants:
6162 frame count of the input frame starting from 0
6165 presentation timestamp of the input frame expressed in time base units
6168 frame rate of the input video, NAN if the input frame rate is unknown
6171 timestamp expressed in seconds, NAN if the input timestamp is unknown
6174 time base of the input video
6177 @subsection Examples
6181 Set the hue to 90 degrees and the saturation to 1.0:
6187 Same command but expressing the hue in radians:
6193 Rotate hue and make the saturation swing between 0
6194 and 2 over a period of 1 second:
6196 hue="H=2*PI*t: s=sin(2*PI*t)+1"
6200 Apply a 3 seconds saturation fade-in effect starting at 0:
6205 The general fade-in expression can be written as:
6207 hue="s=min(0\, max((t-START)/DURATION\, 1))"
6211 Apply a 3 seconds saturation fade-out effect starting at 5 seconds:
6213 hue="s=max(0\, min(1\, (8-t)/3))"
6216 The general fade-out expression can be written as:
6218 hue="s=max(0\, min(1\, (START+DURATION-t)/DURATION))"
6223 @subsection Commands
6225 This filter supports the following commands:
6231 Modify the hue and/or the saturation and/or brightness of the input video.
6232 The command accepts the same syntax of the corresponding option.
6234 If the specified expression is not valid, it is kept at its current
6240 Detect video interlacing type.
6242 This filter tries to detect if the input frames as interlaced, progressive,
6243 top or bottom field first. It will also try and detect fields that are
6244 repeated between adjacent frames (a sign of telecine).
6246 Single frame detection considers only immediately adjacent frames when classifying each frame.
6247 Multiple frame detection incorporates the classification history of previous frames.
6249 The filter will log these metadata values:
6252 @item single.current_frame
6253 Detected type of current frame using single-frame detection. One of:
6254 ``tff'' (top field first), ``bff'' (bottom field first),
6255 ``progressive'', or ``undetermined''
6258 Cumulative number of frames detected as top field first using single-frame detection.
6261 Cumulative number of frames detected as top field first using multiple-frame detection.
6264 Cumulative number of frames detected as bottom field first using single-frame detection.
6266 @item multiple.current_frame
6267 Detected type of current frame using multiple-frame detection. One of:
6268 ``tff'' (top field first), ``bff'' (bottom field first),
6269 ``progressive'', or ``undetermined''
6272 Cumulative number of frames detected as bottom field first using multiple-frame detection.
6274 @item single.progressive
6275 Cumulative number of frames detected as progressive using single-frame detection.
6277 @item multiple.progressive
6278 Cumulative number of frames detected as progressive using multiple-frame detection.
6280 @item single.undetermined
6281 Cumulative number of frames that could not be classified using single-frame detection.
6283 @item multiple.undetermined
6284 Cumulative number of frames that could not be classified using multiple-frame detection.
6286 @item repeated.current_frame
6287 Which field in the current frame is repeated from the last. One of ``neither'', ``top'', or ``bottom''.
6289 @item repeated.neither
6290 Cumulative number of frames with no repeated field.
6293 Cumulative number of frames with the top field repeated from the previous frame's top field.
6295 @item repeated.bottom
6296 Cumulative number of frames with the bottom field repeated from the previous frame's bottom field.
6299 The filter accepts the following options:
6303 Set interlacing threshold.
6305 Set progressive threshold.
6307 Threshold for repeated field detection.
6309 Number of frames after which a given frame's contribution to the
6310 statistics is halved (i.e., it contributes only 0.5 to it's
6311 classification). The default of 0 means that all frames seen are given
6312 full weight of 1.0 forever.
6313 @item analyze_interlaced_flag
6314 When this is not 0 then idet will use the specified number of frames to determine
6315 if the interlaced flag is accurate, it will not count undetermined frames.
6316 If the flag is found to be accurate it will be used without any further
6317 computations, if it is found to be inaccurate it will be cleared without any
6318 further computations. This allows inserting the idet filter as a low computational
6319 method to clean up the interlaced flag
6324 Deinterleave or interleave fields.
6326 This filter allows one to process interlaced images fields without
6327 deinterlacing them. Deinterleaving splits the input frame into 2
6328 fields (so called half pictures). Odd lines are moved to the top
6329 half of the output image, even lines to the bottom half.
6330 You can process (filter) them independently and then re-interleave them.
6332 The filter accepts the following options:
6336 @item chroma_mode, c
6338 Available values for @var{luma_mode}, @var{chroma_mode} and
6339 @var{alpha_mode} are:
6345 @item deinterleave, d
6346 Deinterleave fields, placing one above the other.
6349 Interleave fields. Reverse the effect of deinterleaving.
6351 Default value is @code{none}.
6354 @item chroma_swap, cs
6355 @item alpha_swap, as
6356 Swap luma/chroma/alpha fields. Exchange even & odd lines. Default value is @code{0}.
6361 Simple interlacing filter from progressive contents. This interleaves upper (or
6362 lower) lines from odd frames with lower (or upper) lines from even frames,
6363 halving the frame rate and preserving image height.
6366 Original Original New Frame
6367 Frame 'j' Frame 'j+1' (tff)
6368 ========== =========== ==================
6369 Line 0 --------------------> Frame 'j' Line 0
6370 Line 1 Line 1 ----> Frame 'j+1' Line 1
6371 Line 2 ---------------------> Frame 'j' Line 2
6372 Line 3 Line 3 ----> Frame 'j+1' Line 3
6374 New Frame + 1 will be generated by Frame 'j+2' and Frame 'j+3' and so on
6377 It accepts the following optional parameters:
6381 This determines whether the interlaced frame is taken from the even
6382 (tff - default) or odd (bff) lines of the progressive frame.
6385 Enable (default) or disable the vertical lowpass filter to avoid twitter
6386 interlacing and reduce moire patterns.
6391 Deinterlace input video by applying Donald Graft's adaptive kernel
6392 deinterling. Work on interlaced parts of a video to produce
6395 The description of the accepted parameters follows.
6399 Set the threshold which affects the filter's tolerance when
6400 determining if a pixel line must be processed. It must be an integer
6401 in the range [0,255] and defaults to 10. A value of 0 will result in
6402 applying the process on every pixels.
6405 Paint pixels exceeding the threshold value to white if set to 1.
6409 Set the fields order. Swap fields if set to 1, leave fields alone if
6413 Enable additional sharpening if set to 1. Default is 0.
6416 Enable twoway sharpening if set to 1. Default is 0.
6419 @subsection Examples
6423 Apply default values:
6425 kerndeint=thresh=10:map=0:order=0:sharp=0:twoway=0
6429 Enable additional sharpening:
6435 Paint processed pixels in white:
6441 @section lenscorrection
6443 Correct radial lens distortion
6445 This filter can be used to correct for radial distortion as can result from the use
6446 of wide angle lenses, and thereby re-rectify the image. To find the right parameters
6447 one can use tools available for example as part of opencv or simply trial-and-error.
6448 To use opencv use the calibration sample (under samples/cpp) from the opencv sources
6449 and extract the k1 and k2 coefficients from the resulting matrix.
6451 Note that effectively the same filter is available in the open-source tools Krita and
6452 Digikam from the KDE project.
6454 In contrast to the @ref{vignette} filter, which can also be used to compensate lens errors,
6455 this filter corrects the distortion of the image, whereas @ref{vignette} corrects the
6456 brightness distribution, so you may want to use both filters together in certain
6457 cases, though you will have to take care of ordering, i.e. whether vignetting should
6458 be applied before or after lens correction.
6462 The filter accepts the following options:
6466 Relative x-coordinate of the focal point of the image, and thereby the center of the
6467 distortion. This value has a range [0,1] and is expressed as fractions of the image
6470 Relative y-coordinate of the focal point of the image, and thereby the center of the
6471 distortion. This value has a range [0,1] and is expressed as fractions of the image
6474 Coefficient of the quadratic correction term. 0.5 means no correction.
6476 Coefficient of the double quadratic correction term. 0.5 means no correction.
6479 The formula that generates the correction is:
6481 @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)
6483 where @var{r_0} is halve of the image diagonal and @var{r_src} and @var{r_tgt} are the
6484 distances from the focal point in the source and target images, respectively.
6489 Apply a 3D LUT to an input video.
6491 The filter accepts the following options:
6495 Set the 3D LUT file name.
6497 Currently supported formats:
6509 Select interpolation mode.
6511 Available values are:
6515 Use values from the nearest defined point.
6517 Interpolate values using the 8 points defining a cube.
6519 Interpolate values using a tetrahedron.
6523 @section lut, lutrgb, lutyuv
6525 Compute a look-up table for binding each pixel component input value
6526 to an output value, and apply it to the input video.
6528 @var{lutyuv} applies a lookup table to a YUV input video, @var{lutrgb}
6529 to an RGB input video.
6531 These filters accept the following parameters:
6534 set first pixel component expression
6536 set second pixel component expression
6538 set third pixel component expression
6540 set fourth pixel component expression, corresponds to the alpha component
6543 set red component expression
6545 set green component expression
6547 set blue component expression
6549 alpha component expression
6552 set Y/luminance component expression
6554 set U/Cb component expression
6556 set V/Cr component expression
6559 Each of them specifies the expression to use for computing the lookup table for
6560 the corresponding pixel component values.
6562 The exact component associated to each of the @var{c*} options depends on the
6565 The @var{lut} filter requires either YUV or RGB pixel formats in input,
6566 @var{lutrgb} requires RGB pixel formats in input, and @var{lutyuv} requires YUV.
6568 The expressions can contain the following constants and functions:
6573 The input width and height.
6576 The input value for the pixel component.
6579 The input value, clipped to the @var{minval}-@var{maxval} range.
6582 The maximum value for the pixel component.
6585 The minimum value for the pixel component.
6588 The negated value for the pixel component value, clipped to the
6589 @var{minval}-@var{maxval} range; it corresponds to the expression
6590 "maxval-clipval+minval".
6593 The computed value in @var{val}, clipped to the
6594 @var{minval}-@var{maxval} range.
6596 @item gammaval(gamma)
6597 The computed gamma correction value of the pixel component value,
6598 clipped to the @var{minval}-@var{maxval} range. It corresponds to the
6600 "pow((clipval-minval)/(maxval-minval)\,@var{gamma})*(maxval-minval)+minval"
6604 All expressions default to "val".
6606 @subsection Examples
6612 lutrgb="r=maxval+minval-val:g=maxval+minval-val:b=maxval+minval-val"
6613 lutyuv="y=maxval+minval-val:u=maxval+minval-val:v=maxval+minval-val"
6616 The above is the same as:
6618 lutrgb="r=negval:g=negval:b=negval"
6619 lutyuv="y=negval:u=negval:v=negval"
6629 Remove chroma components, turning the video into a graytone image:
6631 lutyuv="u=128:v=128"
6635 Apply a luma burning effect:
6641 Remove green and blue components:
6647 Set a constant alpha channel value on input:
6649 format=rgba,lutrgb=a="maxval-minval/2"
6653 Correct luminance gamma by a factor of 0.5:
6655 lutyuv=y=gammaval(0.5)
6659 Discard least significant bits of luma:
6661 lutyuv=y='bitand(val, 128+64+32)'
6665 @section mergeplanes
6667 Merge color channel components from several video streams.
6669 The filter accepts up to 4 input streams, and merge selected input
6670 planes to the output video.
6672 This filter accepts the following options:
6675 Set input to output plane mapping. Default is @code{0}.
6677 The mappings is specified as a bitmap. It should be specified as a
6678 hexadecimal number in the form 0xAa[Bb[Cc[Dd]]]. 'Aa' describes the
6679 mapping for the first plane of the output stream. 'A' sets the number of
6680 the input stream to use (from 0 to 3), and 'a' the plane number of the
6681 corresponding input to use (from 0 to 3). The rest of the mappings is
6682 similar, 'Bb' describes the mapping for the output stream second
6683 plane, 'Cc' describes the mapping for the output stream third plane and
6684 'Dd' describes the mapping for the output stream fourth plane.
6687 Set output pixel format. Default is @code{yuva444p}.
6690 @subsection Examples
6694 Merge three gray video streams of same width and height into single video stream:
6696 [a0][a1][a2]mergeplanes=0x001020:yuv444p
6700 Merge 1st yuv444p stream and 2nd gray video stream into yuva444p video stream:
6702 [a0][a1]mergeplanes=0x00010210:yuva444p
6706 Swap Y and A plane in yuva444p stream:
6708 format=yuva444p,mergeplanes=0x03010200:yuva444p
6712 Swap U and V plane in yuv420p stream:
6714 format=yuv420p,mergeplanes=0x000201:yuv420p
6718 Cast a rgb24 clip to yuv444p:
6720 format=rgb24,mergeplanes=0x000102:yuv444p
6726 Apply motion-compensation deinterlacing.
6728 It needs one field per frame as input and must thus be used together
6729 with yadif=1/3 or equivalent.
6731 This filter accepts the following options:
6734 Set the deinterlacing mode.
6736 It accepts one of the following values:
6741 use iterative motion estimation
6743 like @samp{slow}, but use multiple reference frames.
6745 Default value is @samp{fast}.
6748 Set the picture field parity assumed for the input video. It must be
6749 one of the following values:
6753 assume top field first
6755 assume bottom field first
6758 Default value is @samp{bff}.
6761 Set per-block quantization parameter (QP) used by the internal
6764 Higher values should result in a smoother motion vector field but less
6765 optimal individual vectors. Default value is 1.
6770 Drop frames that do not differ greatly from the previous frame in
6771 order to reduce frame rate.
6773 The main use of this filter is for very-low-bitrate encoding
6774 (e.g. streaming over dialup modem), but it could in theory be used for
6775 fixing movies that were inverse-telecined incorrectly.
6777 A description of the accepted options follows.
6781 Set the maximum number of consecutive frames which can be dropped (if
6782 positive), or the minimum interval between dropped frames (if
6783 negative). If the value is 0, the frame is dropped unregarding the
6784 number of previous sequentially dropped frames.
6791 Set the dropping threshold values.
6793 Values for @option{hi} and @option{lo} are for 8x8 pixel blocks and
6794 represent actual pixel value differences, so a threshold of 64
6795 corresponds to 1 unit of difference for each pixel, or the same spread
6796 out differently over the block.
6798 A frame is a candidate for dropping if no 8x8 blocks differ by more
6799 than a threshold of @option{hi}, and if no more than @option{frac} blocks (1
6800 meaning the whole image) differ by more than a threshold of @option{lo}.
6802 Default value for @option{hi} is 64*12, default value for @option{lo} is
6803 64*5, and default value for @option{frac} is 0.33.
6811 It accepts an integer in input; if non-zero it negates the
6812 alpha component (if available). The default value in input is 0.
6816 Force libavfilter not to use any of the specified pixel formats for the
6817 input to the next filter.
6819 It accepts the following parameters:
6823 A '|'-separated list of pixel format names, such as
6824 apix_fmts=yuv420p|monow|rgb24".
6828 @subsection Examples
6832 Force libavfilter to use a format different from @var{yuv420p} for the
6833 input to the vflip filter:
6835 noformat=pix_fmts=yuv420p,vflip
6839 Convert the input video to any of the formats not contained in the list:
6841 noformat=yuv420p|yuv444p|yuv410p
6847 Add noise on video input frame.
6849 The filter accepts the following options:
6857 Set noise seed for specific pixel component or all pixel components in case
6858 of @var{all_seed}. Default value is @code{123457}.
6860 @item all_strength, alls
6861 @item c0_strength, c0s
6862 @item c1_strength, c1s
6863 @item c2_strength, c2s
6864 @item c3_strength, c3s
6865 Set noise strength for specific pixel component or all pixel components in case
6866 @var{all_strength}. Default value is @code{0}. Allowed range is [0, 100].
6868 @item all_flags, allf
6873 Set pixel component flags or set flags for all components if @var{all_flags}.
6874 Available values for component flags are:
6877 averaged temporal noise (smoother)
6879 mix random noise with a (semi)regular pattern
6881 temporal noise (noise pattern changes between frames)
6883 uniform noise (gaussian otherwise)
6887 @subsection Examples
6889 Add temporal and uniform noise to input video:
6891 noise=alls=20:allf=t+u
6896 Pass the video source unchanged to the output.
6900 Apply a video transform using libopencv.
6902 To enable this filter, install the libopencv library and headers and
6903 configure FFmpeg with @code{--enable-libopencv}.
6905 It accepts the following parameters:
6910 The name of the libopencv filter to apply.
6913 The parameters to pass to the libopencv filter. If not specified, the default
6918 Refer to the official libopencv documentation for more precise
6920 @url{http://docs.opencv.org/master/modules/imgproc/doc/filtering.html}
6922 Several libopencv filters are supported; see the following subsections.
6927 Dilate an image by using a specific structuring element.
6928 It corresponds to the libopencv function @code{cvDilate}.
6930 It accepts the parameters: @var{struct_el}|@var{nb_iterations}.
6932 @var{struct_el} represents a structuring element, and has the syntax:
6933 @var{cols}x@var{rows}+@var{anchor_x}x@var{anchor_y}/@var{shape}
6935 @var{cols} and @var{rows} represent the number of columns and rows of
6936 the structuring element, @var{anchor_x} and @var{anchor_y} the anchor
6937 point, and @var{shape} the shape for the structuring element. @var{shape}
6938 must be "rect", "cross", "ellipse", or "custom".
6940 If the value for @var{shape} is "custom", it must be followed by a
6941 string of the form "=@var{filename}". The file with name
6942 @var{filename} is assumed to represent a binary image, with each
6943 printable character corresponding to a bright pixel. When a custom
6944 @var{shape} is used, @var{cols} and @var{rows} are ignored, the number
6945 or columns and rows of the read file are assumed instead.
6947 The default value for @var{struct_el} is "3x3+0x0/rect".
6949 @var{nb_iterations} specifies the number of times the transform is
6950 applied to the image, and defaults to 1.
6954 # Use the default values
6957 # Dilate using a structuring element with a 5x5 cross, iterating two times
6958 ocv=filter_name=dilate:filter_params=5x5+2x2/cross|2
6960 # Read the shape from the file diamond.shape, iterating two times.
6961 # The file diamond.shape may contain a pattern of characters like this
6967 # The specified columns and rows are ignored
6968 # but the anchor point coordinates are not
6969 ocv=dilate:0x0+2x2/custom=diamond.shape|2
6974 Erode an image by using a specific structuring element.
6975 It corresponds to the libopencv function @code{cvErode}.
6977 It accepts the parameters: @var{struct_el}:@var{nb_iterations},
6978 with the same syntax and semantics as the @ref{dilate} filter.
