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, 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 in output the top half of the video is mirrored
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 the 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", a filter with no
112 output pads is called a "sink".
114 @anchor{Filtergraph syntax}
115 @section Filtergraph syntax
117 A filtergraph can be represented using a textual representation, which is
118 recognized by the @option{-filter}/@option{-vf} and @option{-filter_complex}
119 options in @command{ffmpeg} and @option{-vf} in @command{ffplay}, and by the
120 @code{avfilter_graph_parse()}/@code{avfilter_graph_parse2()} function defined in
121 @file{libavfilter/avfilter.h}.
123 A filterchain consists of a sequence of connected filters, each one
124 connected to the previous one in the sequence. A filterchain is
125 represented by a list of ","-separated filter descriptions.
127 A filtergraph consists of a sequence of filterchains. A sequence of
128 filterchains is represented by a list of ";"-separated filterchain
131 A filter is represented by a string of the form:
132 [@var{in_link_1}]...[@var{in_link_N}]@var{filter_name}=@var{arguments}[@var{out_link_1}]...[@var{out_link_M}]
134 @var{filter_name} is the name of the filter class of which the
135 described filter is an instance of, and has to be the name of one of
136 the filter classes registered in the program.
137 The name of the filter class is optionally followed by a string
140 @var{arguments} is a string which contains the parameters used to
141 initialize the filter instance. It may have one of the following forms:
145 A ':'-separated list of @var{key=value} pairs.
148 A ':'-separated list of @var{value}. In this case, the keys are assumed to be
149 the option names in the order they are declared. E.g. the @code{fade} filter
150 declares three options in this order -- @option{type}, @option{start_frame} and
151 @option{nb_frames}. Then the parameter list @var{in:0:30} means that the value
152 @var{in} is assigned to the option @option{type}, @var{0} to
153 @option{start_frame} and @var{30} to @option{nb_frames}.
156 A ':'-separated list of mixed direct @var{value} and long @var{key=value}
157 pairs. The direct @var{value} must precede the @var{key=value} pairs, and
158 follow the same constraints order of the previous point. The following
159 @var{key=value} pairs can be set in any preferred order.
163 If the option value itself is a list of items (e.g. the @code{format} filter
164 takes a list of pixel formats), the items in the list are usually separated by
167 The list of arguments can be quoted using the character "'" as initial
168 and ending mark, and the character '\' for escaping the characters
169 within the quoted text; otherwise the argument string is considered
170 terminated when the next special character (belonging to the set
171 "[]=;,") is encountered.
173 The name and arguments of the filter are optionally preceded and
174 followed by a list of link labels.
175 A link label allows to name a link and associate it to a filter output
176 or input pad. The preceding labels @var{in_link_1}
177 ... @var{in_link_N}, are associated to the filter input pads,
178 the following labels @var{out_link_1} ... @var{out_link_M}, are
179 associated to the output pads.
181 When two link labels with the same name are found in the
182 filtergraph, a link between the corresponding input and output pad is
185 If an output pad is not labelled, it is linked by default to the first
186 unlabelled input pad of the next filter in the filterchain.
187 For example in the filterchain:
189 nullsrc, split[L1], [L2]overlay, nullsink
191 the split filter instance has two output pads, and the overlay filter
192 instance two input pads. The first output pad of split is labelled
193 "L1", the first input pad of overlay is labelled "L2", and the second
194 output pad of split is linked to the second input pad of overlay,
195 which are both unlabelled.
197 In a complete filterchain all the unlabelled filter input and output
198 pads must be connected. A filtergraph is considered valid if all the
199 filter input and output pads of all the filterchains are connected.
201 Libavfilter will automatically insert @ref{scale} filters where format
202 conversion is required. It is possible to specify swscale flags
203 for those automatically inserted scalers by prepending
204 @code{sws_flags=@var{flags};}
205 to the filtergraph description.
207 Follows a BNF description for the filtergraph syntax:
209 @var{NAME} ::= sequence of alphanumeric characters and '_'
210 @var{LINKLABEL} ::= "[" @var{NAME} "]"
211 @var{LINKLABELS} ::= @var{LINKLABEL} [@var{LINKLABELS}]
212 @var{FILTER_ARGUMENTS} ::= sequence of chars (eventually quoted)
213 @var{FILTER} ::= [@var{LINKLABELS}] @var{NAME} ["=" @var{FILTER_ARGUMENTS}] [@var{LINKLABELS}]
214 @var{FILTERCHAIN} ::= @var{FILTER} [,@var{FILTERCHAIN}]
215 @var{FILTERGRAPH} ::= [sws_flags=@var{flags};] @var{FILTERCHAIN} [;@var{FILTERGRAPH}]
218 @section Notes on filtergraph escaping
220 Some filter arguments require the use of special characters, typically
221 @code{:} to separate key=value pairs in a named options list. In this
222 case the user should perform a first level escaping when specifying
223 the filter arguments. For example, consider the following literal
224 string to be embedded in the @ref{drawtext} filter arguments:
226 this is a 'string': may contain one, or more, special characters
229 Since @code{:} is special for the filter arguments syntax, it needs to
230 be escaped, so you get:
232 text=this is a \'string\'\: may contain one, or more, special characters
235 A second level of escaping is required when embedding the filter
236 arguments in a filtergraph description, in order to escape all the
237 filtergraph special characters. Thus the example above becomes:
239 drawtext=text=this is a \\\'string\\\'\\: may contain one\, or more\, special characters
242 Finally an additional level of escaping may be needed when writing the
243 filtergraph description in a shell command, which depends on the
244 escaping rules of the adopted shell. For example, assuming that
245 @code{\} is special and needs to be escaped with another @code{\}, the
246 previous string will finally result in:
248 -vf "drawtext=text=this is a \\\\\\'string\\\\\\'\\\\: may contain one\\, or more\\, special characters"
251 Sometimes, it might be more convenient to employ quoting in place of
252 escaping. For example the string:
254 Caesar: tu quoque, Brute, fili mi
257 Can be quoted in the filter arguments as:
259 text='Caesar: tu quoque, Brute, fili mi'
262 And finally inserted in a filtergraph like:
264 drawtext=text=\'Caesar: tu quoque\, Brute\, fili mi\'
267 See the ``Quoting and escaping'' section in the ffmpeg-utils manual
268 for more information about the escaping and quoting rules adopted by
271 @chapter Timeline editing
273 Some filters support a generic @option{enable} option. For the filters
274 supporting timeline editing, this option can be set to an expression which is
275 evaluated before sending a frame to the filter. If the evaluation is non-zero,
276 the filter will be enabled, otherwise the frame will be sent unchanged to the
277 next filter in the filtergraph.
279 The expression accepts the following values:
282 timestamp expressed in seconds, NAN if the input timestamp is unknown
285 sequential number of the input frame, starting from 0
288 the position in the file of the input frame, NAN if unknown
291 Additionally, these filters support an @option{enable} command that can be used
292 to re-define the expression.
294 Like any other filtering option, the @option{enable} option follows the same
297 For example, to enable a blur filter (@ref{smartblur}) from 10 seconds to 3
298 minutes, and a @ref{curves} filter starting at 3 seconds:
300 smartblur = enable='between(t,10,3*60)',
301 curves = enable='gte(t,3)' : preset=cross_process
304 @c man end FILTERGRAPH DESCRIPTION
306 @chapter Audio Filters
307 @c man begin AUDIO FILTERS
309 When you configure your FFmpeg build, you can disable any of the
310 existing filters using @code{--disable-filters}.
311 The configure output will show the audio filters included in your
314 Below is a description of the currently available audio filters.
318 Convert the input audio format to the specified formats.
320 @emph{This filter is deprecated. Use @ref{aformat} instead.}
322 The filter accepts a string of the form:
323 "@var{sample_format}:@var{channel_layout}".
325 @var{sample_format} specifies the sample format, and can be a string or the
326 corresponding numeric value defined in @file{libavutil/samplefmt.h}. Use 'p'
327 suffix for a planar sample format.
329 @var{channel_layout} specifies the channel layout, and can be a string
330 or the corresponding number value defined in @file{libavutil/channel_layout.h}.
332 The special parameter "auto", signifies that the filter will
333 automatically select the output format depending on the output filter.
339 Convert input to float, planar, stereo:
345 Convert input to unsigned 8-bit, automatically select out channel layout:
353 Delay one or more audio channels.
355 Samples in delayed channel are filled with silence.
357 The filter accepts the following option:
361 Set list of delays in milliseconds for each channel separated by '|'.
362 At least one delay greater than 0 should be provided.
363 Unused delays will be silently ignored. If number of given delays is
364 smaller than number of channels all remaining channels will not be delayed.
371 Delay first channel by 1.5 seconds, the third channel by 0.5 seconds and leave
372 the second channel (and any other channels that may be present) unchanged.
380 Apply echoing to the input audio.
382 Echoes are reflected sound and can occur naturally amongst mountains
383 (and sometimes large buildings) when talking or shouting; digital echo
384 effects emulate this behaviour and are often used to help fill out the
385 sound of a single instrument or vocal. The time difference between the
386 original signal and the reflection is the @code{delay}, and the
387 loudness of the reflected signal is the @code{decay}.
388 Multiple echoes can have different delays and decays.
390 A description of the accepted parameters follows.
394 Set input gain of reflected signal. Default is @code{0.6}.
397 Set output gain of reflected signal. Default is @code{0.3}.
400 Set list of time intervals in milliseconds between original signal and reflections
401 separated by '|'. Allowed range for each @code{delay} is @code{(0 - 90000.0]}.
402 Default is @code{1000}.
405 Set list of loudnesses of reflected signals separated by '|'.
406 Allowed range for each @code{decay} is @code{(0 - 1.0]}.
407 Default is @code{0.5}.
414 Make it sound as if there are twice as many instruments as are actually playing:
416 aecho=0.8:0.88:60:0.4
420 If delay is very short, then it sound like a (metallic) robot playing music:
426 A longer delay will sound like an open air concert in the mountains:
428 aecho=0.8:0.9:1000:0.3
432 Same as above but with one more mountain:
434 aecho=0.8:0.9:1000|1800:0.3|0.25
440 Apply fade-in/out effect to input audio.
442 A description of the accepted parameters follows.
446 Specify the effect type, can be either @code{in} for fade-in, or
447 @code{out} for a fade-out effect. Default is @code{in}.
449 @item start_sample, ss
450 Specify the number of the start sample for starting to apply the fade
451 effect. Default is 0.
454 Specify the number of samples for which the fade effect has to last. At
455 the end of the fade-in effect the output audio will have the same
456 volume as the input audio, at the end of the fade-out transition
457 the output audio will be silence. Default is 44100.
460 Specify time for starting to apply the fade effect. Default is 0.
461 The accepted syntax is:
463 [-]HH[:MM[:SS[.m...]]]
466 See also the function @code{av_parse_time()}.
467 If set this option is used instead of @var{start_sample} one.
470 Specify the duration for which the fade effect has to last. Default is 0.
471 The accepted syntax is:
473 [-]HH[:MM[:SS[.m...]]]
476 See also the function @code{av_parse_time()}.
477 At the end of the fade-in effect the output audio will have the same
478 volume as the input audio, at the end of the fade-out transition
479 the output audio will be silence.
480 If set this option is used instead of @var{nb_samples} one.
483 Set curve for fade transition.
485 It accepts the following values:
488 select triangular, linear slope (default)
490 select quarter of sine wave
492 select half of sine wave
494 select exponential sine wave
498 select inverted parabola
514 Fade in first 15 seconds of audio:
520 Fade out last 25 seconds of a 900 seconds audio:
522 afade=t=out:st=875:d=25
529 Set output format constraints for the input audio. The framework will
530 negotiate the most appropriate format to minimize conversions.
532 The filter accepts the following named parameters:
536 A '|'-separated list of requested sample formats.
539 A '|'-separated list of requested sample rates.
541 @item channel_layouts
542 A '|'-separated list of requested channel layouts.
544 See @ref{channel layout syntax,,the Channel Layout section in the ffmpeg-utils(1) manual,ffmpeg-utils}
545 for the required syntax.
548 If a parameter is omitted, all values are allowed.
550 For example to force the output to either unsigned 8-bit or signed 16-bit stereo:
552 aformat=sample_fmts=u8|s16:channel_layouts=stereo
557 Apply a two-pole all-pass filter with central frequency (in Hz)
558 @var{frequency}, and filter-width @var{width}.
559 An all-pass filter changes the audio's frequency to phase relationship
560 without changing its frequency to amplitude relationship.
562 The filter accepts the following options:
569 Set method to specify band-width of filter.
582 Specify the band-width of a filter in width_type units.
587 Merge two or more audio streams into a single multi-channel stream.
589 The filter accepts the following options:
594 Set the number of inputs. Default is 2.
598 If the channel layouts of the inputs are disjoint, and therefore compatible,
599 the channel layout of the output will be set accordingly and the channels
600 will be reordered as necessary. If the channel layouts of the inputs are not
601 disjoint, the output will have all the channels of the first input then all
602 the channels of the second input, in that order, and the channel layout of
603 the output will be the default value corresponding to the total number of
606 For example, if the first input is in 2.1 (FL+FR+LF) and the second input
607 is FC+BL+BR, then the output will be in 5.1, with the channels in the
608 following order: a1, a2, b1, a3, b2, b3 (a1 is the first channel of the
609 first input, b1 is the first channel of the second input).
611 On the other hand, if both input are in stereo, the output channels will be
612 in the default order: a1, a2, b1, b2, and the channel layout will be
613 arbitrarily set to 4.0, which may or may not be the expected value.
615 All inputs must have the same sample rate, and format.
617 If inputs do not have the same duration, the output will stop with the
624 Merge two mono files into a stereo stream:
626 amovie=left.wav [l] ; amovie=right.mp3 [r] ; [l] [r] amerge
630 Multiple merges assuming 1 video stream and 6 audio streams in @file{input.mkv}:
632 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
638 Mixes multiple audio inputs into a single output.
642 ffmpeg -i INPUT1 -i INPUT2 -i INPUT3 -filter_complex amix=inputs=3:duration=first:dropout_transition=3 OUTPUT
644 will mix 3 input audio streams to a single output with the same duration as the
645 first input and a dropout transition time of 3 seconds.
647 The filter accepts the following named parameters:
651 Number of inputs. If unspecified, it defaults to 2.
654 How to determine the end-of-stream.
658 Duration of longest input. (default)
661 Duration of shortest input.
664 Duration of first input.
668 @item dropout_transition
669 Transition time, in seconds, for volume renormalization when an input
670 stream ends. The default value is 2 seconds.
676 Pass the audio source unchanged to the output.
680 Pad the end of a audio stream with silence, this can be used together with
681 -shortest to extend audio streams to the same length as the video stream.
684 Add a phasing effect to the input audio.
686 A phaser filter creates series of peaks and troughs in the frequency spectrum.
687 The position of the peaks and troughs are modulated so that they vary over time, creating a sweeping effect.
689 A description of the accepted parameters follows.
693 Set input gain. Default is 0.4.
696 Set output gain. Default is 0.74
699 Set delay in milliseconds. Default is 3.0.
702 Set decay. Default is 0.4.
705 Set modulation speed in Hz. Default is 0.5.
708 Set modulation type. Default is triangular.
710 It accepts the following values:
720 Resample the input audio to the specified parameters, using the
721 libswresample library. If none are specified then the filter will
722 automatically convert between its input and output.
724 This filter is also able to stretch/squeeze the audio data to make it match
725 the timestamps or to inject silence / cut out audio to make it match the
726 timestamps, do a combination of both or do neither.
728 The filter accepts the syntax
729 [@var{sample_rate}:]@var{resampler_options}, where @var{sample_rate}
730 expresses a sample rate and @var{resampler_options} is a list of
731 @var{key}=@var{value} pairs, separated by ":". See the
732 ffmpeg-resampler manual for the complete list of supported options.
738 Resample the input audio to 44100Hz:
744 Stretch/squeeze samples to the given timestamps, with a maximum of 1000
745 samples per second compensation:
751 @section asetnsamples
753 Set the number of samples per each output audio frame.
755 The last output packet may contain a different number of samples, as
756 the filter will flush all the remaining samples when the input audio
759 The filter accepts the following options:
763 @item nb_out_samples, n
764 Set the number of frames per each output audio frame. The number is
765 intended as the number of samples @emph{per each channel}.
766 Default value is 1024.
769 If set to 1, the filter will pad the last audio frame with zeroes, so
770 that the last frame will contain the same number of samples as the
771 previous ones. Default value is 1.
774 For example, to set the number of per-frame samples to 1234 and
775 disable padding for the last frame, use:
777 asetnsamples=n=1234:p=0
782 Set the sample rate without altering the PCM data.
783 This will result in a change of speed and pitch.
785 The filter accepts the following options:
789 Set the output sample rate. Default is 44100 Hz.
794 Show a line containing various information for each input audio frame.
795 The input audio is not modified.
797 The shown line contains a sequence of key/value pairs of the form
798 @var{key}:@var{value}.
800 A description of each shown parameter follows:
804 sequential number of the input frame, starting from 0
807 Presentation timestamp of the input frame, in time base units; the time base
808 depends on the filter input pad, and is usually 1/@var{sample_rate}.
811 presentation timestamp of the input frame in seconds
814 position of the frame in the input stream, -1 if this information in
815 unavailable and/or meaningless (for example in case of synthetic audio)
824 sample rate for the audio frame
827 number of samples (per channel) in the frame
830 Adler-32 checksum (printed in hexadecimal) of the audio data. For planar audio
831 the data is treated as if all the planes were concatenated.
833 @item plane_checksums
834 A list of Adler-32 checksums for each data plane.
839 Display time domain statistical information about the audio channels.
840 Statistics are calculated and displayed for each audio channel and,
841 where applicable, an overall figure is also given.
843 The filter accepts the following option:
846 Short window length in seconds, used for peak and trough RMS measurement.
847 Default is @code{0.05} (50 miliseconds). Allowed range is @code{[0.1 - 10]}.
850 A description of each shown parameter follows:
854 Mean amplitude displacement from zero.
857 Minimal sample level.
860 Maximal sample level.
864 Standard peak and RMS level measured in dBFS.
868 Peak and trough values for RMS level measured over a short window.
871 Standard ratio of peak to RMS level (note: not in dB).
874 Flatness (i.e. consecutive samples with the same value) of the signal at its peak levels
875 (i.e. either @var{Min level} or @var{Max level}).
878 Number of occasions (not the number of samples) that the signal attained either
879 @var{Min level} or @var{Max level}.
884 Forward two audio streams and control the order the buffers are forwarded.
886 The filter accepts the following options:
890 Set the expression deciding which stream should be
891 forwarded next: if the result is negative, the first stream is forwarded; if
892 the result is positive or zero, the second stream is forwarded. It can use
893 the following variables:
897 number of buffers forwarded so far on each stream
899 number of samples forwarded so far on each stream
901 current timestamp of each stream
904 The default value is @code{t1-t2}, which means to always forward the stream
905 that has a smaller timestamp.
910 Stress-test @code{amerge} by randomly sending buffers on the wrong
911 input, while avoiding too much of a desynchronization:
913 amovie=file.ogg [a] ; amovie=file.mp3 [b] ;
914 [a] [b] astreamsync=(2*random(1))-1+tanh(5*(t1-t2)) [a2] [b2] ;
920 Synchronize audio data with timestamps by squeezing/stretching it and/or
921 dropping samples/adding silence when needed.
923 This filter is not built by default, please use @ref{aresample} to do squeezing/stretching.
925 The filter accepts the following named parameters:
929 Enable stretching/squeezing the data to make it match the timestamps. Disabled
930 by default. When disabled, time gaps are covered with silence.
933 Minimum difference between timestamps and audio data (in seconds) to trigger
934 adding/dropping samples. Default value is 0.1. If you get non-perfect sync with
935 this filter, try setting this parameter to 0.
938 Maximum compensation in samples per second. Relevant only with compensate=1.
942 Assume the first pts should be this value. The time base is 1 / sample rate.
943 This allows for padding/trimming at the start of stream. By default, no
944 assumption is made about the first frame's expected pts, so no padding or
945 trimming is done. For example, this could be set to 0 to pad the beginning with
946 silence if an audio stream starts after the video stream or to trim any samples
947 with a negative pts due to encoder delay.
955 The filter accepts exactly one parameter, the audio tempo. If not
956 specified then the filter will assume nominal 1.0 tempo. Tempo must
957 be in the [0.5, 2.0] range.
963 Slow down audio to 80% tempo:
969 To speed up audio to 125% tempo:
977 Trim the input so that the output contains one continuous subpart of the input.
979 This filter accepts the following options:
982 Specify time of the start of the kept section, i.e. the audio sample
983 with the timestamp @var{start} will be the first sample in the output.
986 Specify time of the first audio sample that will be dropped, i.e. the
987 audio sample immediately preceding the one with the timestamp @var{end} will be
988 the last sample in the output.
991 Same as @var{start}, except this option sets the start timestamp in samples
995 Same as @var{end}, except this option sets the end timestamp in samples instead
999 Specify maximum duration of the output.
1002 Number of the first sample that should be passed to output.
1005 Number of the first sample that should be dropped.
1008 @option{start}, @option{end}, @option{duration} are expressed as time
1009 duration specifications, check the "Time duration" section in the
1010 ffmpeg-utils manual.
1012 Note that the first two sets of the start/end options and the @option{duration}
1013 option look at the frame timestamp, while the _sample options simply count the
1014 samples that pass through the filter. So start/end_pts and start/end_sample will
1015 give different results when the timestamps are wrong, inexact or do not start at
1016 zero. Also note that this filter does not modify the timestamps. If you wish
1017 that the output timestamps start at zero, insert the asetpts filter after the
1020 If multiple start or end options are set, this filter tries to be greedy and
1021 keep all samples that match at least one of the specified constraints. To keep
1022 only the part that matches all the constraints at once, chain multiple atrim
1025 The defaults are such that all the input is kept. So it is possible to set e.g.
1026 just the end values to keep everything before the specified time.
1031 drop everything except the second minute of input
1033 ffmpeg -i INPUT -af atrim=60:120
1037 keep only the first 1000 samples
1039 ffmpeg -i INPUT -af atrim=end_sample=1000
1046 Apply a two-pole Butterworth band-pass filter with central
1047 frequency @var{frequency}, and (3dB-point) band-width width.
1048 The @var{csg} option selects a constant skirt gain (peak gain = Q)
1049 instead of the default: constant 0dB peak gain.
1050 The filter roll off at 6dB per octave (20dB per decade).
1052 The filter accepts the following options:
1056 Set the filter's central frequency. Default is @code{3000}.
1059 Constant skirt gain if set to 1. Defaults to 0.
1062 Set method to specify band-width of filter.
1075 Specify the band-width of a filter in width_type units.
1080 Apply a two-pole Butterworth band-reject filter with central
1081 frequency @var{frequency}, and (3dB-point) band-width @var{width}.
1082 The filter roll off at 6dB per octave (20dB per decade).
1084 The filter accepts the following options:
1088 Set the filter's central frequency. Default is @code{3000}.
1091 Set method to specify band-width of filter.
1104 Specify the band-width of a filter in width_type units.
1109 Boost or cut the bass (lower) frequencies of the audio using a two-pole
1110 shelving filter with a response similar to that of a standard
1111 hi-fi's tone-controls. This is also known as shelving equalisation (EQ).
1113 The filter accepts the following options:
1117 Give the gain at 0 Hz. Its useful range is about -20
1118 (for a large cut) to +20 (for a large boost).
1119 Beware of clipping when using a positive gain.
1122 Set the filter's central frequency and so can be used
1123 to extend or reduce the frequency range to be boosted or cut.
1124 The default value is @code{100} Hz.
1127 Set method to specify band-width of filter.
1140 Determine how steep is the filter's shelf transition.
1145 Apply a biquad IIR filter with the given coefficients.
1146 Where @var{b0}, @var{b1}, @var{b2} and @var{a0}, @var{a1}, @var{a2}
1147 are the numerator and denominator coefficients respectively.
1151 Remap input channels to new locations.
1153 This filter accepts the following named parameters:
1155 @item channel_layout
1156 Channel layout of the output stream.
1159 Map channels from input to output. The argument is a '|'-separated list of
1160 mappings, each in the @code{@var{in_channel}-@var{out_channel}} or
1161 @var{in_channel} form. @var{in_channel} can be either the name of the input
1162 channel (e.g. FL for front left) or its index in the input channel layout.
1163 @var{out_channel} is the name of the output channel or its index in the output
1164 channel layout. If @var{out_channel} is not given then it is implicitly an
1165 index, starting with zero and increasing by one for each mapping.
1168 If no mapping is present, the filter will implicitly map input channels to
1169 output channels preserving index.
1171 For example, assuming a 5.1+downmix input MOV file
1173 ffmpeg -i in.mov -filter 'channelmap=map=DL-FL|DR-FR' out.wav
1175 will create an output WAV file tagged as stereo from the downmix channels of
1178 To fix a 5.1 WAV improperly encoded in AAC's native channel order
1180 ffmpeg -i in.wav -filter 'channelmap=1|2|0|5|3|4:channel_layout=5.1' out.wav
1183 @section channelsplit
1185 Split each channel in input audio stream into a separate output stream.
1187 This filter accepts the following named parameters:
1189 @item channel_layout
1190 Channel layout of the input stream. Default is "stereo".
1193 For example, assuming a stereo input MP3 file
1195 ffmpeg -i in.mp3 -filter_complex channelsplit out.mkv
1197 will create an output Matroska file with two audio streams, one containing only
1198 the left channel and the other the right channel.
1200 To split a 5.1 WAV file into per-channel files
1202 ffmpeg -i in.wav -filter_complex
1203 'channelsplit=channel_layout=5.1[FL][FR][FC][LFE][SL][SR]'
1204 -map '[FL]' front_left.wav -map '[FR]' front_right.wav -map '[FC]'
1205 front_center.wav -map '[LFE]' lfe.wav -map '[SL]' side_left.wav -map '[SR]'
1211 Compress or expand audio dynamic range.
1213 A description of the accepted options follows.
1218 Set list of times in seconds for each channel over which the instantaneous
1219 level of the input signal is averaged to determine its volume.
1220 @option{attacks} refers to increase of volume and @option{decays} refers
1221 to decrease of volume.
1222 For most situations, the attack time (response to the audio getting louder)
1223 should be shorter than the decay time because the human ear is more sensitive
1224 to sudden loud audio than sudden soft audio.
1225 Typical value for attack is @code{0.3} seconds and for decay @code{0.8}
1229 Set list of points for transfer function, specified in dB relative to maximum
1230 possible signal amplitude.
1231 Each key points list need to be defined using the following syntax:
1232 @code{x0/y0 x1/y1 x2/y2 ...}.
1234 The input values must be in strictly increasing order but the transfer
1235 function does not have to be monotonically rising.
1236 The point @code{0/0} is assumed but may be overridden (by @code{0/out-dBn}).
1237 Typical values for the transfer function are @code{-70/-70 -60/-20}.
1240 Set amount for which the points at where adjacent line segments on the
1241 transfer function meet will be rounded. Defaults is @code{0.01}.
1244 Set additional gain in dB to be applied at all points on the transfer function
1245 and allows easy adjustment of the overall gain.
1246 Default is @code{0}.
1249 Set initial volume in dB to be assumed for each channel when filtering starts.
1250 This permits the user to supply a nominal level initially, so that,
1251 for example, a very large gain is not applied to initial signal levels before
1252 the companding has begun to operate. A typical value for audio which is
1253 initially quiet is -90 dB. Default is @code{0}.
1256 Set delay in seconds. Default is @code{0}. The input audio
1257 is analysed immediately, but audio is delayed before being fed to the
1258 volume adjuster. Specifying a delay approximately equal to the attack/decay
1259 times allows the filter to effectively operate in predictive rather than
1263 @subsection Examples
1266 Make music with both quiet and loud passages suitable for listening
1267 in a noisy environment:
1269 compand=.3 .3:1 1:-90/-60 -60/-40 -40/-30 -20/-20:6:0:-90:0.2
1273 Noise-gate for when the noise is at a lower level than the signal:
1275 compand=.1 .1:.2 .2:-900/-900 -50.1/-900 -50/-50:.01:0:-90:.1
1279 Here is another noise-gate, this time for when the noise is at a higher level
1280 than the signal (making it, in some ways, similar to squelch):
1282 compand=.1 .1:.1 .1:-45.1/-45.1 -45/-900 0/-900:.01:45:-90:.1
1288 Make audio easier to listen to on headphones.
1290 This filter adds `cues' to 44.1kHz stereo (i.e. audio CD format) audio
1291 so that when listened to on headphones the stereo image is moved from
1292 inside your head (standard for headphones) to outside and in front of
1293 the listener (standard for speakers).
1299 Apply a two-pole peaking equalisation (EQ) filter. With this
1300 filter, the signal-level at and around a selected frequency can
1301 be increased or decreased, whilst (unlike bandpass and bandreject
1302 filters) that at all other frequencies is unchanged.
1304 In order to produce complex equalisation curves, this filter can
1305 be given several times, each with a different central frequency.