6982 Smooth the input video.
6984 The filter takes the following parameters:
6985 @var{type}|@var{param1}|@var{param2}|@var{param3}|@var{param4}.
6987 @var{type} is the type of smooth filter to apply, and must be one of
6988 the following values: "blur", "blur_no_scale", "median", "gaussian",
6989 or "bilateral". The default value is "gaussian".
6991 The meaning of @var{param1}, @var{param2}, @var{param3}, and @var{param4}
6992 depend on the smooth type. @var{param1} and
6993 @var{param2} accept integer positive values or 0. @var{param3} and
6994 @var{param4} accept floating point values.
6996 The default value for @var{param1} is 3. The default value for the
6997 other parameters is 0.
6999 These parameters correspond to the parameters assigned to the
7000 libopencv function @code{cvSmooth}.
7005 Overlay one video on top of another.
7007 It takes two inputs and has one output. The first input is the "main"
7008 video on which the second input is overlaid.
7010 It accepts the following parameters:
7012 A description of the accepted options follows.
7017 Set the expression for the x and y coordinates of the overlaid video
7018 on the main video. Default value is "0" for both expressions. In case
7019 the expression is invalid, it is set to a huge value (meaning that the
7020 overlay will not be displayed within the output visible area).
7023 The action to take when EOF is encountered on the secondary input; it accepts
7024 one of the following values:
7028 Repeat the last frame (the default).
7032 Pass the main input through.
7036 Set when the expressions for @option{x}, and @option{y} are evaluated.
7038 It accepts the following values:
7041 only evaluate expressions once during the filter initialization or
7042 when a command is processed
7045 evaluate expressions for each incoming frame
7048 Default value is @samp{frame}.
7051 If set to 1, force the output to terminate when the shortest input
7052 terminates. Default value is 0.
7055 Set the format for the output video.
7057 It accepts the following values:
7072 Default value is @samp{yuv420}.
7074 @item rgb @emph{(deprecated)}
7075 If set to 1, force the filter to accept inputs in the RGB
7076 color space. Default value is 0. This option is deprecated, use
7077 @option{format} instead.
7080 If set to 1, force the filter to draw the last overlay frame over the
7081 main input until the end of the stream. A value of 0 disables this
7082 behavior. Default value is 1.
7085 The @option{x}, and @option{y} expressions can contain the following
7091 The main input width and height.
7095 The overlay input width and height.
7099 The computed values for @var{x} and @var{y}. They are evaluated for
7104 horizontal and vertical chroma subsample values of the output
7105 format. For example for the pixel format "yuv422p" @var{hsub} is 2 and
7109 the number of input frame, starting from 0
7112 the position in the file of the input frame, NAN if unknown
7115 The timestamp, expressed in seconds. It's NAN if the input timestamp is unknown.
7119 Note that the @var{n}, @var{pos}, @var{t} variables are available only
7120 when evaluation is done @emph{per frame}, and will evaluate to NAN
7121 when @option{eval} is set to @samp{init}.
7123 Be aware that frames are taken from each input video in timestamp
7124 order, hence, if their initial timestamps differ, it is a good idea
7125 to pass the two inputs through a @var{setpts=PTS-STARTPTS} filter to
7126 have them begin in the same zero timestamp, as the example for
7127 the @var{movie} filter does.
7129 You can chain together more overlays but you should test the
7130 efficiency of such approach.
7132 @subsection Commands
7134 This filter supports the following commands:
7138 Modify the x and y of the overlay input.
7139 The command accepts the same syntax of the corresponding option.
7141 If the specified expression is not valid, it is kept at its current
7145 @subsection Examples
7149 Draw the overlay at 10 pixels from the bottom right corner of the main
7152 overlay=main_w-overlay_w-10:main_h-overlay_h-10
7155 Using named options the example above becomes:
7157 overlay=x=main_w-overlay_w-10:y=main_h-overlay_h-10
7161 Insert a transparent PNG logo in the bottom left corner of the input,
7162 using the @command{ffmpeg} tool with the @code{-filter_complex} option:
7164 ffmpeg -i input -i logo -filter_complex 'overlay=10:main_h-overlay_h-10' output
7168 Insert 2 different transparent PNG logos (second logo on bottom
7169 right corner) using the @command{ffmpeg} tool:
7171 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
7175 Add a transparent color layer on top of the main video; @code{WxH}
7176 must specify the size of the main input to the overlay filter:
7178 color=color=red@@.3:size=WxH [over]; [in][over] overlay [out]
7182 Play an original video and a filtered version (here with the deshake
7183 filter) side by side using the @command{ffplay} tool:
7185 ffplay input.avi -vf 'split[a][b]; [a]pad=iw*2:ih[src]; [b]deshake[filt]; [src][filt]overlay=w'
7188 The above command is the same as:
7190 ffplay input.avi -vf 'split[b], pad=iw*2[src], [b]deshake, [src]overlay=w'
7194 Make a sliding overlay appearing from the left to the right top part of the
7195 screen starting since time 2:
7197 overlay=x='if(gte(t,2), -w+(t-2)*20, NAN)':y=0
7201 Compose output by putting two input videos side to side:
7203 ffmpeg -i left.avi -i right.avi -filter_complex "
7204 nullsrc=size=200x100 [background];
7205 [0:v] setpts=PTS-STARTPTS, scale=100x100 [left];
7206 [1:v] setpts=PTS-STARTPTS, scale=100x100 [right];
7207 [background][left] overlay=shortest=1 [background+left];
7208 [background+left][right] overlay=shortest=1:x=100 [left+right]
7213 Mask 10-20 seconds of a video by applying the delogo filter to a section
7215 ffmpeg -i test.avi -codec:v:0 wmv2 -ar 11025 -b:v 9000k
7216 -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]'
7221 Chain several overlays in cascade:
7223 nullsrc=s=200x200 [bg];
7224 testsrc=s=100x100, split=4 [in0][in1][in2][in3];
7225 [in0] lutrgb=r=0, [bg] overlay=0:0 [mid0];
7226 [in1] lutrgb=g=0, [mid0] overlay=100:0 [mid1];
7227 [in2] lutrgb=b=0, [mid1] overlay=0:100 [mid2];
7228 [in3] null, [mid2] overlay=100:100 [out0]
7235 Apply Overcomplete Wavelet denoiser.
7237 The filter accepts the following options:
7243 Larger depth values will denoise lower frequency components more, but
7244 slow down filtering.
7246 Must be an int in the range 8-16, default is @code{8}.
7248 @item luma_strength, ls
7251 Must be a double value in the range 0-1000, default is @code{1.0}.
7253 @item chroma_strength, cs
7254 Set chroma strength.
7256 Must be a double value in the range 0-1000, default is @code{1.0}.
7261 Add paddings to the input image, and place the original input at the
7262 provided @var{x}, @var{y} coordinates.
7264 It accepts the following parameters:
7269 Specify an expression for the size of the output image with the
7270 paddings added. If the value for @var{width} or @var{height} is 0, the
7271 corresponding input size is used for the output.
7273 The @var{width} expression can reference the value set by the
7274 @var{height} expression, and vice versa.
7276 The default value of @var{width} and @var{height} is 0.
7280 Specify the offsets to place the input image at within the padded area,
7281 with respect to the top/left border of the output image.
7283 The @var{x} expression can reference the value set by the @var{y}
7284 expression, and vice versa.
7286 The default value of @var{x} and @var{y} is 0.
7289 Specify the color of the padded area. For the syntax of this option,
7290 check the "Color" section in the ffmpeg-utils manual.
7292 The default value of @var{color} is "black".
7295 The value for the @var{width}, @var{height}, @var{x}, and @var{y}
7296 options are expressions containing the following constants:
7301 The input video width and height.
7305 These are the same as @var{in_w} and @var{in_h}.
7309 The output width and height (the size of the padded area), as
7310 specified by the @var{width} and @var{height} expressions.
7314 These are the same as @var{out_w} and @var{out_h}.
7318 The x and y offsets as specified by the @var{x} and @var{y}
7319 expressions, or NAN if not yet specified.
7322 same as @var{iw} / @var{ih}
7325 input sample aspect ratio
7328 input display aspect ratio, it is the same as (@var{iw} / @var{ih}) * @var{sar}
7332 The horizontal and vertical chroma subsample values. For example for the
7333 pixel format "yuv422p" @var{hsub} is 2 and @var{vsub} is 1.
7336 @subsection Examples
7340 Add paddings with the color "violet" to the input video. The output video
7341 size is 640x480, and the top-left corner of the input video is placed at
7344 pad=640:480:0:40:violet
7347 The example above is equivalent to the following command:
7349 pad=width=640:height=480:x=0:y=40:color=violet
7353 Pad the input to get an output with dimensions increased by 3/2,
7354 and put the input video at the center of the padded area:
7356 pad="3/2*iw:3/2*ih:(ow-iw)/2:(oh-ih)/2"
7360 Pad the input to get a squared output with size equal to the maximum
7361 value between the input width and height, and put the input video at
7362 the center of the padded area:
7364 pad="max(iw\,ih):ow:(ow-iw)/2:(oh-ih)/2"
7368 Pad the input to get a final w/h ratio of 16:9:
7370 pad="ih*16/9:ih:(ow-iw)/2:(oh-ih)/2"
7374 In case of anamorphic video, in order to set the output display aspect
7375 correctly, it is necessary to use @var{sar} in the expression,
7376 according to the relation:
7378 (ih * X / ih) * sar = output_dar
7379 X = output_dar / sar
7382 Thus the previous example needs to be modified to:
7384 pad="ih*16/9/sar:ih:(ow-iw)/2:(oh-ih)/2"
7388 Double the output size and put the input video in the bottom-right
7389 corner of the output padded area:
7391 pad="2*iw:2*ih:ow-iw:oh-ih"
7398 Generate one palette for a whole video stream.
7400 It accepts the following options:
7404 Set the maximum number of colors to quantize in the palette.
7405 Note: the palette will still contain 256 colors; the unused palette entries
7408 @item reserve_transparent
7409 Create a palette of 255 colors maximum and reserve the last one for
7410 transparency. Reserving the transparency color is useful for GIF optimization.
7411 If not set, the maximum of colors in the palette will be 256. You probably want
7412 to disable this option for a standalone image.
7416 Set statistics mode.
7418 It accepts the following values:
7421 Compute full frame histograms.
7423 Compute histograms only for the part that differs from previous frame. This
7424 might be relevant to give more importance to the moving part of your input if
7425 the background is static.
7428 Default value is @var{full}.
7431 The filter also exports the frame metadata @code{lavfi.color_quant_ratio}
7432 (@code{nb_color_in / nb_color_out}) which you can use to evaluate the degree of
7433 color quantization of the palette. This information is also visible at
7434 @var{info} logging level.
7436 @subsection Examples
7440 Generate a representative palette of a given video using @command{ffmpeg}:
7442 ffmpeg -i input.mkv -vf palettegen palette.png
7448 Use a palette to downsample an input video stream.
7450 The filter takes two inputs: one video stream and a palette. The palette must
7451 be a 256 pixels image.
7453 It accepts the following options:
7457 Select dithering mode. Available algorithms are:
7460 Ordered 8x8 bayer dithering (deterministic)
7462 Dithering as defined by Paul Heckbert in 1982 (simple error diffusion).
7463 Note: this dithering is sometimes considered "wrong" and is included as a
7465 @item floyd_steinberg
7466 Floyd and Steingberg dithering (error diffusion)
7468 Frankie Sierra dithering v2 (error diffusion)
7470 Frankie Sierra dithering v2 "Lite" (error diffusion)
7473 Default is @var{sierra2_4a}.
7476 When @var{bayer} dithering is selected, this option defines the scale of the
7477 pattern (how much the crosshatch pattern is visible). A low value means more
7478 visible pattern for less banding, and higher value means less visible pattern
7479 at the cost of more banding.
7481 The option must be an integer value in the range [0,5]. Default is @var{2}.
7484 If set, define the zone to process
7488 Only the changing rectangle will be reprocessed. This is similar to GIF
7489 cropping/offsetting compression mechanism. This option can be useful for speed
7490 if only a part of the image is changing, and has use cases such as limiting the
7491 scope of the error diffusal @option{dither} to the rectangle that bounds the
7492 moving scene (it leads to more deterministic output if the scene doesn't change
7493 much, and as a result less moving noise and better GIF compression).
7496 Default is @var{none}.
7499 @subsection Examples
7503 Use a palette (generated for example with @ref{palettegen}) to encode a GIF
7504 using @command{ffmpeg}:
7506 ffmpeg -i input.mkv -i palette.png -lavfi paletteuse output.gif
7510 @section perspective
7512 Correct perspective of video not recorded perpendicular to the screen.
7514 A description of the accepted parameters follows.
7525 Set coordinates expression for top left, top right, bottom left and bottom right corners.
7526 Default values are @code{0:0:W:0:0:H:W:H} with which perspective will remain unchanged.
7527 If the @code{sense} option is set to @code{source}, then the specified points will be sent
7528 to the corners of the destination. If the @code{sense} option is set to @code{destination},
7529 then the corners of the source will be sent to the specified coordinates.
7531 The expressions can use the following variables:
7536 the width and height of video frame.
7540 Set interpolation for perspective correction.
7542 It accepts the following values:
7548 Default value is @samp{linear}.
7551 Set interpretation of coordinate options.
7553 It accepts the following values:
7557 Send point in the source specified by the given coordinates to
7558 the corners of the destination.
7560 @item 1, destination
7562 Send the corners of the source to the point in the destination specified
7563 by the given coordinates.
7565 Default value is @samp{source}.
7571 Delay interlaced video by one field time so that the field order changes.
7573 The intended use is to fix PAL movies that have been captured with the
7574 opposite field order to the film-to-video transfer.
7576 A description of the accepted parameters follows.
7582 It accepts the following values:
7585 Capture field order top-first, transfer bottom-first.
7586 Filter will delay the bottom field.
7589 Capture field order bottom-first, transfer top-first.
7590 Filter will delay the top field.
7593 Capture and transfer with the same field order. This mode only exists
7594 for the documentation of the other options to refer to, but if you
7595 actually select it, the filter will faithfully do nothing.
7598 Capture field order determined automatically by field flags, transfer
7600 Filter selects among @samp{t} and @samp{b} modes on a frame by frame
7601 basis using field flags. If no field information is available,
7602 then this works just like @samp{u}.
7605 Capture unknown or varying, transfer opposite.
7606 Filter selects among @samp{t} and @samp{b} on a frame by frame basis by
7607 analyzing the images and selecting the alternative that produces best
7608 match between the fields.
7611 Capture top-first, transfer unknown or varying.
7612 Filter selects among @samp{t} and @samp{p} using image analysis.
7615 Capture bottom-first, transfer unknown or varying.
7616 Filter selects among @samp{b} and @samp{p} using image analysis.
7619 Capture determined by field flags, transfer unknown or varying.
7620 Filter selects among @samp{t}, @samp{b} and @samp{p} using field flags and
7621 image analysis. If no field information is available, then this works just
7622 like @samp{U}. This is the default mode.
7625 Both capture and transfer unknown or varying.
7626 Filter selects among @samp{t}, @samp{b} and @samp{p} using image analysis only.
7630 @section pixdesctest
7632 Pixel format descriptor test filter, mainly useful for internal
7633 testing. The output video should be equal to the input video.
7637 format=monow, pixdesctest
7640 can be used to test the monowhite pixel format descriptor definition.
7644 Enable the specified chain of postprocessing subfilters using libpostproc. This
7645 library should be automatically selected with a GPL build (@code{--enable-gpl}).
7646 Subfilters must be separated by '/' and can be disabled by prepending a '-'.
7647 Each subfilter and some options have a short and a long name that can be used
7648 interchangeably, i.e. dr/dering are the same.
7650 The filters accept the following options:
7654 Set postprocessing subfilters string.
7657 All subfilters share common options to determine their scope:
7661 Honor the quality commands for this subfilter.
7664 Do chrominance filtering, too (default).
7667 Do luminance filtering only (no chrominance).
7670 Do chrominance filtering only (no luminance).
7673 These options can be appended after the subfilter name, separated by a '|'.
7675 Available subfilters are:
7678 @item hb/hdeblock[|difference[|flatness]]
7679 Horizontal deblocking filter
7682 Difference factor where higher values mean more deblocking (default: @code{32}).
7684 Flatness threshold where lower values mean more deblocking (default: @code{39}).
7687 @item vb/vdeblock[|difference[|flatness]]
7688 Vertical deblocking filter
7691 Difference factor where higher values mean more deblocking (default: @code{32}).
7693 Flatness threshold where lower values mean more deblocking (default: @code{39}).
7696 @item ha/hadeblock[|difference[|flatness]]
7697 Accurate horizontal deblocking filter
7700 Difference factor where higher values mean more deblocking (default: @code{32}).
7702 Flatness threshold where lower values mean more deblocking (default: @code{39}).
7705 @item va/vadeblock[|difference[|flatness]]
7706 Accurate vertical deblocking filter
7709 Difference factor where higher values mean more deblocking (default: @code{32}).
7711 Flatness threshold where lower values mean more deblocking (default: @code{39}).
7715 The horizontal and vertical deblocking filters share the difference and
7716 flatness values so you cannot set different horizontal and vertical
7721 Experimental horizontal deblocking filter
7724 Experimental vertical deblocking filter
7729 @item tn/tmpnoise[|threshold1[|threshold2[|threshold3]]], temporal noise reducer
7732 larger -> stronger filtering
7734 larger -> stronger filtering
7736 larger -> stronger filtering
7739 @item al/autolevels[:f/fullyrange], automatic brightness / contrast correction
7742 Stretch luminance to @code{0-255}.
7745 @item lb/linblenddeint
7746 Linear blend deinterlacing filter that deinterlaces the given block by
7747 filtering all lines with a @code{(1 2 1)} filter.
7749 @item li/linipoldeint
7750 Linear interpolating deinterlacing filter that deinterlaces the given block by
7751 linearly interpolating every second line.
7753 @item ci/cubicipoldeint
7754 Cubic interpolating deinterlacing filter deinterlaces the given block by
7755 cubically interpolating every second line.
7757 @item md/mediandeint
7758 Median deinterlacing filter that deinterlaces the given block by applying a
7759 median filter to every second line.
7761 @item fd/ffmpegdeint
7762 FFmpeg deinterlacing filter that deinterlaces the given block by filtering every
7763 second line with a @code{(-1 4 2 4 -1)} filter.
7766 Vertically applied FIR lowpass deinterlacing filter that deinterlaces the given
7767 block by filtering all lines with a @code{(-1 2 6 2 -1)} filter.
7769 @item fq/forceQuant[|quantizer]
7770 Overrides the quantizer table from the input with the constant quantizer you
7778 Default pp filter combination (@code{hb|a,vb|a,dr|a})
7781 Fast pp filter combination (@code{h1|a,v1|a,dr|a})
7784 High quality pp filter combination (@code{ha|a|128|7,va|a,dr|a})
7787 @subsection Examples
7791 Apply horizontal and vertical deblocking, deringing and automatic
7792 brightness/contrast:
7798 Apply default filters without brightness/contrast correction:
7804 Apply default filters and temporal denoiser:
7806 pp=default/tmpnoise|1|2|3
7810 Apply deblocking on luminance only, and switch vertical deblocking on or off
7811 automatically depending on available CPU time:
7818 Apply Postprocessing filter 7. It is variant of the @ref{spp} filter,
7819 similar to spp = 6 with 7 point DCT, where only the center sample is
7822 The filter accepts the following options:
7826 Force a constant quantization parameter. It accepts an integer in range
7827 0 to 63. If not set, the filter will use the QP from the video stream
7831 Set thresholding mode. Available modes are:
7835 Set hard thresholding.
7837 Set soft thresholding (better de-ringing effect, but likely blurrier).
7839 Set medium thresholding (good results, default).
7845 Obtain the average, maximum and minimum PSNR (Peak Signal to Noise
7846 Ratio) between two input videos.
7848 This filter takes in input two input videos, the first input is
7849 considered the "main" source and is passed unchanged to the
7850 output. The second input is used as a "reference" video for computing
7853 Both video inputs must have the same resolution and pixel format for
7854 this filter to work correctly. Also it assumes that both inputs
7855 have the same number of frames, which are compared one by one.
7857 The obtained average PSNR is printed through the logging system.
7859 The filter stores the accumulated MSE (mean squared error) of each
7860 frame, and at the end of the processing it is averaged across all frames
7861 equally, and the following formula is applied to obtain the PSNR:
7864 PSNR = 10*log10(MAX^2/MSE)
7867 Where MAX is the average of the maximum values of each component of the
7870 The description of the accepted parameters follows.
7874 If specified the filter will use the named file to save the PSNR of
7875 each individual frame.
7878 The file printed if @var{stats_file} is selected, contains a sequence of
7879 key/value pairs of the form @var{key}:@var{value} for each compared
7882 A description of each shown parameter follows:
7886 sequential number of the input frame, starting from 1
7889 Mean Square Error pixel-by-pixel average difference of the compared
7890 frames, averaged over all the image components.
7892 @item mse_y, mse_u, mse_v, mse_r, mse_g, mse_g, mse_a
7893 Mean Square Error pixel-by-pixel average difference of the compared
7894 frames for the component specified by the suffix.
7896 @item psnr_y, psnr_u, psnr_v, psnr_r, psnr_g, psnr_b, psnr_a
7897 Peak Signal to Noise ratio of the compared frames for the component
7898 specified by the suffix.
7903 movie=ref_movie.mpg, setpts=PTS-STARTPTS [main];
7904 [main][ref] psnr="stats_file=stats.log" [out]
7907 On this example the input file being processed is compared with the
7908 reference file @file{ref_movie.mpg}. The PSNR of each individual frame
7909 is stored in @file{stats.log}.
7914 Pulldown reversal (inverse telecine) filter, capable of handling mixed
7915 hard-telecine, 24000/1001 fps progressive, and 30000/1001 fps progressive
7918 The pullup filter is designed to take advantage of future context in making
7919 its decisions. This filter is stateless in the sense that it does not lock
7920 onto a pattern to follow, but it instead looks forward to the following
7921 fields in order to identify matches and rebuild progressive frames.
7923 To produce content with an even framerate, insert the fps filter after
7924 pullup, use @code{fps=24000/1001} if the input frame rate is 29.97fps,
7925 @code{fps=24} for 30fps and the (rare) telecined 25fps input.
7927 The filter accepts the following options:
7934 These options set the amount of "junk" to ignore at the left, right, top, and
7935 bottom of the image, respectively. Left and right are in units of 8 pixels,
7936 while top and bottom are in units of 2 lines.
7937 The default is 8 pixels on each side.
7940 Set the strict breaks. Setting this option to 1 will reduce the chances of
7941 filter generating an occasional mismatched frame, but it may also cause an
7942 excessive number of frames to be dropped during high motion sequences.
7943 Conversely, setting it to -1 will make filter match fields more easily.
7944 This may help processing of video where there is slight blurring between
7945 the fields, but may also cause there to be interlaced frames in the output.
7946 Default value is @code{0}.
7949 Set the metric plane to use. It accepts the following values:
7955 Use chroma blue plane.
7958 Use chroma red plane.
7961 This option may be set to use chroma plane instead of the default luma plane
7962 for doing filter's computations. This may improve accuracy on very clean
7963 source material, but more likely will decrease accuracy, especially if there
7964 is chroma noise (rainbow effect) or any grayscale video.
7965 The main purpose of setting @option{mp} to a chroma plane is to reduce CPU
7966 load and make pullup usable in realtime on slow machines.
7969 For best results (without duplicated frames in the output file) it is
7970 necessary to change the output frame rate. For example, to inverse
7971 telecine NTSC input:
7973 ffmpeg -i input -vf pullup -r 24000/1001 ...
7978 Change video quantization parameters (QP).
7980 The filter accepts the following option:
7984 Set expression for quantization parameter.
7987 The expression is evaluated through the eval API and can contain, among others,
7988 the following constants:
7992 1 if index is not 129, 0 otherwise.
7995 Sequentional index starting from -129 to 128.
7998 @subsection Examples
8008 @section removegrain
8010 The removegrain filter is a spatial denoiser for progressive video.
8014 Set mode for the first plane.
8017 Set mode for the second plane.
8020 Set mode for the third plane.
8023 Set mode for the fourth plane.
8026 Range of mode is from 0 to 24. Description of each mode follows:
8030 Leave input plane unchanged. Default.
8033 Clips the pixel with the minimum and maximum of the 8 neighbour pixels.
8036 Clips the pixel with the second minimum and maximum of the 8 neighbour pixels.
8039 Clips the pixel with the third minimum and maximum of the 8 neighbour pixels.
8042 Clips the pixel with the fourth minimum and maximum of the 8 neighbour pixels.
8043 This is equivalent to a median filter.
8046 Line-sensitive clipping giving the minimal change.
8049 Line-sensitive clipping, intermediate.
8052 Line-sensitive clipping, intermediate.
8055 Line-sensitive clipping, intermediate.
8058 Line-sensitive clipping on a line where the neighbours pixels are the closest.
8061 Replaces the target pixel with the closest neighbour.
8064 [1 2 1] horizontal and vertical kernel blur.
8070 Bob mode, interpolates top field from the line where the neighbours
8071 pixels are the closest.
8074 Bob mode, interpolates bottom field from the line where the neighbours
8075 pixels are the closest.
8078 Bob mode, interpolates top field. Same as 13 but with a more complicated
8079 interpolation formula.
8082 Bob mode, interpolates bottom field. Same as 14 but with a more complicated
8083 interpolation formula.
8086 Clips the pixel with the minimum and maximum of respectively the maximum and
8087 minimum of each pair of opposite neighbour pixels.
8090 Line-sensitive clipping using opposite neighbours whose greatest distance from
8091 the current pixel is minimal.
8094 Replaces the pixel with the average of its 8 neighbours.
8097 Averages the 9 pixels ([1 1 1] horizontal and vertical blur).
8100 Clips pixels using the averages of opposite neighbour.
8103 Same as mode 21 but simpler and faster.
8106 Small edge and halo removal, but reputed useless.
8114 Suppress a TV station logo, using an image file to determine which
8115 pixels comprise the logo. It works by filling in the pixels that
8116 comprise the logo with neighboring pixels.
8118 The filter accepts the following options:
8122 Set the filter bitmap file, which can be any image format supported by
8123 libavformat. The width and height of the image file must match those of the
8124 video stream being processed.
8127 Pixels in the provided bitmap image with a value of zero are not
8128 considered part of the logo, non-zero pixels are considered part of
8129 the logo. If you use white (255) for the logo and black (0) for the
8130 rest, you will be safe. For making the filter bitmap, it is
8131 recommended to take a screen capture of a black frame with the logo
8132 visible, and then using a threshold filter followed by the erode
8133 filter once or twice.
8135 If needed, little splotches can be fixed manually. Remember that if
8136 logo pixels are not covered, the filter quality will be much
8137 reduced. Marking too many pixels as part of the logo does not hurt as
8138 much, but it will increase the amount of blurring needed to cover over
8139 the image and will destroy more information than necessary, and extra
8140 pixels will slow things down on a large logo.
8142 @section repeatfields
8144 This filter uses the repeat_field flag from the Video ES headers and hard repeats
8145 fields based on its value.
8149 Rotate video by an arbitrary angle expressed in radians.
8151 The filter accepts the following options:
8153 A description of the optional parameters follows.
8156 Set an expression for the angle by which to rotate the input video
8157 clockwise, expressed as a number of radians. A negative value will
8158 result in a counter-clockwise rotation. By default it is set to "0".
8160 This expression is evaluated for each frame.
8163 Set the output width expression, default value is "iw".
8164 This expression is evaluated just once during configuration.
8167 Set the output height expression, default value is "ih".
8168 This expression is evaluated just once during configuration.
8171 Enable bilinear interpolation if set to 1, a value of 0 disables
8172 it. Default value is 1.
8175 Set the color used to fill the output area not covered by the rotated
8176 image. For the general syntax of this option, check the "Color" section in the
8177 ffmpeg-utils manual. If the special value "none" is selected then no
8178 background is printed (useful for example if the background is never shown).
8180 Default value is "black".
8183 The expressions for the angle and the output size can contain the
8184 following constants and functions:
8188 sequential number of the input frame, starting from 0. It is always NAN
8189 before the first frame is filtered.
8192 time in seconds of the input frame, it is set to 0 when the filter is
8193 configured. It is always NAN before the first frame is filtered.
8197 horizontal and vertical chroma subsample values. For example for the
8198 pixel format "yuv422p" @var{hsub} is 2 and @var{vsub} is 1.
8202 the input video width and height
8206 the output width and height, that is the size of the padded area as
8207 specified by the @var{width} and @var{height} expressions
8211 the minimal width/height required for completely containing the input
8212 video rotated by @var{a} radians.
8214 These are only available when computing the @option{out_w} and
8215 @option{out_h} expressions.
8218 @subsection Examples
8222 Rotate the input by PI/6 radians clockwise:
8228 Rotate the input by PI/6 radians counter-clockwise:
8234 Rotate the input by 45 degrees clockwise:
8240 Apply a constant rotation with period T, starting from an angle of PI/3:
8242 rotate=PI/3+2*PI*t/T
8246 Make the input video rotation oscillating with a period of T
8247 seconds and an amplitude of A radians:
8249 rotate=A*sin(2*PI/T*t)
8253 Rotate the video, output size is chosen so that the whole rotating
8254 input video is always completely contained in the output:
8256 rotate='2*PI*t:ow=hypot(iw,ih):oh=ow'
8260 Rotate the video, reduce the output size so that no background is ever
8263 rotate=2*PI*t:ow='min(iw,ih)/sqrt(2)':oh=ow:c=none
8267 @subsection Commands
8269 The filter supports the following commands:
8273 Set the angle expression.
8274 The command accepts the same syntax of the corresponding option.
8276 If the specified expression is not valid, it is kept at its current
8282 Apply Shape Adaptive Blur.
8284 The filter accepts the following options:
8287 @item luma_radius, lr
8288 Set luma blur filter strength, must be a value in range 0.1-4.0, default
8289 value is 1.0. A greater value will result in a more blurred image, and
8290 in slower processing.
8292 @item luma_pre_filter_radius, lpfr
8293 Set luma pre-filter radius, must be a value in the 0.1-2.0 range, default
8296 @item luma_strength, ls
8297 Set luma maximum difference between pixels to still be considered, must
8298 be a value in the 0.1-100.0 range, default value is 1.0.
8300 @item chroma_radius, cr
8301 Set chroma blur filter strength, must be a value in range 0.1-4.0. A
8302 greater value will result in a more blurred image, and in slower
8305 @item chroma_pre_filter_radius, cpfr
8306 Set chroma pre-filter radius, must be a value in the 0.1-2.0 range.
8308 @item chroma_strength, cs
8309 Set chroma maximum difference between pixels to still be considered,
8310 must be a value in the 0.1-100.0 range.
8313 Each chroma option value, if not explicitly specified, is set to the
8314 corresponding luma option value.
8319 Scale (resize) the input video, using the libswscale library.
8321 The scale filter forces the output display aspect ratio to be the same
8322 of the input, by changing the output sample aspect ratio.
8324 If the input image format is different from the format requested by
8325 the next filter, the scale filter will convert the input to the
8329 The filter accepts the following options, or any of the options
8330 supported by the libswscale scaler.
8332 See @ref{scaler_options,,the ffmpeg-scaler manual,ffmpeg-scaler} for
8333 the complete list of scaler options.
8338 Set the output video dimension expression. Default value is the input
8341 If the value is 0, the input width is used for the output.
8343 If one of the values is -1, the scale filter will use a value that
8344 maintains the aspect ratio of the input image, calculated from the
8345 other specified dimension. If both of them are -1, the input size is
8348 If one of the values is -n with n > 1, the scale filter will also use a value
8349 that maintains the aspect ratio of the input image, calculated from the other
8350 specified dimension. After that it will, however, make sure that the calculated
8351 dimension is divisible by n and adjust the value if necessary.
8353 See below for the list of accepted constants for use in the dimension
8357 Set the interlacing mode. It accepts the following values:
8361 Force interlaced aware scaling.
8364 Do not apply interlaced scaling.
8367 Select interlaced aware scaling depending on whether the source frames
8368 are flagged as interlaced or not.
8371 Default value is @samp{0}.
8374 Set libswscale scaling flags. See
8375 @ref{sws_flags,,the ffmpeg-scaler manual,ffmpeg-scaler} for the
8376 complete list of values. If not explicitly specified the filter applies
8380 Set the video size. For the syntax of this option, check the
8381 @ref{video size syntax,,"Video size" section in the ffmpeg-utils manual,ffmpeg-utils}.
8383 @item in_color_matrix
8384 @item out_color_matrix
8385 Set in/output YCbCr color space type.
8387 This allows the autodetected value to be overridden as well as allows forcing
8388 a specific value used for the output and encoder.
8390 If not specified, the color space type depends on the pixel format.
8396 Choose automatically.
8399 Format conforming to International Telecommunication Union (ITU)
8400 Recommendation BT.709.
8403 Set color space conforming to the United States Federal Communications
8404 Commission (FCC) Code of Federal Regulations (CFR) Title 47 (2003) 73.682 (a).
8407 Set color space conforming to:
8411 ITU Radiocommunication Sector (ITU-R) Recommendation BT.601
8414 ITU-R Rec. BT.470-6 (1998) Systems B, B1, and G
8417 Society of Motion Picture and Television Engineers (SMPTE) ST 170:2004
8422 Set color space conforming to SMPTE ST 240:1999.
8427 Set in/output YCbCr sample range.
8429 This allows the autodetected value to be overridden as well as allows forcing
8430 a specific value used for the output and encoder. If not specified, the
8431 range depends on the pixel format. Possible values:
8435 Choose automatically.
8438 Set full range (0-255 in case of 8-bit luma).
8441 Set "MPEG" range (16-235 in case of 8-bit luma).
8444 @item force_original_aspect_ratio
8445 Enable decreasing or increasing output video width or height if necessary to
8446 keep the original aspect ratio. Possible values:
8450 Scale the video as specified and disable this feature.
8453 The output video dimensions will automatically be decreased if needed.
8456 The output video dimensions will automatically be increased if needed.
8460 One useful instance of this option is that when you know a specific device's
8461 maximum allowed resolution, you can use this to limit the output video to
8462 that, while retaining the aspect ratio. For example, device A allows
8463 1280x720 playback, and your video is 1920x800. Using this option (set it to
8464 decrease) and specifying 1280x720 to the command line makes the output
8467 Please note that this is a different thing than specifying -1 for @option{w}
8468 or @option{h}, you still need to specify the output resolution for this option
8473 The values of the @option{w} and @option{h} options are expressions
8474 containing the following constants:
8479 The input width and height
8483 These are the same as @var{in_w} and @var{in_h}.
8487 The output (scaled) width and height
8491 These are the same as @var{out_w} and @var{out_h}
8494 The same as @var{iw} / @var{ih}
8497 input sample aspect ratio
8500 The input display aspect ratio. Calculated from @code{(iw / ih) * sar}.
8504 horizontal and vertical input chroma subsample values. For example for the
8505 pixel format "yuv422p" @var{hsub} is 2 and @var{vsub} is 1.
8509 horizontal and vertical output chroma subsample values. For example for the
8510 pixel format "yuv422p" @var{hsub} is 2 and @var{vsub} is 1.
8513 @subsection Examples
8517 Scale the input video to a size of 200x100
8522 This is equivalent to:
8533 Specify a size abbreviation for the output size:
8538 which can also be written as:
8544 Scale the input to 2x:
8550 The above is the same as:
8556 Scale the input to 2x with forced interlaced scaling:
8558 scale=2*iw:2*ih:interl=1
8562 Scale the input to half size:
8568 Increase the width, and set the height to the same size:
8581 Increase the height, and set the width to 3/2 of the height:
8583 scale=w=3/2*oh:h=3/5*ih
8587 Increase the size, making the size a multiple of the chroma
8590 scale="trunc(3/2*iw/hsub)*hsub:trunc(3/2*ih/vsub)*vsub"
8594 Increase the width to a maximum of 500 pixels,
8595 keeping the same aspect ratio as the input:
8597 scale=w='min(500\, iw*3/2):h=-1'
8601 @section separatefields
8603 The @code{separatefields} takes a frame-based video input and splits
8604 each frame into its components fields, producing a new half height clip
8605 with twice the frame rate and twice the frame count.
8607 This filter use field-dominance information in frame to decide which
8608 of each pair of fields to place first in the output.
8609 If it gets it wrong use @ref{setfield} filter before @code{separatefields} filter.
8611 @section setdar, setsar
8613 The @code{setdar} filter sets the Display Aspect Ratio for the filter
8616 This is done by changing the specified Sample (aka Pixel) Aspect
8617 Ratio, according to the following equation:
8619 @var{DAR} = @var{HORIZONTAL_RESOLUTION} / @var{VERTICAL_RESOLUTION} * @var{SAR}
8622 Keep in mind that the @code{setdar} filter does not modify the pixel
8623 dimensions of the video frame. Also, the display aspect ratio set by
8624 this filter may be changed by later filters in the filterchain,
8625 e.g. in case of scaling or if another "setdar" or a "setsar" filter is
8628 The @code{setsar} filter sets the Sample (aka Pixel) Aspect Ratio for
8629 the filter output video.