1307 The filter accepts the following options:
1311 Set the filter's central frequency in Hz.
1314 Set method to specify band-width of filter.
1327 Specify the band-width of a filter in width_type units.
1330 Set the required gain or attenuation in dB.
1331 Beware of clipping when using a positive gain.
1336 Apply a high-pass filter with 3dB point frequency.
1337 The filter can be either single-pole, or double-pole (the default).
1338 The filter roll off at 6dB per pole per octave (20dB per pole per decade).
1340 The filter accepts the following options:
1344 Set frequency in Hz. Default is 3000.
1347 Set number of poles. Default is 2.
1350 Set method to specify band-width of filter.
1363 Specify the band-width of a filter in width_type units.
1364 Applies only to double-pole filter.
1365 The default is 0.707q and gives a Butterworth response.
1370 Join multiple input streams into one multi-channel stream.
1372 The filter accepts the following named parameters:
1376 Number of input streams. Defaults to 2.
1378 @item channel_layout
1379 Desired output channel layout. Defaults to stereo.
1382 Map channels from inputs to output. The argument is a '|'-separated list of
1383 mappings, each in the @code{@var{input_idx}.@var{in_channel}-@var{out_channel}}
1384 form. @var{input_idx} is the 0-based index of the input stream. @var{in_channel}
1385 can be either the name of the input channel (e.g. FL for front left) or its
1386 index in the specified input stream. @var{out_channel} is the name of the output
1390 The filter will attempt to guess the mappings when those are not specified
1391 explicitly. It does so by first trying to find an unused matching input channel
1392 and if that fails it picks the first unused input channel.
1394 E.g. to join 3 inputs (with properly set channel layouts)
1396 ffmpeg -i INPUT1 -i INPUT2 -i INPUT3 -filter_complex join=inputs=3 OUTPUT
1399 To build a 5.1 output from 6 single-channel streams:
1401 ffmpeg -i fl -i fr -i fc -i sl -i sr -i lfe -filter_complex
1402 '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'
1408 Load a LADSPA (Linux Audio Developer's Simple Plugin API) plugin.
1410 To enable compilation of this filter you need to configure FFmpeg with
1411 @code{--enable-ladspa}.
1415 Specifies the name of LADSPA plugin library to load. If the environment
1416 variable @env{LADSPA_PATH} is defined, the LADSPA plugin is searched in
1417 each one of the directories specified by the colon separated list in
1418 @env{LADSPA_PATH}, otherwise in the standard LADSPA paths, which are in
1419 this order: @file{HOME/.ladspa/lib/}, @file{/usr/local/lib/ladspa/},
1420 @file{/usr/lib/ladspa/}.
1423 Specifies the plugin within the library. Some libraries contain only
1424 one plugin, but others contain many of them. If this is not set filter
1425 will list all available plugins within the specified library.
1428 Set the '|' separated list of controls which are zero or more floating point
1429 values that determine the behavior of the loaded plugin (for example delay,
1431 Controls need to be defined using the following syntax:
1432 c0=@var{value0}|c1=@var{value1}|c2=@var{value2}|..., where
1433 @var{valuei} is the value set on the @var{i}-th control.
1434 If @option{controls} is set to @code{help}, all available controls and
1435 their valid ranges are printed.
1437 @item sample_rate, s
1438 Specify the sample rate, default to 44100. Only used if plugin have
1442 Set the number of samples per channel per each output frame, default
1443 is 1024. Only used if plugin have zero inputs.
1446 Set the minimum duration of the sourced audio. See the function
1447 @code{av_parse_time()} for the accepted format, also check the "Time duration"
1448 section in the ffmpeg-utils manual.
1449 Note that the resulting duration may be greater than the specified duration,
1450 as the generated audio is always cut at the end of a complete frame.
1451 If not specified, or the expressed duration is negative, the audio is
1452 supposed to be generated forever.
1453 Only used if plugin have zero inputs.
1457 @subsection Examples
1461 List all available plugins within amp (LADSPA example plugin) library:
1467 List all available controls and their valid ranges for @code{vcf_notch}
1468 plugin from @code{VCF} library:
1470 ladspa=f=vcf:p=vcf_notch:c=help
1474 Simulate low quality audio equipment using @code{Computer Music Toolkit} (CMT)
1477 ladspa=file=cmt:plugin=lofi:controls=c0=22|c1=12|c2=12
1481 Add reverberation to the audio using TAP-plugins
1482 (Tom's Audio Processing plugins):
1484 ladspa=file=tap_reverb:tap_reverb
1488 Generate white noise, with 0.2 amplitude:
1490 ladspa=file=cmt:noise_source_white:c=c0=.2
1494 Generate 20 bpm clicks using plugin @code{C* Click - Metronome} from the
1495 @code{C* Audio Plugin Suite} (CAPS) library:
1497 ladspa=file=caps:Click:c=c1=20'
1501 Apply @code{C* Eq10X2 - Stereo 10-band equaliser} effect:
1503 ladspa=caps:Eq10X2:c=c0=-48|c9=-24|c3=12|c4=2
1507 @subsection Commands
1509 This filter supports the following commands:
1512 Modify the @var{N}-th control value.
1514 If the specified value is not valid, it is ignored and prior one is kept.
1519 Apply a low-pass filter with 3dB point frequency.
1520 The filter can be either single-pole or double-pole (the default).
1521 The filter roll off at 6dB per pole per octave (20dB per pole per decade).
1523 The filter accepts the following options:
1527 Set frequency in Hz. Default is 500.
1530 Set number of poles. Default is 2.
1533 Set method to specify band-width of filter.
1546 Specify the band-width of a filter in width_type units.
1547 Applies only to double-pole filter.
1548 The default is 0.707q and gives a Butterworth response.
1553 Mix channels with specific gain levels. The filter accepts the output
1554 channel layout followed by a set of channels definitions.
1556 This filter is also designed to remap efficiently the channels of an audio
1559 The filter accepts parameters of the form:
1560 "@var{l}:@var{outdef}:@var{outdef}:..."
1564 output channel layout or number of channels
1567 output channel specification, of the form:
1568 "@var{out_name}=[@var{gain}*]@var{in_name}[+[@var{gain}*]@var{in_name}...]"
1571 output channel to define, either a channel name (FL, FR, etc.) or a channel
1572 number (c0, c1, etc.)
1575 multiplicative coefficient for the channel, 1 leaving the volume unchanged
1578 input channel to use, see out_name for details; it is not possible to mix
1579 named and numbered input channels
1582 If the `=' in a channel specification is replaced by `<', then the gains for
1583 that specification will be renormalized so that the total is 1, thus
1584 avoiding clipping noise.
1586 @subsection Mixing examples
1588 For example, if you want to down-mix from stereo to mono, but with a bigger
1589 factor for the left channel:
1591 pan=1:c0=0.9*c0+0.1*c1
1594 A customized down-mix to stereo that works automatically for 3-, 4-, 5- and
1595 7-channels surround:
1597 pan=stereo: FL < FL + 0.5*FC + 0.6*BL + 0.6*SL : FR < FR + 0.5*FC + 0.6*BR + 0.6*SR
1600 Note that @command{ffmpeg} integrates a default down-mix (and up-mix) system
1601 that should be preferred (see "-ac" option) unless you have very specific
1604 @subsection Remapping examples
1606 The channel remapping will be effective if, and only if:
1609 @item gain coefficients are zeroes or ones,
1610 @item only one input per channel output,
1613 If all these conditions are satisfied, the filter will notify the user ("Pure
1614 channel mapping detected"), and use an optimized and lossless method to do the
1617 For example, if you have a 5.1 source and want a stereo audio stream by
1618 dropping the extra channels:
1620 pan="stereo: c0=FL : c1=FR"
1623 Given the same source, you can also switch front left and front right channels
1624 and keep the input channel layout:
1626 pan="5.1: c0=c1 : c1=c0 : c2=c2 : c3=c3 : c4=c4 : c5=c5"
1629 If the input is a stereo audio stream, you can mute the front left channel (and
1630 still keep the stereo channel layout) with:
1635 Still with a stereo audio stream input, you can copy the right channel in both
1636 front left and right:
1638 pan="stereo: c0=FR : c1=FR"
1643 ReplayGain scanner filter. This filter takes an audio stream as an input and
1644 outputs it unchanged.
1645 At end of filtering it displays @code{track_gain} and @code{track_peak}.
1649 Convert the audio sample format, sample rate and channel layout. This filter is
1650 not meant to be used directly.
1652 @section silencedetect
1654 Detect silence in an audio stream.
1656 This filter logs a message when it detects that the input audio volume is less
1657 or equal to a noise tolerance value for a duration greater or equal to the
1658 minimum detected noise duration.
1660 The printed times and duration are expressed in seconds.
1662 The filter accepts the following options:
1666 Set silence duration until notification (default is 2 seconds).
1669 Set noise tolerance. Can be specified in dB (in case "dB" is appended to the
1670 specified value) or amplitude ratio. Default is -60dB, or 0.001.
1673 @subsection Examples
1677 Detect 5 seconds of silence with -50dB noise tolerance:
1679 silencedetect=n=-50dB:d=5
1683 Complete example with @command{ffmpeg} to detect silence with 0.0001 noise
1684 tolerance in @file{silence.mp3}:
1686 ffmpeg -i silence.mp3 -af silencedetect=noise=0.0001 -f null -
1692 Boost or cut treble (upper) frequencies of the audio using a two-pole
1693 shelving filter with a response similar to that of a standard
1694 hi-fi's tone-controls. This is also known as shelving equalisation (EQ).
1696 The filter accepts the following options:
1700 Give the gain at whichever is the lower of ~22 kHz and the
1701 Nyquist frequency. Its useful range is about -20 (for a large cut)
1702 to +20 (for a large boost). Beware of clipping when using a positive gain.
1705 Set the filter's central frequency and so can be used
1706 to extend or reduce the frequency range to be boosted or cut.
1707 The default value is @code{3000} Hz.
1710 Set method to specify band-width of filter.
1723 Determine how steep is the filter's shelf transition.
1728 Adjust the input audio volume.
1730 The filter accepts the following options:
1735 Expresses how the audio volume will be increased or decreased.
1737 Output values are clipped to the maximum value.
1739 The output audio volume is given by the relation:
1741 @var{output_volume} = @var{volume} * @var{input_volume}
1744 Default value for @var{volume} is 1.0.
1747 Set the mathematical precision.
1749 This determines which input sample formats will be allowed, which affects the
1750 precision of the volume scaling.
1754 8-bit fixed-point; limits input sample format to U8, S16, and S32.
1756 32-bit floating-point; limits input sample format to FLT. (default)
1758 64-bit floating-point; limits input sample format to DBL.
1762 @subsection Examples
1766 Halve the input audio volume:
1770 volume=volume=-6.0206dB
1773 In all the above example the named key for @option{volume} can be
1774 omitted, for example like in:
1780 Increase input audio power by 6 decibels using fixed-point precision:
1782 volume=volume=6dB:precision=fixed
1786 @section volumedetect
1788 Detect the volume of the input video.
1790 The filter has no parameters. The input is not modified. Statistics about
1791 the volume will be printed in the log when the input stream end is reached.
1793 In particular it will show the mean volume (root mean square), maximum
1794 volume (on a per-sample basis), and the beginning of a histogram of the
1795 registered volume values (from the maximum value to a cumulated 1/1000 of
1798 All volumes are in decibels relative to the maximum PCM value.
1800 @subsection Examples
1802 Here is an excerpt of the output:
1804 [Parsed_volumedetect_0 @ 0xa23120] mean_volume: -27 dB
1805 [Parsed_volumedetect_0 @ 0xa23120] max_volume: -4 dB
1806 [Parsed_volumedetect_0 @ 0xa23120] histogram_4db: 6
1807 [Parsed_volumedetect_0 @ 0xa23120] histogram_5db: 62
1808 [Parsed_volumedetect_0 @ 0xa23120] histogram_6db: 286
1809 [Parsed_volumedetect_0 @ 0xa23120] histogram_7db: 1042
1810 [Parsed_volumedetect_0 @ 0xa23120] histogram_8db: 2551
1811 [Parsed_volumedetect_0 @ 0xa23120] histogram_9db: 4609
1812 [Parsed_volumedetect_0 @ 0xa23120] histogram_10db: 8409
1818 The mean square energy is approximately -27 dB, or 10^-2.7.
1820 The largest sample is at -4 dB, or more precisely between -4 dB and -5 dB.
1822 There are 6 samples at -4 dB, 62 at -5 dB, 286 at -6 dB, etc.
1825 In other words, raising the volume by +4 dB does not cause any clipping,
1826 raising it by +5 dB causes clipping for 6 samples, etc.
1828 @c man end AUDIO FILTERS
1830 @chapter Audio Sources
1831 @c man begin AUDIO SOURCES
1833 Below is a description of the currently available audio sources.
1837 Buffer audio frames, and make them available to the filter chain.
1839 This source is mainly intended for a programmatic use, in particular
1840 through the interface defined in @file{libavfilter/asrc_abuffer.h}.
1842 It accepts the following named parameters:
1847 Timebase which will be used for timestamps of submitted frames. It must be
1848 either a floating-point number or in @var{numerator}/@var{denominator} form.
1851 The sample rate of the incoming audio buffers.
1854 The sample format of the incoming audio buffers.
1855 Either a sample format name or its corresponging integer representation from
1856 the enum AVSampleFormat in @file{libavutil/samplefmt.h}
1858 @item channel_layout
1859 The channel layout of the incoming audio buffers.
1860 Either a channel layout name from channel_layout_map in
1861 @file{libavutil/channel_layout.c} or its corresponding integer representation
1862 from the AV_CH_LAYOUT_* macros in @file{libavutil/channel_layout.h}
1865 The number of channels of the incoming audio buffers.
1866 If both @var{channels} and @var{channel_layout} are specified, then they
1871 @subsection Examples
1874 abuffer=sample_rate=44100:sample_fmt=s16p:channel_layout=stereo
1877 will instruct the source to accept planar 16bit signed stereo at 44100Hz.
1878 Since the sample format with name "s16p" corresponds to the number
1879 6 and the "stereo" channel layout corresponds to the value 0x3, this is
1882 abuffer=sample_rate=44100:sample_fmt=6:channel_layout=0x3
1887 Generate an audio signal specified by an expression.
1889 This source accepts in input one or more expressions (one for each
1890 channel), which are evaluated and used to generate a corresponding
1893 This source accepts the following options:
1897 Set the '|'-separated expressions list for each separate channel. In case the
1898 @option{channel_layout} option is not specified, the selected channel layout
1899 depends on the number of provided expressions.
1901 @item channel_layout, c
1902 Set the channel layout. The number of channels in the specified layout
1903 must be equal to the number of specified expressions.
1906 Set the minimum duration of the sourced audio. See the function
1907 @code{av_parse_time()} for the accepted format.
1908 Note that the resulting duration may be greater than the specified
1909 duration, as the generated audio is always cut at the end of a
1912 If not specified, or the expressed duration is negative, the audio is
1913 supposed to be generated forever.
1916 Set the number of samples per channel per each output frame,
1919 @item sample_rate, s
1920 Specify the sample rate, default to 44100.
1923 Each expression in @var{exprs} can contain the following constants:
1927 number of the evaluated sample, starting from 0
1930 time of the evaluated sample expressed in seconds, starting from 0
1937 @subsection Examples
1947 Generate a sin signal with frequency of 440 Hz, set sample rate to
1950 aevalsrc="sin(440*2*PI*t):s=8000"
1954 Generate a two channels signal, specify the channel layout (Front
1955 Center + Back Center) explicitly:
1957 aevalsrc="sin(420*2*PI*t)|cos(430*2*PI*t):c=FC|BC"
1961 Generate white noise:
1963 aevalsrc="-2+random(0)"
1967 Generate an amplitude modulated signal:
1969 aevalsrc="sin(10*2*PI*t)*sin(880*2*PI*t)"
1973 Generate 2.5 Hz binaural beats on a 360 Hz carrier:
1975 aevalsrc="0.1*sin(2*PI*(360-2.5/2)*t) | 0.1*sin(2*PI*(360+2.5/2)*t)"
1982 Null audio source, return unprocessed audio frames. It is mainly useful
1983 as a template and to be employed in analysis / debugging tools, or as
1984 the source for filters which ignore the input data (for example the sox
1987 This source accepts the following options:
1991 @item channel_layout, cl
1993 Specify the channel layout, and can be either an integer or a string
1994 representing a channel layout. The default value of @var{channel_layout}
1997 Check the channel_layout_map definition in
1998 @file{libavutil/channel_layout.c} for the mapping between strings and
1999 channel layout values.
2001 @item sample_rate, r
2002 Specify the sample rate, and defaults to 44100.
2005 Set the number of samples per requested frames.
2009 @subsection Examples
2013 Set the sample rate to 48000 Hz and the channel layout to AV_CH_LAYOUT_MONO.
2015 anullsrc=r=48000:cl=4
2019 Do the same operation with a more obvious syntax:
2021 anullsrc=r=48000:cl=mono
2025 All the parameters need to be explicitly defined.
2029 Synthesize a voice utterance using the libflite library.
2031 To enable compilation of this filter you need to configure FFmpeg with
2032 @code{--enable-libflite}.
2034 Note that the flite library is not thread-safe.
2036 The filter accepts the following options:
2041 If set to 1, list the names of the available voices and exit
2042 immediately. Default value is 0.
2045 Set the maximum number of samples per frame. Default value is 512.
2048 Set the filename containing the text to speak.
2051 Set the text to speak.
2054 Set the voice to use for the speech synthesis. Default value is
2055 @code{kal}. See also the @var{list_voices} option.
2058 @subsection Examples
2062 Read from file @file{speech.txt}, and synthetize the text using the
2063 standard flite voice:
2065 flite=textfile=speech.txt
2069 Read the specified text selecting the @code{slt} voice:
2071 flite=text='So fare thee well, poor devil of a Sub-Sub, whose commentator I am':voice=slt
2075 Input text to ffmpeg:
2077 ffmpeg -f lavfi -i flite=text='So fare thee well, poor devil of a Sub-Sub, whose commentator I am':voice=slt
2081 Make @file{ffplay} speak the specified text, using @code{flite} and
2082 the @code{lavfi} device:
2084 ffplay -f lavfi flite=text='No more be grieved for which that thou hast done.'
2088 For more information about libflite, check:
2089 @url{http://www.speech.cs.cmu.edu/flite/}
2093 Generate an audio signal made of a sine wave with amplitude 1/8.
2095 The audio signal is bit-exact.
2097 The filter accepts the following options:
2102 Set the carrier frequency. Default is 440 Hz.
2104 @item beep_factor, b
2105 Enable a periodic beep every second with frequency @var{beep_factor} times
2106 the carrier frequency. Default is 0, meaning the beep is disabled.
2108 @item sample_rate, r
2109 Specify the sample rate, default is 44100.
2112 Specify the duration of the generated audio stream.
2114 @item samples_per_frame
2115 Set the number of samples per output frame, default is 1024.
2118 @subsection Examples
2123 Generate a simple 440 Hz sine wave:
2129 Generate a 220 Hz sine wave with a 880 Hz beep each second, for 5 seconds:
2133 sine=frequency=220:beep_factor=4:duration=5
2138 @c man end AUDIO SOURCES
2140 @chapter Audio Sinks
2141 @c man begin AUDIO SINKS
2143 Below is a description of the currently available audio sinks.
2145 @section abuffersink
2147 Buffer audio frames, and make them available to the end of filter chain.
2149 This sink is mainly intended for programmatic use, in particular
2150 through the interface defined in @file{libavfilter/buffersink.h}
2151 or the options system.
2153 It accepts a pointer to an AVABufferSinkContext structure, which
2154 defines the incoming buffers' formats, to be passed as the opaque
2155 parameter to @code{avfilter_init_filter} for initialization.
2159 Null audio sink, do absolutely nothing with the input audio. It is
2160 mainly useful as a template and to be employed in analysis / debugging
2163 @c man end AUDIO SINKS
2165 @chapter Video Filters
2166 @c man begin VIDEO FILTERS
2168 When you configure your FFmpeg build, you can disable any of the
2169 existing filters using @code{--disable-filters}.
2170 The configure output will show the video filters included in your
2173 Below is a description of the currently available video filters.
2175 @section alphaextract
2177 Extract the alpha component from the input as a grayscale video. This
2178 is especially useful with the @var{alphamerge} filter.
2182 Add or replace the alpha component of the primary input with the
2183 grayscale value of a second input. This is intended for use with
2184 @var{alphaextract} to allow the transmission or storage of frame
2185 sequences that have alpha in a format that doesn't support an alpha
2188 For example, to reconstruct full frames from a normal YUV-encoded video
2189 and a separate video created with @var{alphaextract}, you might use:
2191 movie=in_alpha.mkv [alpha]; [in][alpha] alphamerge [out]
2194 Since this filter is designed for reconstruction, it operates on frame
2195 sequences without considering timestamps, and terminates when either
2196 input reaches end of stream. This will cause problems if your encoding
2197 pipeline drops frames. If you're trying to apply an image as an
2198 overlay to a video stream, consider the @var{overlay} filter instead.
2202 Same as the @ref{subtitles} filter, except that it doesn't require libavcodec
2203 and libavformat to work. On the other hand, it is limited to ASS (Advanced
2204 Substation Alpha) subtitles files.
2208 Compute the bounding box for the non-black pixels in the input frame
2211 This filter computes the bounding box containing all the pixels with a
2212 luminance value greater than the minimum allowed value.
2213 The parameters describing the bounding box are printed on the filter
2216 The filter accepts the following option:
2220 Set the minimal luminance value. Default is @code{16}.
2223 @section blackdetect
2225 Detect video intervals that are (almost) completely black. Can be
2226 useful to detect chapter transitions, commercials, or invalid
2227 recordings. Output lines contains the time for the start, end and
2228 duration of the detected black interval expressed in seconds.
2230 In order to display the output lines, you need to set the loglevel at
2231 least to the AV_LOG_INFO value.
2233 The filter accepts the following options:
2236 @item black_min_duration, d
2237 Set the minimum detected black duration expressed in seconds. It must
2238 be a non-negative floating point number.
2240 Default value is 2.0.
2242 @item picture_black_ratio_th, pic_th
2243 Set the threshold for considering a picture "black".
2244 Express the minimum value for the ratio:
2246 @var{nb_black_pixels} / @var{nb_pixels}
2249 for which a picture is considered black.
2250 Default value is 0.98.
2252 @item pixel_black_th, pix_th
2253 Set the threshold for considering a pixel "black".
2255 The threshold expresses the maximum pixel luminance value for which a
2256 pixel is considered "black". The provided value is scaled according to
2257 the following equation:
2259 @var{absolute_threshold} = @var{luminance_minimum_value} + @var{pixel_black_th} * @var{luminance_range_size}
2262 @var{luminance_range_size} and @var{luminance_minimum_value} depend on
2263 the input video format, the range is [0-255] for YUV full-range
2264 formats and [16-235] for YUV non full-range formats.
2266 Default value is 0.10.
2269 The following example sets the maximum pixel threshold to the minimum
2270 value, and detects only black intervals of 2 or more seconds:
2272 blackdetect=d=2:pix_th=0.00
2277 Detect frames that are (almost) completely black. Can be useful to
2278 detect chapter transitions or commercials. Output lines consist of
2279 the frame number of the detected frame, the percentage of blackness,
2280 the position in the file if known or -1 and the timestamp in seconds.
2282 In order to display the output lines, you need to set the loglevel at
2283 least to the AV_LOG_INFO value.
2285 The filter accepts the following options:
2290 Set the percentage of the pixels that have to be below the threshold, defaults
2293 @item threshold, thresh
2294 Set the threshold below which a pixel value is considered black, defaults to
2301 Blend two video frames into each other.
2303 It takes two input streams and outputs one stream, the first input is the
2304 "top" layer and second input is "bottom" layer.
2305 Output terminates when shortest input terminates.
2307 A description of the accepted options follows.
2315 Set blend mode for specific pixel component or all pixel components in case
2316 of @var{all_mode}. Default value is @code{normal}.
2318 Available values for component modes are:
2351 Set blend opacity for specific pixel component or all pixel components in case
2352 of @var{all_opacity}. Only used in combination with pixel component blend modes.
2359 Set blend expression for specific pixel component or all pixel components in case
2360 of @var{all_expr}. Note that related mode options will be ignored if those are set.
2362 The expressions can use the following variables:
2366 The sequential number of the filtered frame, starting from @code{0}.
2370 the coordinates of the current sample
2374 the width and height of currently filtered plane
2378 Width and height scale depending on the currently filtered plane. It is the
2379 ratio between the corresponding luma plane number of pixels and the current
2380 plane ones. E.g. for YUV4:2:0 the values are @code{1,1} for the luma plane, and
2381 @code{0.5,0.5} for chroma planes.
2384 Time of the current frame, expressed in seconds.
2387 Value of pixel component at current location for first video frame (top layer).
2390 Value of pixel component at current location for second video frame (bottom layer).
2394 Force termination when the shortest input terminates. Default is @code{0}.
2396 Continue applying the last bottom frame after the end of the stream. A value of
2397 @code{0} disable the filter after the last frame of the bottom layer is reached.
2398 Default is @code{1}.
2401 @subsection Examples
2405 Apply transition from bottom layer to top layer in first 10 seconds:
2407 blend=all_expr='A*(if(gte(T,10),1,T/10))+B*(1-(if(gte(T,10),1,T/10)))'
2411 Apply 1x1 checkerboard effect:
2413 blend=all_expr='if(eq(mod(X,2),mod(Y,2)),A,B)'
2417 Apply uncover left effect:
2419 blend=all_expr='if(gte(N*SW+X,W),A,B)'
2423 Apply uncover down effect:
2425 blend=all_expr='if(gte(Y-N*SH,0),A,B)'
2429 Apply uncover up-left effect:
2431 blend=all_expr='if(gte(T*SH*40+Y,H)*gte((T*40*SW+X)*W/H,W),A,B)'
2437 Apply boxblur algorithm to the input video.
2439 The filter accepts the following options:
2443 @item luma_radius, lr
2444 @item luma_power, lp
2445 @item chroma_radius, cr
2446 @item chroma_power, cp
2447 @item alpha_radius, ar
2448 @item alpha_power, ap
2452 A description of the accepted options follows.
2455 @item luma_radius, lr
2456 @item chroma_radius, cr
2457 @item alpha_radius, ar
2458 Set an expression for the box radius in pixels used for blurring the
2459 corresponding input plane.
2461 The radius value must be a non-negative number, and must not be
2462 greater than the value of the expression @code{min(w,h)/2} for the
2463 luma and alpha planes, and of @code{min(cw,ch)/2} for the chroma
2466 Default value for @option{luma_radius} is "2". If not specified,
2467 @option{chroma_radius} and @option{alpha_radius} default to the
2468 corresponding value set for @option{luma_radius}.
2470 The expressions can contain the following constants:
2474 the input width and height in pixels
2478 the input chroma image width and height in pixels
2482 horizontal and vertical chroma subsample values. For example for the
2483 pixel format "yuv422p" @var{hsub} is 2 and @var{vsub} is 1.
2486 @item luma_power, lp
2487 @item chroma_power, cp
2488 @item alpha_power, ap
2489 Specify how many times the boxblur filter is applied to the
2490 corresponding plane.
2492 Default value for @option{luma_power} is 2. If not specified,
2493 @option{chroma_power} and @option{alpha_power} default to the
2494 corresponding value set for @option{luma_power}.
2496 A value of 0 will disable the effect.
2499 @subsection Examples
2503 Apply a boxblur filter with luma, chroma, and alpha radius
2506 boxblur=luma_radius=2:luma_power=1
2511 Set luma radius to 2, alpha and chroma radius to 0:
2513 boxblur=2:1:cr=0:ar=0
2517 Set luma and chroma radius to a fraction of the video dimension:
2519 boxblur=luma_radius=min(h\,w)/10:luma_power=1:chroma_radius=min(cw\,ch)/10:chroma_power=1
2523 @section colorbalance
2524 Modify intensity of primary colors (red, green and blue) of input frames.
2526 The filter allows an input frame to be adjusted in the shadows, midtones or highlights
2527 regions for the red-cyan, green-magenta or blue-yellow balance.
2529 A positive adjustment value shifts the balance towards the primary color, a negative
2530 value towards the complementary color.