8631 Note that as a consequence of the application of this filter, the
8632 output display aspect ratio will change according to the equation
8635 Keep in mind that the sample aspect ratio set by the @code{setsar}
8636 filter may be changed by later filters in the filterchain, e.g. if
8637 another "setsar" or a "setdar" filter is applied.
8639 It accepts the following parameters:
8642 @item r, ratio, dar (@code{setdar} only), sar (@code{setsar} only)
8643 Set the aspect ratio used by the filter.
8645 The parameter can be a floating point number string, an expression, or
8646 a string of the form @var{num}:@var{den}, where @var{num} and
8647 @var{den} are the numerator and denominator of the aspect ratio. If
8648 the parameter is not specified, it is assumed the value "0".
8649 In case the form "@var{num}:@var{den}" is used, the @code{:} character
8653 Set the maximum integer value to use for expressing numerator and
8654 denominator when reducing the expressed aspect ratio to a rational.
8655 Default value is @code{100}.
8659 The parameter @var{sar} is an expression containing
8660 the following constants:
8664 These are approximated values for the mathematical constants e
8665 (Euler's number), pi (Greek pi), and phi (the golden ratio).
8668 The input width and height.
8671 These are the same as @var{w} / @var{h}.
8674 The input sample aspect ratio.
8677 The input display aspect ratio. It is the same as
8678 (@var{w} / @var{h}) * @var{sar}.
8681 Horizontal and vertical chroma subsample values. For example, for the
8682 pixel format "yuv422p" @var{hsub} is 2 and @var{vsub} is 1.
8685 @subsection Examples
8690 To change the display aspect ratio to 16:9, specify one of the following:
8698 To change the sample aspect ratio to 10:11, specify:
8704 To set a display aspect ratio of 16:9, and specify a maximum integer value of
8705 1000 in the aspect ratio reduction, use the command:
8707 setdar=ratio=16/9:max=1000
8715 Force field for the output video frame.
8717 The @code{setfield} filter marks the interlace type field for the
8718 output frames. It does not change the input frame, but only sets the
8719 corresponding property, which affects how the frame is treated by
8720 following filters (e.g. @code{fieldorder} or @code{yadif}).
8722 The filter accepts the following options:
8727 Available values are:
8731 Keep the same field property.
8734 Mark the frame as bottom-field-first.
8737 Mark the frame as top-field-first.
8740 Mark the frame as progressive.
8746 Show a line containing various information for each input video frame.
8747 The input video is not modified.
8749 The shown line contains a sequence of key/value pairs of the form
8750 @var{key}:@var{value}.
8752 The following values are shown in the output:
8756 The (sequential) number of the input frame, starting from 0.
8759 The Presentation TimeStamp of the input frame, expressed as a number of
8760 time base units. The time base unit depends on the filter input pad.
8763 The Presentation TimeStamp of the input frame, expressed as a number of
8767 The position of the frame in the input stream, or -1 if this information is
8768 unavailable and/or meaningless (for example in case of synthetic video).
8771 The pixel format name.
8774 The sample aspect ratio of the input frame, expressed in the form
8775 @var{num}/@var{den}.
8778 The size of the input frame. For the syntax of this option, check the
8779 @ref{video size syntax,,"Video size" section in the ffmpeg-utils manual,ffmpeg-utils}.
8782 The type of interlaced mode ("P" for "progressive", "T" for top field first, "B"
8783 for bottom field first).
8786 This is 1 if the frame is a key frame, 0 otherwise.
8789 The picture type of the input frame ("I" for an I-frame, "P" for a
8790 P-frame, "B" for a B-frame, or "?" for an unknown type).
8791 Also refer to the documentation of the @code{AVPictureType} enum and of
8792 the @code{av_get_picture_type_char} function defined in
8793 @file{libavutil/avutil.h}.
8796 The Adler-32 checksum (printed in hexadecimal) of all the planes of the input frame.
8798 @item plane_checksum
8799 The Adler-32 checksum (printed in hexadecimal) of each plane of the input frame,
8800 expressed in the form "[@var{c0} @var{c1} @var{c2} @var{c3}]".
8803 @section showpalette
8805 Displays the 256 colors palette of each frame. This filter is only relevant for
8806 @var{pal8} pixel format frames.
8808 It accepts the following option:
8812 Set the size of the box used to represent one palette color entry. Default is
8813 @code{30} (for a @code{30x30} pixel box).
8816 @section shuffleplanes
8818 Reorder and/or duplicate video planes.
8820 It accepts the following parameters:
8825 The index of the input plane to be used as the first output plane.
8828 The index of the input plane to be used as the second output plane.
8831 The index of the input plane to be used as the third output plane.
8834 The index of the input plane to be used as the fourth output plane.
8838 The first plane has the index 0. The default is to keep the input unchanged.
8840 Swap the second and third planes of the input:
8842 ffmpeg -i INPUT -vf shuffleplanes=0:2:1:3 OUTPUT
8845 @anchor{signalstats}
8846 @section signalstats
8847 Evaluate various visual metrics that assist in determining issues associated
8848 with the digitization of analog video media.
8850 By default the filter will log these metadata values:
8854 Display the minimal Y value contained within the input frame. Expressed in
8858 Display the Y value at the 10% percentile within the input frame. Expressed in
8862 Display the average Y value within the input frame. Expressed in range of
8866 Display the Y value at the 90% percentile within the input frame. Expressed in
8870 Display the maximum Y value contained within the input frame. Expressed in
8874 Display the minimal U value contained within the input frame. Expressed in
8878 Display the U value at the 10% percentile within the input frame. Expressed in
8882 Display the average U value within the input frame. Expressed in range of
8886 Display the U value at the 90% percentile within the input frame. Expressed in
8890 Display the maximum U value contained within the input frame. Expressed in
8894 Display the minimal V value contained within the input frame. Expressed in
8898 Display the V value at the 10% percentile within the input frame. Expressed in
8902 Display the average V value within the input frame. Expressed in range of
8906 Display the V value at the 90% percentile within the input frame. Expressed in
8910 Display the maximum V value contained within the input frame. Expressed in
8914 Display the minimal saturation value contained within the input frame.
8915 Expressed in range of [0-~181.02].
8918 Display the saturation value at the 10% percentile within the input frame.
8919 Expressed in range of [0-~181.02].
8922 Display the average saturation value within the input frame. Expressed in range
8926 Display the saturation value at the 90% percentile within the input frame.
8927 Expressed in range of [0-~181.02].
8930 Display the maximum saturation value contained within the input frame.
8931 Expressed in range of [0-~181.02].
8934 Display the median value for hue within the input frame. Expressed in range of
8938 Display the average value for hue within the input frame. Expressed in range of
8942 Display the average of sample value difference between all values of the Y
8943 plane in the current frame and corresponding values of the previous input frame.
8944 Expressed in range of [0-255].
8947 Display the average of sample value difference between all values of the U
8948 plane in the current frame and corresponding values of the previous input frame.
8949 Expressed in range of [0-255].
8952 Display the average of sample value difference between all values of the V
8953 plane in the current frame and corresponding values of the previous input frame.
8954 Expressed in range of [0-255].
8957 The filter accepts the following options:
8963 @option{stat} specify an additional form of image analysis.
8964 @option{out} output video with the specified type of pixel highlighted.
8966 Both options accept the following values:
8970 Identify @var{temporal outliers} pixels. A @var{temporal outlier} is a pixel
8971 unlike the neighboring pixels of the same field. Examples of temporal outliers
8972 include the results of video dropouts, head clogs, or tape tracking issues.
8975 Identify @var{vertical line repetition}. Vertical line repetition includes
8976 similar rows of pixels within a frame. In born-digital video vertical line
8977 repetition is common, but this pattern is uncommon in video digitized from an
8978 analog source. When it occurs in video that results from the digitization of an
8979 analog source it can indicate concealment from a dropout compensator.
8982 Identify pixels that fall outside of legal broadcast range.
8986 Set the highlight color for the @option{out} option. The default color is
8990 @subsection Examples
8994 Output data of various video metrics:
8996 ffprobe -f lavfi movie=example.mov,signalstats="stat=tout+vrep+brng" -show_frames
9000 Output specific data about the minimum and maximum values of the Y plane per frame:
9002 ffprobe -f lavfi movie=example.mov,signalstats -show_entries frame_tags=lavfi.signalstats.YMAX,lavfi.signalstats.YMIN
9006 Playback video while highlighting pixels that are outside of broadcast range in red.
9008 ffplay example.mov -vf signalstats="out=brng:color=red"
9012 Playback video with signalstats metadata drawn over the frame.
9014 ffplay example.mov -vf signalstats=stat=brng+vrep+tout,drawtext=fontfile=FreeSerif.ttf:textfile=signalstat_drawtext.txt
9017 The contents of signalstat_drawtext.txt used in the command are:
9020 Y (%@{metadata:lavfi.signalstats.YMIN@}-%@{metadata:lavfi.signalstats.YMAX@})
9021 U (%@{metadata:lavfi.signalstats.UMIN@}-%@{metadata:lavfi.signalstats.UMAX@})
9022 V (%@{metadata:lavfi.signalstats.VMIN@}-%@{metadata:lavfi.signalstats.VMAX@})
9023 saturation maximum: %@{metadata:lavfi.signalstats.SATMAX@}
9031 Blur the input video without impacting the outlines.
9033 It accepts the following options:
9036 @item luma_radius, lr
9037 Set the luma radius. The option value must be a float number in
9038 the range [0.1,5.0] that specifies the variance of the gaussian filter
9039 used to blur the image (slower if larger). Default value is 1.0.
9041 @item luma_strength, ls
9042 Set the luma strength. The option value must be a float number
9043 in the range [-1.0,1.0] that configures the blurring. A value included
9044 in [0.0,1.0] will blur the image whereas a value included in
9045 [-1.0,0.0] will sharpen the image. Default value is 1.0.
9047 @item luma_threshold, lt
9048 Set the luma threshold used as a coefficient to determine
9049 whether a pixel should be blurred or not. The option value must be an
9050 integer in the range [-30,30]. A value of 0 will filter all the image,
9051 a value included in [0,30] will filter flat areas and a value included
9052 in [-30,0] will filter edges. Default value is 0.
9054 @item chroma_radius, cr
9055 Set the chroma radius. The option value must be a float number in
9056 the range [0.1,5.0] that specifies the variance of the gaussian filter
9057 used to blur the image (slower if larger). Default value is 1.0.
9059 @item chroma_strength, cs
9060 Set the chroma strength. The option value must be a float number
9061 in the range [-1.0,1.0] that configures the blurring. A value included
9062 in [0.0,1.0] will blur the image whereas a value included in
9063 [-1.0,0.0] will sharpen the image. Default value is 1.0.
9065 @item chroma_threshold, ct
9066 Set the chroma threshold used as a coefficient to determine
9067 whether a pixel should be blurred or not. The option value must be an
9068 integer in the range [-30,30]. A value of 0 will filter all the image,
9069 a value included in [0,30] will filter flat areas and a value included
9070 in [-30,0] will filter edges. Default value is 0.
9073 If a chroma option is not explicitly set, the corresponding luma value
9078 Obtain the SSIM (Structural SImilarity Metric) between two input videos.
9080 This filter takes in input two input videos, the first input is
9081 considered the "main" source and is passed unchanged to the
9082 output. The second input is used as a "reference" video for computing
9085 Both video inputs must have the same resolution and pixel format for
9086 this filter to work correctly. Also it assumes that both inputs
9087 have the same number of frames, which are compared one by one.
9089 The filter stores the calculated SSIM of each frame.
9091 The description of the accepted parameters follows.
9095 If specified the filter will use the named file to save the SSIM of
9096 each individual frame.
9099 The file printed if @var{stats_file} is selected, contains a sequence of
9100 key/value pairs of the form @var{key}:@var{value} for each compared
9103 A description of each shown parameter follows:
9107 sequential number of the input frame, starting from 1
9109 @item Y, U, V, R, G, B
9110 SSIM of the compared frames for the component specified by the suffix.
9113 SSIM of the compared frames for the whole frame.
9116 Same as above but in dB representation.
9121 movie=ref_movie.mpg, setpts=PTS-STARTPTS [main];
9122 [main][ref] ssim="stats_file=stats.log" [out]
9125 On this example the input file being processed is compared with the
9126 reference file @file{ref_movie.mpg}. The SSIM of each individual frame
9127 is stored in @file{stats.log}.
9129 Another example with both psnr and ssim at same time:
9131 ffmpeg -i main.mpg -i ref.mpg -lavfi "ssim;[0:v][1:v]psnr" -f null -
9136 Convert between different stereoscopic image formats.
9138 The filters accept the following options:
9142 Set stereoscopic image format of input.
9144 Available values for input image formats are:
9147 side by side parallel (left eye left, right eye right)
9150 side by side crosseye (right eye left, left eye right)
9153 side by side parallel with half width resolution
9154 (left eye left, right eye right)
9157 side by side crosseye with half width resolution
9158 (right eye left, left eye right)
9161 above-below (left eye above, right eye below)
9164 above-below (right eye above, left eye below)
9167 above-below with half height resolution
9168 (left eye above, right eye below)
9171 above-below with half height resolution
9172 (right eye above, left eye below)
9175 alternating frames (left eye first, right eye second)
9178 alternating frames (right eye first, left eye second)
9180 Default value is @samp{sbsl}.
9184 Set stereoscopic image format of output.
9186 Available values for output image formats are all the input formats as well as:
9189 anaglyph red/blue gray
9190 (red filter on left eye, blue filter on right eye)
9193 anaglyph red/green gray
9194 (red filter on left eye, green filter on right eye)
9197 anaglyph red/cyan gray
9198 (red filter on left eye, cyan filter on right eye)
9201 anaglyph red/cyan half colored
9202 (red filter on left eye, cyan filter on right eye)
9205 anaglyph red/cyan color
9206 (red filter on left eye, cyan filter on right eye)
9209 anaglyph red/cyan color optimized with the least squares projection of dubois
9210 (red filter on left eye, cyan filter on right eye)
9213 anaglyph green/magenta gray
9214 (green filter on left eye, magenta filter on right eye)
9217 anaglyph green/magenta half colored
9218 (green filter on left eye, magenta filter on right eye)
9221 anaglyph green/magenta colored
9222 (green filter on left eye, magenta filter on right eye)
9225 anaglyph green/magenta color optimized with the least squares projection of dubois
9226 (green filter on left eye, magenta filter on right eye)
9229 anaglyph yellow/blue gray
9230 (yellow filter on left eye, blue filter on right eye)
9233 anaglyph yellow/blue half colored
9234 (yellow filter on left eye, blue filter on right eye)
9237 anaglyph yellow/blue colored
9238 (yellow filter on left eye, blue filter on right eye)
9241 anaglyph yellow/blue color optimized with the least squares projection of dubois
9242 (yellow filter on left eye, blue filter on right eye)
9245 interleaved rows (left eye has top row, right eye starts on next row)
9248 interleaved rows (right eye has top row, left eye starts on next row)
9251 mono output (left eye only)
9254 mono output (right eye only)
9257 Default value is @samp{arcd}.
9260 @subsection Examples
9264 Convert input video from side by side parallel to anaglyph yellow/blue dubois:
9270 Convert input video from above bellow (left eye above, right eye below) to side by side crosseye.
9279 Apply a simple postprocessing filter that compresses and decompresses the image
9280 at several (or - in the case of @option{quality} level @code{6} - all) shifts
9281 and average the results.
9283 The filter accepts the following options:
9287 Set quality. This option defines the number of levels for averaging. It accepts
9288 an integer in the range 0-6. If set to @code{0}, the filter will have no
9289 effect. A value of @code{6} means the higher quality. For each increment of
9290 that value the speed drops by a factor of approximately 2. Default value is
9294 Force a constant quantization parameter. If not set, the filter will use the QP
9295 from the video stream (if available).
9298 Set thresholding mode. Available modes are:
9302 Set hard thresholding (default).
9304 Set soft thresholding (better de-ringing effect, but likely blurrier).
9308 Enable the use of the QP from the B-Frames if set to @code{1}. Using this
9309 option may cause flicker since the B-Frames have often larger QP. Default is
9310 @code{0} (not enabled).
9316 Draw subtitles on top of input video using the libass library.
9318 To enable compilation of this filter you need to configure FFmpeg with
9319 @code{--enable-libass}. This filter also requires a build with libavcodec and
9320 libavformat to convert the passed subtitles file to ASS (Advanced Substation
9321 Alpha) subtitles format.
9323 The filter accepts the following options:
9327 Set the filename of the subtitle file to read. It must be specified.
9330 Specify the size of the original video, the video for which the ASS file
9331 was composed. For the syntax of this option, check the
9332 @ref{video size syntax,,"Video size" section in the ffmpeg-utils manual,ffmpeg-utils}.
9333 Due to a misdesign in ASS aspect ratio arithmetic, this is necessary to
9334 correctly scale the fonts if the aspect ratio has been changed.
9337 Set subtitles input character encoding. @code{subtitles} filter only. Only
9338 useful if not UTF-8.
9340 @item stream_index, si
9341 Set subtitles stream index. @code{subtitles} filter only.
9344 Override default style or script info parameters of the subtitles. It accepts a
9345 string containing ASS style format @code{KEY=VALUE} couples separated by ",".
9348 If the first key is not specified, it is assumed that the first value
9349 specifies the @option{filename}.
9351 For example, to render the file @file{sub.srt} on top of the input
9352 video, use the command:
9357 which is equivalent to:
9359 subtitles=filename=sub.srt
9362 To render the default subtitles stream from file @file{video.mkv}, use:
9367 To render the second subtitles stream from that file, use:
9369 subtitles=video.mkv:si=1
9372 To make the subtitles stream from @file{sub.srt} appear in transparent green
9373 @code{DejaVu Serif}, use:
9375 subtitles=sub.srt:force_style='FontName=DejaVu Serif,PrimaryColour=&HAA00FF00'
9380 Scale the input by 2x and smooth using the Super2xSaI (Scale and
9381 Interpolate) pixel art scaling algorithm.
9383 Useful for enlarging pixel art images without reducing sharpness.
9390 Apply telecine process to the video.
9392 This filter accepts the following options:
9401 The default value is @code{top}.
9405 A string of numbers representing the pulldown pattern you wish to apply.
9406 The default value is @code{23}.
9410 Some typical patterns:
9415 24p: 2332 (preferred)
9422 24p: 222222222223 ("Euro pulldown")
9428 Select the most representative frame in a given sequence of consecutive frames.
9430 The filter accepts the following options:
9434 Set the frames batch size to analyze; in a set of @var{n} frames, the filter
9435 will pick one of them, and then handle the next batch of @var{n} frames until
9436 the end. Default is @code{100}.