2532 The filter accepts the following options:
2538 Adjust red, green and blue shadows (darkest pixels).
2543 Adjust red, green and blue midtones (medium pixels).
2548 Adjust red, green and blue highlights (brightest pixels).
2550 Allowed ranges for options are @code{[-1.0, 1.0]}. Defaults are @code{0}.
2553 @subsection Examples
2557 Add red color cast to shadows:
2563 @section colorchannelmixer
2565 Adjust video input frames by re-mixing color channels.
2567 This filter modifies a color channel by adding the values associated to
2568 the other channels of the same pixels. For example if the value to
2569 modify is red, the output value will be:
2571 @var{red}=@var{red}*@var{rr} + @var{blue}*@var{rb} + @var{green}*@var{rg} + @var{alpha}*@var{ra}
2574 The filter accepts the following options:
2581 Adjust contribution of input red, green, blue and alpha channels for output red channel.
2582 Default is @code{1} for @var{rr}, and @code{0} for @var{rg}, @var{rb} and @var{ra}.
2588 Adjust contribution of input red, green, blue and alpha channels for output green channel.
2589 Default is @code{1} for @var{gg}, and @code{0} for @var{gr}, @var{gb} and @var{ga}.
2595 Adjust contribution of input red, green, blue and alpha channels for output blue channel.
2596 Default is @code{1} for @var{bb}, and @code{0} for @var{br}, @var{bg} and @var{ba}.
2602 Adjust contribution of input red, green, blue and alpha channels for output alpha channel.
2603 Default is @code{1} for @var{aa}, and @code{0} for @var{ar}, @var{ag} and @var{ab}.
2605 Allowed ranges for options are @code{[-2.0, 2.0]}.
2608 @subsection Examples
2612 Convert source to grayscale:
2614 colorchannelmixer=.3:.4:.3:0:.3:.4:.3:0:.3:.4:.3
2617 Simulate sepia tones:
2619 colorchannelmixer=.393:.769:.189:0:.349:.686:.168:0:.272:.534:.131
2623 @section colormatrix
2625 Convert color matrix.
2627 The filter accepts the following options:
2632 Specify the source and destination color matrix. Both values must be
2635 The accepted values are:
2651 For example to convert from BT.601 to SMPTE-240M, use the command:
2653 colormatrix=bt601:smpte240m
2658 Copy the input source unchanged to the output. Mainly useful for
2663 Crop the input video to given dimensions.
2665 The filter accepts the following options:
2669 Width of the output video. It defaults to @code{iw}.
2670 This expression is evaluated only once during the filter
2674 Height of the output video. It defaults to @code{ih}.
2675 This expression is evaluated only once during the filter
2679 Horizontal position, in the input video, of the left edge of the output video.
2680 It defaults to @code{(in_w-out_w)/2}.
2681 This expression is evaluated per-frame.
2684 Vertical position, in the input video, of the top edge of the output video.
2685 It defaults to @code{(in_h-out_h)/2}.
2686 This expression is evaluated per-frame.
2689 If set to 1 will force the output display aspect ratio
2690 to be the same of the input, by changing the output sample aspect
2691 ratio. It defaults to 0.
2694 The @var{out_w}, @var{out_h}, @var{x}, @var{y} parameters are
2695 expressions containing the following constants:
2700 the computed values for @var{x} and @var{y}. They are evaluated for
2705 the input width and height
2709 same as @var{in_w} and @var{in_h}
2713 the output (cropped) width and height
2717 same as @var{out_w} and @var{out_h}
2720 same as @var{iw} / @var{ih}
2723 input sample aspect ratio
2726 input display aspect ratio, it is the same as (@var{iw} / @var{ih}) * @var{sar}
2730 horizontal and vertical chroma subsample values. For example for the
2731 pixel format "yuv422p" @var{hsub} is 2 and @var{vsub} is 1.
2734 the number of input frame, starting from 0
2737 the position in the file of the input frame, NAN if unknown
2740 timestamp expressed in seconds, NAN if the input timestamp is unknown
2744 The expression for @var{out_w} may depend on the value of @var{out_h},
2745 and the expression for @var{out_h} may depend on @var{out_w}, but they
2746 cannot depend on @var{x} and @var{y}, as @var{x} and @var{y} are
2747 evaluated after @var{out_w} and @var{out_h}.
2749 The @var{x} and @var{y} parameters specify the expressions for the
2750 position of the top-left corner of the output (non-cropped) area. They
2751 are evaluated for each frame. If the evaluated value is not valid, it
2752 is approximated to the nearest valid value.
2754 The expression for @var{x} may depend on @var{y}, and the expression
2755 for @var{y} may depend on @var{x}.
2757 @subsection Examples
2761 Crop area with size 100x100 at position (12,34).
2766 Using named options, the example above becomes:
2768 crop=w=100:h=100:x=12:y=34
2772 Crop the central input area with size 100x100:
2778 Crop the central input area with size 2/3 of the input video:
2780 crop=2/3*in_w:2/3*in_h
2784 Crop the input video central square:
2791 Delimit the rectangle with the top-left corner placed at position
2792 100:100 and the right-bottom corner corresponding to the right-bottom
2793 corner of the input image:
2795 crop=in_w-100:in_h-100:100:100
2799 Crop 10 pixels from the left and right borders, and 20 pixels from
2800 the top and bottom borders
2802 crop=in_w-2*10:in_h-2*20
2806 Keep only the bottom right quarter of the input image:
2808 crop=in_w/2:in_h/2:in_w/2:in_h/2
2812 Crop height for getting Greek harmony:
2814 crop=in_w:1/PHI*in_w
2818 Appply trembling effect:
2820 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)
2824 Apply erratic camera effect depending on timestamp:
2826 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)"
2830 Set x depending on the value of y:
2832 crop=in_w/2:in_h/2:y:10+10*sin(n/10)
2838 Auto-detect crop size.
2840 Calculate necessary cropping parameters and prints the recommended
2841 parameters through the logging system. The detected dimensions
2842 correspond to the non-black area of the input video.
2844 The filter accepts the following options:
2849 Set higher black value threshold, which can be optionally specified
2850 from nothing (0) to everything (255). An intensity value greater
2851 to the set value is considered non-black. Default value is 24.
2854 Set the value for which the width/height should be divisible by. The
2855 offset is automatically adjusted to center the video. Use 2 to get
2856 only even dimensions (needed for 4:2:2 video). 16 is best when
2857 encoding to most video codecs. Default value is 16.
2859 @item reset_count, reset
2860 Set the counter that determines after how many frames cropdetect will
2861 reset the previously detected largest video area and start over to
2862 detect the current optimal crop area. Default value is 0.
2864 This can be useful when channel logos distort the video area. 0
2865 indicates never reset and return the largest area encountered during
2872 Apply color adjustments using curves.
2874 This filter is similar to the Adobe Photoshop and GIMP curves tools. Each
2875 component (red, green and blue) has its values defined by @var{N} key points
2876 tied from each other using a smooth curve. The x-axis represents the pixel
2877 values from the input frame, and the y-axis the new pixel values to be set for
2880 By default, a component curve is defined by the two points @var{(0;0)} and
2881 @var{(1;1)}. This creates a straight line where each original pixel value is
2882 "adjusted" to its own value, which means no change to the image.
2884 The filter allows you to redefine these two points and add some more. A new
2885 curve (using a natural cubic spline interpolation) will be define to pass
2886 smoothly through all these new coordinates. The new defined points needs to be
2887 strictly increasing over the x-axis, and their @var{x} and @var{y} values must
2888 be in the @var{[0;1]} interval. If the computed curves happened to go outside
2889 the vector spaces, the values will be clipped accordingly.
2891 If there is no key point defined in @code{x=0}, the filter will automatically
2892 insert a @var{(0;0)} point. In the same way, if there is no key point defined
2893 in @code{x=1}, the filter will automatically insert a @var{(1;1)} point.
2895 The filter accepts the following options:
2899 Select one of the available color presets. This option can be used in addition
2900 to the @option{r}, @option{g}, @option{b} parameters; in this case, the later
2901 options takes priority on the preset values.
2902 Available presets are:
2905 @item color_negative
2908 @item increase_contrast
2910 @item linear_contrast
2911 @item medium_contrast
2913 @item strong_contrast
2916 Default is @code{none}.
2918 Set the master key points. These points will define a second pass mapping. It
2919 is sometimes called a "luminance" or "value" mapping. It can be used with
2920 @option{r}, @option{g}, @option{b} or @option{all} since it acts like a
2921 post-processing LUT.
2923 Set the key points for the red component.
2925 Set the key points for the green component.
2927 Set the key points for the blue component.
2929 Set the key points for all components (not including master).
2930 Can be used in addition to the other key points component
2931 options. In this case, the unset component(s) will fallback on this
2932 @option{all} setting.
2934 Specify a Photoshop curves file (@code{.asv}) to import the settings from.
2937 To avoid some filtergraph syntax conflicts, each key points list need to be
2938 defined using the following syntax: @code{x0/y0 x1/y1 x2/y2 ...}.
2940 @subsection Examples
2944 Increase slightly the middle level of blue:
2946 curves=blue='0.5/0.58'
2952 curves=r='0/0.11 .42/.51 1/0.95':g='0.50/0.48':b='0/0.22 .49/.44 1/0.8'
2954 Here we obtain the following coordinates for each components:
2957 @code{(0;0.11) (0.42;0.51) (1;0.95)}
2959 @code{(0;0) (0.50;0.48) (1;1)}
2961 @code{(0;0.22) (0.49;0.44) (1;0.80)}
2965 The previous example can also be achieved with the associated built-in preset:
2967 curves=preset=vintage
2977 Use a Photoshop preset and redefine the points of the green component:
2979 curves=psfile='MyCurvesPresets/purple.asv':green='0.45/0.53'
2985 Denoise frames using 2D DCT (frequency domain filtering).
2987 This filter is not designed for real time and can be extremely slow.
2989 The filter accepts the following options:
2993 Set the noise sigma constant.
2995 This @var{sigma} defines a hard threshold of @code{3 * sigma}; every DCT
2996 coefficient (absolute value) below this threshold with be dropped.
2998 If you need a more advanced filtering, see @option{expr}.
3000 Default is @code{0}.
3003 Set number overlapping pixels for each block. Each block is of size
3004 @code{16x16}. Since the filter can be slow, you may want to reduce this value,
3005 at the cost of a less effective filter and the risk of various artefacts.
3007 If the overlapping value doesn't allow to process the whole input width or
3008 height, a warning will be displayed and according borders won't be denoised.
3010 Default value is @code{15}.
3013 Set the coefficient factor expression.
3015 For each coefficient of a DCT block, this expression will be evaluated as a
3016 multiplier value for the coefficient.
3018 If this is option is set, the @option{sigma} option will be ignored.
3020 The absolute value of the coefficient can be accessed through the @var{c}
3024 @subsection Examples
3026 Apply a denoise with a @option{sigma} of @code{4.5}:
3031 The same operation can be achieved using the expression system:
3033 dctdnoiz=e='gte(c, 4.5*3)'
3039 Drop duplicated frames at regular intervals.
3041 The filter accepts the following options:
3045 Set the number of frames from which one will be dropped. Setting this to
3046 @var{N} means one frame in every batch of @var{N} frames will be dropped.
3047 Default is @code{5}.
3050 Set the threshold for duplicate detection. If the difference metric for a frame
3051 is less than or equal to this value, then it is declared as duplicate. Default
3055 Set scene change threshold. Default is @code{15}.
3059 Set the size of the x and y-axis blocks used during metric calculations.
3060 Larger blocks give better noise suppression, but also give worse detection of
3061 small movements. Must be a power of two. Default is @code{32}.
3064 Mark main input as a pre-processed input and activate clean source input
3065 stream. This allows the input to be pre-processed with various filters to help
3066 the metrics calculation while keeping the frame selection lossless. When set to
3067 @code{1}, the first stream is for the pre-processed input, and the second
3068 stream is the clean source from where the kept frames are chosen. Default is
3072 Set whether or not chroma is considered in the metric calculations. Default is
3078 Suppress a TV station logo by a simple interpolation of the surrounding
3079 pixels. Just set a rectangle covering the logo and watch it disappear
3080 (and sometimes something even uglier appear - your mileage may vary).
3082 This filter accepts the following options:
3087 Specify the top left corner coordinates of the logo. They must be
3092 Specify the width and height of the logo to clear. They must be
3096 Specify the thickness of the fuzzy edge of the rectangle (added to
3097 @var{w} and @var{h}). The default value is 4.
3100 When set to 1, a green rectangle is drawn on the screen to simplify
3101 finding the right @var{x}, @var{y}, @var{w}, and @var{h} parameters.
3102 The default value is 0.
3104 The rectangle is drawn on the outermost pixels which will be (partly)
3105 replaced with interpolated values. The values of the next pixels
3106 immediately outside this rectangle in each direction will be used to
3107 compute the interpolated pixel values inside the rectangle.
3111 @subsection Examples
3115 Set a rectangle covering the area with top left corner coordinates 0,0
3116 and size 100x77, setting a band of size 10:
3118 delogo=x=0:y=0:w=100:h=77:band=10
3125 Attempt to fix small changes in horizontal and/or vertical shift. This
3126 filter helps remove camera shake from hand-holding a camera, bumping a
3127 tripod, moving on a vehicle, etc.
3129 The filter accepts the following options:
3137 Specify a rectangular area where to limit the search for motion
3139 If desired the search for motion vectors can be limited to a
3140 rectangular area of the frame defined by its top left corner, width
3141 and height. These parameters have the same meaning as the drawbox
3142 filter which can be used to visualise the position of the bounding
3145 This is useful when simultaneous movement of subjects within the frame
3146 might be confused for camera motion by the motion vector search.
3148 If any or all of @var{x}, @var{y}, @var{w} and @var{h} are set to -1
3149 then the full frame is used. This allows later options to be set
3150 without specifying the bounding box for the motion vector search.
3152 Default - search the whole frame.
3156 Specify the maximum extent of movement in x and y directions in the
3157 range 0-64 pixels. Default 16.
3160 Specify how to generate pixels to fill blanks at the edge of the
3161 frame. Available values are:
3164 Fill zeroes at blank locations
3166 Original image at blank locations
3168 Extruded edge value at blank locations
3170 Mirrored edge at blank locations
3172 Default value is @samp{mirror}.
3175 Specify the blocksize to use for motion search. Range 4-128 pixels,
3179 Specify the contrast threshold for blocks. Only blocks with more than
3180 the specified contrast (difference between darkest and lightest
3181 pixels) will be considered. Range 1-255, default 125.
3184 Specify the search strategy. Available values are:
3187 Set exhaustive search
3189 Set less exhaustive search.
3191 Default value is @samp{exhaustive}.
3194 If set then a detailed log of the motion search is written to the
3198 If set to 1, specify using OpenCL capabilities, only available if
3199 FFmpeg was configured with @code{--enable-opencl}. Default value is 0.
3205 Draw a colored box on the input image.
3207 This filter accepts the following options:
3212 The expressions which specify the top left corner coordinates of the box. Default to 0.
3216 The expressions which specify the width and height of the box, if 0 they are interpreted as
3217 the input width and height. Default to 0.
3220 Specify the color of the box to write. For the general syntax of this option,
3221 check the "Color" section in the ffmpeg-utils manual. If the special
3222 value @code{invert} is used, the box edge color is the same as the
3223 video with inverted luma.
3226 The expression which sets the thickness of the box edge. Default value is @code{3}.
3228 See below for the list of accepted constants.
3231 The parameters for @var{x}, @var{y}, @var{w} and @var{h} and @var{t} are expressions containing the
3232 following constants:
3236 The input display aspect ratio, it is the same as (@var{w} / @var{h}) * @var{sar}.
3240 horizontal and vertical chroma subsample values. For example for the
3241 pixel format "yuv422p" @var{hsub} is 2 and @var{vsub} is 1.
3245 The input width and height.
3248 The input sample aspect ratio.
3252 The x and y offset coordinates where the box is drawn.
3256 The width and height of the drawn box.
3259 The thickness of the drawn box.
3261 These constants allow the @var{x}, @var{y}, @var{w}, @var{h} and @var{t} expressions to refer to
3262 each other, so you may for example specify @code{y=x/dar} or @code{h=w/dar}.
3266 @subsection Examples
3270 Draw a black box around the edge of the input image:
3276 Draw a box with color red and an opacity of 50%:
3278 drawbox=10:20:200:60:red@@0.5
3281 The previous example can be specified as:
3283 drawbox=x=10:y=20:w=200:h=60:color=red@@0.5
3287 Fill the box with pink color:
3289 drawbox=x=10:y=10:w=100:h=100:color=pink@@0.5:t=max
3293 Draw a 2-pixel red 2.40:1 mask:
3295 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
3301 Draw a grid on the input image.
3303 This filter accepts the following options:
3308 The expressions which specify the coordinates of some point of grid intersection (meant to configure offset). Both default to 0.
3312 The expressions which specify the width and height of the grid cell, if 0 they are interpreted as the
3313 input width and height, respectively, minus @code{thickness}, so image gets
3314 framed. Default to 0.
3317 Specify the color of the grid. For the general syntax of this option,
3318 check the "Color" section in the ffmpeg-utils manual. If the special
3319 value @code{invert} is used, the grid color is the same as the
3320 video with inverted luma.
3323 The expression which sets the thickness of the grid line. Default value is @code{1}.
3325 See below for the list of accepted constants.
3328 The parameters for @var{x}, @var{y}, @var{w} and @var{h} and @var{t} are expressions containing the
3329 following constants:
3333 The input display aspect ratio, it is the same as (@var{w} / @var{h}) * @var{sar}.
3337 horizontal and vertical chroma subsample values. For example for the
3338 pixel format "yuv422p" @var{hsub} is 2 and @var{vsub} is 1.
3342 The input grid cell width and height.
3345 The input sample aspect ratio.
3349 The x and y coordinates of some point of grid intersection (meant to configure offset).
3353 The width and height of the drawn cell.
3356 The thickness of the drawn cell.
3358 These constants allow the @var{x}, @var{y}, @var{w}, @var{h} and @var{t} expressions to refer to
3359 each other, so you may for example specify @code{y=x/dar} or @code{h=w/dar}.
3363 @subsection Examples
3367 Draw a grid with cell 100x100 pixels, thickness 2 pixels, with color red and an opacity of 50%:
3369 drawgrid=width=100:height=100:thickness=2:color=red@@0.5
3373 Draw a white 3x3 grid with an opacity of 50%:
3375 drawgrid=w=iw/3:h=ih/3:t=2:c=white@@0.5
3382 Draw text string or text from specified file on top of video using the
3383 libfreetype library.
3385 To enable compilation of this filter you need to configure FFmpeg with
3386 @code{--enable-libfreetype}.
3390 The description of the accepted parameters follows.
3395 Used to draw a box around text using background color.
3396 Value should be either 1 (enable) or 0 (disable).
3397 The default value of @var{box} is 0.
3400 The color to be used for drawing box around text. For the syntax of this
3401 option, check the "Color" section in the ffmpeg-utils manual.
3403 The default value of @var{boxcolor} is "white".
3406 Select how the @var{text} is expanded. Can be either @code{none},
3407 @code{strftime} (deprecated) or
3408 @code{normal} (default). See the @ref{drawtext_expansion, Text expansion} section
3412 If true, check and fix text coords to avoid clipping.
3415 The color to be used for drawing fonts. For the syntax of this option, check
3416 the "Color" section in the ffmpeg-utils manual.
3418 The default value of @var{fontcolor} is "black".
3421 The font file to be used for drawing text. Path must be included.
3422 This parameter is mandatory.
3425 The font size to be used for drawing text.
3426 The default value of @var{fontsize} is 16.
3429 Flags to be used for loading the fonts.
3431 The flags map the corresponding flags supported by libfreetype, and are
3432 a combination of the following values:
3439 @item vertical_layout
3440 @item force_autohint
3443 @item ignore_global_advance_width
3445 @item ignore_transform
3451 Default value is "render".
3453 For more information consult the documentation for the FT_LOAD_*
3457 The color to be used for drawing a shadow behind the drawn text. For the
3458 syntax of this option, check the "Color" section in the ffmpeg-utils manual.
3460 The default value of @var{shadowcolor} is "black".
3464 The x and y offsets for the text shadow position with respect to the
3465 position of the text. They can be either positive or negative
3466 values. Default value for both is "0".
3469 The starting frame number for the n/frame_num variable. The default value
3473 The size in number of spaces to use for rendering the tab.
3477 Set the initial timecode representation in "hh:mm:ss[:;.]ff"
3478 format. It can be used with or without text parameter. @var{timecode_rate}
3479 option must be specified.
3481 @item timecode_rate, rate, r
3482 Set the timecode frame rate (timecode only).
3485 The text string to be drawn. The text must be a sequence of UTF-8
3487 This parameter is mandatory if no file is specified with the parameter
3491 A text file containing text to be drawn. The text must be a sequence
3492 of UTF-8 encoded characters.
3494 This parameter is mandatory if no text string is specified with the
3495 parameter @var{text}.
3497 If both @var{text} and @var{textfile} are specified, an error is thrown.
3500 If set to 1, the @var{textfile} will be reloaded before each frame.
3501 Be sure to update it atomically, or it may be read partially, or even fail.
3505 The expressions which specify the offsets where text will be drawn
3506 within the video frame. They are relative to the top/left border of the
3509 The default value of @var{x} and @var{y} is "0".
3511 See below for the list of accepted constants and functions.
3514 The parameters for @var{x} and @var{y} are expressions containing the
3515 following constants and functions:
3519 input display aspect ratio, it is the same as (@var{w} / @var{h}) * @var{sar}
3523 horizontal and vertical chroma subsample values. For example for the
3524 pixel format "yuv422p" @var{hsub} is 2 and @var{vsub} is 1.
3527 the height of each text line
3535 @item max_glyph_a, ascent
3536 the maximum distance from the baseline to the highest/upper grid
3537 coordinate used to place a glyph outline point, for all the rendered
3539 It is a positive value, due to the grid's orientation with the Y axis
3542 @item max_glyph_d, descent
3543 the maximum distance from the baseline to the lowest grid coordinate
3544 used to place a glyph outline point, for all the rendered glyphs.
3545 This is a negative value, due to the grid's orientation, with the Y axis
3549 maximum glyph height, that is the maximum height for all the glyphs
3550 contained in the rendered text, it is equivalent to @var{ascent} -
3554 maximum glyph width, that is the maximum width for all the glyphs
3555 contained in the rendered text
3558 the number of input frame, starting from 0
3560 @item rand(min, max)
3561 return a random number included between @var{min} and @var{max}
3564 input sample aspect ratio
3567 timestamp expressed in seconds, NAN if the input timestamp is unknown
3570 the height of the rendered text
3573 the width of the rendered text
3577 the x and y offset coordinates where the text is drawn.
3579 These parameters allow the @var{x} and @var{y} expressions to refer
3580 each other, so you can for example specify @code{y=x/dar}.
3583 If libavfilter was built with @code{--enable-fontconfig}, then
3584 @option{fontfile} can be a fontconfig pattern or omitted.
3586 @anchor{drawtext_expansion}
3587 @subsection Text expansion
3589 If @option{expansion} is set to @code{strftime},
3590 the filter recognizes strftime() sequences in the provided text and
3591 expands them accordingly. Check the documentation of strftime(). This
3592 feature is deprecated.
3594 If @option{expansion} is set to @code{none}, the text is printed verbatim.
3596 If @option{expansion} is set to @code{normal} (which is the default),
3597 the following expansion mechanism is used.
3599 The backslash character '\', followed by any character, always expands to
3600 the second character.
3602 Sequence of the form @code{%@{...@}} are expanded. The text between the
3603 braces is a function name, possibly followed by arguments separated by ':'.
3604 If the arguments contain special characters or delimiters (':' or '@}'),
3605 they should be escaped.
3607 Note that they probably must also be escaped as the value for the
3608 @option{text} option in the filter argument string and as the filter
3609 argument in the filtergraph description, and possibly also for the shell,
3610 that makes up to four levels of escaping; using a text file avoids these
3613 The following functions are available:
3618 The expression evaluation result.
3620 It must take one argument specifying the expression to be evaluated,
3621 which accepts the same constants and functions as the @var{x} and
3622 @var{y} values. Note that not all constants should be used, for
3623 example the text size is not known when evaluating the expression, so
3624 the constants @var{text_w} and @var{text_h} will have an undefined
3628 The time at which the filter is running, expressed in UTC.
3629 It can accept an argument: a strftime() format string.
3632 The time at which the filter is running, expressed in the local time zone.
3633 It can accept an argument: a strftime() format string.
3636 Frame metadata. It must take one argument specifying metadata key.
3639 The frame number, starting from 0.
3642 A 1 character description of the current picture type.
3645 The timestamp of the current frame, in seconds, with microsecond accuracy.
3649 @subsection Examples
3653 Draw "Test Text" with font FreeSerif, using the default values for the
3654 optional parameters.
3657 drawtext="fontfile=/usr/share/fonts/truetype/freefont/FreeSerif.ttf: text='Test Text'"
3661 Draw 'Test Text' with font FreeSerif of size 24 at position x=100
3662 and y=50 (counting from the top-left corner of the screen), text is
3663 yellow with a red box around it. Both the text and the box have an
3667 drawtext="fontfile=/usr/share/fonts/truetype/freefont/FreeSerif.ttf: text='Test Text':\
3668 x=100: y=50: fontsize=24: fontcolor=yellow@@0.2: box=1: boxcolor=red@@0.2"
3671 Note that the double quotes are not necessary if spaces are not used
3672 within the parameter list.
3675 Show the text at the center of the video frame:
3677 drawtext="fontsize=30:fontfile=FreeSerif.ttf:text='hello world':x=(w-text_w)/2:y=(h-text_h-line_h)/2"
3681 Show a text line sliding from right to left in the last row of the video
3682 frame. The file @file{LONG_LINE} is assumed to contain a single line
3685 drawtext="fontsize=15:fontfile=FreeSerif.ttf:text=LONG_LINE:y=h-line_h:x=-50*t"
3689 Show the content of file @file{CREDITS} off the bottom of the frame and scroll up.
3691 drawtext="fontsize=20:fontfile=FreeSerif.ttf:textfile=CREDITS:y=h-20*t"
3695 Draw a single green letter "g", at the center of the input video.
3696 The glyph baseline is placed at half screen height.
3698 drawtext="fontsize=60:fontfile=FreeSerif.ttf:fontcolor=green:text=g:x=(w-max_glyph_w)/2:y=h/2-ascent"
3702 Show text for 1 second every 3 seconds:
3704 drawtext="fontfile=FreeSerif.ttf:fontcolor=white:x=100:y=x/dar:enable=lt(mod(t\,3)\,1):text='blink'"
3708 Use fontconfig to set the font. Note that the colons need to be escaped.
3710 drawtext='fontfile=Linux Libertine O-40\:style=Semibold:text=FFmpeg'
3714 Print the date of a real-time encoding (see strftime(3)):
3716 drawtext='fontfile=FreeSans.ttf:text=%@{localtime:%a %b %d %Y@}'
3721 For more information about libfreetype, check:
3722 @url{http://www.freetype.org/}.
3724 For more information about fontconfig, check:
3725 @url{http://freedesktop.org/software/fontconfig/fontconfig-user.html}.
3729 Detect and draw edges. The filter uses the Canny Edge Detection algorithm.
3731 The filter accepts the following options:
3736 Set low and high threshold values used by the Canny thresholding
3739 The high threshold selects the "strong" edge pixels, which are then
3740 connected through 8-connectivity with the "weak" edge pixels selected
3741 by the low threshold.
3743 @var{low} and @var{high} threshold values must be choosen in the range
3744 [0,1], and @var{low} should be lesser or equal to @var{high}.
3746 Default value for @var{low} is @code{20/255}, and default value for @var{high}
3752 edgedetect=low=0.1:high=0.4
3755 @section extractplanes
3757 Extract color channel components from input video stream into
3758 separate grayscale video streams.
3760 The filter accepts the following option:
3764 Set plane(s) to extract.
3766 Available values for planes are:
3777 Choosing planes not available in the input will result in an error.
3778 That means you cannot select @code{r}, @code{g}, @code{b} planes
3779 with @code{y}, @code{u}, @code{v} planes at same time.
3782 @subsection Examples
3786 Extract luma, u and v color channel component from input video frame
3787 into 3 grayscale outputs:
3789 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
3795 Apply a posterize effect using the ELBG (Enhanced LBG) algorithm.
3797 For each input image, the filter will compute the optimal mapping from
3798 the input to the output given the codebook length, that is the number
3799 of distinct output colors.
3801 This filter accepts the following options.
3804 @item codebook_length, l
3805 Set codebook length. The value must be a positive integer, and
3806 represents the number of distinct output colors. Default value is 256.
3809 Set the maximum number of iterations to apply for computing the optimal
3810 mapping. The higher the value the better the result and the higher the
3811 computation time. Default value is 1.
3814 Set a random seed, must be an integer included between 0 and
3815 UINT32_MAX. If not specified, or if explicitly set to -1, the filter
3816 will try to use a good random seed on a best effort basis.
3821 Apply fade-in/out effect to input video.
3823 This filter accepts the following options:
3827 The effect type -- can be either "in" for fade-in, or "out" for a fade-out
3829 Default is @code{in}.
3831 @item start_frame, s
3832 Specify the number of the start frame for starting to apply the fade
3833 effect. Default is 0.
3836 The number of frames for which the fade effect has to last. At the end of the
3837 fade-in effect the output video will have the same intensity as the input video,
3838 at the end of the fade-out transition the output video will be filled with the
3839 selected @option{color}.
3843 If set to 1, fade only alpha channel, if one exists on the input.
3846 @item start_time, st
3847 Specify the timestamp (in seconds) of the frame to start to apply the fade
3848 effect. If both start_frame and start_time are specified, the fade will start at
3849 whichever comes last. Default is 0.