9439 Since the filter keeps track of the whole frames sequence, a bigger @var{n}
9440 value will result in a higher memory usage, so a high value is not recommended.
9442 @subsection Examples
9446 Extract one picture each 50 frames:
9452 Complete example of a thumbnail creation with @command{ffmpeg}:
9454 ffmpeg -i in.avi -vf thumbnail,scale=300:200 -frames:v 1 out.png
9460 Tile several successive frames together.
9462 The filter accepts the following options:
9467 Set the grid size (i.e. the number of lines and columns). For the syntax of
9468 this option, check the
9469 @ref{video size syntax,,"Video size" section in the ffmpeg-utils manual,ffmpeg-utils}.
9472 Set the maximum number of frames to render in the given area. It must be less
9473 than or equal to @var{w}x@var{h}. The default value is @code{0}, meaning all
9474 the area will be used.
9477 Set the outer border margin in pixels.
9480 Set the inner border thickness (i.e. the number of pixels between frames). For
9481 more advanced padding options (such as having different values for the edges),
9482 refer to the pad video filter.
9485 Specify the color of the unused area. For the syntax of this option, check the
9486 "Color" section in the ffmpeg-utils manual. The default value of @var{color}
9490 @subsection Examples
9494 Produce 8x8 PNG tiles of all keyframes (@option{-skip_frame nokey}) in a movie:
9496 ffmpeg -skip_frame nokey -i file.avi -vf 'scale=128:72,tile=8x8' -an -vsync 0 keyframes%03d.png
9498 The @option{-vsync 0} is necessary to prevent @command{ffmpeg} from
9499 duplicating each output frame to accommodate the originally detected frame
9503 Display @code{5} pictures in an area of @code{3x2} frames,
9504 with @code{7} pixels between them, and @code{2} pixels of initial margin, using
9505 mixed flat and named options:
9507 tile=3x2:nb_frames=5:padding=7:margin=2
9513 Perform various types of temporal field interlacing.
9515 Frames are counted starting from 1, so the first input frame is
9518 The filter accepts the following options:
9523 Specify the mode of the interlacing. This option can also be specified
9524 as a value alone. See below for a list of values for this option.
9526 Available values are:
9530 Move odd frames into the upper field, even into the lower field,
9531 generating a double height frame at half frame rate.
9535 Frame 1 Frame 2 Frame 3 Frame 4
9537 11111 22222 33333 44444
9538 11111 22222 33333 44444
9539 11111 22222 33333 44444
9540 11111 22222 33333 44444
9554 Only output even frames, odd frames are dropped, generating a frame with
9555 unchanged height at half frame rate.
9560 Frame 1 Frame 2 Frame 3 Frame 4
9562 11111 22222 33333 44444
9563 11111 22222 33333 44444
9564 11111 22222 33333 44444
9565 11111 22222 33333 44444
9575 Only output odd frames, even frames are dropped, generating a frame with
9576 unchanged height at half frame rate.
9581 Frame 1 Frame 2 Frame 3 Frame 4
9583 11111 22222 33333 44444
9584 11111 22222 33333 44444
9585 11111 22222 33333 44444
9586 11111 22222 33333 44444
9596 Expand each frame to full height, but pad alternate lines with black,
9597 generating a frame with double height at the same input frame rate.
9602 Frame 1 Frame 2 Frame 3 Frame 4
9604 11111 22222 33333 44444
9605 11111 22222 33333 44444
9606 11111 22222 33333 44444
9607 11111 22222 33333 44444
9610 11111 ..... 33333 .....
9611 ..... 22222 ..... 44444
9612 11111 ..... 33333 .....
9613 ..... 22222 ..... 44444
9614 11111 ..... 33333 .....
9615 ..... 22222 ..... 44444
9616 11111 ..... 33333 .....
9617 ..... 22222 ..... 44444
9621 @item interleave_top, 4
9622 Interleave the upper field from odd frames with the lower field from
9623 even frames, generating a frame with unchanged height at half frame rate.
9628 Frame 1 Frame 2 Frame 3 Frame 4
9630 11111<- 22222 33333<- 44444
9631 11111 22222<- 33333 44444<-
9632 11111<- 22222 33333<- 44444
9633 11111 22222<- 33333 44444<-
9643 @item interleave_bottom, 5
9644 Interleave the lower field from odd frames with the upper field from
9645 even frames, generating a frame with unchanged height at half frame rate.
9650 Frame 1 Frame 2 Frame 3 Frame 4
9652 11111 22222<- 33333 44444<-
9653 11111<- 22222 33333<- 44444
9654 11111 22222<- 33333 44444<-
9655 11111<- 22222 33333<- 44444
9665 @item interlacex2, 6
9666 Double frame rate with unchanged height. Frames are inserted each
9667 containing the second temporal field from the previous input frame and
9668 the first temporal field from the next input frame. This mode relies on
9669 the top_field_first flag. Useful for interlaced video displays with no
9670 field synchronisation.
9675 Frame 1 Frame 2 Frame 3 Frame 4
9677 11111 22222 33333 44444
9678 11111 22222 33333 44444
9679 11111 22222 33333 44444
9680 11111 22222 33333 44444
9683 11111 22222 22222 33333 33333 44444 44444
9684 11111 11111 22222 22222 33333 33333 44444
9685 11111 22222 22222 33333 33333 44444 44444
9686 11111 11111 22222 22222 33333 33333 44444
9692 Numeric values are deprecated but are accepted for backward
9693 compatibility reasons.
9695 Default mode is @code{merge}.
9698 Specify flags influencing the filter process.
9700 Available value for @var{flags} is:
9703 @item low_pass_filter, vlfp
9704 Enable vertical low-pass filtering in the filter.
9705 Vertical low-pass filtering is required when creating an interlaced
9706 destination from a progressive source which contains high-frequency
9707 vertical detail. Filtering will reduce interlace 'twitter' and Moire
9710 Vertical low-pass filtering can only be enabled for @option{mode}
9711 @var{interleave_top} and @var{interleave_bottom}.
9718 Transpose rows with columns in the input video and optionally flip it.
9720 It accepts the following parameters:
9725 Specify the transposition direction.
9727 Can assume the following values:
9729 @item 0, 4, cclock_flip
9730 Rotate by 90 degrees counterclockwise and vertically flip (default), that is:
9738 Rotate by 90 degrees clockwise, that is:
9746 Rotate by 90 degrees counterclockwise, that is:
9753 @item 3, 7, clock_flip
9754 Rotate by 90 degrees clockwise and vertically flip, that is:
9762 For values between 4-7, the transposition is only done if the input
9763 video geometry is portrait and not landscape. These values are
9764 deprecated, the @code{passthrough} option should be used instead.
9766 Numerical values are deprecated, and should be dropped in favor of
9770 Do not apply the transposition if the input geometry matches the one
9771 specified by the specified value. It accepts the following values:
9774 Always apply transposition.
9776 Preserve portrait geometry (when @var{height} >= @var{width}).
9778 Preserve landscape geometry (when @var{width} >= @var{height}).
9781 Default value is @code{none}.
9784 For example to rotate by 90 degrees clockwise and preserve portrait
9787 transpose=dir=1:passthrough=portrait
9790 The command above can also be specified as:
9792 transpose=1:portrait
9796 Trim the input so that the output contains one continuous subpart of the input.
9798 It accepts the following parameters:
9801 Specify the time of the start of the kept section, i.e. the frame with the
9802 timestamp @var{start} will be the first frame in the output.
9805 Specify the time of the first frame that will be dropped, i.e. the frame
9806 immediately preceding the one with the timestamp @var{end} will be the last
9807 frame in the output.
9810 This is the same as @var{start}, except this option sets the start timestamp
9811 in timebase units instead of seconds.
9814 This is the same as @var{end}, except this option sets the end timestamp
9815 in timebase units instead of seconds.
9818 The maximum duration of the output in seconds.
9821 The number of the first frame that should be passed to the output.
9824 The number of the first frame that should be dropped.
9827 @option{start}, @option{end}, and @option{duration} are expressed as time
9828 duration specifications; see
9829 @ref{time duration syntax,,the Time duration section in the ffmpeg-utils(1) manual,ffmpeg-utils}
9830 for the accepted syntax.
9832 Note that the first two sets of the start/end options and the @option{duration}
9833 option look at the frame timestamp, while the _frame variants simply count the
9834 frames that pass through the filter. Also note that this filter does not modify
9835 the timestamps. If you wish for the output timestamps to start at zero, insert a
9836 setpts filter after the trim filter.
9838 If multiple start or end options are set, this filter tries to be greedy and
9839 keep all the frames that match at least one of the specified constraints. To keep
9840 only the part that matches all the constraints at once, chain multiple trim
9843 The defaults are such that all the input is kept. So it is possible to set e.g.
9844 just the end values to keep everything before the specified time.
9849 Drop everything except the second minute of input:
9851 ffmpeg -i INPUT -vf trim=60:120
9855 Keep only the first second:
9857 ffmpeg -i INPUT -vf trim=duration=1
9866 Sharpen or blur the input video.
9868 It accepts the following parameters:
9871 @item luma_msize_x, lx
9872 Set the luma matrix horizontal size. It must be an odd integer between
9873 3 and 63. The default value is 5.
9875 @item luma_msize_y, ly
9876 Set the luma matrix vertical size. It must be an odd integer between 3
9877 and 63. The default value is 5.
9879 @item luma_amount, la
9880 Set the luma effect strength. It must be a floating point number, reasonable
9881 values lay between -1.5 and 1.5.
9883 Negative values will blur the input video, while positive values will
9884 sharpen it, a value of zero will disable the effect.
9886 Default value is 1.0.
9888 @item chroma_msize_x, cx
9889 Set the chroma matrix horizontal size. It must be an odd integer
9890 between 3 and 63. The default value is 5.
9892 @item chroma_msize_y, cy
9893 Set the chroma matrix vertical size. It must be an odd integer
9894 between 3 and 63. The default value is 5.
9896 @item chroma_amount, ca
9897 Set the chroma effect strength. It must be a floating point number, reasonable
9898 values lay between -1.5 and 1.5.
9900 Negative values will blur the input video, while positive values will
9901 sharpen it, a value of zero will disable the effect.
9903 Default value is 0.0.
9906 If set to 1, specify using OpenCL capabilities, only available if
9907 FFmpeg was configured with @code{--enable-opencl}. Default value is 0.
9911 All parameters are optional and default to the equivalent of the
9912 string '5:5:1.0:5:5:0.0'.
9914 @subsection Examples
9918 Apply strong luma sharpen effect:
9920 unsharp=luma_msize_x=7:luma_msize_y=7:luma_amount=2.5
9924 Apply a strong blur of both luma and chroma parameters:
9926 unsharp=7:7:-2:7:7:-2
9932 Apply ultra slow/simple postprocessing filter that compresses and decompresses
9933 the image at several (or - in the case of @option{quality} level @code{8} - all)
9934 shifts and average the results.
9936 The way this differs from the behavior of spp is that uspp actually encodes &
9937 decodes each case with libavcodec Snow, whereas spp uses a simplified intra only 8x8
9938 DCT similar to MJPEG.
9940 The filter accepts the following options:
9944 Set quality. This option defines the number of levels for averaging. It accepts
9945 an integer in the range 0-8. If set to @code{0}, the filter will have no
9946 effect. A value of @code{8} means the higher quality. For each increment of
9947 that value the speed drops by a factor of approximately 2. Default value is
9951 Force a constant quantization parameter. If not set, the filter will use the QP
9952 from the video stream (if available).
9955 @anchor{vidstabdetect}
9956 @section vidstabdetect
9958 Analyze video stabilization/deshaking. Perform pass 1 of 2, see
9959 @ref{vidstabtransform} for pass 2.
9961 This filter generates a file with relative translation and rotation
9962 transform information about subsequent frames, which is then used by
9963 the @ref{vidstabtransform} filter.
9965 To enable compilation of this filter you need to configure FFmpeg with
9966 @code{--enable-libvidstab}.
9968 This filter accepts the following options:
9972 Set the path to the file used to write the transforms information.
9973 Default value is @file{transforms.trf}.
9976 Set how shaky the video is and how quick the camera is. It accepts an
9977 integer in the range 1-10, a value of 1 means little shakiness, a
9978 value of 10 means strong shakiness. Default value is 5.
9981 Set the accuracy of the detection process. It must be a value in the
9982 range 1-15. A value of 1 means low accuracy, a value of 15 means high
9983 accuracy. Default value is 15.
9986 Set stepsize of the search process. The region around minimum is
9987 scanned with 1 pixel resolution. Default value is 6.
9990 Set minimum contrast. Below this value a local measurement field is
9991 discarded. Must be a floating point value in the range 0-1. Default
9995 Set reference frame number for tripod mode.
9997 If enabled, the motion of the frames is compared to a reference frame
9998 in the filtered stream, identified by the specified number. The idea
9999 is to compensate all movements in a more-or-less static scene and keep
10000 the camera view absolutely still.
10002 If set to 0, it is disabled. The frames are counted starting from 1.
10005 Show fields and transforms in the resulting frames. It accepts an
10006 integer in the range 0-2. Default value is 0, which disables any
10010 @subsection Examples
10014 Use default values:
10020 Analyze strongly shaky movie and put the results in file
10021 @file{mytransforms.trf}:
10023 vidstabdetect=shakiness=10:accuracy=15:result="mytransforms.trf"
10027 Visualize the result of internal transformations in the resulting
10030 vidstabdetect=show=1
10034 Analyze a video with medium shakiness using @command{ffmpeg}:
10036 ffmpeg -i input -vf vidstabdetect=shakiness=5:show=1 dummy.avi
10040 @anchor{vidstabtransform}
10041 @section vidstabtransform
10043 Video stabilization/deshaking: pass 2 of 2,
10044 see @ref{vidstabdetect} for pass 1.
10046 Read a file with transform information for each frame and
10047 apply/compensate them. Together with the @ref{vidstabdetect}
10048 filter this can be used to deshake videos. See also
10049 @url{http://public.hronopik.de/vid.stab}. It is important to also use
10050 the @ref{unsharp} filter, see below.
10052 To enable compilation of this filter you need to configure FFmpeg with
10053 @code{--enable-libvidstab}.
10055 @subsection Options
10059 Set path to the file used to read the transforms. Default value is
10060 @file{transforms.trf}.
10063 Set the number of frames (value*2 + 1) used for lowpass filtering the
10064 camera movements. Default value is 10.
10066 For example a number of 10 means that 21 frames are used (10 in the
10067 past and 10 in the future) to smoothen the motion in the video. A
10068 larger value leads to a smoother video, but limits the acceleration of
10069 the camera (pan/tilt movements). 0 is a special case where a static
10070 camera is simulated.
10073 Set the camera path optimization algorithm.
10075 Accepted values are:
10078 gaussian kernel low-pass filter on camera motion (default)
10080 averaging on transformations
10084 Set maximal number of pixels to translate frames. Default value is -1,
10088 Set maximal angle in radians (degree*PI/180) to rotate frames. Default
10089 value is -1, meaning no limit.
10092 Specify how to deal with borders that may be visible due to movement
10095 Available values are:
10098 keep image information from previous frame (default)
10100 fill the border black
10104 Invert transforms if set to 1. Default value is 0.
10107 Consider transforms as relative to previous frame if set to 1,
10108 absolute if set to 0. Default value is 0.
10111 Set percentage to zoom. A positive value will result in a zoom-in
10112 effect, a negative value in a zoom-out effect. Default value is 0 (no
10116 Set optimal zooming to avoid borders.
10118 Accepted values are:
10123 optimal static zoom value is determined (only very strong movements
10124 will lead to visible borders) (default)
10126 optimal adaptive zoom value is determined (no borders will be
10127 visible), see @option{zoomspeed}
10130 Note that the value given at zoom is added to the one calculated here.
10133 Set percent to zoom maximally each frame (enabled when
10134 @option{optzoom} is set to 2). Range is from 0 to 5, default value is
10138 Specify type of interpolation.
10140 Available values are:
10145 linear only horizontal
10147 linear in both directions (default)
10149 cubic in both directions (slow)
10153 Enable virtual tripod mode if set to 1, which is equivalent to
10154 @code{relative=0:smoothing=0}. Default value is 0.
10156 Use also @code{tripod} option of @ref{vidstabdetect}.
10159 Increase log verbosity if set to 1. Also the detected global motions
10160 are written to the temporary file @file{global_motions.trf}. Default
10164 @subsection Examples
10168 Use @command{ffmpeg} for a typical stabilization with default values:
10170 ffmpeg -i inp.mpeg -vf vidstabtransform,unsharp=5:5:0.8:3:3:0.4 inp_stabilized.mpeg
10173 Note the use of the @ref{unsharp} filter which is always recommended.
10176 Zoom in a bit more and load transform data from a given file:
10178 vidstabtransform=zoom=5:input="mytransforms.trf"
10182 Smoothen the video even more:
10184 vidstabtransform=smoothing=30
10190 Flip the input video vertically.
10192 For example, to vertically flip a video with @command{ffmpeg}:
10194 ffmpeg -i in.avi -vf "vflip" out.avi
10200 Make or reverse a natural vignetting effect.
10202 The filter accepts the following options:
10206 Set lens angle expression as a number of radians.
10208 The value is clipped in the @code{[0,PI/2]} range.
10210 Default value: @code{"PI/5"}
10214 Set center coordinates expressions. Respectively @code{"w/2"} and @code{"h/2"}
10218 Set forward/backward mode.
10220 Available modes are:
10223 The larger the distance from the central point, the darker the image becomes.
10226 The larger the distance from the central point, the brighter the image becomes.
10227 This can be used to reverse a vignette effect, though there is no automatic
10228 detection to extract the lens @option{angle} and other settings (yet). It can
10229 also be used to create a burning effect.
10232 Default value is @samp{forward}.
10235 Set evaluation mode for the expressions (@option{angle}, @option{x0}, @option{y0}).
10237 It accepts the following values:
10240 Evaluate expressions only once during the filter initialization.
10243 Evaluate expressions for each incoming frame. This is way slower than the
10244 @samp{init} mode since it requires all the scalers to be re-computed, but it
10245 allows advanced dynamic expressions.