3852 The number of seconds for which the fade effect has to last. At the end of the
3853 fade-in effect the output video will have the same intensity as the input video,
3854 at the end of the fade-out transition the output video will be filled with the
3855 selected @option{color}.
3856 If both duration and nb_frames are specified, duration is used. Default is 0.
3859 Specify the color of the fade. Default is "black".
3862 @subsection Examples
3866 Fade in first 30 frames of video:
3871 The command above is equivalent to:
3877 Fade out last 45 frames of a 200-frame video:
3880 fade=type=out:start_frame=155:nb_frames=45
3884 Fade in first 25 frames and fade out last 25 frames of a 1000-frame video:
3886 fade=in:0:25, fade=out:975:25
3890 Make first 5 frames yellow, then fade in from frame 5-24:
3892 fade=in:5:20:color=yellow
3896 Fade in alpha over first 25 frames of video:
3898 fade=in:0:25:alpha=1
3902 Make first 5.5 seconds black, then fade in for 0.5 seconds:
3904 fade=t=in:st=5.5:d=0.5
3911 Extract a single field from an interlaced image using stride
3912 arithmetic to avoid wasting CPU time. The output frames are marked as
3915 The filter accepts the following options:
3919 Specify whether to extract the top (if the value is @code{0} or
3920 @code{top}) or the bottom field (if the value is @code{1} or
3926 Field matching filter for inverse telecine. It is meant to reconstruct the
3927 progressive frames from a telecined stream. The filter does not drop duplicated
3928 frames, so to achieve a complete inverse telecine @code{fieldmatch} needs to be
3929 followed by a decimation filter such as @ref{decimate} in the filtergraph.
3931 The separation of the field matching and the decimation is notably motivated by
3932 the possibility of inserting a de-interlacing filter fallback between the two.
3933 If the source has mixed telecined and real interlaced content,
3934 @code{fieldmatch} will not be able to match fields for the interlaced parts.
3935 But these remaining combed frames will be marked as interlaced, and thus can be
3936 de-interlaced by a later filter such as @ref{yadif} before decimation.
3938 In addition to the various configuration options, @code{fieldmatch} can take an
3939 optional second stream, activated through the @option{ppsrc} option. If
3940 enabled, the frames reconstruction will be based on the fields and frames from
3941 this second stream. This allows the first input to be pre-processed in order to
3942 help the various algorithms of the filter, while keeping the output lossless
3943 (assuming the fields are matched properly). Typically, a field-aware denoiser,
3944 or brightness/contrast adjustments can help.
3946 Note that this filter uses the same algorithms as TIVTC/TFM (AviSynth project)
3947 and VIVTC/VFM (VapourSynth project). The later is a light clone of TFM from
3948 which @code{fieldmatch} is based on. While the semantic and usage are very
3949 close, some behaviour and options names can differ.
3951 The filter accepts the following options:
3955 Specify the assumed field order of the input stream. Available values are:
3959 Auto detect parity (use FFmpeg's internal parity value).
3961 Assume bottom field first.
3963 Assume top field first.
3966 Note that it is sometimes recommended not to trust the parity announced by the
3969 Default value is @var{auto}.
3972 Set the matching mode or strategy to use. @option{pc} mode is the safest in the
3973 sense that it won't risk creating jerkiness due to duplicate frames when
3974 possible, but if there are bad edits or blended fields it will end up
3975 outputting combed frames when a good match might actually exist. On the other
3976 hand, @option{pcn_ub} mode is the most risky in terms of creating jerkiness,
3977 but will almost always find a good frame if there is one. The other values are
3978 all somewhere in between @option{pc} and @option{pcn_ub} in terms of risking
3979 jerkiness and creating duplicate frames versus finding good matches in sections
3980 with bad edits, orphaned fields, blended fields, etc.
3982 More details about p/c/n/u/b are available in @ref{p/c/n/u/b meaning} section.
3984 Available values are:
3988 2-way matching (p/c)
3990 2-way matching, and trying 3rd match if still combed (p/c + n)
3992 2-way matching, and trying 3rd match (same order) if still combed (p/c + u)
3994 2-way matching, trying 3rd match if still combed, and trying 4th/5th matches if
3995 still combed (p/c + n + u/b)
3997 3-way matching (p/c/n)
3999 3-way matching, and trying 4th/5th matches if all 3 of the original matches are
4000 detected as combed (p/c/n + u/b)
4003 The parenthesis at the end indicate the matches that would be used for that
4004 mode assuming @option{order}=@var{tff} (and @option{field} on @var{auto} or
4007 In terms of speed @option{pc} mode is by far the fastest and @option{pcn_ub} is
4010 Default value is @var{pc_n}.
4013 Mark the main input stream as a pre-processed input, and enable the secondary
4014 input stream as the clean source to pick the fields from. See the filter
4015 introduction for more details. It is similar to the @option{clip2} feature from
4018 Default value is @code{0} (disabled).
4021 Set the field to match from. It is recommended to set this to the same value as
4022 @option{order} unless you experience matching failures with that setting. In
4023 certain circumstances changing the field that is used to match from can have a
4024 large impact on matching performance. Available values are:
4028 Automatic (same value as @option{order}).
4030 Match from the bottom field.
4032 Match from the top field.
4035 Default value is @var{auto}.
4038 Set whether or not chroma is included during the match comparisons. In most
4039 cases it is recommended to leave this enabled. You should set this to @code{0}
4040 only if your clip has bad chroma problems such as heavy rainbowing or other
4041 artifacts. Setting this to @code{0} could also be used to speed things up at
4042 the cost of some accuracy.
4044 Default value is @code{1}.
4048 These define an exclusion band which excludes the lines between @option{y0} and
4049 @option{y1} from being included in the field matching decision. An exclusion
4050 band can be used to ignore subtitles, a logo, or other things that may
4051 interfere with the matching. @option{y0} sets the starting scan line and
4052 @option{y1} sets the ending line; all lines in between @option{y0} and
4053 @option{y1} (including @option{y0} and @option{y1}) will be ignored. Setting
4054 @option{y0} and @option{y1} to the same value will disable the feature.
4055 @option{y0} and @option{y1} defaults to @code{0}.
4058 Set the scene change detection threshold as a percentage of maximum change on
4059 the luma plane. Good values are in the @code{[8.0, 14.0]} range. Scene change
4060 detection is only relevant in case @option{combmatch}=@var{sc}. The range for
4061 @option{scthresh} is @code{[0.0, 100.0]}.
4063 Default value is @code{12.0}.
4066 When @option{combatch} is not @var{none}, @code{fieldmatch} will take into
4067 account the combed scores of matches when deciding what match to use as the
4068 final match. Available values are:
4072 No final matching based on combed scores.
4074 Combed scores are only used when a scene change is detected.
4076 Use combed scores all the time.
4079 Default is @var{sc}.
4082 Force @code{fieldmatch} to calculate the combed metrics for certain matches and
4083 print them. This setting is known as @option{micout} in TFM/VFM vocabulary.
4084 Available values are:
4088 No forced calculation.
4090 Force p/c/n calculations.
4092 Force p/c/n/u/b calculations.
4095 Default value is @var{none}.
4098 This is the area combing threshold used for combed frame detection. This
4099 essentially controls how "strong" or "visible" combing must be to be detected.
4100 Larger values mean combing must be more visible and smaller values mean combing
4101 can be less visible or strong and still be detected. Valid settings are from
4102 @code{-1} (every pixel will be detected as combed) to @code{255} (no pixel will
4103 be detected as combed). This is basically a pixel difference value. A good
4104 range is @code{[8, 12]}.
4106 Default value is @code{9}.
4109 Sets whether or not chroma is considered in the combed frame decision. Only
4110 disable this if your source has chroma problems (rainbowing, etc.) that are
4111 causing problems for the combed frame detection with chroma enabled. Actually,
4112 using @option{chroma}=@var{0} is usually more reliable, except for the case
4113 where there is chroma only combing in the source.
4115 Default value is @code{0}.
4119 Respectively set the x-axis and y-axis size of the window used during combed
4120 frame detection. This has to do with the size of the area in which
4121 @option{combpel} pixels are required to be detected as combed for a frame to be
4122 declared combed. See the @option{combpel} parameter description for more info.
4123 Possible values are any number that is a power of 2 starting at 4 and going up
4126 Default value is @code{16}.
4129 The number of combed pixels inside any of the @option{blocky} by
4130 @option{blockx} size blocks on the frame for the frame to be detected as
4131 combed. While @option{cthresh} controls how "visible" the combing must be, this
4132 setting controls "how much" combing there must be in any localized area (a
4133 window defined by the @option{blockx} and @option{blocky} settings) on the
4134 frame. Minimum value is @code{0} and maximum is @code{blocky x blockx} (at
4135 which point no frames will ever be detected as combed). This setting is known
4136 as @option{MI} in TFM/VFM vocabulary.
4138 Default value is @code{80}.
4141 @anchor{p/c/n/u/b meaning}
4142 @subsection p/c/n/u/b meaning
4144 @subsubsection p/c/n
4146 We assume the following telecined stream:
4149 Top fields: 1 2 2 3 4
4150 Bottom fields: 1 2 3 4 4
4153 The numbers correspond to the progressive frame the fields relate to. Here, the
4154 first two frames are progressive, the 3rd and 4th are combed, and so on.
4156 When @code{fieldmatch} is configured to run a matching from bottom
4157 (@option{field}=@var{bottom}) this is how this input stream get transformed:
4162 B 1 2 3 4 4 <-- matching reference
4171 As a result of the field matching, we can see that some frames get duplicated.
4172 To perform a complete inverse telecine, you need to rely on a decimation filter
4173 after this operation. See for instance the @ref{decimate} filter.
4175 The same operation now matching from top fields (@option{field}=@var{top})
4180 T 1 2 2 3 4 <-- matching reference
4190 In these examples, we can see what @var{p}, @var{c} and @var{n} mean;
4191 basically, they refer to the frame and field of the opposite parity:
4194 @item @var{p} matches the field of the opposite parity in the previous frame
4195 @item @var{c} matches the field of the opposite parity in the current frame
4196 @item @var{n} matches the field of the opposite parity in the next frame
4201 The @var{u} and @var{b} matching are a bit special in the sense that they match
4202 from the opposite parity flag. In the following examples, we assume that we are
4203 currently matching the 2nd frame (Top:2, bottom:2). According to the match, a
4204 'x' is placed above and below each matched fields.
4206 With bottom matching (@option{field}=@var{bottom}):
4211 Top 1 2 2 1 2 2 1 2 2 1 2 2 1 2 2
4212 Bottom 1 2 3 1 2 3 1 2 3 1 2 3 1 2 3
4220 With top matching (@option{field}=@var{top}):
4225 Top 1 2 2 1 2 2 1 2 2 1 2 2 1 2 2
4226 Bottom 1 2 3 1 2 3 1 2 3 1 2 3 1 2 3
4234 @subsection Examples
4236 Simple IVTC of a top field first telecined stream:
4238 fieldmatch=order=tff:combmatch=none, decimate
4241 Advanced IVTC, with fallback on @ref{yadif} for still combed frames:
4243 fieldmatch=order=tff:combmatch=full, yadif=deint=interlaced, decimate
4248 Transform the field order of the input video.
4250 This filter accepts the following options:
4255 Output field order. Valid values are @var{tff} for top field first or @var{bff}
4256 for bottom field first.
4259 Default value is @samp{tff}.
4261 Transformation is achieved by shifting the picture content up or down
4262 by one line, and filling the remaining line with appropriate picture content.
4263 This method is consistent with most broadcast field order converters.
4265 If the input video is not flagged as being interlaced, or it is already
4266 flagged as being of the required output field order then this filter does
4267 not alter the incoming video.
4269 This filter is very useful when converting to or from PAL DV material,
4270 which is bottom field first.
4274 ffmpeg -i in.vob -vf "fieldorder=bff" out.dv
4279 Buffer input images and send them when they are requested.
4281 This filter is mainly useful when auto-inserted by the libavfilter
4284 The filter does not take parameters.
4289 Convert the input video to one of the specified pixel formats.
4290 Libavfilter will try to pick one that is supported for the input to
4293 This filter accepts the following parameters:
4297 A '|'-separated list of pixel format names, for example
4298 "pix_fmts=yuv420p|monow|rgb24".
4302 @subsection Examples
4306 Convert the input video to the format @var{yuv420p}
4308 format=pix_fmts=yuv420p
4311 Convert the input video to any of the formats in the list
4313 format=pix_fmts=yuv420p|yuv444p|yuv410p
4320 Convert the video to specified constant frame rate by duplicating or dropping
4321 frames as necessary.
4323 This filter accepts the following named parameters:
4327 Desired output frame rate. The default is @code{25}.
4332 Possible values are:
4335 zero round towards 0
4339 round towards -infinity
4341 round towards +infinity
4345 The default is @code{near}.
4348 Assume the first PTS should be the given value, in seconds. This allows for
4349 padding/trimming at the start of stream. By default, no assumption is made
4350 about the first frame's expected PTS, so no padding or trimming is done.
4351 For example, this could be set to 0 to pad the beginning with duplicates of
4352 the first frame if a video stream starts after the audio stream or to trim any
4353 frames with a negative PTS.
4357 Alternatively, the options can be specified as a flat string:
4358 @var{fps}[:@var{round}].
4360 See also the @ref{setpts} filter.
4362 @subsection Examples
4366 A typical usage in order to set the fps to 25:
4372 Sets the fps to 24, using abbreviation and rounding method to round to nearest:
4374 fps=fps=film:round=near
4380 Select one frame every N-th frame.
4382 This filter accepts the following option:
4385 Select frame after every @code{step} frames.
4386 Allowed values are positive integers higher than 0. Default value is @code{1}.
4392 Apply a frei0r effect to the input video.
4394 To enable compilation of this filter you need to install the frei0r
4395 header and configure FFmpeg with @code{--enable-frei0r}.
4397 This filter accepts the following options:
4402 The name to the frei0r effect to load. If the environment variable
4403 @env{FREI0R_PATH} is defined, the frei0r effect is searched in each one of the
4404 directories specified by the colon separated list in @env{FREIOR_PATH},
4405 otherwise in the standard frei0r paths, which are in this order:
4406 @file{HOME/.frei0r-1/lib/}, @file{/usr/local/lib/frei0r-1/},
4407 @file{/usr/lib/frei0r-1/}.
4410 A '|'-separated list of parameters to pass to the frei0r effect.
4414 A frei0r effect parameter can be a boolean (whose values are specified
4415 with "y" and "n"), a double, a color (specified by the syntax
4416 @var{R}/@var{G}/@var{B}, (@var{R}, @var{G}, and @var{B} being float
4417 numbers from 0.0 to 1.0) or by a color description specified in the "Color"
4418 section in the ffmpeg-utils manual), a position (specified by the syntax @var{X}/@var{Y},
4419 @var{X} and @var{Y} being float numbers) and a string.
4421 The number and kind of parameters depend on the loaded effect. If an
4422 effect parameter is not specified the default value is set.
4424 @subsection Examples
4428 Apply the distort0r effect, set the first two double parameters:
4430 frei0r=filter_name=distort0r:filter_params=0.5|0.01
4434 Apply the colordistance effect, take a color as first parameter:
4436 frei0r=colordistance:0.2/0.3/0.4
4437 frei0r=colordistance:violet
4438 frei0r=colordistance:0x112233
4442 Apply the perspective effect, specify the top left and top right image
4445 frei0r=perspective:0.2/0.2|0.8/0.2
4449 For more information see:
4450 @url{http://frei0r.dyne.org}
4454 The filter accepts the following options:
4458 Set the luminance expression.
4460 Set the chrominance blue expression.
4462 Set the chrominance red expression.
4464 Set the alpha expression.
4466 Set the red expression.
4468 Set the green expression.
4470 Set the blue expression.
4473 The colorspace is selected according to the specified options. If one
4474 of the @option{lum_expr}, @option{cb_expr}, or @option{cr_expr}
4475 options is specified, the filter will automatically select a YCbCr
4476 colorspace. If one of the @option{red_expr}, @option{green_expr}, or
4477 @option{blue_expr} options is specified, it will select an RGB
4480 If one of the chrominance expression is not defined, it falls back on the other
4481 one. If no alpha expression is specified it will evaluate to opaque value.
4482 If none of chrominance expressions are specified, they will evaluate
4483 to the luminance expression.
4485 The expressions can use the following variables and functions:
4489 The sequential number of the filtered frame, starting from @code{0}.
4493 The coordinates of the current sample.
4497 The width and height of the image.
4501 Width and height scale depending on the currently filtered plane. It is the
4502 ratio between the corresponding luma plane number of pixels and the current
4503 plane ones. E.g. for YUV4:2:0 the values are @code{1,1} for the luma plane, and
4504 @code{0.5,0.5} for chroma planes.
4507 Time of the current frame, expressed in seconds.
4510 Return the value of the pixel at location (@var{x},@var{y}) of the current
4514 Return the value of the pixel at location (@var{x},@var{y}) of the luminance
4518 Return the value of the pixel at location (@var{x},@var{y}) of the
4519 blue-difference chroma plane. Return 0 if there is no such plane.
4522 Return the value of the pixel at location (@var{x},@var{y}) of the
4523 red-difference chroma plane. Return 0 if there is no such plane.
4528 Return the value of the pixel at location (@var{x},@var{y}) of the
4529 red/green/blue component. Return 0 if there is no such component.
4532 Return the value of the pixel at location (@var{x},@var{y}) of the alpha
4533 plane. Return 0 if there is no such plane.
4536 For functions, if @var{x} and @var{y} are outside the area, the value will be
4537 automatically clipped to the closer edge.
4539 @subsection Examples
4543 Flip the image horizontally:
4549 Generate a bidimensional sine wave, with angle @code{PI/3} and a
4550 wavelength of 100 pixels:
4552 geq=128 + 100*sin(2*(PI/100)*(cos(PI/3)*(X-50*T) + sin(PI/3)*Y)):128:128
4556 Generate a fancy enigmatic moving light:
4558 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
4562 Generate a quick emboss effect:
4564 format=gray,geq=lum_expr='(p(X,Y)+(256-p(X-4,Y-4)))/2'
4568 Modify RGB components depending on pixel position:
4570 geq=r='X/W*r(X,Y)':g='(1-X/W)*g(X,Y)':b='(H-Y)/H*b(X,Y)'
4576 Fix the banding artifacts that are sometimes introduced into nearly flat
4577 regions by truncation to 8bit color depth.
4578 Interpolate the gradients that should go where the bands are, and
4581 This filter is designed for playback only. Do not use it prior to
4582 lossy compression, because compression tends to lose the dither and
4583 bring back the bands.
4585 This filter accepts the following options:
4590 The maximum amount by which the filter will change any one pixel. Also the
4591 threshold for detecting nearly flat regions. Acceptable values range from .51 to
4592 64, default value is 1.2, out-of-range values will be clipped to the valid
4596 The neighborhood to fit the gradient to. A larger radius makes for smoother
4597 gradients, but also prevents the filter from modifying the pixels near detailed
4598 regions. Acceptable values are 8-32, default value is 16, out-of-range values
4599 will be clipped to the valid range.
4603 Alternatively, the options can be specified as a flat string:
4604 @var{strength}[:@var{radius}]
4606 @subsection Examples
4610 Apply the filter with a @code{3.5} strength and radius of @code{8}:
4616 Specify radius, omitting the strength (which will fall-back to the default
4627 Apply a Hald CLUT to a video stream.
4629 First input is the video stream to process, and second one is the Hald CLUT.
4630 The Hald CLUT input can be a simple picture or a complete video stream.
4632 The filter accepts the following options:
4636 Force termination when the shortest input terminates. Default is @code{0}.
4638 Continue applying the last CLUT after the end of the stream. A value of
4639 @code{0} disable the filter after the last frame of the CLUT is reached.
4640 Default is @code{1}.
4643 @code{haldclut} also has the same interpolation options as @ref{lut3d} (both
4644 filters share the same internals).
4646 More information about the Hald CLUT can be found on Eskil Steenberg's website
4647 (Hald CLUT author) at @url{http://www.quelsolaar.com/technology/clut.html}.
4649 @subsection Workflow examples
4651 @subsubsection Hald CLUT video stream
4653 Generate an identity Hald CLUT stream altered with various effects:
4655 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
4658 Note: make sure you use a lossless codec.
4660 Then use it with @code{haldclut} to apply it on some random stream:
4662 ffmpeg -f lavfi -i mandelbrot -i clut.nut -filter_complex '[0][1] haldclut' -t 20 mandelclut.mkv
4665 The Hald CLUT will be applied to the 10 first seconds (duration of
4666 @file{clut.nut}), then the latest picture of that CLUT stream will be applied
4667 to the remaining frames of the @code{mandelbrot} stream.
4669 @subsubsection Hald CLUT with preview
4671 A Hald CLUT is supposed to be a squared image of @code{Level*Level*Level} by
4672 @code{Level*Level*Level} pixels. For a given Hald CLUT, FFmpeg will select the
4673 biggest possible square starting at the top left of the picture. The remaining
4674 padding pixels (bottom or right) will be ignored. This area can be used to add
4675 a preview of the Hald CLUT.
4677 Typically, the following generated Hald CLUT will be supported by the
4678 @code{haldclut} filter:
4681 ffmpeg -f lavfi -i @ref{haldclutsrc}=8 -vf "
4682 pad=iw+320 [padded_clut];
4683 smptebars=s=320x256, split [a][b];
4684 [padded_clut][a] overlay=W-320:h, curves=color_negative [main];
4685 [main][b] overlay=W-320" -frames:v 1 clut.png
4688 It contains the original and a preview of the effect of the CLUT: SMPTE color
4689 bars are displayed on the right-top, and below the same color bars processed by
4692 Then, the effect of this Hald CLUT can be visualized with:
4694 ffplay input.mkv -vf "movie=clut.png, [in] haldclut"
4699 Flip the input video horizontally.
4701 For example to horizontally flip the input video with @command{ffmpeg}:
4703 ffmpeg -i in.avi -vf "hflip" out.avi
4707 This filter applies a global color histogram equalization on a
4710 It can be used to correct video that has a compressed range of pixel
4711 intensities. The filter redistributes the pixel intensities to
4712 equalize their distribution across the intensity range. It may be
4713 viewed as an "automatically adjusting contrast filter". This filter is
4714 useful only for correcting degraded or poorly captured source
4717 The filter accepts the following options:
4721 Determine the amount of equalization to be applied. As the strength
4722 is reduced, the distribution of pixel intensities more-and-more
4723 approaches that of the input frame. The value must be a float number
4724 in the range [0,1] and defaults to 0.200.
4727 Set the maximum intensity that can generated and scale the output
4728 values appropriately. The strength should be set as desired and then
4729 the intensity can be limited if needed to avoid washing-out. The value
4730 must be a float number in the range [0,1] and defaults to 0.210.
4733 Set the antibanding level. If enabled the filter will randomly vary
4734 the luminance of output pixels by a small amount to avoid banding of
4735 the histogram. Possible values are @code{none}, @code{weak} or
4736 @code{strong}. It defaults to @code{none}.
4741 Compute and draw a color distribution histogram for the input video.
4743 The computed histogram is a representation of distribution of color components
4746 The filter accepts the following options:
4752 It accepts the following values:
4755 standard histogram that display color components distribution in an image.
4756 Displays color graph for each color component. Shows distribution
4757 of the Y, U, V, A or R, G, B components, depending on input format,
4758 in current frame. Bellow each graph is color component scale meter.
4761 chroma values in vectorscope, if brighter more such chroma values are
4762 distributed in an image.
4763 Displays chroma values (U/V color placement) in two dimensional graph
4764 (which is called a vectorscope). It can be used to read of the hue and
4765 saturation of the current frame. At a same time it is a histogram.
4766 The whiter a pixel in the vectorscope, the more pixels of the input frame
4767 correspond to that pixel (that is the more pixels have this chroma value).
4768 The V component is displayed on the horizontal (X) axis, with the leftmost
4769 side being V = 0 and the rightmost side being V = 255.
4770 The U component is displayed on the vertical (Y) axis, with the top
4771 representing U = 0 and the bottom representing U = 255.
4773 The position of a white pixel in the graph corresponds to the chroma value
4774 of a pixel of the input clip. So the graph can be used to read of the
4775 hue (color flavor) and the saturation (the dominance of the hue in the color).
4776 As the hue of a color changes, it moves around the square. At the center of
4777 the square, the saturation is zero, which means that the corresponding pixel
4778 has no color. If you increase the amount of a specific color, while leaving
4779 the other colors unchanged, the saturation increases, and you move towards
4780 the edge of the square.
4783 chroma values in vectorscope, similar as @code{color} but actual chroma values
4787 per row/column color component graph. In row mode graph in the left side represents
4788 color component value 0 and right side represents value = 255. In column mode top
4789 side represents color component value = 0 and bottom side represents value = 255.
4791 Default value is @code{levels}.
4794 Set height of level in @code{levels}. Default value is @code{200}.
4795 Allowed range is [50, 2048].
4798 Set height of color scale in @code{levels}. Default value is @code{12}.
4799 Allowed range is [0, 40].
4802 Set step for @code{waveform} mode. Smaller values are useful to find out how much
4803 of same luminance values across input rows/columns are distributed.
4804 Default value is @code{10}. Allowed range is [1, 255].
4807 Set mode for @code{waveform}. Can be either @code{row}, or @code{column}.
4808 Default is @code{row}.
4810 @item waveform_mirror
4811 Set mirroring mode for @code{waveform}. @code{0} means unmirrored, @code{1}
4812 means mirrored. In mirrored mode, higher values will be represented on the left
4813 side for @code{row} mode and at the top for @code{column} mode. Default is
4814 @code{0} (unmirrored).
4817 Set display mode for @code{waveform} and @code{levels}.
4818 It accepts the following values:
4821 Display separate graph for the color components side by side in
4822 @code{row} waveform mode or one below other in @code{column} waveform mode
4823 for @code{waveform} histogram mode. For @code{levels} histogram mode
4824 per color component graphs are placed one bellow other.
4826 This display mode in @code{waveform} histogram mode makes it easy to spot
4827 color casts in the highlights and shadows of an image, by comparing the
4828 contours of the top and the bottom of each waveform.
4829 Since whites, grays, and blacks are characterized by
4830 exactly equal amounts of red, green, and blue, neutral areas of the
4831 picture should display three waveforms of roughly equal width/height.
4832 If not, the correction is easy to make by making adjustments to level the
4836 Presents information that's identical to that in the @code{parade}, except
4837 that the graphs representing color components are superimposed directly
4840 This display mode in @code{waveform} histogram mode can make it easier to spot
4841 the relative differences or similarities in overlapping areas of the color
4842 components that are supposed to be identical, such as neutral whites, grays,
4845 Default is @code{parade}.
4848 Set mode for @code{levels}. Can be either @code{linear}, or @code{logarithmic}.
4849 Default is @code{linear}.
4852 @subsection Examples
4857 Calculate and draw histogram:
4859 ffplay -i input -vf histogram
4867 High precision/quality 3d denoise filter. This filter aims to reduce
4868 image noise producing smooth images and making still images really
4869 still. It should enhance compressibility.
4871 It accepts the following optional parameters:
4875 a non-negative float number which specifies spatial luma strength,
4878 @item chroma_spatial
4879 a non-negative float number which specifies spatial chroma strength,
4880 defaults to 3.0*@var{luma_spatial}/4.0
4883 a float number which specifies luma temporal strength, defaults to
4884 6.0*@var{luma_spatial}/4.0
4887 a float number which specifies chroma temporal strength, defaults to
4888 @var{luma_tmp}*@var{chroma_spatial}/@var{luma_spatial}
4893 Modify the hue and/or the saturation of the input.
4895 This filter accepts the following options:
4899 Specify the hue angle as a number of degrees. It accepts an expression,
4900 and defaults to "0".
4903 Specify the saturation in the [-10,10] range. It accepts an expression and
4907 Specify the hue angle as a number of radians. It accepts an
4908 expression, and defaults to "0".
4911 Specify the brightness in the [-10,10] range. It accepts an expression and
4915 @option{h} and @option{H} are mutually exclusive, and can't be
4916 specified at the same time.
4918 The @option{b}, @option{h}, @option{H} and @option{s} option values are
4919 expressions containing the following constants:
4923 frame count of the input frame starting from 0
4926 presentation timestamp of the input frame expressed in time base units
4929 frame rate of the input video, NAN if the input frame rate is unknown
4932 timestamp expressed in seconds, NAN if the input timestamp is unknown
4935 time base of the input video
4938 @subsection Examples
4942 Set the hue to 90 degrees and the saturation to 1.0:
4948 Same command but expressing the hue in radians:
4954 Rotate hue and make the saturation swing between 0
4955 and 2 over a period of 1 second:
4957 hue="H=2*PI*t: s=sin(2*PI*t)+1"
4961 Apply a 3 seconds saturation fade-in effect starting at 0:
4966 The general fade-in expression can be written as:
4968 hue="s=min(0\, max((t-START)/DURATION\, 1))"
4972 Apply a 3 seconds saturation fade-out effect starting at 5 seconds:
4974 hue="s=max(0\, min(1\, (8-t)/3))"
4977 The general fade-out expression can be written as:
4979 hue="s=max(0\, min(1\, (START+DURATION-t)/DURATION))"
4984 @subsection Commands
4986 This filter supports the following commands:
4992 Modify the hue and/or the saturation and/or brightness of the input video.
4993 The command accepts the same syntax of the corresponding option.
4995 If the specified expression is not valid, it is kept at its current
5001 Detect video interlacing type.
5003 This filter tries to detect if the input is interlaced or progressive,
5004 top or bottom field first.