10248 Default value is @samp{init}.
10251 Set dithering to reduce the circular banding effects. Default is @code{1}
10255 Set vignette aspect. This setting allows one to adjust the shape of the vignette.
10256 Setting this value to the SAR of the input will make a rectangular vignetting
10257 following the dimensions of the video.
10259 Default is @code{1/1}.
10262 @subsection Expressions
10264 The @option{alpha}, @option{x0} and @option{y0} expressions can contain the
10265 following parameters.
10270 input width and height
10273 the number of input frame, starting from 0
10276 the PTS (Presentation TimeStamp) time of the filtered video frame, expressed in
10277 @var{TB} units, NAN if undefined
10280 frame rate of the input video, NAN if the input frame rate is unknown
10283 the PTS (Presentation TimeStamp) of the filtered video frame,
10284 expressed in seconds, NAN if undefined
10287 time base of the input video
10291 @subsection Examples
10295 Apply simple strong vignetting effect:
10301 Make a flickering vignetting:
10303 vignette='PI/4+random(1)*PI/50':eval=frame
10310 Deinterlace the input video ("w3fdif" stands for "Weston 3 Field
10311 Deinterlacing Filter").
10313 Based on the process described by Martin Weston for BBC R&D, and
10314 implemented based on the de-interlace algorithm written by Jim
10315 Easterbrook for BBC R&D, the Weston 3 field deinterlacing filter
10316 uses filter coefficients calculated by BBC R&D.
10318 There are two sets of filter coefficients, so called "simple":
10319 and "complex". Which set of filter coefficients is used can
10320 be set by passing an optional parameter:
10324 Set the interlacing filter coefficients. Accepts one of the following values:
10328 Simple filter coefficient set.
10330 More-complex filter coefficient set.
10332 Default value is @samp{complex}.
10335 Specify which frames to deinterlace. Accept one of the following values:
10339 Deinterlace all frames,
10341 Only deinterlace frames marked as interlaced.
10344 Default value is @samp{all}.
10348 Apply the xBR high-quality magnification filter which is designed for pixel
10349 art. It follows a set of edge-detection rules, see
10350 @url{http://www.libretro.com/forums/viewtopic.php?f=6&t=134}.
10352 It accepts the following option:
10356 Set the scaling dimension: @code{2} for @code{2xBR}, @code{3} for
10357 @code{3xBR} and @code{4} for @code{4xBR}.
10358 Default is @code{3}.
10364 Deinterlace the input video ("yadif" means "yet another deinterlacing
10367 It accepts the following parameters:
10373 The interlacing mode to adopt. It accepts one of the following values:
10376 @item 0, send_frame
10377 Output one frame for each frame.
10378 @item 1, send_field
10379 Output one frame for each field.
10380 @item 2, send_frame_nospatial
10381 Like @code{send_frame}, but it skips the spatial interlacing check.
10382 @item 3, send_field_nospatial
10383 Like @code{send_field}, but it skips the spatial interlacing check.
10386 The default value is @code{send_frame}.
10389 The picture field parity assumed for the input interlaced video. It accepts one
10390 of the following values:
10394 Assume the top field is first.
10396 Assume the bottom field is first.
10398 Enable automatic detection of field parity.
10401 The default value is @code{auto}.
10402 If the interlacing is unknown or the decoder does not export this information,
10403 top field first will be assumed.
10406 Specify which frames to deinterlace. Accept one of the following
10411 Deinterlace all frames.
10412 @item 1, interlaced
10413 Only deinterlace frames marked as interlaced.
10416 The default value is @code{all}.
10421 Apply Zoom & Pan effect.
10423 This filter accepts the following options:
10427 Set the zoom expression. Default is 1.
10431 Set the x and y expression. Default is 0.
10434 Set the duration expression in number of frames.
10435 This sets for how many number of frames effect will last for
10436 single input image.
10439 Set the output image size, default is 'hd720'.
10442 Each expression can contain the following constants:
10461 Output frame count.
10465 Last calculated 'x' and 'y' position from 'x' and 'y' expression
10466 for current input frame.
10470 'x' and 'y' of last output frame of previous input frame or 0 when there was
10471 not yet such frame (first input frame).
10474 Last calculated zoom from 'z' expression for current input frame.
10477 Last calculated zoom of last output frame of previous input frame.
10480 Number of output frames for current input frame. Calculated from 'd' expression
10481 for each input frame.
10484 number of output frames created for previous input frame
10487 Rational number: input width / input height
10490 sample aspect ratio
10493 display aspect ratio
10497 @subsection Examples
10501 Zoom-in up to 1.5 and pan at same time to some spot near center of picture:
10503 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
10507 Zoom-in up to 1.5 and pan always at center of picture:
10509 zoompan=z='min(zoom+0.0015,1.5)':d=700:x='iw/2-(iw/zoom/2)':y='ih/2-(ih/zoom/2)'
10513 @c man end VIDEO FILTERS
10515 @chapter Video Sources
10516 @c man begin VIDEO SOURCES
10518 Below is a description of the currently available video sources.
10522 Buffer video frames, and make them available to the filter chain.
10524 This source is mainly intended for a programmatic use, in particular
10525 through the interface defined in @file{libavfilter/vsrc_buffer.h}.
10527 It accepts the following parameters:
10532 Specify the size (width and height) of the buffered video frames. For the
10533 syntax of this option, check the
10534 @ref{video size syntax,,"Video size" section in the ffmpeg-utils manual,ffmpeg-utils}.
10537 The input video width.
10540 The input video height.
10543 A string representing the pixel format of the buffered video frames.
10544 It may be a number corresponding to a pixel format, or a pixel format
10548 Specify the timebase assumed by the timestamps of the buffered frames.
10551 Specify the frame rate expected for the video stream.
10553 @item pixel_aspect, sar
10554 The sample (pixel) aspect ratio of the input video.
10557 Specify the optional parameters to be used for the scale filter which
10558 is automatically inserted when an input change is detected in the
10559 input size or format.
10564 buffer=width=320:height=240:pix_fmt=yuv410p:time_base=1/24:sar=1
10567 will instruct the source to accept video frames with size 320x240 and
10568 with format "yuv410p", assuming 1/24 as the timestamps timebase and
10569 square pixels (1:1 sample aspect ratio).
10570 Since the pixel format with name "yuv410p" corresponds to the number 6
10571 (check the enum AVPixelFormat definition in @file{libavutil/pixfmt.h}),
10572 this example corresponds to:
10574 buffer=size=320x240:pixfmt=6:time_base=1/24:pixel_aspect=1/1
10577 Alternatively, the options can be specified as a flat string, but this
10578 syntax is deprecated:
10580 @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}]
10584 Create a pattern generated by an elementary cellular automaton.
10586 The initial state of the cellular automaton can be defined through the
10587 @option{filename}, and @option{pattern} options. If such options are
10588 not specified an initial state is created randomly.
10590 At each new frame a new row in the video is filled with the result of
10591 the cellular automaton next generation. The behavior when the whole
10592 frame is filled is defined by the @option{scroll} option.
10594 This source accepts the following options:
10598 Read the initial cellular automaton state, i.e. the starting row, from
10599 the specified file.
10600 In the file, each non-whitespace character is considered an alive
10601 cell, a newline will terminate the row, and further characters in the
10602 file will be ignored.
10605 Read the initial cellular automaton state, i.e. the starting row, from
10606 the specified string.
10608 Each non-whitespace character in the string is considered an alive
10609 cell, a newline will terminate the row, and further characters in the
10610 string will be ignored.
10613 Set the video rate, that is the number of frames generated per second.
10616 @item random_fill_ratio, ratio
10617 Set the random fill ratio for the initial cellular automaton row. It
10618 is a floating point number value ranging from 0 to 1, defaults to
10621 This option is ignored when a file or a pattern is specified.
10623 @item random_seed, seed
10624 Set the seed for filling randomly the initial row, must be an integer
10625 included between 0 and UINT32_MAX. If not specified, or if explicitly
10626 set to -1, the filter will try to use a good random seed on a best
10630 Set the cellular automaton rule, it is a number ranging from 0 to 255.
10631 Default value is 110.
10634 Set the size of the output video. For the syntax of this option, check the
10635 @ref{video size syntax,,"Video size" section in the ffmpeg-utils manual,ffmpeg-utils}.
10637 If @option{filename} or @option{pattern} is specified, the size is set
10638 by default to the width of the specified initial state row, and the
10639 height is set to @var{width} * PHI.
10641 If @option{size} is set, it must contain the width of the specified
10642 pattern string, and the specified pattern will be centered in the
10645 If a filename or a pattern string is not specified, the size value
10646 defaults to "320x518" (used for a randomly generated initial state).
10649 If set to 1, scroll the output upward when all the rows in the output
10650 have been already filled. If set to 0, the new generated row will be
10651 written over the top row just after the bottom row is filled.
10654 @item start_full, full
10655 If set to 1, completely fill the output with generated rows before
10656 outputting the first frame.
10657 This is the default behavior, for disabling set the value to 0.
10660 If set to 1, stitch the left and right row edges together.
10661 This is the default behavior, for disabling set the value to 0.
10664 @subsection Examples
10668 Read the initial state from @file{pattern}, and specify an output of
10671 cellauto=f=pattern:s=200x400
10675 Generate a random initial row with a width of 200 cells, with a fill
10678 cellauto=ratio=2/3:s=200x200
10682 Create a pattern generated by rule 18 starting by a single alive cell
10683 centered on an initial row with width 100:
10685 cellauto=p=@@:s=100x400:full=0:rule=18
10689 Specify a more elaborated initial pattern:
10691 cellauto=p='@@@@ @@ @@@@':s=100x400:full=0:rule=18
10696 @section mandelbrot
10698 Generate a Mandelbrot set fractal, and progressively zoom towards the
10699 point specified with @var{start_x} and @var{start_y}.
10701 This source accepts the following options:
10706 Set the terminal pts value. Default value is 400.
10709 Set the terminal scale value.
10710 Must be a floating point value. Default value is 0.3.
10713 Set the inner coloring mode, that is the algorithm used to draw the
10714 Mandelbrot fractal internal region.
10716 It shall assume one of the following values:
10721 Show time until convergence.
10723 Set color based on point closest to the origin of the iterations.
10728 Default value is @var{mincol}.
10731 Set the bailout value. Default value is 10.0.
10734 Set the maximum of iterations performed by the rendering
10735 algorithm. Default value is 7189.
10738 Set outer coloring mode.
10739 It shall assume one of following values:
10741 @item iteration_count
10742 Set iteration cound mode.
10743 @item normalized_iteration_count
10744 set normalized iteration count mode.
10746 Default value is @var{normalized_iteration_count}.
10749 Set frame rate, expressed as number of frames per second. Default
10753 Set frame size. For the syntax of this option, check the "Video
10754 size" section in the ffmpeg-utils manual. Default value is "640x480".
10757 Set the initial scale value. Default value is 3.0.
10760 Set the initial x position. Must be a floating point value between
10761 -100 and 100. Default value is -0.743643887037158704752191506114774.
10764 Set the initial y position. Must be a floating point value between
10765 -100 and 100. Default value is -0.131825904205311970493132056385139.
10770 Generate various test patterns, as generated by the MPlayer test filter.
10772 The size of the generated video is fixed, and is 256x256.
10773 This source is useful in particular for testing encoding features.
10775 This source accepts the following options:
10780 Specify the frame rate of the sourced video, as the number of frames
10781 generated per second. It has to be a string in the format
10782 @var{frame_rate_num}/@var{frame_rate_den}, an integer number, a floating point
10783 number or a valid video frame rate abbreviation. The default value is
10787 Set the duration of the sourced video. See
10788 @ref{time duration syntax,,the Time duration section in the ffmpeg-utils(1) manual,ffmpeg-utils}
10789 for the accepted syntax.
10791 If not specified, or the expressed duration is negative, the video is
10792 supposed to be generated forever.
10796 Set the number or the name of the test to perform. Supported tests are:
10812 Default value is "all", which will cycle through the list of all tests.
10817 mptestsrc=t=dc_luma
10820 will generate a "dc_luma" test pattern.
10822 @section frei0r_src
10824 Provide a frei0r source.
10826 To enable compilation of this filter you need to install the frei0r
10827 header and configure FFmpeg with @code{--enable-frei0r}.
10829 This source accepts the following parameters:
10834 The size of the video to generate. For the syntax of this option, check the
10835 @ref{video size syntax,,"Video size" section in the ffmpeg-utils manual,ffmpeg-utils}.
10838 The framerate of the generated video. It may be a string of the form
10839 @var{num}/@var{den} or a frame rate abbreviation.
10842 The name to the frei0r source to load. For more information regarding frei0r and
10843 how to set the parameters, read the @ref{frei0r} section in the video filters
10846 @item filter_params
10847 A '|'-separated list of parameters to pass to the frei0r source.
10851 For example, to generate a frei0r partik0l source with size 200x200
10852 and frame rate 10 which is overlaid on the overlay filter main input:
10854 frei0r_src=size=200x200:framerate=10:filter_name=partik0l:filter_params=1234 [overlay]; [in][overlay] overlay
10859 Generate a life pattern.
10861 This source is based on a generalization of John Conway's life game.
10863 The sourced input represents a life grid, each pixel represents a cell
10864 which can be in one of two possible states, alive or dead. Every cell
10865 interacts with its eight neighbours, which are the cells that are
10866 horizontally, vertically, or diagonally adjacent.
10868 At each interaction the grid evolves according to the adopted rule,
10869 which specifies the number of neighbor alive cells which will make a
10870 cell stay alive or born. The @option{rule} option allows one to specify
10873 This source accepts the following options:
10877 Set the file from which to read the initial grid state. In the file,
10878 each non-whitespace character is considered an alive cell, and newline
10879 is used to delimit the end of each row.
10881 If this option is not specified, the initial grid is generated
10885 Set the video rate, that is the number of frames generated per second.
10888 @item random_fill_ratio, ratio
10889 Set the random fill ratio for the initial random grid. It is a
10890 floating point number value ranging from 0 to 1, defaults to 1/PHI.
10891 It is ignored when a file is specified.
10893 @item random_seed, seed
10894 Set the seed for filling the initial random grid, must be an integer
10895 included between 0 and UINT32_MAX. If not specified, or if explicitly
10896 set to -1, the filter will try to use a good random seed on a best
10902 A rule can be specified with a code of the kind "S@var{NS}/B@var{NB}",
10903 where @var{NS} and @var{NB} are sequences of numbers in the range 0-8,
10904 @var{NS} specifies the number of alive neighbor cells which make a
10905 live cell stay alive, and @var{NB} the number of alive neighbor cells
10906 which make a dead cell to become alive (i.e. to "born").
10907 "s" and "b" can be used in place of "S" and "B", respectively.
10909 Alternatively a rule can be specified by an 18-bits integer. The 9
10910 high order bits are used to encode the next cell state if it is alive
10911 for each number of neighbor alive cells, the low order bits specify
10912 the rule for "borning" new cells. Higher order bits encode for an
10913 higher number of neighbor cells.
10914 For example the number 6153 = @code{(12<<9)+9} specifies a stay alive
10915 rule of 12 and a born rule of 9, which corresponds to "S23/B03".
10917 Default value is "S23/B3", which is the original Conway's game of life
10918 rule, and will keep a cell alive if it has 2 or 3 neighbor alive
10919 cells, and will born a new cell if there are three alive cells around
10923 Set the size of the output video. For the syntax of this option, check the
10924 @ref{video size syntax,,"Video size" section in the ffmpeg-utils manual,ffmpeg-utils}.
10926 If @option{filename} is specified, the size is set by default to the
10927 same size of the input file. If @option{size} is set, it must contain
10928 the size specified in the input file, and the initial grid defined in
10929 that file is centered in the larger resulting area.
10931 If a filename is not specified, the size value defaults to "320x240"
10932 (used for a randomly generated initial grid).
10935 If set to 1, stitch the left and right grid edges together, and the
10936 top and bottom edges also. Defaults to 1.
10939 Set cell mold speed. If set, a dead cell will go from @option{death_color} to
10940 @option{mold_color} with a step of @option{mold}. @option{mold} can have a
10941 value from 0 to 255.
10944 Set the color of living (or new born) cells.
10947 Set the color of dead cells. If @option{mold} is set, this is the first color
10948 used to represent a dead cell.
10951 Set mold color, for definitely dead and moldy cells.
10953 For the syntax of these 3 color options, check the "Color" section in the
10954 ffmpeg-utils manual.
10957 @subsection Examples
10961 Read a grid from @file{pattern}, and center it on a grid of size
10964 life=f=pattern:s=300x300
10968 Generate a random grid of size 200x200, with a fill ratio of 2/3:
10970 life=ratio=2/3:s=200x200
10974 Specify a custom rule for evolving a randomly generated grid:
10980 Full example with slow death effect (mold) using @command{ffplay}:
10982 ffplay -f lavfi life=s=300x200:mold=10:r=60:ratio=0.1:death_color=#C83232:life_color=#00ff00,scale=1200:800:flags=16
10987 @anchor{haldclutsrc}
10989 @anchor{rgbtestsrc}
10991 @anchor{smptehdbars}
10993 @section color, haldclutsrc, nullsrc, rgbtestsrc, smptebars, smptehdbars, testsrc
10995 The @code{color} source provides an uniformly colored input.
10997 The @code{haldclutsrc} source provides an identity Hald CLUT. See also
10998 @ref{haldclut} filter.
11000 The @code{nullsrc} source returns unprocessed video frames. It is
11001 mainly useful to be employed in analysis / debugging tools, or as the
11002 source for filters which ignore the input data.
11004 The @code{rgbtestsrc} source generates an RGB test pattern useful for
11005 detecting RGB vs BGR issues. You should see a red, green and blue
11006 stripe from top to bottom.
11008 The @code{smptebars} source generates a color bars pattern, based on
11009 the SMPTE Engineering Guideline EG 1-1990.
11011 The @code{smptehdbars} source generates a color bars pattern, based on
11012 the SMPTE RP 219-2002.
11014 The @code{testsrc} source generates a test video pattern, showing a
11015 color pattern, a scrolling gradient and a timestamp. This is mainly
11016 intended for testing purposes.
11018 The sources accept the following parameters:
11023 Specify the color of the source, only available in the @code{color}
11024 source. For the syntax of this option, check the "Color" section in the
11025 ffmpeg-utils manual.
11028 Specify the level of the Hald CLUT, only available in the @code{haldclutsrc}
11029 source. A level of @code{N} generates a picture of @code{N*N*N} by @code{N*N*N}
11030 pixels to be used as identity matrix for 3D lookup tables. Each component is
11031 coded on a @code{1/(N*N)} scale.