5006 The filter accepts the following options:
5010 Set interlacing threshold.
5012 Set progressive threshold.
5017 Deinterleave or interleave fields.
5019 This filter allows to process interlaced images fields without
5020 deinterlacing them. Deinterleaving splits the input frame into 2
5021 fields (so called half pictures). Odd lines are moved to the top
5022 half of the output image, even lines to the bottom half.
5023 You can process (filter) them independently and then re-interleave them.
5025 The filter accepts the following options:
5029 @item chroma_mode, c
5031 Available values for @var{luma_mode}, @var{chroma_mode} and
5032 @var{alpha_mode} are:
5038 @item deinterleave, d
5039 Deinterleave fields, placing one above the other.
5042 Interleave fields. Reverse the effect of deinterleaving.
5044 Default value is @code{none}.
5047 @item chroma_swap, cs
5048 @item alpha_swap, as
5049 Swap luma/chroma/alpha fields. Exchange even & odd lines. Default value is @code{0}.
5054 Simple interlacing filter from progressive contents. This interleaves upper (or
5055 lower) lines from odd frames with lower (or upper) lines from even frames,
5056 halving the frame rate and preserving image height.
5059 Original Original New Frame
5060 Frame 'j' Frame 'j+1' (tff)
5061 ========== =========== ==================
5062 Line 0 --------------------> Frame 'j' Line 0
5063 Line 1 Line 1 ----> Frame 'j+1' Line 1
5064 Line 2 ---------------------> Frame 'j' Line 2
5065 Line 3 Line 3 ----> Frame 'j+1' Line 3
5067 New Frame + 1 will be generated by Frame 'j+2' and Frame 'j+3' and so on
5070 It accepts the following optional parameters:
5074 determines whether the interlaced frame is taken from the even (tff - default)
5075 or odd (bff) lines of the progressive frame.
5078 Enable (default) or disable the vertical lowpass filter to avoid twitter
5079 interlacing and reduce moire patterns.
5084 Deinterlace input video by applying Donald Graft's adaptive kernel
5085 deinterling. Work on interlaced parts of a video to produce
5088 The description of the accepted parameters follows.
5092 Set the threshold which affects the filter's tolerance when
5093 determining if a pixel line must be processed. It must be an integer
5094 in the range [0,255] and defaults to 10. A value of 0 will result in
5095 applying the process on every pixels.
5098 Paint pixels exceeding the threshold value to white if set to 1.
5102 Set the fields order. Swap fields if set to 1, leave fields alone if
5106 Enable additional sharpening if set to 1. Default is 0.
5109 Enable twoway sharpening if set to 1. Default is 0.
5112 @subsection Examples
5116 Apply default values:
5118 kerndeint=thresh=10:map=0:order=0:sharp=0:twoway=0
5122 Enable additional sharpening:
5128 Paint processed pixels in white:
5137 Apply a 3D LUT to an input video.
5139 The filter accepts the following options:
5143 Set the 3D LUT file name.
5145 Currently supported formats:
5157 Select interpolation mode.
5159 Available values are:
5163 Use values from the nearest defined point.
5165 Interpolate values using the 8 points defining a cube.
5167 Interpolate values using a tetrahedron.
5171 @section lut, lutrgb, lutyuv
5173 Compute a look-up table for binding each pixel component input value
5174 to an output value, and apply it to input video.
5176 @var{lutyuv} applies a lookup table to a YUV input video, @var{lutrgb}
5177 to an RGB input video.
5179 These filters accept the following options:
5182 set first pixel component expression
5184 set second pixel component expression
5186 set third pixel component expression
5188 set fourth pixel component expression, corresponds to the alpha component
5191 set red component expression
5193 set green component expression
5195 set blue component expression
5197 alpha component expression
5200 set Y/luminance component expression
5202 set U/Cb component expression
5204 set V/Cr component expression
5207 Each of them specifies the expression to use for computing the lookup table for
5208 the corresponding pixel component values.
5210 The exact component associated to each of the @var{c*} options depends on the
5213 The @var{lut} filter requires either YUV or RGB pixel formats in input,
5214 @var{lutrgb} requires RGB pixel formats in input, and @var{lutyuv} requires YUV.
5216 The expressions can contain the following constants and functions:
5221 the input width and height
5224 input value for the pixel component
5227 the input value clipped in the @var{minval}-@var{maxval} range
5230 maximum value for the pixel component
5233 minimum value for the pixel component
5236 the negated value for the pixel component value clipped in the
5237 @var{minval}-@var{maxval} range , it corresponds to the expression
5238 "maxval-clipval+minval"
5241 the computed value in @var{val} clipped in the
5242 @var{minval}-@var{maxval} range
5244 @item gammaval(gamma)
5245 the computed gamma correction value of the pixel component value
5246 clipped in the @var{minval}-@var{maxval} range, corresponds to the
5248 "pow((clipval-minval)/(maxval-minval)\,@var{gamma})*(maxval-minval)+minval"
5252 All expressions default to "val".
5254 @subsection Examples
5260 lutrgb="r=maxval+minval-val:g=maxval+minval-val:b=maxval+minval-val"
5261 lutyuv="y=maxval+minval-val:u=maxval+minval-val:v=maxval+minval-val"
5264 The above is the same as:
5266 lutrgb="r=negval:g=negval:b=negval"
5267 lutyuv="y=negval:u=negval:v=negval"
5277 Remove chroma components, turns the video into a graytone image:
5279 lutyuv="u=128:v=128"
5283 Apply a luma burning effect:
5289 Remove green and blue components:
5295 Set a constant alpha channel value on input:
5297 format=rgba,lutrgb=a="maxval-minval/2"
5301 Correct luminance gamma by a 0.5 factor:
5303 lutyuv=y=gammaval(0.5)
5307 Discard least significant bits of luma:
5309 lutyuv=y='bitand(val, 128+64+32)'
5313 @section mergeplanes
5315 Merge color channel components from several video streams.
5317 The filter accepts up to 4 input streams, and merge selected input
5318 planes to the output video.
5320 This filter accepts the following options:
5323 Set input to output plane mapping. Default is @code{0}.
5325 The mappings is specified as a bitmap. It should be specified as a
5326 hexadecimal number in the form 0xAa[Bb[Cc[Dd]]]. 'Aa' describes the
5327 mapping for the first plane of the output stream. 'A' sets the number of
5328 the input stream to use (from 0 to 3), and 'a' the plane number of the
5329 corresponding input to use (from 0 to 3). The rest of the mappings is
5330 similar, 'Bb' describes the mapping for the output stream second
5331 plane, 'Cc' describes the mapping for the output stream third plane and
5332 'Dd' describes the mapping for the output stream fourth plane.
5335 Set output pixel format. Default is @code{yuva444p}.
5338 @subsection Examples
5342 Merge three gray video streams of same width and height into single video stream:
5344 [a0][a1][a2]mergeplanes=0x001020:yuv444p
5348 Merge 1st yuv444p stream and 2nd gray video stream into yuva444p video stream:
5350 [a0][a1]mergeplanes=0x00010210:yuva444p
5354 Swap Y and A plane in yuva444p stream:
5356 format=yuva444p,mergeplanes=0x03010200:yuva444p
5360 Swap U and V plane in yuv420p stream:
5362 format=yuv420p,mergeplanes=0x000201:yuv420p
5366 Cast a rgb24 clip to yuv444p:
5368 format=rgb24,mergeplanes=0x000102:yuv444p
5374 Apply motion-compensation deinterlacing.
5376 It needs one field per frame as input and must thus be used together
5377 with yadif=1/3 or equivalent.
5379 This filter accepts the following options:
5382 Set the deinterlacing mode.
5384 It accepts one of the following values:
5389 use iterative motion estimation
5391 like @samp{slow}, but use multiple reference frames.
5393 Default value is @samp{fast}.
5396 Set the picture field parity assumed for the input video. It must be
5397 one of the following values:
5401 assume top field first
5403 assume bottom field first
5406 Default value is @samp{bff}.
5409 Set per-block quantization parameter (QP) used by the internal
5412 Higher values should result in a smoother motion vector field but less
5413 optimal individual vectors. Default value is 1.
5418 Apply an MPlayer filter to the input video.
5420 This filter provides a wrapper around some of the filters of
5423 This wrapper is considered experimental. Some of the wrapped filters
5424 may not work properly and we may drop support for them, as they will
5425 be implemented natively into FFmpeg. Thus you should avoid
5426 depending on them when writing portable scripts.
5428 The filter accepts the parameters:
5429 @var{filter_name}[:=]@var{filter_params}
5431 @var{filter_name} is the name of a supported MPlayer filter,
5432 @var{filter_params} is a string containing the parameters accepted by
5435 The list of the currently supported filters follows:
5446 The parameter syntax and behavior for the listed filters are the same
5447 of the corresponding MPlayer filters. For detailed instructions check
5448 the "VIDEO FILTERS" section in the MPlayer manual.
5450 @subsection Examples
5454 Adjust gamma, brightness, contrast:
5460 See also mplayer(1), @url{http://www.mplayerhq.hu/}.
5464 Drop frames that do not differ greatly from the previous frame in
5465 order to reduce frame rate.
5467 The main use of this filter is for very-low-bitrate encoding
5468 (e.g. streaming over dialup modem), but it could in theory be used for
5469 fixing movies that were inverse-telecined incorrectly.
5471 A description of the accepted options follows.
5475 Set the maximum number of consecutive frames which can be dropped (if
5476 positive), or the minimum interval between dropped frames (if
5477 negative). If the value is 0, the frame is dropped unregarding the
5478 number of previous sequentially dropped frames.
5485 Set the dropping threshold values.
5487 Values for @option{hi} and @option{lo} are for 8x8 pixel blocks and
5488 represent actual pixel value differences, so a threshold of 64
5489 corresponds to 1 unit of difference for each pixel, or the same spread
5490 out differently over the block.
5492 A frame is a candidate for dropping if no 8x8 blocks differ by more
5493 than a threshold of @option{hi}, and if no more than @option{frac} blocks (1
5494 meaning the whole image) differ by more than a threshold of @option{lo}.
5496 Default value for @option{hi} is 64*12, default value for @option{lo} is
5497 64*5, and default value for @option{frac} is 0.33.
5505 This filter accepts an integer in input, if non-zero it negates the
5506 alpha component (if available). The default value in input is 0.
5510 Force libavfilter not to use any of the specified pixel formats for the
5511 input to the next filter.
5513 This filter accepts the following parameters:
5517 A '|'-separated list of pixel format names, for example
5518 "pix_fmts=yuv420p|monow|rgb24".
5522 @subsection Examples
5526 Force libavfilter to use a format different from @var{yuv420p} for the
5527 input to the vflip filter:
5529 noformat=pix_fmts=yuv420p,vflip
5533 Convert the input video to any of the formats not contained in the list:
5535 noformat=yuv420p|yuv444p|yuv410p
5541 Add noise on video input frame.
5543 The filter accepts the following options:
5551 Set noise seed for specific pixel component or all pixel components in case
5552 of @var{all_seed}. Default value is @code{123457}.
5554 @item all_strength, alls
5555 @item c0_strength, c0s
5556 @item c1_strength, c1s
5557 @item c2_strength, c2s
5558 @item c3_strength, c3s
5559 Set noise strength for specific pixel component or all pixel components in case
5560 @var{all_strength}. Default value is @code{0}. Allowed range is [0, 100].
5562 @item all_flags, allf
5567 Set pixel component flags or set flags for all components if @var{all_flags}.
5568 Available values for component flags are:
5571 averaged temporal noise (smoother)
5573 mix random noise with a (semi)regular pattern
5575 temporal noise (noise pattern changes between frames)
5577 uniform noise (gaussian otherwise)
5581 @subsection Examples
5583 Add temporal and uniform noise to input video:
5585 noise=alls=20:allf=t+u
5590 Pass the video source unchanged to the output.
5594 Apply video transform using libopencv.
5596 To enable this filter install libopencv library and headers and
5597 configure FFmpeg with @code{--enable-libopencv}.
5599 This filter accepts the following parameters:
5604 The name of the libopencv filter to apply.
5607 The parameters to pass to the libopencv filter. If not specified the default
5612 Refer to the official libopencv documentation for more precise
5614 @url{http://opencv.willowgarage.com/documentation/c/image_filtering.html}
5616 Follows the list of supported libopencv filters.
5621 Dilate an image by using a specific structuring element.
5622 This filter corresponds to the libopencv function @code{cvDilate}.
5624 It accepts the parameters: @var{struct_el}|@var{nb_iterations}.
5626 @var{struct_el} represents a structuring element, and has the syntax:
5627 @var{cols}x@var{rows}+@var{anchor_x}x@var{anchor_y}/@var{shape}
5629 @var{cols} and @var{rows} represent the number of columns and rows of
5630 the structuring element, @var{anchor_x} and @var{anchor_y} the anchor
5631 point, and @var{shape} the shape for the structuring element, and
5632 can be one of the values "rect", "cross", "ellipse", "custom".
5634 If the value for @var{shape} is "custom", it must be followed by a
5635 string of the form "=@var{filename}". The file with name
5636 @var{filename} is assumed to represent a binary image, with each
5637 printable character corresponding to a bright pixel. When a custom
5638 @var{shape} is used, @var{cols} and @var{rows} are ignored, the number
5639 or columns and rows of the read file are assumed instead.
5641 The default value for @var{struct_el} is "3x3+0x0/rect".
5643 @var{nb_iterations} specifies the number of times the transform is
5644 applied to the image, and defaults to 1.
5646 Follow some example:
5648 # use the default values
5651 # dilate using a structuring element with a 5x5 cross, iterate two times
5652 ocv=filter_name=dilate:filter_params=5x5+2x2/cross|2
5654 # read the shape from the file diamond.shape, iterate two times
5655 # the file diamond.shape may contain a pattern of characters like this:
5661 # the specified cols and rows are ignored (but not the anchor point coordinates)
5662 ocv=dilate:0x0+2x2/custom=diamond.shape|2
5667 Erode an image by using a specific structuring element.
5668 This filter corresponds to the libopencv function @code{cvErode}.
5670 The filter accepts the parameters: @var{struct_el}:@var{nb_iterations},
5671 with the same syntax and semantics as the @ref{dilate} filter.
5675 Smooth the input video.
5677 The filter takes the following parameters:
5678 @var{type}|@var{param1}|@var{param2}|@var{param3}|@var{param4}.
5680 @var{type} is the type of smooth filter to apply, and can be one of
5681 the following values: "blur", "blur_no_scale", "median", "gaussian",
5682 "bilateral". The default value is "gaussian".
5684 @var{param1}, @var{param2}, @var{param3}, and @var{param4} are
5685 parameters whose meanings depend on smooth type. @var{param1} and
5686 @var{param2} accept integer positive values or 0, @var{param3} and
5687 @var{param4} accept float values.
5689 The default value for @var{param1} is 3, the default value for the
5690 other parameters is 0.
5692 These parameters correspond to the parameters assigned to the
5693 libopencv function @code{cvSmooth}.
5698 Overlay one video on top of another.
5700 It takes two inputs and one output, the first input is the "main"
5701 video on which the second input is overlayed.
5703 This filter accepts the following parameters:
5705 A description of the accepted options follows.
5710 Set the expression for the x and y coordinates of the overlayed video
5711 on the main video. Default value is "0" for both expressions. In case
5712 the expression is invalid, it is set to a huge value (meaning that the
5713 overlay will not be displayed within the output visible area).
5716 Set when the expressions for @option{x}, and @option{y} are evaluated.
5718 It accepts the following values:
5721 only evaluate expressions once during the filter initialization or
5722 when a command is processed
5725 evaluate expressions for each incoming frame
5728 Default value is @samp{frame}.
5731 If set to 1, force the output to terminate when the shortest input
5732 terminates. Default value is 0.
5735 Set the format for the output video.
5737 It accepts the following values:
5749 Default value is @samp{yuv420}.
5751 @item rgb @emph{(deprecated)}
5752 If set to 1, force the filter to accept inputs in the RGB
5753 color space. Default value is 0. This option is deprecated, use
5754 @option{format} instead.
5757 If set to 1, force the filter to draw the last overlay frame over the
5758 main input until the end of the stream. A value of 0 disables this
5759 behavior. Default value is 1.
5762 The @option{x}, and @option{y} expressions can contain the following
5768 main input width and height
5772 overlay input width and height
5776 the computed values for @var{x} and @var{y}. They are evaluated for
5781 horizontal and vertical chroma subsample values of the output
5782 format. For example for the pixel format "yuv422p" @var{hsub} is 2 and
5786 the number of input frame, starting from 0
5789 the position in the file of the input frame, NAN if unknown
5792 timestamp expressed in seconds, NAN if the input timestamp is unknown
5795 Note that the @var{n}, @var{pos}, @var{t} variables are available only
5796 when evaluation is done @emph{per frame}, and will evaluate to NAN
5797 when @option{eval} is set to @samp{init}.
5799 Be aware that frames are taken from each input video in timestamp
5800 order, hence, if their initial timestamps differ, it is a good idea
5801 to pass the two inputs through a @var{setpts=PTS-STARTPTS} filter to
5802 have them begin in the same zero timestamp, as it does the example for
5803 the @var{movie} filter.
5805 You can chain together more overlays but you should test the
5806 efficiency of such approach.
5808 @subsection Commands
5810 This filter supports the following commands:
5814 Modify the x and y of the overlay input.
5815 The command accepts the same syntax of the corresponding option.
5817 If the specified expression is not valid, it is kept at its current
5821 @subsection Examples
5825 Draw the overlay at 10 pixels from the bottom right corner of the main
5828 overlay=main_w-overlay_w-10:main_h-overlay_h-10
5831 Using named options the example above becomes:
5833 overlay=x=main_w-overlay_w-10:y=main_h-overlay_h-10
5837 Insert a transparent PNG logo in the bottom left corner of the input,
5838 using the @command{ffmpeg} tool with the @code{-filter_complex} option:
5840 ffmpeg -i input -i logo -filter_complex 'overlay=10:main_h-overlay_h-10' output
5844 Insert 2 different transparent PNG logos (second logo on bottom
5845 right corner) using the @command{ffmpeg} tool:
5847 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
5851 Add a transparent color layer on top of the main video, @code{WxH}
5852 must specify the size of the main input to the overlay filter:
5854 color=color=red@@.3:size=WxH [over]; [in][over] overlay [out]
5858 Play an original video and a filtered version (here with the deshake
5859 filter) side by side using the @command{ffplay} tool:
5861 ffplay input.avi -vf 'split[a][b]; [a]pad=iw*2:ih[src]; [b]deshake[filt]; [src][filt]overlay=w'
5864 The above command is the same as:
5866 ffplay input.avi -vf 'split[b], pad=iw*2[src], [b]deshake, [src]overlay=w'
5870 Make a sliding overlay appearing from the left to the right top part of the
5871 screen starting since time 2:
5873 overlay=x='if(gte(t,2), -w+(t-2)*20, NAN)':y=0
5877 Compose output by putting two input videos side to side:
5879 ffmpeg -i left.avi -i right.avi -filter_complex "
5880 nullsrc=size=200x100 [background];
5881 [0:v] setpts=PTS-STARTPTS, scale=100x100 [left];
5882 [1:v] setpts=PTS-STARTPTS, scale=100x100 [right];
5883 [background][left] overlay=shortest=1 [background+left];
5884 [background+left][right] overlay=shortest=1:x=100 [left+right]
5889 Chain several overlays in cascade:
5891 nullsrc=s=200x200 [bg];
5892 testsrc=s=100x100, split=4 [in0][in1][in2][in3];
5893 [in0] lutrgb=r=0, [bg] overlay=0:0 [mid0];
5894 [in1] lutrgb=g=0, [mid0] overlay=100:0 [mid1];
5895 [in2] lutrgb=b=0, [mid1] overlay=0:100 [mid2];
5896 [in3] null, [mid2] overlay=100:100 [out0]
5903 Apply Overcomplete Wavelet denoiser.
5905 The filter accepts the following options:
5911 Larger depth values will denoise lower frequency components more, but
5912 slow down filtering.
5914 Must be an int in the range 8-16, default is @code{8}.
5916 @item luma_strength, ls
5919 Must be a double value in the range 0-1000, default is @code{1.0}.
5921 @item chroma_strength, cs
5922 Set chroma strength.
5924 Must be a double value in the range 0-1000, default is @code{1.0}.
5929 Add paddings to the input image, and place the original input at the
5930 given coordinates @var{x}, @var{y}.
5932 This filter accepts the following parameters:
5937 Specify an expression for the size of the output image with the
5938 paddings added. If the value for @var{width} or @var{height} is 0, the
5939 corresponding input size is used for the output.
5941 The @var{width} expression can reference the value set by the
5942 @var{height} expression, and vice versa.
5944 The default value of @var{width} and @var{height} is 0.
5948 Specify an expression for the offsets where to place the input image
5949 in the padded area with respect to the top/left border of the output
5952 The @var{x} expression can reference the value set by the @var{y}
5953 expression, and vice versa.
5955 The default value of @var{x} and @var{y} is 0.
5958 Specify the color of the padded area. For the syntax of this option,
5959 check the "Color" section in the ffmpeg-utils manual.
5961 The default value of @var{color} is "black".
5964 The value for the @var{width}, @var{height}, @var{x}, and @var{y}
5965 options are expressions containing the following constants:
5970 the input video width and height
5974 same as @var{in_w} and @var{in_h}
5978 the output width and height, that is the size of the padded area as
5979 specified by the @var{width} and @var{height} expressions
5983 same as @var{out_w} and @var{out_h}
5987 x and y offsets as specified by the @var{x} and @var{y}
5988 expressions, or NAN if not yet specified
5991 same as @var{iw} / @var{ih}
5994 input sample aspect ratio
5997 input display aspect ratio, it is the same as (@var{iw} / @var{ih}) * @var{sar}
6001 horizontal and vertical chroma subsample values. For example for the
6002 pixel format "yuv422p" @var{hsub} is 2 and @var{vsub} is 1.
6005 @subsection Examples
6009 Add paddings with color "violet" to the input video. Output video
6010 size is 640x480, the top-left corner of the input video is placed at
6013 pad=640:480:0:40:violet
6016 The example above is equivalent to the following command:
6018 pad=width=640:height=480:x=0:y=40:color=violet
6022 Pad the input to get an output with dimensions increased by 3/2,
6023 and put the input video at the center of the padded area:
6025 pad="3/2*iw:3/2*ih:(ow-iw)/2:(oh-ih)/2"
6029 Pad the input to get a squared output with size equal to the maximum
6030 value between the input width and height, and put the input video at
6031 the center of the padded area:
6033 pad="max(iw\,ih):ow:(ow-iw)/2:(oh-ih)/2"
6037 Pad the input to get a final w/h ratio of 16:9:
6039 pad="ih*16/9:ih:(ow-iw)/2:(oh-ih)/2"
6043 In case of anamorphic video, in order to set the output display aspect
6044 correctly, it is necessary to use @var{sar} in the expression,
6045 according to the relation:
6047 (ih * X / ih) * sar = output_dar
6048 X = output_dar / sar
6051 Thus the previous example needs to be modified to:
6053 pad="ih*16/9/sar:ih:(ow-iw)/2:(oh-ih)/2"
6057 Double output size and put the input video in the bottom-right
6058 corner of the output padded area:
6060 pad="2*iw:2*ih:ow-iw:oh-ih"
6064 @section perspective
6066 Correct perspective of video not recorded perpendicular to the screen.
6068 A description of the accepted parameters follows.
6079 Set coordinates expression for top left, top right, bottom left and bottom right corners.
6080 Default values are @code{0:0:W:0:0:H:W:H} with which perspective will remain unchanged.
6082 The expressions can use the following variables:
6087 the width and height of video frame.
6091 Set interpolation for perspective correction.
6093 It accepts the following values:
6099 Default value is @samp{linear}.
6104 Delay interlaced video by one field time so that the field order changes.
6106 The intended use is to fix PAL movies that have been captured with the
6107 opposite field order to the film-to-video transfer.
6109 A description of the accepted parameters follows.
6115 It accepts the following values:
6118 Capture field order top-first, transfer bottom-first.
6119 Filter will delay the bottom field.
6122 Capture field order bottom-first, transfer top-first.
6123 Filter will delay the top field.
6126 Capture and transfer with the same field order. This mode only exists
6127 for the documentation of the other options to refer to, but if you
6128 actually select it, the filter will faithfully do nothing.
6131 Capture field order determined automatically by field flags, transfer
6133 Filter selects among @samp{t} and @samp{b} modes on a frame by frame
6134 basis using field flags. If no field information is available,
6135 then this works just like @samp{u}.
6138 Capture unknown or varying, transfer opposite.
6139 Filter selects among @samp{t} and @samp{b} on a frame by frame basis by
6140 analyzing the images and selecting the alternative that produces best
6141 match between the fields.
6144 Capture top-first, transfer unknown or varying.
6145 Filter selects among @samp{t} and @samp{p} using image analysis.
6148 Capture bottom-first, transfer unknown or varying.
6149 Filter selects among @samp{b} and @samp{p} using image analysis.
6152 Capture determined by field flags, transfer unknown or varying.
6153 Filter selects among @samp{t}, @samp{b} and @samp{p} using field flags and
6154 image analysis. If no field information is available, then this works just
6155 like @samp{U}. This is the default mode.
6158 Both capture and transfer unknown or varying.
6159 Filter selects among @samp{t}, @samp{b} and @samp{p} using image analysis only.
6163 @section pixdesctest
6165 Pixel format descriptor test filter, mainly useful for internal
6166 testing. The output video should be equal to the input video.
6170 format=monow, pixdesctest
6173 can be used to test the monowhite pixel format descriptor definition.
6177 Enable the specified chain of postprocessing subfilters using libpostproc. This
6178 library should be automatically selected with a GPL build (@code{--enable-gpl}).
6179 Subfilters must be separated by '/' and can be disabled by prepending a '-'.
6180 Each subfilter and some options have a short and a long name that can be used
6181 interchangeably, i.e. dr/dering are the same.
6183 The filters accept the following options:
6187 Set postprocessing subfilters string.
6190 All subfilters share common options to determine their scope:
6194 Honor the quality commands for this subfilter.
6197 Do chrominance filtering, too (default).
6200 Do luminance filtering only (no chrominance).
6203 Do chrominance filtering only (no luminance).
6206 These options can be appended after the subfilter name, separated by a '|'.
6208 Available subfilters are:
6211 @item hb/hdeblock[|difference[|flatness]]
6212 Horizontal deblocking filter
6215 Difference factor where higher values mean more deblocking (default: @code{32}).
6217 Flatness threshold where lower values mean more deblocking (default: @code{39}).
6220 @item vb/vdeblock[|difference[|flatness]]
6221 Vertical deblocking filter
6224 Difference factor where higher values mean more deblocking (default: @code{32}).
6226 Flatness threshold where lower values mean more deblocking (default: @code{39}).
6229 @item ha/hadeblock[|difference[|flatness]]
6230 Accurate horizontal deblocking filter
6233 Difference factor where higher values mean more deblocking (default: @code{32}).
6235 Flatness threshold where lower values mean more deblocking (default: @code{39}).
6238 @item va/vadeblock[|difference[|flatness]]
6239 Accurate vertical deblocking filter
6242 Difference factor where higher values mean more deblocking (default: @code{32}).
6244 Flatness threshold where lower values mean more deblocking (default: @code{39}).
6248 The horizontal and vertical deblocking filters share the difference and
6249 flatness values so you cannot set different horizontal and vertical
6254 Experimental horizontal deblocking filter
6257 Experimental vertical deblocking filter
6262 @item tn/tmpnoise[|threshold1[|threshold2[|threshold3]]], temporal noise reducer
6265 larger -> stronger filtering
6267 larger -> stronger filtering
6269 larger -> stronger filtering
6272 @item al/autolevels[:f/fullyrange], automatic brightness / contrast correction
6275 Stretch luminance to @code{0-255}.
6278 @item lb/linblenddeint
6279 Linear blend deinterlacing filter that deinterlaces the given block by
6280 filtering all lines with a @code{(1 2 1)} filter.
6282 @item li/linipoldeint
6283 Linear interpolating deinterlacing filter that deinterlaces the given block by
6284 linearly interpolating every second line.
6286 @item ci/cubicipoldeint
6287 Cubic interpolating deinterlacing filter deinterlaces the given block by
6288 cubically interpolating every second line.
6290 @item md/mediandeint
6291 Median deinterlacing filter that deinterlaces the given block by applying a
6292 median filter to every second line.
6294 @item fd/ffmpegdeint
6295 FFmpeg deinterlacing filter that deinterlaces the given block by filtering every
6296 second line with a @code{(-1 4 2 4 -1)} filter.
6299 Vertically applied FIR lowpass deinterlacing filter that deinterlaces the given
6300 block by filtering all lines with a @code{(-1 2 6 2 -1)} filter.
6302 @item fq/forceQuant[|quantizer]
6303 Overrides the quantizer table from the input with the constant quantizer you
6311 Default pp filter combination (@code{hb|a,vb|a,dr|a})
6314 Fast pp filter combination (@code{h1|a,v1|a,dr|a})
6317 High quality pp filter combination (@code{ha|a|128|7,va|a,dr|a})
6320 @subsection Examples
6324 Apply horizontal and vertical deblocking, deringing and automatic
6325 brightness/contrast:
6331 Apply default filters without brightness/contrast correction:
6337 Apply default filters and temporal denoiser:
6339 pp=default/tmpnoise|1|2|3
6343 Apply deblocking on luminance only, and switch vertical deblocking on or off
6344 automatically depending on available CPU time:
6352 Obtain the average, maximum and minimum PSNR (Peak Signal to Noise
6353 Ratio) between two input videos.
6355 This filter takes in input two input videos, the first input is
6356 considered the "main" source and is passed unchanged to the
6357 output. The second input is used as a "reference" video for computing
6360 Both video inputs must have the same resolution and pixel format for
6361 this filter to work correctly. Also it assumes that both inputs
6362 have the same number of frames, which are compared one by one.
6364 The obtained average PSNR is printed through the logging system.
6366 The filter stores the accumulated MSE (mean squared error) of each
6367 frame, and at the end of the processing it is averaged across all frames
6368 equally, and the following formula is applied to obtain the PSNR:
6371 PSNR = 10*log10(MAX^2/MSE)
6374 Where MAX is the average of the maximum values of each component of the
6377 The description of the accepted parameters follows.
6381 If specified the filter will use the named file to save the PSNR of
6382 each individual frame.
6385 The file printed if @var{stats_file} is selected, contains a sequence of
6386 key/value pairs of the form @var{key}:@var{value} for each compared
6389 A description of each shown parameter follows:
6393 sequential number of the input frame, starting from 1
6396 Mean Square Error pixel-by-pixel average difference of the compared
6397 frames, averaged over all the image components.
6399 @item mse_y, mse_u, mse_v, mse_r, mse_g, mse_g, mse_a
6400 Mean Square Error pixel-by-pixel average difference of the compared
6401 frames for the component specified by the suffix.
6403 @item psnr_y, psnr_u, psnr_v, psnr_r, psnr_g, psnr_b, psnr_a
6404 Peak Signal to Noise ratio of the compared frames for the component
6405 specified by the suffix.
6410 movie=ref_movie.mpg, setpts=PTS-STARTPTS [main];
6411 [main][ref] psnr="stats_file=stats.log" [out]
6414 On this example the input file being processed is compared with the
6415 reference file @file{ref_movie.mpg}. The PSNR of each individual frame
6416 is stored in @file{stats.log}.
6420 Pulldown reversal (inverse telecine) filter, capable of handling mixed
6421 hard-telecine, 24000/1001 fps progressive, and 30000/1001 fps progressive
6424 The pullup filter is designed to take advantage of future context in making
6425 its decisions. This filter is stateless in the sense that it does not lock
6426 onto a pattern to follow, but it instead looks forward to the following
6427 fields in order to identify matches and rebuild progressive frames.
6429 To produce content with an even framerate, insert the fps filter after
6430 pullup, use @code{fps=24000/1001} if the input frame rate is 29.97fps,
6431 @code{fps=24} for 30fps and the (rare) telecined 25fps input.
6433 The filter accepts the following options:
6440 These options set the amount of "junk" to ignore at the left, right, top, and
6441 bottom of the image, respectively. Left and right are in units of 8 pixels,
6442 while top and bottom are in units of 2 lines.
6443 The default is 8 pixels on each side.
6446 Set the strict breaks. Setting this option to 1 will reduce the chances of
6447 filter generating an occasional mismatched frame, but it may also cause an
6448 excessive number of frames to be dropped during high motion sequences.
6449 Conversely, setting it to -1 will make filter match fields more easily.
6450 This may help processing of video where there is slight blurring between
6451 the fields, but may also cause there to be interlaced frames in the output.
6452 Default value is @code{0}.
6455 Set the metric plane to use. It accepts the following values:
6461 Use chroma blue plane.
6464 Use chroma red plane.
6467 This option may be set to use chroma plane instead of the default luma plane
6468 for doing filter's computations. This may improve accuracy on very clean
6469 source material, but more likely will decrease accuracy, especially if there
6470 is chroma noise (rainbow effect) or any grayscale video.
6471 The main purpose of setting @option{mp} to a chroma plane is to reduce CPU
6472 load and make pullup usable in realtime on slow machines.
6475 For best results (without duplicated frames in the output file) it is
6476 necessary to change the output frame rate. For example, to inverse
6477 telecine NTSC input:
6479 ffmpeg -i input -vf pullup -r 24000/1001 ...
6484 Suppress a TV station logo, using an image file to determine which
6485 pixels comprise the logo. It works by filling in the pixels that
6486 comprise the logo with neighboring pixels.
6488 The filter accepts the following options:
6492 Set the filter bitmap file, which can be any image format supported by
6493 libavformat. The width and height of the image file must match those of the
6494 video stream being processed.
6497 Pixels in the provided bitmap image with a value of zero are not
6498 considered part of the logo, non-zero pixels are considered part of
6499 the logo. If you use white (255) for the logo and black (0) for the
6500 rest, you will be safe. For making the filter bitmap, it is
6501 recommended to take a screen capture of a black frame with the logo
6502 visible, and then using a threshold filter followed by the erode
6503 filter once or twice.
6505 If needed, little splotches can be fixed manually. Remember that if
6506 logo pixels are not covered, the filter quality will be much
6507 reduced. Marking too many pixels as part of the logo does not hurt as
6508 much, but it will increase the amount of blurring needed to cover over
6509 the image and will destroy more information than necessary, and extra
6510 pixels will slow things down on a large logo.
6514 Rotate video by an arbitrary angle expressed in radians.
6516 The filter accepts the following options:
6518 A description of the optional parameters follows.
6521 Set an expression for the angle by which to rotate the input video
6522 clockwise, expressed as a number of radians. A negative value will
6523 result in a counter-clockwise rotation. By default it is set to "0".
6525 This expression is evaluated for each frame.
6528 Set the output width expression, default value is "iw".
6529 This expression is evaluated just once during configuration.
6532 Set the output height expression, default value is "ih".
6533 This expression is evaluated just once during configuration.
6536 Enable bilinear interpolation if set to 1, a value of 0 disables
6537 it. Default value is 1.
6540 Set the color used to fill the output area not covered by the rotated
6541 image. For the generalsyntax of this option, check the "Color" section in the
6542 ffmpeg-utils manual. If the special value "none" is selected then no
6543 background is printed (useful for example if the background is never shown).
6545 Default value is "black".
6548 The expressions for the angle and the output size can contain the
6549 following constants and functions:
6553 sequential number of the input frame, starting from 0. It is always NAN
6554 before the first frame is filtered.
6557 time in seconds of the input frame, it is set to 0 when the filter is
6558 configured. It is always NAN before the first frame is filtered.
6562 horizontal and vertical chroma subsample values. For example for the
6563 pixel format "yuv422p" @var{hsub} is 2 and @var{vsub} is 1.
6567 the input video width and heigth
6571 the output width and heigth, that is the size of the padded area as
6572 specified by the @var{width} and @var{height} expressions
6576 the minimal width/height required for completely containing the input
6577 video rotated by @var{a} radians.
6579 These are only available when computing the @option{out_w} and
6580 @option{out_h} expressions.
6583 @subsection Examples
6587 Rotate the input by PI/6 radians clockwise:
6593 Rotate the input by PI/6 radians counter-clockwise:
6599 Apply a constant rotation with period T, starting from an angle of PI/3:
6601 rotate=PI/3+2*PI*t/T
6605 Make the input video rotation oscillating with a period of T
6606 seconds and an amplitude of A radians:
6608 rotate=A*sin(2*PI/T*t)
6612 Rotate the video, output size is choosen so that the whole rotating
6613 input video is always completely contained in the output:
6615 rotate='2*PI*t:ow=hypot(iw,ih):oh=ow'
6619 Rotate the video, reduce the output size so that no background is ever
6622 rotate=2*PI*t:ow='min(iw,ih)/sqrt(2)':oh=ow:c=none
6626 @subsection Commands
6628 The filter supports the following commands:
6632 Set the angle expression.
6633 The command accepts the same syntax of the corresponding option.
6635 If the specified expression is not valid, it is kept at its current
6641 Apply Shape Adaptive Blur.
6643 The filter accepts the following options:
6646 @item luma_radius, lr
6647 Set luma blur filter strength, must be a value in range 0.1-4.0, default
6648 value is 1.0. A greater value will result in a more blurred image, and
6649 in slower processing.
6651 @item luma_pre_filter_radius, lpfr
6652 Set luma pre-filter radius, must be a value in the 0.1-2.0 range, default
6655 @item luma_strength, ls
6656 Set luma maximum difference between pixels to still be considered, must
6657 be a value in the 0.1-100.0 range, default value is 1.0.
6659 @item chroma_radius, cr
6660 Set chroma blur filter strength, must be a value in range 0.1-4.0. A
6661 greater value will result in a more blurred image, and in slower
6664 @item chroma_pre_filter_radius, cpfr
6665 Set chroma pre-filter radius, must be a value in the 0.1-2.0 range.
6667 @item chroma_strength, cs
6668 Set chroma maximum difference between pixels to still be considered,
6669 must be a value in the 0.1-100.0 range.
6672 Each chroma option value, if not explicitly specified, is set to the
6673 corresponding luma option value.
6678 Scale (resize) the input video, using the libswscale library.
6680 The scale filter forces the output display aspect ratio to be the same
6681 of the input, by changing the output sample aspect ratio.
6683 If the input image format is different from the format requested by
6684 the next filter, the scale filter will convert the input to the
6688 The filter accepts the following options, or any of the options
6689 supported by the libswscale scaler.
6691 See @ref{scaler_options,,the ffmpeg-scaler manual,ffmpeg-scaler} for
6692 the complete list of scaler options.
6697 Set the output video dimension expression. Default value is the input
6700 If the value is 0, the input width is used for the output.
6702 If one of the values is -1, the scale filter will use a value that
6703 maintains the aspect ratio of the input image, calculated from the
6704 other specified dimension. If both of them are -1, the input size is
6707 See below for the list of accepted constants for use in the dimension
6711 Set the interlacing mode. It accepts the following values:
6715 Force interlaced aware scaling.
6718 Do not apply interlaced scaling.
6721 Select interlaced aware scaling depending on whether the source frames
6722 are flagged as interlaced or not.
6725 Default value is @samp{0}.
6728 Set libswscale scaling flags. See
6729 @ref{sws_flags,,the ffmpeg-scaler manual,ffmpeg-scaler} for the
6730 complete list of values. If not explictly specified the filter applies
6734 Set the video size. For the syntax of this option, check the "Video size"
6735 section in the ffmpeg-utils manual.
6737 @item in_color_matrix
6738 @item out_color_matrix
6739 Set in/output YCbCr color space type.
6741 This allows the autodetected value to be overridden as well as allows forcing
6742 a specific value used for the output and encoder.
6744 If not specified, the color space type depends on the pixel format.
6750 Choose automatically.
6753 Format conforming to International Telecommunication Union (ITU)
6754 Recommendation BT.709.
6757 Set color space conforming to the United States Federal Communications
6758 Commission (FCC) Code of Federal Regulations (CFR) Title 47 (2003) 73.682 (a).
6761 Set color space conforming to:
6765 ITU Radiocommunication Sector (ITU-R) Recommendation BT.601
6768 ITU-R Rec. BT.470-6 (1998) Systems B, B1, and G
6771 Society of Motion Picture and Television Engineers (SMPTE) ST 170:2004
6776 Set color space conforming to SMPTE ST 240:1999.
6781 Set in/output YCbCr sample range.
6783 This allows the autodetected value to be overridden as well as allows forcing
6784 a specific value used for the output and encoder. If not specified, the
6785 range depends on the pixel format. Possible values:
6789 Choose automatically.
6792 Set full range (0-255 in case of 8-bit luma).
6795 Set "MPEG" range (16-235 in case of 8-bit luma).
6798 @item force_original_aspect_ratio
6799 Enable decreasing or increasing output video width or height if necessary to
6800 keep the original aspect ratio. Possible values:
6804 Scale the video as specified and disable this feature.
6807 The output video dimensions will automatically be decreased if needed.
6810 The output video dimensions will automatically be increased if needed.
6814 One useful instance of this option is that when you know a specific device's
6815 maximum allowed resolution, you can use this to limit the output video to
6816 that, while retaining the aspect ratio. For example, device A allows
6817 1280x720 playback, and your video is 1920x800. Using this option (set it to
6818 decrease) and specifying 1280x720 to the command line makes the output
6821 Please note that this is a different thing than specifying -1 for @option{w}
6822 or @option{h}, you still need to specify the output resolution for this option
6827 The values of the @option{w} and @option{h} options are expressions
6828 containing the following constants:
6833 the input width and height
6837 same as @var{in_w} and @var{in_h}
6841 the output (scaled) width and height
6845 same as @var{out_w} and @var{out_h}
6848 same as @var{iw} / @var{ih}
6851 input sample aspect ratio
6854 input display aspect ratio. Calculated from @code{(iw / ih) * sar}.
6858 horizontal and vertical input chroma subsample values. For example for the
6859 pixel format "yuv422p" @var{hsub} is 2 and @var{vsub} is 1.
6863 horizontal and vertical output chroma subsample values. For example for the
6864 pixel format "yuv422p" @var{hsub} is 2 and @var{vsub} is 1.
6867 @subsection Examples
6871 Scale the input video to a size of 200x100:
6876 This is equivalent to:
6887 Specify a size abbreviation for the output size:
6892 which can also be written as:
6898 Scale the input to 2x:
6904 The above is the same as:
6910 Scale the input to 2x with forced interlaced scaling:
6912 scale=2*iw:2*ih:interl=1
6916 Scale the input to half size:
6922 Increase the width, and set the height to the same size:
6928 Seek for Greek harmony:
6935 Increase the height, and set the width to 3/2 of the height:
6937 scale=w=3/2*oh:h=3/5*ih
6941 Increase the size, but make the size a multiple of the chroma
6944 scale="trunc(3/2*iw/hsub)*hsub:trunc(3/2*ih/vsub)*vsub"
6948 Increase the width to a maximum of 500 pixels, keep the same input
6951 scale=w='min(500\, iw*3/2):h=-1'
6955 @section separatefields
6957 The @code{separatefields} takes a frame-based video input and splits
6958 each frame into its components fields, producing a new half height clip
6959 with twice the frame rate and twice the frame count.
6961 This filter use field-dominance information in frame to decide which
6962 of each pair of fields to place first in the output.
6963 If it gets it wrong use @ref{setfield} filter before @code{separatefields} filter.
6965 @section setdar, setsar
6967 The @code{setdar} filter sets the Display Aspect Ratio for the filter
6970 This is done by changing the specified Sample (aka Pixel) Aspect
6971 Ratio, according to the following equation:
6973 @var{DAR} = @var{HORIZONTAL_RESOLUTION} / @var{VERTICAL_RESOLUTION} * @var{SAR}
6976 Keep in mind that the @code{setdar} filter does not modify the pixel
6977 dimensions of the video frame. Also the display aspect ratio set by
6978 this filter may be changed by later filters in the filterchain,
6979 e.g. in case of scaling or if another "setdar" or a "setsar" filter is
6982 The @code{setsar} filter sets the Sample (aka Pixel) Aspect Ratio for
6983 the filter output video.
6985 Note that as a consequence of the application of this filter, the
6986 output display aspect ratio will change according to the equation
6989 Keep in mind that the sample aspect ratio set by the @code{setsar}
6990 filter may be changed by later filters in the filterchain, e.g. if
6991 another "setsar" or a "setdar" filter is applied.
6993 The filters accept the following options:
6996 @item r, ratio, dar (@code{setdar} only), sar (@code{setsar} only)
6997 Set the aspect ratio used by the filter.
6999 The parameter can be a floating point number string, an expression, or
7000 a string of the form @var{num}:@var{den}, where @var{num} and
7001 @var{den} are the numerator and denominator of the aspect ratio. If
7002 the parameter is not specified, it is assumed the value "0".
7003 In case the form "@var{num}:@var{den}" is used, the @code{:} character
7007 Set the maximum integer value to use for expressing numerator and
7008 denominator when reducing the expressed aspect ratio to a rational.
7009 Default value is @code{100}.
7013 The parameter @var{sar} is an expression containing
7014 the following constants:
7018 the corresponding mathematical approximated values for e
7019 (euler number), pi (greek PI), phi (golden ratio)
7022 the input width and height
7025 same as @var{w} / @var{h}
7028 input sample aspect ratio
7031 input display aspect ratio, it is the same as (@var{w} / @var{h}) * @var{sar}
7034 horizontal and vertical chroma subsample values. For example for the
7035 pixel format "yuv422p" @var{hsub} is 2 and @var{vsub} is 1.
7038 @subsection Examples
7043 To change the display aspect ratio to 16:9, specify one of the following:
7051 To change the sample aspect ratio to 10:11, specify:
7057 To set a display aspect ratio of 16:9, and specify a maximum integer value of
7058 1000 in the aspect ratio reduction, use the command:
7060 setdar=ratio=16/9:max=1000
7068 Force field for the output video frame.
7070 The @code{setfield} filter marks the interlace type field for the
7071 output frames. It does not change the input frame, but only sets the
7072 corresponding property, which affects how the frame is treated by
7073 following filters (e.g. @code{fieldorder} or @code{yadif}).
7075 The filter accepts the following options:
7080 Available values are:
7084 Keep the same field property.
7087 Mark the frame as bottom-field-first.
7090 Mark the frame as top-field-first.
7093 Mark the frame as progressive.
7099 Show a line containing various information for each input video frame.
7100 The input video is not modified.
7102 The shown line contains a sequence of key/value pairs of the form
7103 @var{key}:@var{value}.
7105 A description of each shown parameter follows:
7109 sequential number of the input frame, starting from 0
7112 Presentation TimeStamp of the input frame, expressed as a number of
7113 time base units. The time base unit depends on the filter input pad.
7116 Presentation TimeStamp of the input frame, expressed as a number of
7120 position of the frame in the input stream, -1 if this information in
7121 unavailable and/or meaningless (for example in case of synthetic video)
7127 sample aspect ratio of the input frame, expressed in the form
7131 size of the input frame. For the syntax of this option, check the "Video size"
7132 section in the ffmpeg-utils manual.
7135 interlaced mode ("P" for "progressive", "T" for top field first, "B"
7136 for bottom field first)
7139 1 if the frame is a key frame, 0 otherwise
7142 picture type of the input frame ("I" for an I-frame, "P" for a
7143 P-frame, "B" for a B-frame, "?" for unknown type).
7144 Check also the documentation of the @code{AVPictureType} enum and of
7145 the @code{av_get_picture_type_char} function defined in
7146 @file{libavutil/avutil.h}.
7149 Adler-32 checksum (printed in hexadecimal) of all the planes of the input frame
7151 @item plane_checksum
7152 Adler-32 checksum (printed in hexadecimal) of each plane of the input frame,
7153 expressed in the form "[@var{c0} @var{c1} @var{c2} @var{c3}]"
7159 Blur the input video without impacting the outlines.
7161 The filter accepts the following options:
7164 @item luma_radius, lr
7165 Set the luma radius. The option value must be a float number in
7166 the range [0.1,5.0] that specifies the variance of the gaussian filter
7167 used to blur the image (slower if larger). Default value is 1.0.
7169 @item luma_strength, ls
7170 Set the luma strength. The option value must be a float number
7171 in the range [-1.0,1.0] that configures the blurring. A value included
7172 in [0.0,1.0] will blur the image whereas a value included in
7173 [-1.0,0.0] will sharpen the image. Default value is 1.0.
7175 @item luma_threshold, lt
7176 Set the luma threshold used as a coefficient to determine
7177 whether a pixel should be blurred or not. The option value must be an
7178 integer in the range [-30,30]. A value of 0 will filter all the image,
7179 a value included in [0,30] will filter flat areas and a value included
7180 in [-30,0] will filter edges. Default value is 0.
7182 @item chroma_radius, cr
7183 Set the chroma radius. The option value must be a float number in
7184 the range [0.1,5.0] that specifies the variance of the gaussian filter
7185 used to blur the image (slower if larger). Default value is 1.0.
7187 @item chroma_strength, cs
7188 Set the chroma strength. The option value must be a float number
7189 in the range [-1.0,1.0] that configures the blurring. A value included
7190 in [0.0,1.0] will blur the image whereas a value included in
7191 [-1.0,0.0] will sharpen the image. Default value is 1.0.
7193 @item chroma_threshold, ct
7194 Set the chroma threshold used as a coefficient to determine
7195 whether a pixel should be blurred or not. The option value must be an
7196 integer in the range [-30,30]. A value of 0 will filter all the image,
7197 a value included in [0,30] will filter flat areas and a value included
7198 in [-30,0] will filter edges. Default value is 0.
7201 If a chroma option is not explicitly set, the corresponding luma value
7206 Convert between different stereoscopic image formats.
7208 The filters accept the following options:
7212 Set stereoscopic image format of input.
7214 Available values for input image formats are:
7217 side by side parallel (left eye left, right eye right)
7220 side by side crosseye (right eye left, left eye right)
7223 side by side parallel with half width resolution
7224 (left eye left, right eye right)
7227 side by side crosseye with half width resolution
7228 (right eye left, left eye right)
7231 above-below (left eye above, right eye below)
7234 above-below (right eye above, left eye below)
7237 above-below with half height resolution
7238 (left eye above, right eye below)
7241 above-below with half height resolution
7242 (right eye above, left eye below)
7245 alternating frames (left eye first, right eye second)
7248 alternating frames (right eye first, left eye second)
7250 Default value is @samp{sbsl}.
7254 Set stereoscopic image format of output.
7256 Available values for output image formats are all the input formats as well as:
7259 anaglyph red/blue gray
7260 (red filter on left eye, blue filter on right eye)
7263 anaglyph red/green gray
7264 (red filter on left eye, green filter on right eye)
7267 anaglyph red/cyan gray
7268 (red filter on left eye, cyan filter on right eye)
7271 anaglyph red/cyan half colored
7272 (red filter on left eye, cyan filter on right eye)
7275 anaglyph red/cyan color
7276 (red filter on left eye, cyan filter on right eye)
7279 anaglyph red/cyan color optimized with the least squares projection of dubois
7280 (red filter on left eye, cyan filter on right eye)
7283 anaglyph green/magenta gray
7284 (green filter on left eye, magenta filter on right eye)
7287 anaglyph green/magenta half colored
7288 (green filter on left eye, magenta filter on right eye)
7291 anaglyph green/magenta colored
7292 (green filter on left eye, magenta filter on right eye)
7295 anaglyph green/magenta color optimized with the least squares projection of dubois
7296 (green filter on left eye, magenta filter on right eye)
7299 anaglyph yellow/blue gray
7300 (yellow filter on left eye, blue filter on right eye)
7303 anaglyph yellow/blue half colored
7304 (yellow filter on left eye, blue filter on right eye)
7307 anaglyph yellow/blue colored
7308 (yellow filter on left eye, blue filter on right eye)
7311 anaglyph yellow/blue color optimized with the least squares projection of dubois
7312 (yellow filter on left eye, blue filter on right eye)
7315 interleaved rows (left eye has top row, right eye starts on next row)
7318 interleaved rows (right eye has top row, left eye starts on next row)
7321 mono output (left eye only)
7324 mono output (right eye only)
7327 Default value is @samp{arcd}.
7330 @subsection Examples
7334 Convert input video from side by side parallel to anaglyph yellow/blue dubois:
7340 Convert input video from above bellow (left eye above, right eye below) to side by side crosseye.
7348 Apply a simple postprocessing filter that compresses and decompresses the image
7349 at several (or - in the case of @option{quality} level @code{6} - all) shifts
7350 and average the results.
7352 The filter accepts the following options:
7356 Set quality. This option defines the number of levels for averaging. It accepts
7357 an integer in the range 0-6. If set to @code{0}, the filter will have no
7358 effect. A value of @code{6} means the higher quality. For each increment of
7359 that value the speed drops by a factor of approximately 2. Default value is
7363 Force a constant quantization parameter. If not set, the filter will use the QP
7364 from the video stream (if available).
7367 Set thresholding mode. Available modes are:
7371 Set hard thresholding (default).
7373 Set soft thresholding (better de-ringing effect, but likely blurrier).
7377 Enable the use of the QP from the B-Frames if set to @code{1}. Using this
7378 option may cause flicker since the B-Frames have often larger QP. Default is
7379 @code{0} (not enabled).
7385 Draw subtitles on top of input video using the libass library.
7387 To enable compilation of this filter you need to configure FFmpeg with
7388 @code{--enable-libass}. This filter also requires a build with libavcodec and
7389 libavformat to convert the passed subtitles file to ASS (Advanced Substation
7390 Alpha) subtitles format.
7392 The filter accepts the following options:
7396 Set the filename of the subtitle file to read. It must be specified.
7399 Specify the size of the original video, the video for which the ASS file
7400 was composed. For the syntax of this option, check the "Video size" section in
7401 the ffmpeg-utils manual. Due to a misdesign in ASS aspect ratio arithmetic,
7402 this is necessary to correctly scale the fonts if the aspect ratio has been
7406 Set subtitles input character encoding. @code{subtitles} filter only. Only
7407 useful if not UTF-8.
7410 If the first key is not specified, it is assumed that the first value
7411 specifies the @option{filename}.
7413 For example, to render the file @file{sub.srt} on top of the input
7414 video, use the command:
7419 which is equivalent to:
7421 subtitles=filename=sub.srt
7426 Scale the input by 2x and smooth using the Super2xSaI (Scale and
7427 Interpolate) pixel art scaling algorithm.
7429 Useful for enlarging pixel art images without reducing sharpness.
7436 Apply telecine process to the video.
7438 This filter accepts the following options:
7447 The default value is @code{top}.
7451 A string of numbers representing the pulldown pattern you wish to apply.
7452 The default value is @code{23}.
7456 Some typical patterns:
7461 24p: 2332 (preferred)
7468 24p: 222222222223 ("Euro pulldown")
7474 Select the most representative frame in a given sequence of consecutive frames.
7476 The filter accepts the following options:
7480 Set the frames batch size to analyze; in a set of @var{n} frames, the filter
7481 will pick one of them, and then handle the next batch of @var{n} frames until
7482 the end. Default is @code{100}.
7485 Since the filter keeps track of the whole frames sequence, a bigger @var{n}
7486 value will result in a higher memory usage, so a high value is not recommended.
7488 @subsection Examples
7492 Extract one picture each 50 frames:
7498 Complete example of a thumbnail creation with @command{ffmpeg}:
7500 ffmpeg -i in.avi -vf thumbnail,scale=300:200 -frames:v 1 out.png
7506 Tile several successive frames together.
7508 The filter accepts the following options:
7513 Set the grid size (i.e. the number of lines and columns). For the syntax of
7514 this option, check the "Video size" section in the ffmpeg-utils manual.
7517 Set the maximum number of frames to render in the given area. It must be less
7518 than or equal to @var{w}x@var{h}. The default value is @code{0}, meaning all
7519 the area will be used.
7522 Set the outer border margin in pixels.
7525 Set the inner border thickness (i.e. the number of pixels between frames). For
7526 more advanced padding options (such as having different values for the edges),
7527 refer to the pad video filter.
7530 Specify the color of the unused areaFor the syntax of this option, check the
7531 "Color" section in the ffmpeg-utils manual. The default value of @var{color}
7535 @subsection Examples
7539 Produce 8x8 PNG tiles of all keyframes (@option{-skip_frame nokey}) in a movie:
7541 ffmpeg -skip_frame nokey -i file.avi -vf 'scale=128:72,tile=8x8' -an -vsync 0 keyframes%03d.png
7543 The @option{-vsync 0} is necessary to prevent @command{ffmpeg} from
7544 duplicating each output frame to accomodate the originally detected frame
7548 Display @code{5} pictures in an area of @code{3x2} frames,
7549 with @code{7} pixels between them, and @code{2} pixels of initial margin, using
7550 mixed flat and named options:
7552 tile=3x2:nb_frames=5:padding=7:margin=2
7558 Perform various types of temporal field interlacing.
7560 Frames are counted starting from 1, so the first input frame is
7563 The filter accepts the following options:
7568 Specify the mode of the interlacing. This option can also be specified
7569 as a value alone. See below for a list of values for this option.