11034 Specify the size of the sourced video. For the syntax of this option, check the
11035 @ref{video size syntax,,"Video size" section in the ffmpeg-utils manual,ffmpeg-utils}.
11036 The default value is @code{320x240}.
11038 This option is not available with the @code{haldclutsrc} filter.
11041 Specify the frame rate of the sourced video, as the number of frames
11042 generated per second. It has to be a string in the format
11043 @var{frame_rate_num}/@var{frame_rate_den}, an integer number, a floating point
11044 number or a valid video frame rate abbreviation. The default value is
11048 Set the sample aspect ratio of the sourced video.
11051 Set the duration of the sourced video. See
11052 @ref{time duration syntax,,the Time duration section in the ffmpeg-utils(1) manual,ffmpeg-utils}
11053 for the accepted syntax.
11055 If not specified, or the expressed duration is negative, the video is
11056 supposed to be generated forever.
11059 Set the number of decimals to show in the timestamp, only available in the
11060 @code{testsrc} source.
11062 The displayed timestamp value will correspond to the original
11063 timestamp value multiplied by the power of 10 of the specified
11064 value. Default value is 0.
11067 For example the following:
11069 testsrc=duration=5.3:size=qcif:rate=10
11072 will generate a video with a duration of 5.3 seconds, with size
11073 176x144 and a frame rate of 10 frames per second.
11075 The following graph description will generate a red source
11076 with an opacity of 0.2, with size "qcif" and a frame rate of 10
11079 color=c=red@@0.2:s=qcif:r=10
11082 If the input content is to be ignored, @code{nullsrc} can be used. The
11083 following command generates noise in the luminance plane by employing
11084 the @code{geq} filter:
11086 nullsrc=s=256x256, geq=random(1)*255:128:128
11089 @subsection Commands
11091 The @code{color} source supports the following commands:
11095 Set the color of the created image. Accepts the same syntax of the
11096 corresponding @option{color} option.
11099 @c man end VIDEO SOURCES
11101 @chapter Video Sinks
11102 @c man begin VIDEO SINKS
11104 Below is a description of the currently available video sinks.
11106 @section buffersink
11108 Buffer video frames, and make them available to the end of the filter
11111 This sink is mainly intended for programmatic use, in particular
11112 through the interface defined in @file{libavfilter/buffersink.h}
11113 or the options system.
11115 It accepts a pointer to an AVBufferSinkContext structure, which
11116 defines the incoming buffers' formats, to be passed as the opaque
11117 parameter to @code{avfilter_init_filter} for initialization.
11121 Null video sink: do absolutely nothing with the input video. It is
11122 mainly useful as a template and for use in analysis / debugging
11125 @c man end VIDEO SINKS
11127 @chapter Multimedia Filters
11128 @c man begin MULTIMEDIA FILTERS
11130 Below is a description of the currently available multimedia filters.
11132 @section avectorscope
11134 Convert input audio to a video output, representing the audio vector
11137 The filter is used to measure the difference between channels of stereo
11138 audio stream. A monoaural signal, consisting of identical left and right
11139 signal, results in straight vertical line. Any stereo separation is visible
11140 as a deviation from this line, creating a Lissajous figure.
11141 If the straight (or deviation from it) but horizontal line appears this
11142 indicates that the left and right channels are out of phase.
11144 The filter accepts the following options:
11148 Set the vectorscope mode.
11150 Available values are:
11153 Lissajous rotated by 45 degrees.
11156 Same as above but not rotated.
11159 Default value is @samp{lissajous}.
11162 Set the video size for the output. For the syntax of this option, check the
11163 @ref{video size syntax,,"Video size" section in the ffmpeg-utils manual,ffmpeg-utils}.
11164 Default value is @code{400x400}.
11167 Set the output frame rate. Default value is @code{25}.
11172 Specify the red, green and blue contrast. Default values are @code{40}, @code{160} and @code{80}.
11173 Allowed range is @code{[0, 255]}.
11178 Specify the red, green and blue fade. Default values are @code{15}, @code{10} and @code{5}.
11179 Allowed range is @code{[0, 255]}.
11182 Set the zoom factor. Default value is @code{1}. Allowed range is @code{[1, 10]}.
11185 @subsection Examples
11189 Complete example using @command{ffplay}:
11191 ffplay -f lavfi 'amovie=input.mp3, asplit [a][out1];
11192 [a] avectorscope=zoom=1.3:rc=2:gc=200:bc=10:rf=1:gf=8:bf=7 [out0]'
11198 Concatenate audio and video streams, joining them together one after the
11201 The filter works on segments of synchronized video and audio streams. All
11202 segments must have the same number of streams of each type, and that will
11203 also be the number of streams at output.
11205 The filter accepts the following options:
11210 Set the number of segments. Default is 2.
11213 Set the number of output video streams, that is also the number of video
11214 streams in each segment. Default is 1.
11217 Set the number of output audio streams, that is also the number of audio
11218 streams in each segment. Default is 0.
11221 Activate unsafe mode: do not fail if segments have a different format.
11225 The filter has @var{v}+@var{a} outputs: first @var{v} video outputs, then
11226 @var{a} audio outputs.
11228 There are @var{n}x(@var{v}+@var{a}) inputs: first the inputs for the first
11229 segment, in the same order as the outputs, then the inputs for the second
11232 Related streams do not always have exactly the same duration, for various
11233 reasons including codec frame size or sloppy authoring. For that reason,
11234 related synchronized streams (e.g. a video and its audio track) should be
11235 concatenated at once. The concat filter will use the duration of the longest
11236 stream in each segment (except the last one), and if necessary pad shorter
11237 audio streams with silence.
11239 For this filter to work correctly, all segments must start at timestamp 0.
11241 All corresponding streams must have the same parameters in all segments; the
11242 filtering system will automatically select a common pixel format for video
11243 streams, and a common sample format, sample rate and channel layout for
11244 audio streams, but other settings, such as resolution, must be converted
11245 explicitly by the user.
11247 Different frame rates are acceptable but will result in variable frame rate
11248 at output; be sure to configure the output file to handle it.
11250 @subsection Examples
11254 Concatenate an opening, an episode and an ending, all in bilingual version
11255 (video in stream 0, audio in streams 1 and 2):
11257 ffmpeg -i opening.mkv -i episode.mkv -i ending.mkv -filter_complex \
11258 '[0:0] [0:1] [0:2] [1:0] [1:1] [1:2] [2:0] [2:1] [2:2]
11259 concat=n=3:v=1:a=2 [v] [a1] [a2]' \
11260 -map '[v]' -map '[a1]' -map '[a2]' output.mkv
11264 Concatenate two parts, handling audio and video separately, using the
11265 (a)movie sources, and adjusting the resolution:
11267 movie=part1.mp4, scale=512:288 [v1] ; amovie=part1.mp4 [a1] ;
11268 movie=part2.mp4, scale=512:288 [v2] ; amovie=part2.mp4 [a2] ;
11269 [v1] [v2] concat [outv] ; [a1] [a2] concat=v=0:a=1 [outa]
11271 Note that a desync will happen at the stitch if the audio and video streams
11272 do not have exactly the same duration in the first file.
11279 EBU R128 scanner filter. This filter takes an audio stream as input and outputs
11280 it unchanged. By default, it logs a message at a frequency of 10Hz with the
11281 Momentary loudness (identified by @code{M}), Short-term loudness (@code{S}),
11282 Integrated loudness (@code{I}) and Loudness Range (@code{LRA}).
11284 The filter also has a video output (see the @var{video} option) with a real
11285 time graph to observe the loudness evolution. The graphic contains the logged
11286 message mentioned above, so it is not printed anymore when this option is set,
11287 unless the verbose logging is set. The main graphing area contains the
11288 short-term loudness (3 seconds of analysis), and the gauge on the right is for
11289 the momentary loudness (400 milliseconds).
11291 More information about the Loudness Recommendation EBU R128 on
11292 @url{http://tech.ebu.ch/loudness}.
11294 The filter accepts the following options:
11299 Activate the video output. The audio stream is passed unchanged whether this
11300 option is set or no. The video stream will be the first output stream if
11301 activated. Default is @code{0}.
11304 Set the video size. This option is for video only. For the syntax of this
11306 @ref{video size syntax,,"Video size" section in the ffmpeg-utils manual,ffmpeg-utils}.
11307 Default and minimum resolution is @code{640x480}.
11310 Set the EBU scale meter. Default is @code{9}. Common values are @code{9} and
11311 @code{18}, respectively for EBU scale meter +9 and EBU scale meter +18. Any
11312 other integer value between this range is allowed.
11315 Set metadata injection. If set to @code{1}, the audio input will be segmented
11316 into 100ms output frames, each of them containing various loudness information
11317 in metadata. All the metadata keys are prefixed with @code{lavfi.r128.}.
11319 Default is @code{0}.
11322 Force the frame logging level.
11324 Available values are:
11327 information logging level
11329 verbose logging level
11332 By default, the logging level is set to @var{info}. If the @option{video} or
11333 the @option{metadata} options are set, it switches to @var{verbose}.
11338 Available modes can be cumulated (the option is a @code{flag} type). Possible
11342 Disable any peak mode (default).
11344 Enable sample-peak mode.
11346 Simple peak mode looking for the higher sample value. It logs a message
11347 for sample-peak (identified by @code{SPK}).
11349 Enable true-peak mode.
11351 If enabled, the peak lookup is done on an over-sampled version of the input
11352 stream for better peak accuracy. It logs a message for true-peak.
11353 (identified by @code{TPK}) and true-peak per frame (identified by @code{FTPK}).
11354 This mode requires a build with @code{libswresample}.
11359 @subsection Examples
11363 Real-time graph using @command{ffplay}, with a EBU scale meter +18:
11365 ffplay -f lavfi -i "amovie=input.mp3,ebur128=video=1:meter=18 [out0][out1]"
11369 Run an analysis with @command{ffmpeg}:
11371 ffmpeg -nostats -i input.mp3 -filter_complex ebur128 -f null -
11375 @section interleave, ainterleave
11377 Temporally interleave frames from several inputs.
11379 @code{interleave} works with video inputs, @code{ainterleave} with audio.
11381 These filters read frames from several inputs and send the oldest
11382 queued frame to the output.
11384 Input streams must have a well defined, monotonically increasing frame
11387 In order to submit one frame to output, these filters need to enqueue
11388 at least one frame for each input, so they cannot work in case one
11389 input is not yet terminated and will not receive incoming frames.
11391 For example consider the case when one input is a @code{select} filter
11392 which always drop input frames. The @code{interleave} filter will keep
11393 reading from that input, but it will never be able to send new frames
11394 to output until the input will send an end-of-stream signal.
11396 Also, depending on inputs synchronization, the filters will drop
11397 frames in case one input receives more frames than the other ones, and
11398 the queue is already filled.
11400 These filters accept the following options:
11404 Set the number of different inputs, it is 2 by default.
11407 @subsection Examples
11411 Interleave frames belonging to different streams using @command{ffmpeg}:
11413 ffmpeg -i bambi.avi -i pr0n.mkv -filter_complex "[0:v][1:v] interleave" out.avi
11417 Add flickering blur effect:
11419 select='if(gt(random(0), 0.2), 1, 2)':n=2 [tmp], boxblur=2:2, [tmp] interleave
11423 @section perms, aperms
11425 Set read/write permissions for the output frames.
11427 These filters are mainly aimed at developers to test direct path in the
11428 following filter in the filtergraph.
11430 The filters accept the following options:
11434 Select the permissions mode.
11436 It accepts the following values:
11439 Do nothing. This is the default.
11441 Set all the output frames read-only.
11443 Set all the output frames directly writable.
11445 Make the frame read-only if writable, and writable if read-only.
11447 Set each output frame read-only or writable randomly.
11451 Set the seed for the @var{random} mode, must be an integer included between
11452 @code{0} and @code{UINT32_MAX}. If not specified, or if explicitly set to
11453 @code{-1}, the filter will try to use a good random seed on a best effort
11457 Note: in case of auto-inserted filter between the permission filter and the
11458 following one, the permission might not be received as expected in that
11459 following filter. Inserting a @ref{format} or @ref{aformat} filter before the
11460 perms/aperms filter can avoid this problem.
11462 @section select, aselect
11464 Select frames to pass in output.
11466 This filter accepts the following options:
11471 Set expression, which is evaluated for each input frame.
11473 If the expression is evaluated to zero, the frame is discarded.
11475 If the evaluation result is negative or NaN, the frame is sent to the
11476 first output; otherwise it is sent to the output with index
11477 @code{ceil(val)-1}, assuming that the input index starts from 0.
11479 For example a value of @code{1.2} corresponds to the output with index
11480 @code{ceil(1.2)-1 = 2-1 = 1}, that is the second output.
11483 Set the number of outputs. The output to which to send the selected
11484 frame is based on the result of the evaluation. Default value is 1.
11487 The expression can contain the following constants:
11491 The (sequential) number of the filtered frame, starting from 0.
11494 The (sequential) number of the selected frame, starting from 0.
11496 @item prev_selected_n
11497 The sequential number of the last selected frame. It's NAN if undefined.
11500 The timebase of the input timestamps.
11503 The PTS (Presentation TimeStamp) of the filtered video frame,
11504 expressed in @var{TB} units. It's NAN if undefined.
11507 The PTS of the filtered video frame,
11508 expressed in seconds. It's NAN if undefined.
11511 The PTS of the previously filtered video frame. It's NAN if undefined.
11513 @item prev_selected_pts
11514 The PTS of the last previously filtered video frame. It's NAN if undefined.
11516 @item prev_selected_t
11517 The PTS of the last previously selected video frame. It's NAN if undefined.
11520 The PTS of the first video frame in the video. It's NAN if undefined.
11523 The time of the first video frame in the video. It's NAN if undefined.
11525 @item pict_type @emph{(video only)}
11526 The type of the filtered frame. It can assume one of the following
11538 @item interlace_type @emph{(video only)}
11539 The frame interlace type. It can assume one of the following values:
11542 The frame is progressive (not interlaced).
11544 The frame is top-field-first.
11546 The frame is bottom-field-first.
11549 @item consumed_sample_n @emph{(audio only)}
11550 the number of selected samples before the current frame
11552 @item samples_n @emph{(audio only)}
11553 the number of samples in the current frame
11555 @item sample_rate @emph{(audio only)}
11556 the input sample rate
11559 This is 1 if the filtered frame is a key-frame, 0 otherwise.
11562 the position in the file of the filtered frame, -1 if the information
11563 is not available (e.g. for synthetic video)
11565 @item scene @emph{(video only)}
11566 value between 0 and 1 to indicate a new scene; a low value reflects a low
11567 probability for the current frame to introduce a new scene, while a higher
11568 value means the current frame is more likely to be one (see the example below)
11572 The default value of the select expression is "1".
11574 @subsection Examples
11578 Select all frames in input:
11583 The example above is the same as:
11595 Select only I-frames:
11597 select='eq(pict_type\,I)'
11601 Select one frame every 100:
11603 select='not(mod(n\,100))'
11607 Select only frames contained in the 10-20 time interval:
11609 select=between(t\,10\,20)
11613 Select only I frames contained in the 10-20 time interval:
11615 select=between(t\,10\,20)*eq(pict_type\,I)
11619 Select frames with a minimum distance of 10 seconds:
11621 select='isnan(prev_selected_t)+gte(t-prev_selected_t\,10)'
11625 Use aselect to select only audio frames with samples number > 100:
11627 aselect='gt(samples_n\,100)'
11631 Create a mosaic of the first scenes:
11633 ffmpeg -i video.avi -vf select='gt(scene\,0.4)',scale=160:120,tile -frames:v 1 preview.png
11636 Comparing @var{scene} against a value between 0.3 and 0.5 is generally a sane
11640 Send even and odd frames to separate outputs, and compose them:
11642 select=n=2:e='mod(n, 2)+1' [odd][even]; [odd] pad=h=2*ih [tmp]; [tmp][even] overlay=y=h
11646 @section sendcmd, asendcmd
11648 Send commands to filters in the filtergraph.
11650 These filters read commands to be sent to other filters in the
11653 @code{sendcmd} must be inserted between two video filters,
11654 @code{asendcmd} must be inserted between two audio filters, but apart
11655 from that they act the same way.
11657 The specification of commands can be provided in the filter arguments
11658 with the @var{commands} option, or in a file specified by the
11659 @var{filename} option.
11661 These filters accept the following options:
11664 Set the commands to be read and sent to the other filters.
11666 Set the filename of the commands to be read and sent to the other
11670 @subsection Commands syntax
11672 A commands description consists of a sequence of interval
11673 specifications, comprising a list of commands to be executed when a
11674 particular event related to that interval occurs. The occurring event
11675 is typically the current frame time entering or leaving a given time
11678 An interval is specified by the following syntax:
11680 @var{START}[-@var{END}] @var{COMMANDS};
11683 The time interval is specified by the @var{START} and @var{END} times.
11684 @var{END} is optional and defaults to the maximum time.
11686 The current frame time is considered within the specified interval if
11687 it is included in the interval [@var{START}, @var{END}), that is when
11688 the time is greater or equal to @var{START} and is lesser than
11691 @var{COMMANDS} consists of a sequence of one or more command
11692 specifications, separated by ",", relating to that interval. The
11693 syntax of a command specification is given by:
11695 [@var{FLAGS}] @var{TARGET} @var{COMMAND} @var{ARG}
11698 @var{FLAGS} is optional and specifies the type of events relating to
11699 the time interval which enable sending the specified command, and must
11700 be a non-null sequence of identifier flags separated by "+" or "|" and
11701 enclosed between "[" and "]".
11703 The following flags are recognized:
11706 The command is sent when the current frame timestamp enters the
11707 specified interval. In other words, the command is sent when the
11708 previous frame timestamp was not in the given interval, and the
11712 The command is sent when the current frame timestamp leaves the
11713 specified interval. In other words, the command is sent when the
11714 previous frame timestamp was in the given interval, and the
11718 If @var{FLAGS} is not specified, a default value of @code{[enter]} is
11721 @var{TARGET} specifies the target of the command, usually the name of
11722 the filter class or a specific filter instance name.
11724 @var{COMMAND} specifies the name of the command for the target filter.
11726 @var{ARG} is optional and specifies the optional list of argument for
11727 the given @var{COMMAND}.
11729 Between one interval specification and another, whitespaces, or
11730 sequences of characters starting with @code{#} until the end of line,
11731 are ignored and can be used to annotate comments.