7571 Available values are:
7575 Move odd frames into the upper field, even into the lower field,
7576 generating a double height frame at half frame rate.
7579 Only output even frames, odd frames are dropped, generating a frame with
7580 unchanged height at half frame rate.
7583 Only output odd frames, even frames are dropped, generating a frame with
7584 unchanged height at half frame rate.
7587 Expand each frame to full height, but pad alternate lines with black,
7588 generating a frame with double height at the same input frame rate.
7590 @item interleave_top, 4
7591 Interleave the upper field from odd frames with the lower field from
7592 even frames, generating a frame with unchanged height at half frame rate.
7594 @item interleave_bottom, 5
7595 Interleave the lower field from odd frames with the upper field from
7596 even frames, generating a frame with unchanged height at half frame rate.
7598 @item interlacex2, 6
7599 Double frame rate with unchanged height. Frames are inserted each
7600 containing the second temporal field from the previous input frame and
7601 the first temporal field from the next input frame. This mode relies on
7602 the top_field_first flag. Useful for interlaced video displays with no
7603 field synchronisation.
7606 Numeric values are deprecated but are accepted for backward
7607 compatibility reasons.
7609 Default mode is @code{merge}.
7612 Specify flags influencing the filter process.
7614 Available value for @var{flags} is:
7617 @item low_pass_filter, vlfp
7618 Enable vertical low-pass filtering in the filter.
7619 Vertical low-pass filtering is required when creating an interlaced
7620 destination from a progressive source which contains high-frequency
7621 vertical detail. Filtering will reduce interlace 'twitter' and Moire
7624 Vertical low-pass filtering can only be enabled for @option{mode}
7625 @var{interleave_top} and @var{interleave_bottom}.
7632 Transpose rows with columns in the input video and optionally flip it.
7634 This filter accepts the following options:
7639 Specify the transposition direction.
7641 Can assume the following values:
7643 @item 0, 4, cclock_flip
7644 Rotate by 90 degrees counterclockwise and vertically flip (default), that is:
7652 Rotate by 90 degrees clockwise, that is:
7660 Rotate by 90 degrees counterclockwise, that is:
7667 @item 3, 7, clock_flip
7668 Rotate by 90 degrees clockwise and vertically flip, that is:
7676 For values between 4-7, the transposition is only done if the input
7677 video geometry is portrait and not landscape. These values are
7678 deprecated, the @code{passthrough} option should be used instead.
7680 Numerical values are deprecated, and should be dropped in favor of
7684 Do not apply the transposition if the input geometry matches the one
7685 specified by the specified value. It accepts the following values:
7688 Always apply transposition.
7690 Preserve portrait geometry (when @var{height} >= @var{width}).
7692 Preserve landscape geometry (when @var{width} >= @var{height}).
7695 Default value is @code{none}.
7698 For example to rotate by 90 degrees clockwise and preserve portrait
7701 transpose=dir=1:passthrough=portrait
7704 The command above can also be specified as:
7706 transpose=1:portrait
7710 Trim the input so that the output contains one continuous subpart of the input.
7712 This filter accepts the following options:
7715 Specify time of the start of the kept section, i.e. the frame with the
7716 timestamp @var{start} will be the first frame in the output.
7719 Specify time of the first frame that will be dropped, i.e. the frame
7720 immediately preceding the one with the timestamp @var{end} will be the last
7721 frame in the output.
7724 Same as @var{start}, except this option sets the start timestamp in timebase
7725 units instead of seconds.
7728 Same as @var{end}, except this option sets the end timestamp in timebase units
7732 Specify maximum duration of the output.
7735 Number of the first frame that should be passed to output.
7738 Number of the first frame that should be dropped.
7741 @option{start}, @option{end}, @option{duration} are expressed as time
7742 duration specifications, check the "Time duration" section in the
7743 ffmpeg-utils manual.
7745 Note that the first two sets of the start/end options and the @option{duration}
7746 option look at the frame timestamp, while the _frame variants simply count the
7747 frames that pass through the filter. Also note that this filter does not modify
7748 the timestamps. If you wish that the output timestamps start at zero, insert a
7749 setpts filter after the trim filter.
7751 If multiple start or end options are set, this filter tries to be greedy and
7752 keep all the frames that match at least one of the specified constraints. To keep
7753 only the part that matches all the constraints at once, chain multiple trim
7756 The defaults are such that all the input is kept. So it is possible to set e.g.
7757 just the end values to keep everything before the specified time.
7762 drop everything except the second minute of input
7764 ffmpeg -i INPUT -vf trim=60:120
7768 keep only the first second
7770 ffmpeg -i INPUT -vf trim=duration=1
7778 Sharpen or blur the input video.
7780 It accepts the following parameters:
7783 @item luma_msize_x, lx
7784 Set the luma matrix horizontal size. It must be an odd integer between
7785 3 and 63, default value is 5.
7787 @item luma_msize_y, ly
7788 Set the luma matrix vertical size. It must be an odd integer between 3
7789 and 63, default value is 5.
7791 @item luma_amount, la
7792 Set the luma effect strength. It can be a float number, reasonable
7793 values lay between -1.5 and 1.5.
7795 Negative values will blur the input video, while positive values will
7796 sharpen it, a value of zero will disable the effect.
7798 Default value is 1.0.
7800 @item chroma_msize_x, cx
7801 Set the chroma matrix horizontal size. It must be an odd integer
7802 between 3 and 63, default value is 5.
7804 @item chroma_msize_y, cy
7805 Set the chroma matrix vertical size. It must be an odd integer
7806 between 3 and 63, default value is 5.
7808 @item chroma_amount, ca
7809 Set the chroma effect strength. It can be a float number, reasonable
7810 values lay between -1.5 and 1.5.
7812 Negative values will blur the input video, while positive values will
7813 sharpen it, a value of zero will disable the effect.
7815 Default value is 0.0.
7818 If set to 1, specify using OpenCL capabilities, only available if
7819 FFmpeg was configured with @code{--enable-opencl}. Default value is 0.
7823 All parameters are optional and default to the equivalent of the
7824 string '5:5:1.0:5:5:0.0'.
7826 @subsection Examples
7830 Apply strong luma sharpen effect:
7832 unsharp=luma_msize_x=7:luma_msize_y=7:luma_amount=2.5
7836 Apply strong blur of both luma and chroma parameters:
7838 unsharp=7:7:-2:7:7:-2
7842 @anchor{vidstabdetect}
7843 @section vidstabdetect
7845 Analyze video stabilization/deshaking. Perform pass 1 of 2, see
7846 @ref{vidstabtransform} for pass 2.
7848 This filter generates a file with relative translation and rotation
7849 transform information about subsequent frames, which is then used by
7850 the @ref{vidstabtransform} filter.
7852 To enable compilation of this filter you need to configure FFmpeg with
7853 @code{--enable-libvidstab}.
7855 This filter accepts the following options:
7859 Set the path to the file used to write the transforms information.
7860 Default value is @file{transforms.trf}.
7863 Set how shaky the video is and how quick the camera is. It accepts an
7864 integer in the range 1-10, a value of 1 means little shakiness, a
7865 value of 10 means strong shakiness. Default value is 5.
7868 Set the accuracy of the detection process. It must be a value in the
7869 range 1-15. A value of 1 means low accuracy, a value of 15 means high
7870 accuracy. Default value is 9.
7873 Set stepsize of the search process. The region around minimum is
7874 scanned with 1 pixel resolution. Default value is 6.
7877 Set minimum contrast. Below this value a local measurement field is
7878 discarded. Must be a floating point value in the range 0-1. Default
7882 Set reference frame number for tripod mode.
7884 If enabled, the motion of the frames is compared to a reference frame
7885 in the filtered stream, identified by the specified number. The idea
7886 is to compensate all movements in a more-or-less static scene and keep
7887 the camera view absolutely still.
7889 If set to 0, it is disabled. The frames are counted starting from 1.
7892 Show fields and transforms in the resulting frames. It accepts an
7893 integer in the range 0-2. Default value is 0, which disables any
7897 @subsection Examples
7907 Analyze strongly shaky movie and put the results in file
7908 @file{mytransforms.trf}:
7910 vidstabdetect=shakiness=10:accuracy=15:result="mytransforms.trf"
7914 Visualize the result of internal transformations in the resulting
7917 vidstabdetect=show=1
7921 Analyze a video with medium shakiness using @command{ffmpeg}:
7923 ffmpeg -i input -vf vidstabdetect=shakiness=5:show=1 dummy.avi
7927 @anchor{vidstabtransform}
7928 @section vidstabtransform
7930 Video stabilization/deshaking: pass 2 of 2,
7931 see @ref{vidstabdetect} for pass 1.
7933 Read a file with transform information for each frame and
7934 apply/compensate them. Together with the @ref{vidstabdetect}
7935 filter this can be used to deshake videos. See also
7936 @url{http://public.hronopik.de/vid.stab}. It is important to also use
7937 the unsharp filter, see below.
7939 To enable compilation of this filter you need to configure FFmpeg with
7940 @code{--enable-libvidstab}.
7942 This filter accepts the following options:
7947 path to the file used to read the transforms (default: @file{transforms.trf})
7950 number of frames (value*2 + 1) used for lowpass filtering the camera movements
7951 (default: 10). For example a number of 10 means that 21 frames are used
7952 (10 in the past and 10 in the future) to smoothen the motion in the
7953 video. A larger values leads to a smoother video, but limits the
7954 acceleration of the camera (pan/tilt movements).
7957 maximal number of pixels to translate frames (default: -1 no limit)
7960 maximal angle in radians (degree*PI/180) to rotate frames (default: -1
7964 How to deal with borders that may be visible due to movement
7965 compensation. Available values are:
7969 keep image information from previous frame (default)
7971 fill the border black
7977 keep transforms normal (default)
7983 consider transforms as
7988 relative to previous frame (default)
7992 percentage to zoom (default: 0)
8001 set optimal zooming to avoid borders
8006 optimal static zoom value is determined (only very strong movements will lead to visible borders) (default)
8008 optimal adaptive zoom value is determined (no borders will be visible)
8010 Note that the value given at zoom is added to the one calculated
8014 type of interpolation
8016 Available values are:
8021 linear only horizontal
8023 linear in both directions (default)
8025 cubic in both directions (slow)
8029 virtual tripod mode means that the video is stabilized such that the
8030 camera stays stationary. Use also @code{tripod} option of
8031 @ref{vidstabdetect}.
8036 virtual tripod mode: equivalent to @code{relative=0:smoothing=0}
8041 @subsection Examples
8045 typical call with default default values:
8046 (note the unsharp filter which is always recommended)
8048 ffmpeg -i inp.mpeg -vf vidstabtransform,unsharp=5:5:0.8:3:3:0.4 inp_stabilized.mpeg
8052 zoom in a bit more and load transform data from a given file
8054 vidstabtransform=zoom=5:input="mytransforms.trf"
8058 smoothen the video even more
8060 vidstabtransform=smoothing=30
8067 Flip the input video vertically.
8069 For example, to vertically flip a video with @command{ffmpeg}:
8071 ffmpeg -i in.avi -vf "vflip" out.avi
8076 Make or reverse a natural vignetting effect.
8078 The filter accepts the following options:
8082 Set lens angle expression as a number of radians.
8084 The value is clipped in the @code{[0,PI/2]} range.
8086 Default value: @code{"PI/5"}
8090 Set center coordinates expressions. Respectively @code{"w/2"} and @code{"h/2"}
8094 Set forward/backward mode.
8096 Available modes are:
8099 The larger the distance from the central point, the darker the image becomes.
8102 The larger the distance from the central point, the brighter the image becomes.
8103 This can be used to reverse a vignette effect, though there is no automatic
8104 detection to extract the lens @option{angle} and other settings (yet). It can
8105 also be used to create a burning effect.
8108 Default value is @samp{forward}.
8111 Set evaluation mode for the expressions (@option{angle}, @option{x0}, @option{y0}).
8113 It accepts the following values:
8116 Evaluate expressions only once during the filter initialization.
8119 Evaluate expressions for each incoming frame. This is way slower than the
8120 @samp{init} mode since it requires all the scalers to be re-computed, but it
8121 allows advanced dynamic expressions.
8124 Default value is @samp{init}.
8127 Set dithering to reduce the circular banding effects. Default is @code{1}
8131 Set vignette aspect. This setting allows to adjust the shape of the vignette.
8132 Setting this value to the SAR of the input will make a rectangular vignetting
8133 following the dimensions of the video.
8135 Default is @code{1/1}.
8138 @subsection Expressions
8140 The @option{alpha}, @option{x0} and @option{y0} expressions can contain the
8141 following parameters.
8146 input width and height
8149 the number of input frame, starting from 0
8152 the PTS (Presentation TimeStamp) time of the filtered video frame, expressed in
8153 @var{TB} units, NAN if undefined
8156 frame rate of the input video, NAN if the input frame rate is unknown
8159 the PTS (Presentation TimeStamp) of the filtered video frame,
8160 expressed in seconds, NAN if undefined
8163 time base of the input video
8167 @subsection Examples
8171 Apply simple strong vignetting effect:
8177 Make a flickering vignetting:
8179 vignette='PI/4+random(1)*PI/50':eval=frame
8186 Deinterlace the input video ("w3fdif" stands for "Weston 3 Field
8187 Deinterlacing Filter").
8189 Based on the process described by Martin Weston for BBC R&D, and
8190 implemented based on the de-interlace algorithm written by Jim
8191 Easterbrook for BBC R&D, the Weston 3 field deinterlacing filter
8192 uses filter coefficients calculated by BBC R&D.
8194 There are two sets of filter coefficients, so called "simple":
8195 and "complex". Which set of filter coefficients is used can
8196 be set by passing an optional parameter:
8200 Set the interlacing filter coefficients. Accepts one of the following values:
8204 Simple filter coefficient set.
8206 More-complex filter coefficient set.
8208 Default value is @samp{complex}.
8211 Specify which frames to deinterlace. Accept one of the following values:
8215 Deinterlace all frames,
8217 Only deinterlace frames marked as interlaced.
8220 Default value is @samp{all}.
8226 Deinterlace the input video ("yadif" means "yet another deinterlacing
8229 This filter accepts the following options:
8235 The interlacing mode to adopt, accepts one of the following values:
8239 output 1 frame for each frame
8241 output 1 frame for each field
8242 @item 2, send_frame_nospatial
8243 like @code{send_frame} but skip spatial interlacing check
8244 @item 3, send_field_nospatial
8245 like @code{send_field} but skip spatial interlacing check
8248 Default value is @code{send_frame}.
8251 The picture field parity assumed for the input interlaced video, accepts one of
8252 the following values:
8256 assume top field first
8258 assume bottom field first
8260 enable automatic detection
8263 Default value is @code{auto}.
8264 If interlacing is unknown or decoder does not export this information,
8265 top field first will be assumed.
8268 Specify which frames to deinterlace. Accept one of the following
8273 deinterlace all frames
8275 only deinterlace frames marked as interlaced
8278 Default value is @code{all}.
8281 @c man end VIDEO FILTERS
8283 @chapter Video Sources
8284 @c man begin VIDEO SOURCES
8286 Below is a description of the currently available video sources.
8290 Buffer video frames, and make them available to the filter chain.
8292 This source is mainly intended for a programmatic use, in particular
8293 through the interface defined in @file{libavfilter/vsrc_buffer.h}.
8295 This source accepts the following options:
8300 Specify the size (width and height) of the buffered video frames. For the
8301 syntax of this option, check the "Video size" section in the ffmpeg-utils
8311 A string representing the pixel format of the buffered video frames.
8312 It may be a number corresponding to a pixel format, or a pixel format
8316 Specify the timebase assumed by the timestamps of the buffered frames.
8319 Specify the frame rate expected for the video stream.
8321 @item pixel_aspect, sar
8322 Specify the sample aspect ratio assumed by the video frames.
8325 Specify the optional parameters to be used for the scale filter which
8326 is automatically inserted when an input change is detected in the
8327 input size or format.
8332 buffer=width=320:height=240:pix_fmt=yuv410p:time_base=1/24:sar=1
8335 will instruct the source to accept video frames with size 320x240 and
8336 with format "yuv410p", assuming 1/24 as the timestamps timebase and
8337 square pixels (1:1 sample aspect ratio).
8338 Since the pixel format with name "yuv410p" corresponds to the number 6
8339 (check the enum AVPixelFormat definition in @file{libavutil/pixfmt.h}),
8340 this example corresponds to:
8342 buffer=size=320x240:pixfmt=6:time_base=1/24:pixel_aspect=1/1
8345 Alternatively, the options can be specified as a flat string, but this
8346 syntax is deprecated:
8348 @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}]
8352 Create a pattern generated by an elementary cellular automaton.
8354 The initial state of the cellular automaton can be defined through the
8355 @option{filename}, and @option{pattern} options. If such options are
8356 not specified an initial state is created randomly.
8358 At each new frame a new row in the video is filled with the result of
8359 the cellular automaton next generation. The behavior when the whole
8360 frame is filled is defined by the @option{scroll} option.
8362 This source accepts the following options:
8366 Read the initial cellular automaton state, i.e. the starting row, from
8368 In the file, each non-whitespace character is considered an alive
8369 cell, a newline will terminate the row, and further characters in the
8370 file will be ignored.
8373 Read the initial cellular automaton state, i.e. the starting row, from
8374 the specified string.
8376 Each non-whitespace character in the string is considered an alive
8377 cell, a newline will terminate the row, and further characters in the
8378 string will be ignored.
8381 Set the video rate, that is the number of frames generated per second.
8384 @item random_fill_ratio, ratio
8385 Set the random fill ratio for the initial cellular automaton row. It
8386 is a floating point number value ranging from 0 to 1, defaults to
8389 This option is ignored when a file or a pattern is specified.
8391 @item random_seed, seed
8392 Set the seed for filling randomly the initial row, must be an integer
8393 included between 0 and UINT32_MAX. If not specified, or if explicitly
8394 set to -1, the filter will try to use a good random seed on a best
8398 Set the cellular automaton rule, it is a number ranging from 0 to 255.
8399 Default value is 110.
8402 Set the size of the output video. For the syntax of this option, check
8403 the "Video size" section in the ffmpeg-utils manual.
8405 If @option{filename} or @option{pattern} is specified, the size is set
8406 by default to the width of the specified initial state row, and the
8407 height is set to @var{width} * PHI.
8409 If @option{size} is set, it must contain the width of the specified
8410 pattern string, and the specified pattern will be centered in the
8413 If a filename or a pattern string is not specified, the size value
8414 defaults to "320x518" (used for a randomly generated initial state).
8417 If set to 1, scroll the output upward when all the rows in the output
8418 have been already filled. If set to 0, the new generated row will be
8419 written over the top row just after the bottom row is filled.
8422 @item start_full, full
8423 If set to 1, completely fill the output with generated rows before
8424 outputting the first frame.
8425 This is the default behavior, for disabling set the value to 0.
8428 If set to 1, stitch the left and right row edges together.
8429 This is the default behavior, for disabling set the value to 0.
8432 @subsection Examples
8436 Read the initial state from @file{pattern}, and specify an output of
8439 cellauto=f=pattern:s=200x400
8443 Generate a random initial row with a width of 200 cells, with a fill
8446 cellauto=ratio=2/3:s=200x200
8450 Create a pattern generated by rule 18 starting by a single alive cell
8451 centered on an initial row with width 100:
8453 cellauto=p=@@:s=100x400:full=0:rule=18
8457 Specify a more elaborated initial pattern:
8459 cellauto=p='@@@@ @@ @@@@':s=100x400:full=0:rule=18
8466 Generate a Mandelbrot set fractal, and progressively zoom towards the
8467 point specified with @var{start_x} and @var{start_y}.
8469 This source accepts the following options:
8474 Set the terminal pts value. Default value is 400.
8477 Set the terminal scale value.
8478 Must be a floating point value. Default value is 0.3.
8481 Set the inner coloring mode, that is the algorithm used to draw the
8482 Mandelbrot fractal internal region.
8484 It shall assume one of the following values:
8489 Show time until convergence.
8491 Set color based on point closest to the origin of the iterations.
8496 Default value is @var{mincol}.
8499 Set the bailout value. Default value is 10.0.
8502 Set the maximum of iterations performed by the rendering
8503 algorithm. Default value is 7189.
8506 Set outer coloring mode.
8507 It shall assume one of following values:
8509 @item iteration_count
8510 Set iteration cound mode.
8511 @item normalized_iteration_count
8512 set normalized iteration count mode.
8514 Default value is @var{normalized_iteration_count}.
8517 Set frame rate, expressed as number of frames per second. Default
8521 Set frame size. For the syntax of this option, check the "Video
8522 size" section in the ffmpeg-utils manual. Default value is "640x480".
8525 Set the initial scale value. Default value is 3.0.
8528 Set the initial x position. Must be a floating point value between
8529 -100 and 100. Default value is -0.743643887037158704752191506114774.
8532 Set the initial y position. Must be a floating point value between
8533 -100 and 100. Default value is -0.131825904205311970493132056385139.
8538 Generate various test patterns, as generated by the MPlayer test filter.
8540 The size of the generated video is fixed, and is 256x256.
8541 This source is useful in particular for testing encoding features.
8543 This source accepts the following options:
8548 Specify the frame rate of the sourced video, as the number of frames
8549 generated per second. It has to be a string in the format
8550 @var{frame_rate_num}/@var{frame_rate_den}, an integer number, a float
8551 number or a valid video frame rate abbreviation. The default value is
8555 Set the video duration of the sourced video. The accepted syntax is:
8560 See also the function @code{av_parse_time()}.
8562 If not specified, or the expressed duration is negative, the video is
8563 supposed to be generated forever.
8567 Set the number or the name of the test to perform. Supported tests are:
8582 Default value is "all", which will cycle through the list of all tests.
8585 For example the following:
8590 will generate a "dc_luma" test pattern.
8594 Provide a frei0r source.
8596 To enable compilation of this filter you need to install the frei0r
8597 header and configure FFmpeg with @code{--enable-frei0r}.
8599 This source accepts the following options:
8604 The size of the video to generate. For the syntax of this option, check the
8605 "Video size" section in the ffmpeg-utils manual.
8608 Framerate of the generated video, may be a string of the form
8609 @var{num}/@var{den} or a frame rate abbreviation.
8612 The name to the frei0r source to load. For more information regarding frei0r and
8613 how to set the parameters read the section @ref{frei0r} in the description of
8617 A '|'-separated list of parameters to pass to the frei0r source.
8621 For example, to generate a frei0r partik0l source with size 200x200
8622 and frame rate 10 which is overlayed on the overlay filter main input:
8624 frei0r_src=size=200x200:framerate=10:filter_name=partik0l:filter_params=1234 [overlay]; [in][overlay] overlay
8629 Generate a life pattern.
8631 This source is based on a generalization of John Conway's life game.
8633 The sourced input represents a life grid, each pixel represents a cell
8634 which can be in one of two possible states, alive or dead. Every cell
8635 interacts with its eight neighbours, which are the cells that are
8636 horizontally, vertically, or diagonally adjacent.
8638 At each interaction the grid evolves according to the adopted rule,
8639 which specifies the number of neighbor alive cells which will make a
8640 cell stay alive or born. The @option{rule} option allows to specify
8643 This source accepts the following options:
8647 Set the file from which to read the initial grid state. In the file,
8648 each non-whitespace character is considered an alive cell, and newline
8649 is used to delimit the end of each row.
8651 If this option is not specified, the initial grid is generated
8655 Set the video rate, that is the number of frames generated per second.
8658 @item random_fill_ratio, ratio
8659 Set the random fill ratio for the initial random grid. It is a
8660 floating point number value ranging from 0 to 1, defaults to 1/PHI.
8661 It is ignored when a file is specified.
8663 @item random_seed, seed
8664 Set the seed for filling the initial random grid, must be an integer
8665 included between 0 and UINT32_MAX. If not specified, or if explicitly
8666 set to -1, the filter will try to use a good random seed on a best
8672 A rule can be specified with a code of the kind "S@var{NS}/B@var{NB}",
8673 where @var{NS} and @var{NB} are sequences of numbers in the range 0-8,
8674 @var{NS} specifies the number of alive neighbor cells which make a
8675 live cell stay alive, and @var{NB} the number of alive neighbor cells
8676 which make a dead cell to become alive (i.e. to "born").
8677 "s" and "b" can be used in place of "S" and "B", respectively.
8679 Alternatively a rule can be specified by an 18-bits integer. The 9
8680 high order bits are used to encode the next cell state if it is alive
8681 for each number of neighbor alive cells, the low order bits specify
8682 the rule for "borning" new cells. Higher order bits encode for an
8683 higher number of neighbor cells.
8684 For example the number 6153 = @code{(12<<9)+9} specifies a stay alive
8685 rule of 12 and a born rule of 9, which corresponds to "S23/B03".
8687 Default value is "S23/B3", which is the original Conway's game of life
8688 rule, and will keep a cell alive if it has 2 or 3 neighbor alive
8689 cells, and will born a new cell if there are three alive cells around
8693 Set the size of the output video. For the syntax of this option, check the
8694 "Video size" section in the ffmpeg-utils manual.
8696 If @option{filename} is specified, the size is set by default to the
8697 same size of the input file. If @option{size} is set, it must contain
8698 the size specified in the input file, and the initial grid defined in
8699 that file is centered in the larger resulting area.
8701 If a filename is not specified, the size value defaults to "320x240"
8702 (used for a randomly generated initial grid).
8705 If set to 1, stitch the left and right grid edges together, and the
8706 top and bottom edges also. Defaults to 1.
8709 Set cell mold speed. If set, a dead cell will go from @option{death_color} to
8710 @option{mold_color} with a step of @option{mold}. @option{mold} can have a
8711 value from 0 to 255.
8714 Set the color of living (or new born) cells.
8717 Set the color of dead cells. If @option{mold} is set, this is the first color
8718 used to represent a dead cell.
8721 Set mold color, for definitely dead and moldy cells.
8723 For the syntax of these 3 color options, check the "Color" section in the
8724 ffmpeg-utils manual.
8727 @subsection Examples
8731 Read a grid from @file{pattern}, and center it on a grid of size
8734 life=f=pattern:s=300x300
8738 Generate a random grid of size 200x200, with a fill ratio of 2/3:
8740 life=ratio=2/3:s=200x200
8744 Specify a custom rule for evolving a randomly generated grid:
8750 Full example with slow death effect (mold) using @command{ffplay}:
8752 ffplay -f lavfi life=s=300x200:mold=10:r=60:ratio=0.1:death_color=#C83232:life_color=#00ff00,scale=1200:800:flags=16
8757 @anchor{haldclutsrc}
8761 @anchor{smptehdbars}
8763 @section color, haldclutsrc, nullsrc, rgbtestsrc, smptebars, smptehdbars, testsrc
8765 The @code{color} source provides an uniformly colored input.
8767 The @code{haldclutsrc} source provides an identity Hald CLUT. See also
8768 @ref{haldclut} filter.
8770 The @code{nullsrc} source returns unprocessed video frames. It is
8771 mainly useful to be employed in analysis / debugging tools, or as the
8772 source for filters which ignore the input data.
8774 The @code{rgbtestsrc} source generates an RGB test pattern useful for
8775 detecting RGB vs BGR issues. You should see a red, green and blue
8776 stripe from top to bottom.
8778 The @code{smptebars} source generates a color bars pattern, based on
8779 the SMPTE Engineering Guideline EG 1-1990.
8781 The @code{smptehdbars} source generates a color bars pattern, based on
8782 the SMPTE RP 219-2002.
8784 The @code{testsrc} source generates a test video pattern, showing a
8785 color pattern, a scrolling gradient and a timestamp. This is mainly
8786 intended for testing purposes.
8788 The sources accept the following options:
8793 Specify the color of the source, only available in the @code{color}
8794 source. For the syntax of this option, check the "Color" section in the
8795 ffmpeg-utils manual.
8798 Specify the level of the Hald CLUT, only available in the @code{haldclutsrc}
8799 source. A level of @code{N} generates a picture of @code{N*N*N} by @code{N*N*N}
8800 pixels to be used as identity matrix for 3D lookup tables. Each component is
8801 coded on a @code{1/(N*N)} scale.
8804 Specify the size of the sourced video. For the syntax of this option, check the
8805 "Video size" section in the ffmpeg-utils manual. The default value is
8808 This option is not available with the @code{haldclutsrc} filter.
8811 Specify the frame rate of the sourced video, as the number of frames
8812 generated per second. It has to be a string in the format
8813 @var{frame_rate_num}/@var{frame_rate_den}, an integer number, a float
8814 number or a valid video frame rate abbreviation. The default value is
8818 Set the sample aspect ratio of the sourced video.
8821 Set the video duration of the sourced video. The accepted syntax is:
8823 [-]HH[:MM[:SS[.m...]]]
8826 See also the function @code{av_parse_time()}.
8828 If not specified, or the expressed duration is negative, the video is
8829 supposed to be generated forever.
8832 Set the number of decimals to show in the timestamp, only available in the
8833 @code{testsrc} source.
8835 The displayed timestamp value will correspond to the original
8836 timestamp value multiplied by the power of 10 of the specified
8837 value. Default value is 0.
8840 For example the following:
8842 testsrc=duration=5.3:size=qcif:rate=10
8845 will generate a video with a duration of 5.3 seconds, with size
8846 176x144 and a frame rate of 10 frames per second.
8848 The following graph description will generate a red source
8849 with an opacity of 0.2, with size "qcif" and a frame rate of 10
8852 color=c=red@@0.2:s=qcif:r=10
8855 If the input content is to be ignored, @code{nullsrc} can be used. The
8856 following command generates noise in the luminance plane by employing
8857 the @code{geq} filter:
8859 nullsrc=s=256x256, geq=random(1)*255:128:128
8862 @subsection Commands
8864 The @code{color} source supports the following commands:
8868 Set the color of the created image. Accepts the same syntax of the
8869 corresponding @option{color} option.
8872 @c man end VIDEO SOURCES
8874 @chapter Video Sinks
8875 @c man begin VIDEO SINKS
8877 Below is a description of the currently available video sinks.
8881 Buffer video frames, and make them available to the end of the filter
8884 This sink is mainly intended for a programmatic use, in particular
8885 through the interface defined in @file{libavfilter/buffersink.h}
8886 or the options system.
8888 It accepts a pointer to an AVBufferSinkContext structure, which
8889 defines the incoming buffers' formats, to be passed as the opaque
8890 parameter to @code{avfilter_init_filter} for initialization.
8894 Null video sink, do absolutely nothing with the input video. It is
8895 mainly useful as a template and to be employed in analysis / debugging
8898 @c man end VIDEO SINKS
8900 @chapter Multimedia Filters
8901 @c man begin MULTIMEDIA FILTERS
8903 Below is a description of the currently available multimedia filters.
8905 @section avectorscope
8907 Convert input audio to a video output, representing the audio vector
8910 The filter is used to measure the difference between channels of stereo
8911 audio stream. A monoaural signal, consisting of identical left and right
8912 signal, results in straight vertical line. Any stereo separation is visible
8913 as a deviation from this line, creating a Lissajous figure.
8914 If the straight (or deviation from it) but horizontal line appears this
8915 indicates that the left and right channels are out of phase.
8917 The filter accepts the following options:
8921 Set the vectorscope mode.
8923 Available values are:
8926 Lissajous rotated by 45 degrees.
8929 Same as above but not rotated.
8932 Default value is @samp{lissajous}.
8935 Set the video size for the output. For the syntax of this option, check the "Video size"
8936 section in the ffmpeg-utils manual. Default value is @code{400x400}.
8939 Set the output frame rate. Default value is @code{25}.
8944 Specify the red, green and blue contrast. Default values are @code{40}, @code{160} and @code{80}.
8945 Allowed range is @code{[0, 255]}.
8950 Specify the red, green and blue fade. Default values are @code{15}, @code{10} and @code{5}.
8951 Allowed range is @code{[0, 255]}.
8954 Set the zoom factor. Default value is @code{1}. Allowed range is @code{[1, 10]}.
8957 @subsection Examples
8961 Complete example using @command{ffplay}:
8963 ffplay -f lavfi 'amovie=input.mp3, asplit [a][out1];
8964 [a] avectorscope=zoom=1.3:rc=2:gc=200:bc=10:rf=1:gf=8:bf=7 [out0]'
8970 Concatenate audio and video streams, joining them together one after the
8973 The filter works on segments of synchronized video and audio streams. All
8974 segments must have the same number of streams of each type, and that will
8975 also be the number of streams at output.
8977 The filter accepts the following options:
8982 Set the number of segments. Default is 2.
8985 Set the number of output video streams, that is also the number of video
8986 streams in each segment. Default is 1.
8989 Set the number of output audio streams, that is also the number of video
8990 streams in each segment. Default is 0.
8993 Activate unsafe mode: do not fail if segments have a different format.
8997 The filter has @var{v}+@var{a} outputs: first @var{v} video outputs, then
8998 @var{a} audio outputs.
9000 There are @var{n}x(@var{v}+@var{a}) inputs: first the inputs for the first
9001 segment, in the same order as the outputs, then the inputs for the second
9004 Related streams do not always have exactly the same duration, for various
9005 reasons including codec frame size or sloppy authoring. For that reason,
9006 related synchronized streams (e.g. a video and its audio track) should be
9007 concatenated at once. The concat filter will use the duration of the longest
9008 stream in each segment (except the last one), and if necessary pad shorter
9009 audio streams with silence.
9011 For this filter to work correctly, all segments must start at timestamp 0.
9013 All corresponding streams must have the same parameters in all segments; the
9014 filtering system will automatically select a common pixel format for video
9015 streams, and a common sample format, sample rate and channel layout for
9016 audio streams, but other settings, such as resolution, must be converted
9017 explicitly by the user.
9019 Different frame rates are acceptable but will result in variable frame rate
9020 at output; be sure to configure the output file to handle it.
9022 @subsection Examples
9026 Concatenate an opening, an episode and an ending, all in bilingual version
9027 (video in stream 0, audio in streams 1 and 2):
9029 ffmpeg -i opening.mkv -i episode.mkv -i ending.mkv -filter_complex \
9030 '[0:0] [0:1] [0:2] [1:0] [1:1] [1:2] [2:0] [2:1] [2:2]
9031 concat=n=3:v=1:a=2 [v] [a1] [a2]' \
9032 -map '[v]' -map '[a1]' -map '[a2]' output.mkv
9036 Concatenate two parts, handling audio and video separately, using the
9037 (a)movie sources, and adjusting the resolution:
9039 movie=part1.mp4, scale=512:288 [v1] ; amovie=part1.mp4 [a1] ;
9040 movie=part2.mp4, scale=512:288 [v2] ; amovie=part2.mp4 [a2] ;
9041 [v1] [v2] concat [outv] ; [a1] [a2] concat=v=0:a=1 [outa]
9043 Note that a desync will happen at the stitch if the audio and video streams
9044 do not have exactly the same duration in the first file.
9050 EBU R128 scanner filter. This filter takes an audio stream as input and outputs
9051 it unchanged. By default, it logs a message at a frequency of 10Hz with the
9052 Momentary loudness (identified by @code{M}), Short-term loudness (@code{S}),
9053 Integrated loudness (@code{I}) and Loudness Range (@code{LRA}).
9055 The filter also has a video output (see the @var{video} option) with a real
9056 time graph to observe the loudness evolution. The graphic contains the logged
9057 message mentioned above, so it is not printed anymore when this option is set,
9058 unless the verbose logging is set. The main graphing area contains the
9059 short-term loudness (3 seconds of analysis), and the gauge on the right is for
9060 the momentary loudness (400 milliseconds).
9062 More information about the Loudness Recommendation EBU R128 on
9063 @url{http://tech.ebu.ch/loudness}.
9065 The filter accepts the following options:
9070 Activate the video output. The audio stream is passed unchanged whether this
9071 option is set or no. The video stream will be the first output stream if
9072 activated. Default is @code{0}.
9075 Set the video size. This option is for video only. For the syntax of this
9076 option, check the "Video size" section in the ffmpeg-utils manual. Default
9077 and minimum resolution is @code{640x480}.
9080 Set the EBU scale meter. Default is @code{9}. Common values are @code{9} and
9081 @code{18}, respectively for EBU scale meter +9 and EBU scale meter +18. Any
9082 other integer value between this range is allowed.
9085 Set metadata injection. If set to @code{1}, the audio input will be segmented
9086 into 100ms output frames, each of them containing various loudness information
9087 in metadata. All the metadata keys are prefixed with @code{lavfi.r128.}.
9089 Default is @code{0}.
9092 Force the frame logging level.
9094 Available values are:
9097 information logging level
9099 verbose logging level
9102 By default, the logging level is set to @var{info}. If the @option{video} or
9103 the @option{metadata} options are set, it switches to @var{verbose}.
9106 @subsection Examples
9110 Real-time graph using @command{ffplay}, with a EBU scale meter +18:
9112 ffplay -f lavfi -i "amovie=input.mp3,ebur128=video=1:meter=18 [out0][out1]"
9116 Run an analysis with @command{ffmpeg}:
9118 ffmpeg -nostats -i input.mp3 -filter_complex ebur128 -f null -
9122 @section interleave, ainterleave
9124 Temporally interleave frames from several inputs.
9126 @code{interleave} works with video inputs, @code{ainterleave} with audio.
9128 These filters read frames from several inputs and send the oldest
9129 queued frame to the output.
9131 Input streams must have a well defined, monotonically increasing frame
9134 In order to submit one frame to output, these filters need to enqueue
9135 at least one frame for each input, so they cannot work in case one
9136 input is not yet terminated and will not receive incoming frames.
9138 For example consider the case when one input is a @code{select} filter
9139 which always drop input frames. The @code{interleave} filter will keep
9140 reading from that input, but it will never be able to send new frames
9141 to output until the input will send an end-of-stream signal.
9143 Also, depending on inputs synchronization, the filters will drop
9144 frames in case one input receives more frames than the other ones, and
9145 the queue is already filled.
9147 These filters accept the following options:
9151 Set the number of different inputs, it is 2 by default.
9154 @subsection Examples
9158 Interleave frames belonging to different streams using @command{ffmpeg}:
9160 ffmpeg -i bambi.avi -i pr0n.mkv -filter_complex "[0:v][1:v] interleave" out.avi
9164 Add flickering blur effect:
9166 select='if(gt(random(0), 0.2), 1, 2)':n=2 [tmp], boxblur=2:2, [tmp] interleave
9170 @section perms, aperms
9172 Set read/write permissions for the output frames.
9174 These filters are mainly aimed at developers to test direct path in the
9175 following filter in the filtergraph.
9177 The filters accept the following options:
9181 Select the permissions mode.
9183 It accepts the following values:
9186 Do nothing. This is the default.
9188 Set all the output frames read-only.
9190 Set all the output frames directly writable.
9192 Make the frame read-only if writable, and writable if read-only.
9194 Set each output frame read-only or writable randomly.
9198 Set the seed for the @var{random} mode, must be an integer included between
9199 @code{0} and @code{UINT32_MAX}. If not specified, or if explicitly set to
9200 @code{-1}, the filter will try to use a good random seed on a best effort
9204 Note: in case of auto-inserted filter between the permission filter and the
9205 following one, the permission might not be received as expected in that
9206 following filter. Inserting a @ref{format} or @ref{aformat} filter before the
9207 perms/aperms filter can avoid this problem.
9209 @section select, aselect
9211 Select frames to pass in output.
9213 This filter accepts the following options:
9218 Set expression, which is evaluated for each input frame.
9220 If the expression is evaluated to zero, the frame is discarded.
9222 If the evaluation result is negative or NaN, the frame is sent to the
9223 first output; otherwise it is sent to the output with index
9224 @code{ceil(val)-1}, assuming that the input index starts from 0.
9226 For example a value of @code{1.2} corresponds to the output with index
9227 @code{ceil(1.2)-1 = 2-1 = 1}, that is the second output.
9230 Set the number of outputs. The output to which to send the selected
9231 frame is based on the result of the evaluation. Default value is 1.
9234 The expression can contain the following constants:
9238 the sequential number of the filtered frame, starting from 0
9241 the sequential number of the selected frame, starting from 0
9243 @item prev_selected_n
9244 the sequential number of the last selected frame, NAN if undefined
9247 timebase of the input timestamps
9250 the PTS (Presentation TimeStamp) of the filtered video frame,
9251 expressed in @var{TB} units, NAN if undefined
9254 the PTS (Presentation TimeStamp) of the filtered video frame,
9255 expressed in seconds, NAN if undefined
9258 the PTS of the previously filtered video frame, NAN if undefined
9260 @item prev_selected_pts
9261 the PTS of the last previously filtered video frame, NAN if undefined
9263 @item prev_selected_t
9264 the PTS of the last previously selected video frame, NAN if undefined
9267 the PTS of the first video frame in the video, NAN if undefined
9270 the time of the first video frame in the video, NAN if undefined
9272 @item pict_type @emph{(video only)}
9273 the type of the filtered frame, can assume one of the following
9285 @item interlace_type @emph{(video only)}
9286 the frame interlace type, can assume one of the following values:
9289 the frame is progressive (not interlaced)
9291 the frame is top-field-first
9293 the frame is bottom-field-first
9296 @item consumed_sample_n @emph{(audio only)}
9297 the number of selected samples before the current frame
9299 @item samples_n @emph{(audio only)}
9300 the number of samples in the current frame
9302 @item sample_rate @emph{(audio only)}
9303 the input sample rate
9306 1 if the filtered frame is a key-frame, 0 otherwise
9309 the position in the file of the filtered frame, -1 if the information
9310 is not available (e.g. for synthetic video)
9312 @item scene @emph{(video only)}
9313 value between 0 and 1 to indicate a new scene; a low value reflects a low
9314 probability for the current frame to introduce a new scene, while a higher
9315 value means the current frame is more likely to be one (see the example below)
9319 The default value of the select expression is "1".
9321 @subsection Examples
9325 Select all frames in input:
9330 The example above is the same as:
9342 Select only I-frames:
9344 select='eq(pict_type\,I)'
9348 Select one frame every 100:
9350 select='not(mod(n\,100))'
9354 Select only frames contained in the 10-20 time interval:
9356 select=between(t\,10\,20)
9360 Select only I frames contained in the 10-20 time interval:
9362 select=between(t\,10\,20)*eq(pict_type\,I)
9366 Select frames with a minimum distance of 10 seconds:
9368 select='isnan(prev_selected_t)+gte(t-prev_selected_t\,10)'
9372 Use aselect to select only audio frames with samples number > 100:
9374 aselect='gt(samples_n\,100)'
9378 Create a mosaic of the first scenes:
9380 ffmpeg -i video.avi -vf select='gt(scene\,0.4)',scale=160:120,tile -frames:v 1 preview.png
9383 Comparing @var{scene} against a value between 0.3 and 0.5 is generally a sane
9387 Send even and odd frames to separate outputs, and compose them:
9389 select=n=2:e='mod(n, 2)+1' [odd][even]; [odd] pad=h=2*ih [tmp]; [tmp][even] overlay=y=h
9393 @section sendcmd, asendcmd
9395 Send commands to filters in the filtergraph.
9397 These filters read commands to be sent to other filters in the
9400 @code{sendcmd} must be inserted between two video filters,
9401 @code{asendcmd} must be inserted between two audio filters, but apart
9402 from that they act the same way.
9404 The specification of commands can be provided in the filter arguments
9405 with the @var{commands} option, or in a file specified by the
9406 @var{filename} option.
9408 These filters accept the following options:
9411 Set the commands to be read and sent to the other filters.
9413 Set the filename of the commands to be read and sent to the other
9417 @subsection Commands syntax
9419 A commands description consists of a sequence of interval
9420 specifications, comprising a list of commands to be executed when a
9421 particular event related to that interval occurs. The occurring event
9422 is typically the current frame time entering or leaving a given time
9425 An interval is specified by the following syntax:
9427 @var{START}[-@var{END}] @var{COMMANDS};
9430 The time interval is specified by the @var{START} and @var{END} times.
9431 @var{END} is optional and defaults to the maximum time.
9433 The current frame time is considered within the specified interval if
9434 it is included in the interval [@var{START}, @var{END}), that is when
9435 the time is greater or equal to @var{START} and is lesser than
9438 @var{COMMANDS} consists of a sequence of one or more command
9439 specifications, separated by ",", relating to that interval. The
9440 syntax of a command specification is given by:
9442 [@var{FLAGS}] @var{TARGET} @var{COMMAND} @var{ARG}
9445 @var{FLAGS} is optional and specifies the type of events relating to
9446 the time interval which enable sending the specified command, and must
9447 be a non-null sequence of identifier flags separated by "+" or "|" and
9448 enclosed between "[" and "]".
9450 The following flags are recognized:
9453 The command is sent when the current frame timestamp enters the
9454 specified interval. In other words, the command is sent when the
9455 previous frame timestamp was not in the given interval, and the
9459 The command is sent when the current frame timestamp leaves the
9460 specified interval. In other words, the command is sent when the
9461 previous frame timestamp was in the given interval, and the
9465 If @var{FLAGS} is not specified, a default value of @code{[enter]} is
9468 @var{TARGET} specifies the target of the command, usually the name of
9469 the filter class or a specific filter instance name.
9471 @var{COMMAND} specifies the name of the command for the target filter.
9473 @var{ARG} is optional and specifies the optional list of argument for
9474 the given @var{COMMAND}.
9476 Between one interval specification and another, whitespaces, or
9477 sequences of characters starting with @code{#} until the end of line,
9478 are ignored and can be used to annotate comments.
9480 A simplified BNF description of the commands specification syntax
9483 @var{COMMAND_FLAG} ::= "enter" | "leave"
9484 @var{COMMAND_FLAGS} ::= @var{COMMAND_FLAG} [(+|"|")@var{COMMAND_FLAG}]
9485 @var{COMMAND} ::= ["[" @var{COMMAND_FLAGS} "]"] @var{TARGET} @var{COMMAND} [@var{ARG}]
9486 @var{COMMANDS} ::= @var{COMMAND} [,@var{COMMANDS}]
9487 @var{INTERVAL} ::= @var{START}[-@var{END}] @var{COMMANDS}
9488 @var{INTERVALS} ::= @var{INTERVAL}[;@var{INTERVALS}]
9491 @subsection Examples
9495 Specify audio tempo change at second 4:
9497 asendcmd=c='4.0 atempo tempo 1.5',atempo
9501 Specify a list of drawtext and hue commands in a file.
9503 # show text in the interval 5-10
9504 5.0-10.0 [enter] drawtext reinit 'fontfile=FreeSerif.ttf:text=hello world',
9505 [leave] drawtext reinit 'fontfile=FreeSerif.ttf:text=';
9507 # desaturate the image in the interval 15-20
9508 15.0-20.0 [enter] hue s 0,
9509 [enter] drawtext reinit 'fontfile=FreeSerif.ttf:text=nocolor',
9511 [leave] drawtext reinit 'fontfile=FreeSerif.ttf:text=color';
9513 # apply an exponential saturation fade-out effect, starting from time 25
9514 25 [enter] hue s exp(25-t)
9517 A filtergraph allowing to read and process the above command list
9518 stored in a file @file{test.cmd}, can be specified with:
9520 sendcmd=f=test.cmd,drawtext=fontfile=FreeSerif.ttf:text='',hue
9525 @section setpts, asetpts
9527 Change the PTS (presentation timestamp) of the input frames.
9529 @code{setpts} works on video frames, @code{asetpts} on audio frames.
9531 This filter accepts the following options:
9536 The expression which is evaluated for each frame to construct its timestamp.
9540 The expression is evaluated through the eval API and can contain the following
9545 frame rate, only defined for constant frame-rate video
9548 the presentation timestamp in input
9551 the count of the input frame for video or the number of consumed samples,
9552 not including the current frame for audio, starting from 0.
9554 @item NB_CONSUMED_SAMPLES
9555 the number of consumed samples, not including the current frame (only
9559 the number of samples in the current frame (only audio)
9561 @item SAMPLE_RATE, SR
9565 the PTS of the first frame
9568 the time in seconds of the first frame
9571 tell if the current frame is interlaced
9574 the time in seconds of the current frame
9577 original position in the file of the frame, or undefined if undefined
9578 for the current frame
9584 previous input time in seconds
9590 previous output time in seconds
9593 wallclock (RTC) time in microseconds. This is deprecated, use time(0)
9597 wallclock (RTC) time at the start of the movie in microseconds
9600 timebase of the input timestamps
9604 @subsection Examples
9608 Start counting PTS from zero
9614 Apply fast motion effect:
9620 Apply slow motion effect:
9626 Set fixed rate of 25 frames per second:
9632 Set fixed rate 25 fps with some jitter:
9634 setpts='1/(25*TB) * (N + 0.05 * sin(N*2*PI/25))'
9638 Apply an offset of 10 seconds to the input PTS:
9644 Generate timestamps from a "live source" and rebase onto the current timebase:
9646 setpts='(RTCTIME - RTCSTART) / (TB * 1000000)'
9650 Generate timestamps by counting samples:
9657 @section settb, asettb
9659 Set the timebase to use for the output frames timestamps.
9660 It is mainly useful for testing timebase configuration.
9662 This filter accepts the following options:
9667 The expression which is evaluated into the output timebase.
9671 The value for @option{tb} is an arithmetic expression representing a
9672 rational. The expression can contain the constants "AVTB" (the default
9673 timebase), "intb" (the input timebase) and "sr" (the sample rate,
9674 audio only). Default value is "intb".
9676 @subsection Examples
9680 Set the timebase to 1/25:
9686 Set the timebase to 1/10:
9692 Set the timebase to 1001/1000:
9698 Set the timebase to 2*intb:
9704 Set the default timebase value:
9710 @section showspectrum
9712 Convert input audio to a video output, representing the audio frequency
9715 The filter accepts the following options:
9719 Specify the video size for the output. For the syntax of this option, check
9720 the "Video size" section in the ffmpeg-utils manual. Default value is
9724 Specify if the spectrum should slide along the window. Default value is
9728 Specify display mode.
9730 It accepts the following values:
9733 all channels are displayed in the same row
9735 all channels are displayed in separate rows
9738 Default value is @samp{combined}.
9741 Specify display color mode.
9743 It accepts the following values:
9746 each channel is displayed in a separate color
9748 each channel is is displayed using the same color scheme
9751 Default value is @samp{channel}.
9754 Specify scale used for calculating intensity color values.
9756 It accepts the following values:
9761 square root, default
9768 Default value is @samp{sqrt}.
9771 Set saturation modifier for displayed colors. Negative values provide
9772 alternative color scheme. @code{0} is no saturation at all.
9773 Saturation must be in [-10.0, 10.0] range.
9774 Default value is @code{1}.
9777 Set window function.
9779 It accepts the following values:
9782 No samples pre-processing (do not expect this to be faster)
9791 Default value is @code{hann}.
9794 The usage is very similar to the showwaves filter; see the examples in that
9797 @subsection Examples
9801 Large window with logarithmic color scaling:
9803 showspectrum=s=1280x480:scale=log
9807 Complete example for a colored and sliding spectrum per channel using @command{ffplay}:
9809 ffplay -f lavfi 'amovie=input.mp3, asplit [a][out1];
9810 [a] showspectrum=mode=separate:color=intensity:slide=1:scale=cbrt [out0]'
9816 Convert input audio to a video output, representing the samples waves.
9818 The filter accepts the following options:
9822 Specify the video size for the output. For the syntax of this option, check
9823 the "Video size" section in the ffmpeg-utils manual. Default value
9829 Available values are:
9832 Draw a point for each sample.
9835 Draw a vertical line for each sample.
9838 Default value is @code{point}.
9841 Set the number of samples which are printed on the same column. A
9842 larger value will decrease the frame rate. Must be a positive
9843 integer. This option can be set only if the value for @var{rate}
9844 is not explicitly specified.
9847 Set the (approximate) output frame rate. This is done by setting the
9848 option @var{n}. Default value is "25".
9852 @subsection Examples
9856 Output the input file audio and the corresponding video representation
9859 amovie=a.mp3,asplit[out0],showwaves[out1]
9863 Create a synthetic signal and show it with showwaves, forcing a
9864 frame rate of 30 frames per second:
9866 aevalsrc=sin(1*2*PI*t)*sin(880*2*PI*t):cos(2*PI*200*t),asplit[out0],showwaves=r=30[out1]
9870 @section split, asplit
9872 Split input into several identical outputs.
9874 @code{asplit} works with audio input, @code{split} with video.
9876 The filter accepts a single parameter which specifies the number of outputs. If
9877 unspecified, it defaults to 2.
9879 @subsection Examples
9883 Create two separate outputs from the same input:
9885 [in] split [out0][out1]
9889 To create 3 or more outputs, you need to specify the number of
9892 [in] asplit=3 [out0][out1][out2]
9896 Create two separate outputs from the same input, one cropped and
9899 [in] split [splitout1][splitout2];
9900 [splitout1] crop=100:100:0:0 [cropout];
9901 [splitout2] pad=200:200:100:100 [padout];
9905 Create 5 copies of the input audio with @command{ffmpeg}:
9907 ffmpeg -i INPUT -filter_complex asplit=5 OUTPUT
9913 Receive commands sent through a libzmq client, and forward them to
9914 filters in the filtergraph.
9916 @code{zmq} and @code{azmq} work as a pass-through filters. @code{zmq}
9917 must be inserted between two video filters, @code{azmq} between two
9920 To enable these filters you need to install the libzmq library and
9921 headers and configure FFmpeg with @code{--enable-libzmq}.
9923 For more information about libzmq see:
9924 @url{http://www.zeromq.org/}
9926 The @code{zmq} and @code{azmq} filters work as a libzmq server, which
9927 receives messages sent through a network interface defined by the
9928 @option{bind_address} option.
9930 The received message must be in the form:
9932 @var{TARGET} @var{COMMAND} [@var{ARG}]
9935 @var{TARGET} specifies the target of the command, usually the name of
9936 the filter class or a specific filter instance name.
9938 @var{COMMAND} specifies the name of the command for the target filter.
9940 @var{ARG} is optional and specifies the optional argument list for the
9941 given @var{COMMAND}.
9943 Upon reception, the message is processed and the corresponding command
9944 is injected into the filtergraph. Depending on the result, the filter
9945 will send a reply to the client, adopting the format:
9947 @var{ERROR_CODE} @var{ERROR_REASON}
9951 @var{MESSAGE} is optional.
9953 @subsection Examples
9955 Look at @file{tools/zmqsend} for an example of a zmq client which can
9956 be used to send commands processed by these filters.
9958 Consider the following filtergraph generated by @command{ffplay}
9960 ffplay -dumpgraph 1 -f lavfi "
9961 color=s=100x100:c=red [l];
9962 color=s=100x100:c=blue [r];
9963 nullsrc=s=200x100, zmq [bg];
9964 [bg][l] overlay [bg+l];
9965 [bg+l][r] overlay=x=100 "
9968 To change the color of the left side of the video, the following
9969 command can be used:
9971 echo Parsed_color_0 c yellow | tools/zmqsend
9974 To change the right side:
9976 echo Parsed_color_1 c pink | tools/zmqsend
9979 @c man end MULTIMEDIA FILTERS
9981 @chapter Multimedia Sources
9982 @c man begin MULTIMEDIA SOURCES
9984 Below is a description of the currently available multimedia sources.
9988 This is the same as @ref{movie} source, except it selects an audio
9994 Read audio and/or video stream(s) from a movie container.
9996 This filter accepts the following options:
10000 The name of the resource to read (not necessarily a file but also a device or a
10001 stream accessed through some protocol).
10003 @item format_name, f
10004 Specifies the format assumed for the movie to read, and can be either
10005 the name of a container or an input device. If not specified the
10006 format is guessed from @var{movie_name} or by probing.
10008 @item seek_point, sp
10009 Specifies the seek point in seconds, the frames will be output
10010 starting from this seek point, the parameter is evaluated with
10011 @code{av_strtod} so the numerical value may be suffixed by an IS
10012 postfix. Default value is "0".
10015 Specifies the streams to read. Several streams can be specified,
10016 separated by "+". The source will then have as many outputs, in the
10017 same order. The syntax is explained in the ``Stream specifiers''
10018 section in the ffmpeg manual. Two special names, "dv" and "da" specify
10019 respectively the default (best suited) video and audio stream. Default
10020 is "dv", or "da" if the filter is called as "amovie".
10022 @item stream_index, si
10023 Specifies the index of the video stream to read. If the value is -1,
10024 the best suited video stream will be automatically selected. Default
10025 value is "-1". Deprecated. If the filter is called "amovie", it will select
10026 audio instead of video.
10029 Specifies how many times to read the stream in sequence.
10030 If the value is less than 1, the stream will be read again and again.
10031 Default value is "1".
10033 Note that when the movie is looped the source timestamps are not
10034 changed, so it will generate non monotonically increasing timestamps.
10037 This filter allows to overlay a second video on top of main input of
10038 a filtergraph as shown in this graph:
10040 input -----------> deltapts0 --> overlay --> output
10043 movie --> scale--> deltapts1 -------+
10046 @subsection Examples
10050 Skip 3.2 seconds from the start of the avi file in.avi, and overlay it
10051 on top of the input labelled as "in":
10053 movie=in.avi:seek_point=3.2, scale=180:-1, setpts=PTS-STARTPTS [over];
10054 [in] setpts=PTS-STARTPTS [main];
10055 [main][over] overlay=16:16 [out]
10059 Read from a video4linux2 device, and overlay it on top of the input
10062 movie=/dev/video0:f=video4linux2, scale=180:-1, setpts=PTS-STARTPTS [over];
10063 [in] setpts=PTS-STARTPTS [main];
10064 [main][over] overlay=16:16 [out]
10068 Read the first video stream and the audio stream with id 0x81 from
10069 dvd.vob; the video is connected to the pad named "video" and the audio is
10070 connected to the pad named "audio":
10072 movie=dvd.vob:s=v:0+#0x81 [video] [audio]
10076 @c man end MULTIMEDIA SOURCES