11733 A simplified BNF description of the commands specification syntax
11736 @var{COMMAND_FLAG} ::= "enter" | "leave"
11737 @var{COMMAND_FLAGS} ::= @var{COMMAND_FLAG} [(+|"|")@var{COMMAND_FLAG}]
11738 @var{COMMAND} ::= ["[" @var{COMMAND_FLAGS} "]"] @var{TARGET} @var{COMMAND} [@var{ARG}]
11739 @var{COMMANDS} ::= @var{COMMAND} [,@var{COMMANDS}]
11740 @var{INTERVAL} ::= @var{START}[-@var{END}] @var{COMMANDS}
11741 @var{INTERVALS} ::= @var{INTERVAL}[;@var{INTERVALS}]
11744 @subsection Examples
11748 Specify audio tempo change at second 4:
11750 asendcmd=c='4.0 atempo tempo 1.5',atempo
11754 Specify a list of drawtext and hue commands in a file.
11756 # show text in the interval 5-10
11757 5.0-10.0 [enter] drawtext reinit 'fontfile=FreeSerif.ttf:text=hello world',
11758 [leave] drawtext reinit 'fontfile=FreeSerif.ttf:text=';
11760 # desaturate the image in the interval 15-20
11761 15.0-20.0 [enter] hue s 0,
11762 [enter] drawtext reinit 'fontfile=FreeSerif.ttf:text=nocolor',
11764 [leave] drawtext reinit 'fontfile=FreeSerif.ttf:text=color';
11766 # apply an exponential saturation fade-out effect, starting from time 25
11767 25 [enter] hue s exp(25-t)
11770 A filtergraph allowing to read and process the above command list
11771 stored in a file @file{test.cmd}, can be specified with:
11773 sendcmd=f=test.cmd,drawtext=fontfile=FreeSerif.ttf:text='',hue
11778 @section setpts, asetpts
11780 Change the PTS (presentation timestamp) of the input frames.
11782 @code{setpts} works on video frames, @code{asetpts} on audio frames.
11784 This filter accepts the following options:
11789 The expression which is evaluated for each frame to construct its timestamp.
11793 The expression is evaluated through the eval API and can contain the following
11798 frame rate, only defined for constant frame-rate video
11801 The presentation timestamp in input
11804 The count of the input frame for video or the number of consumed samples,
11805 not including the current frame for audio, starting from 0.
11807 @item NB_CONSUMED_SAMPLES
11808 The number of consumed samples, not including the current frame (only
11811 @item NB_SAMPLES, S
11812 The number of samples in the current frame (only audio)
11814 @item SAMPLE_RATE, SR
11815 The audio sample rate.
11818 The PTS of the first frame.
11821 the time in seconds of the first frame
11824 State whether the current frame is interlaced.
11827 the time in seconds of the current frame
11830 original position in the file of the frame, or undefined if undefined
11831 for the current frame
11834 The previous input PTS.
11837 previous input time in seconds
11840 The previous output PTS.
11843 previous output time in seconds
11846 The wallclock (RTC) time in microseconds. This is deprecated, use time(0)
11850 The wallclock (RTC) time at the start of the movie in microseconds.
11853 The timebase of the input timestamps.
11857 @subsection Examples
11861 Start counting PTS from zero
11863 setpts=PTS-STARTPTS
11867 Apply fast motion effect:
11873 Apply slow motion effect:
11879 Set fixed rate of 25 frames per second:
11885 Set fixed rate 25 fps with some jitter:
11887 setpts='1/(25*TB) * (N + 0.05 * sin(N*2*PI/25))'
11891 Apply an offset of 10 seconds to the input PTS:
11897 Generate timestamps from a "live source" and rebase onto the current timebase:
11899 setpts='(RTCTIME - RTCSTART) / (TB * 1000000)'
11903 Generate timestamps by counting samples:
11910 @section settb, asettb
11912 Set the timebase to use for the output frames timestamps.
11913 It is mainly useful for testing timebase configuration.
11915 It accepts the following parameters:
11920 The expression which is evaluated into the output timebase.
11924 The value for @option{tb} is an arithmetic expression representing a
11925 rational. The expression can contain the constants "AVTB" (the default
11926 timebase), "intb" (the input timebase) and "sr" (the sample rate,
11927 audio only). Default value is "intb".
11929 @subsection Examples
11933 Set the timebase to 1/25:
11939 Set the timebase to 1/10:
11945 Set the timebase to 1001/1000:
11951 Set the timebase to 2*intb:
11957 Set the default timebase value:
11964 Convert input audio to a video output representing
11965 frequency spectrum logarithmically (using constant Q transform with
11966 Brown-Puckette algorithm), with musical tone scale, from E0 to D#10 (10 octaves).
11968 The filter accepts the following options:
11972 Specify transform volume (multiplier) expression. The expression can contain
11975 @item frequency, freq, f
11976 the frequency where transform is evaluated
11977 @item timeclamp, tc
11978 value of timeclamp option
11982 @item a_weighting(f)
11983 A-weighting of equal loudness
11984 @item b_weighting(f)
11985 B-weighting of equal loudness
11986 @item c_weighting(f)
11987 C-weighting of equal loudness
11989 Default value is @code{16}.
11992 Specify transform length expression. The expression can contain variables:
11994 @item frequency, freq, f
11995 the frequency where transform is evaluated
11996 @item timeclamp, tc
11997 value of timeclamp option
11999 Default value is @code{384/f*tc/(384/f+tc)}.
12002 Specify the transform timeclamp. At low frequency, there is trade-off between
12003 accuracy in time domain and frequency domain. If timeclamp is lower,
12004 event in time domain is represented more accurately (such as fast bass drum),
12005 otherwise event in frequency domain is represented more accurately
12006 (such as bass guitar). Acceptable value is [0.1, 1.0]. Default value is @code{0.17}.
12009 Specify the transform coeffclamp. If coeffclamp is lower, transform is
12010 more accurate, otherwise transform is faster. Acceptable value is [0.1, 10.0].
12011 Default value is @code{1.0}.
12014 Specify gamma. Lower gamma makes the spectrum more contrast, higher gamma
12015 makes the spectrum having more range. Acceptable value is [1.0, 7.0].
12016 Default value is @code{3.0}.
12019 Specify gamma of bargraph. Acceptable value is [1.0, 7.0].
12020 Default value is @code{1.0}.
12023 Specify font file for use with freetype. If not specified, use embedded font.
12026 Specify font color expression. This is arithmetic expression that should return
12027 integer value 0xRRGGBB. The expression can contain variables:
12029 @item frequency, freq, f
12030 the frequency where transform is evaluated
12031 @item timeclamp, tc
12032 value of timeclamp option
12037 midi number of frequency f, some midi numbers: E0(16), C1(24), C2(36), A4(69)
12038 @item r(x), g(x), b(x)
12039 red, green, and blue value of intensity x
12041 Default value is @code{st(0, (midi(f)-59.5)/12);
12042 st(1, if(between(ld(0),0,1), 0.5-0.5*cos(2*PI*ld(0)), 0));
12043 r(1-ld(1)) + b(ld(1))}
12046 If set to 1 (the default), the video size is 1920x1080 (full HD),
12047 if set to 0, the video size is 960x540. Use this option to make CPU usage lower.
12050 Specify video fps. Default value is @code{25}.
12053 Specify number of transform per frame, so there are fps*count transforms
12054 per second. Note that audio data rate must be divisible by fps*count.
12055 Default value is @code{6}.
12059 @subsection Examples
12063 Playing audio while showing the spectrum:
12065 ffplay -f lavfi 'amovie=a.mp3, asplit [a][out1]; [a] showcqt [out0]'
12069 Same as above, but with frame rate 30 fps:
12071 ffplay -f lavfi 'amovie=a.mp3, asplit [a][out1]; [a] showcqt=fps=30:count=5 [out0]'
12075 Playing at 960x540 and lower CPU usage:
12077 ffplay -f lavfi 'amovie=a.mp3, asplit [a][out1]; [a] showcqt=fullhd=0:count=3 [out0]'
12081 A1 and its harmonics: A1, A2, (near)E3, A3:
12083 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),
12084 asplit[a][out1]; [a] showcqt [out0]'
12088 Same as above, but with more accuracy in frequency domain (and slower):
12090 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),
12091 asplit[a][out1]; [a] showcqt=timeclamp=0.5 [out0]'
12095 B-weighting of equal loudness
12097 volume=16*b_weighting(f)
12103 tlength=100/f*tc/(100/f+tc)
12107 Custom fontcolor, C-note is colored green, others are colored blue
12109 fontcolor='if(mod(floor(midi(f)+0.5),12), 0x0000FF, g(1))'
12113 Custom gamma, now spectrum is linear to the amplitude.
12120 @section showspectrum
12122 Convert input audio to a video output, representing the audio frequency
12125 The filter accepts the following options:
12129 Specify the video size for the output. For the syntax of this option, check the
12130 @ref{video size syntax,,"Video size" section in the ffmpeg-utils manual,ffmpeg-utils}.
12131 Default value is @code{640x512}.
12134 Specify how the spectrum should slide along the window.
12136 It accepts the following values:
12139 the samples start again on the left when they reach the right
12141 the samples scroll from right to left
12143 frames are only produced when the samples reach the right
12146 Default value is @code{replace}.
12149 Specify display mode.
12151 It accepts the following values:
12154 all channels are displayed in the same row
12156 all channels are displayed in separate rows
12159 Default value is @samp{combined}.
12162 Specify display color mode.
12164 It accepts the following values:
12167 each channel is displayed in a separate color
12169 each channel is is displayed using the same color scheme
12172 Default value is @samp{channel}.
12175 Specify scale used for calculating intensity color values.
12177 It accepts the following values:
12182 square root, default
12189 Default value is @samp{sqrt}.
12192 Set saturation modifier for displayed colors. Negative values provide
12193 alternative color scheme. @code{0} is no saturation at all.
12194 Saturation must be in [-10.0, 10.0] range.
12195 Default value is @code{1}.
12198 Set window function.
12200 It accepts the following values:
12203 No samples pre-processing (do not expect this to be faster)
12212 Default value is @code{hann}.
12215 The usage is very similar to the showwaves filter; see the examples in that
12218 @subsection Examples
12222 Large window with logarithmic color scaling:
12224 showspectrum=s=1280x480:scale=log
12228 Complete example for a colored and sliding spectrum per channel using @command{ffplay}:
12230 ffplay -f lavfi 'amovie=input.mp3, asplit [a][out1];
12231 [a] showspectrum=mode=separate:color=intensity:slide=1:scale=cbrt [out0]'
12235 @section showvolume
12237 Convert input audio volume to a video output.
12239 The filter accepts the following options:
12246 Set border width, allowed range is [0, 5]. Default is 1.
12249 Set channel width, allowed range is [40, 1080]. Default is 400.
12252 Set channel height, allowed range is [1, 100]. Default is 20.
12255 Set fade, allowed range is [1, 255]. Default is 20.
12258 Set volume color expression.
12260 The expression can use the following variables:
12264 Current max volume of channel in dB.
12267 Current channel number, starting from 0.
12271 If set, displays channel names. Default is enabled.
12276 Convert input audio to a video output, representing the samples waves.
12278 The filter accepts the following options:
12282 Specify the video size for the output. For the syntax of this option, check the
12283 @ref{video size syntax,,"Video size" section in the ffmpeg-utils manual,ffmpeg-utils}.
12284 Default value is @code{600x240}.
12289 Available values are:
12292 Draw a point for each sample.
12295 Draw a vertical line for each sample.
12298 Draw a point for each sample and a line between them.
12301 Draw a centered vertical line for each sample.
12304 Default value is @code{point}.
12307 Set the number of samples which are printed on the same column. A
12308 larger value will decrease the frame rate. Must be a positive
12309 integer. This option can be set only if the value for @var{rate}
12310 is not explicitly specified.
12313 Set the (approximate) output frame rate. This is done by setting the
12314 option @var{n}. Default value is "25".
12316 @item split_channels
12317 Set if channels should be drawn separately or overlap. Default value is 0.
12321 @subsection Examples
12325 Output the input file audio and the corresponding video representation
12328 amovie=a.mp3,asplit[out0],showwaves[out1]
12332 Create a synthetic signal and show it with showwaves, forcing a
12333 frame rate of 30 frames per second:
12335 aevalsrc=sin(1*2*PI*t)*sin(880*2*PI*t):cos(2*PI*200*t),asplit[out0],showwaves=r=30[out1]
12339 @section showwavespic
12341 Convert input audio to a single video frame, representing the samples waves.
12343 The filter accepts the following options:
12347 Specify the video size for the output. For the syntax of this option, check the
12348 @ref{video size syntax,,"Video size" section in the ffmpeg-utils manual,ffmpeg-utils}.
12349 Default value is @code{600x240}.
12351 @item split_channels
12352 Set if channels should be drawn separately or overlap. Default value is 0.
12355 @subsection Examples
12359 Extract a channel split representation of the wave form of a whole audio track
12360 in a 1024x800 picture using @command{ffmpeg}:
12362 ffmpeg -i audio.flac -lavfi showwavespic=split_channels=1:s=1024x800 waveform.png
12366 @section split, asplit
12368 Split input into several identical outputs.
12370 @code{asplit} works with audio input, @code{split} with video.
12372 The filter accepts a single parameter which specifies the number of outputs. If
12373 unspecified, it defaults to 2.
12375 @subsection Examples
12379 Create two separate outputs from the same input:
12381 [in] split [out0][out1]
12385 To create 3 or more outputs, you need to specify the number of
12388 [in] asplit=3 [out0][out1][out2]
12392 Create two separate outputs from the same input, one cropped and
12395 [in] split [splitout1][splitout2];
12396 [splitout1] crop=100:100:0:0 [cropout];
12397 [splitout2] pad=200:200:100:100 [padout];
12401 Create 5 copies of the input audio with @command{ffmpeg}:
12403 ffmpeg -i INPUT -filter_complex asplit=5 OUTPUT
12409 Receive commands sent through a libzmq client, and forward them to
12410 filters in the filtergraph.
12412 @code{zmq} and @code{azmq} work as a pass-through filters. @code{zmq}
12413 must be inserted between two video filters, @code{azmq} between two
12416 To enable these filters you need to install the libzmq library and
12417 headers and configure FFmpeg with @code{--enable-libzmq}.
12419 For more information about libzmq see:
12420 @url{http://www.zeromq.org/}
12422 The @code{zmq} and @code{azmq} filters work as a libzmq server, which
12423 receives messages sent through a network interface defined by the
12424 @option{bind_address} option.
12426 The received message must be in the form:
12428 @var{TARGET} @var{COMMAND} [@var{ARG}]
12431 @var{TARGET} specifies the target of the command, usually the name of
12432 the filter class or a specific filter instance name.
12434 @var{COMMAND} specifies the name of the command for the target filter.
12436 @var{ARG} is optional and specifies the optional argument list for the
12437 given @var{COMMAND}.
12439 Upon reception, the message is processed and the corresponding command
12440 is injected into the filtergraph. Depending on the result, the filter
12441 will send a reply to the client, adopting the format:
12443 @var{ERROR_CODE} @var{ERROR_REASON}
12447 @var{MESSAGE} is optional.
12449 @subsection Examples
12451 Look at @file{tools/zmqsend} for an example of a zmq client which can
12452 be used to send commands processed by these filters.
12454 Consider the following filtergraph generated by @command{ffplay}
12456 ffplay -dumpgraph 1 -f lavfi "
12457 color=s=100x100:c=red [l];
12458 color=s=100x100:c=blue [r];
12459 nullsrc=s=200x100, zmq [bg];
12460 [bg][l] overlay [bg+l];
12461 [bg+l][r] overlay=x=100 "
12464 To change the color of the left side of the video, the following
12465 command can be used:
12467 echo Parsed_color_0 c yellow | tools/zmqsend
12470 To change the right side:
12472 echo Parsed_color_1 c pink | tools/zmqsend
12475 @c man end MULTIMEDIA FILTERS
12477 @chapter Multimedia Sources
12478 @c man begin MULTIMEDIA SOURCES
12480 Below is a description of the currently available multimedia sources.
12484 This is the same as @ref{movie} source, except it selects an audio
12490 Read audio and/or video stream(s) from a movie container.
12492 It accepts the following parameters:
12496 The name of the resource to read (not necessarily a file; it can also be a
12497 device or a stream accessed through some protocol).
12499 @item format_name, f
12500 Specifies the format assumed for the movie to read, and can be either
12501 the name of a container or an input device. If not specified, the
12502 format is guessed from @var{movie_name} or by probing.
12504 @item seek_point, sp
12505 Specifies the seek point in seconds. The frames will be output
12506 starting from this seek point. The parameter is evaluated with
12507 @code{av_strtod}, so the numerical value may be suffixed by an IS
12508 postfix. The default value is "0".
12511 Specifies the streams to read. Several streams can be specified,
12512 separated by "+". The source will then have as many outputs, in the
12513 same order. The syntax is explained in the ``Stream specifiers''
12514 section in the ffmpeg manual. Two special names, "dv" and "da" specify
12515 respectively the default (best suited) video and audio stream. Default
12516 is "dv", or "da" if the filter is called as "amovie".
12518 @item stream_index, si
12519 Specifies the index of the video stream to read. If the value is -1,
12520 the most suitable video stream will be automatically selected. The default
12521 value is "-1". Deprecated. If the filter is called "amovie", it will select
12522 audio instead of video.
12525 Specifies how many times to read the stream in sequence.
12526 If the value is less than 1, the stream will be read again and again.
12527 Default value is "1".
12529 Note that when the movie is looped the source timestamps are not
12530 changed, so it will generate non monotonically increasing timestamps.
12533 It allows overlaying a second video on top of the main input of
12534 a filtergraph, as shown in this graph:
12536 input -----------> deltapts0 --> overlay --> output
12539 movie --> scale--> deltapts1 -------+
12541 @subsection Examples
12545 Skip 3.2 seconds from the start of the AVI file in.avi, and overlay it
12546 on top of the input labelled "in":
12548 movie=in.avi:seek_point=3.2, scale=180:-1, setpts=PTS-STARTPTS [over];
12549 [in] setpts=PTS-STARTPTS [main];
12550 [main][over] overlay=16:16 [out]
12554 Read from a video4linux2 device, and overlay it on top of the input
12557 movie=/dev/video0:f=video4linux2, scale=180:-1, setpts=PTS-STARTPTS [over];
12558 [in] setpts=PTS-STARTPTS [main];
12559 [main][over] overlay=16:16 [out]
12563 Read the first video stream and the audio stream with id 0x81 from
12564 dvd.vob; the video is connected to the pad named "video" and the audio is
12565 connected to the pad named "audio":
12567 movie=dvd.vob:s=v:0+#0x81 [video] [audio]
12571 @c man end MULTIMEDIA SOURCES