Recherche avancée

Sur d’autres sites (8428)

• ffmpeg : Is it possible to replace frames in a variable frame-rate video ?

  31 décembre 2022, par Arnon Weinberg

  Machine learning algorithms for video processing typically work on frames (images) rather than video.

  


  In my work, I use ffmpeg to dump a specific scene as a sequence of .png files, process them in some way (denoise, deblur, colorize, annotate, inpainting, etc), output the results into an equal number of .png files, and then update the original video with the new frames.

  


  This works well with constant frame-rate (CFR) video. I dump the images as so (eg, 50-frame sequence starting at 1:47) :

  


  ffmpeg -i input.mp4 -vf "select='gte(t,107)*lt(selected_n,50)'" -vsync passthrough '107+%06d.png'


  


  And then after editing the images, I replace the originals as so (for a 12.5fps CFR video) :

  


  ffmpeg -i input.mp4 -itsoffset 107 -framerate 25/2 -i '107+%06d.png' -filter_complex "[0]overlay=eof_action=pass" -vsync passthrough -c:a copy output.mp4


  


  However, many of the videos I work with are variable frame-rate (VFR), and this has created some challenges.

  


  A simple solution is to convert VFR video to CFR, which ffmpeg wants to do anyway, but I'm wondering if it's possible to avoid this. The reason is that CFR requires either dropping frames - since the purpose of ML video processing is usually to improve the output, I'd like to avoid this - or duplicating frames - but an upscaling algorithm that I'm working with right now uses the previous and next frame for data - if the previous or next frame is a duplicate, then ... no data for upscaling.

  


  With -vsync passthrough, I had hoped that I could simply remove the -framerate option, and preserve the original frames as-is, but the resulting command :

  


  ffmpeg -i input.mp4 -itsoffset 107 -i '107+%06d.png' -filter_complex "[0]overlay=eof_action=pass" -vsync passthrough -c:a copy output.mp4


  


  uses ffmpeg's default of 25fps, and drops a lot of frames. Is there a reliable way to replace frames in VFR video ?

  


• ffmpeg : What is the best practice to keep a live connection/socket with a camera, and save time on ffprobe

  15 mai 2022, par Jeff Strongman

  Today... I used the following command : with subprocess.PIPE and subprocess.Popen in python 3 :

  


  ffmpeg -i udp://{address_of_camera} \
  -vf select='if(eq(pict_type,I),st(1,t),gt(t,ld(1)))' setpts=N/FRAME_RATE/TB \
  -f rawvideo -an -vframes {NUM_WANTED_FRAMES} pipe:`


  


  This command helps me to capture NUM_WANTED_FRAMES frames from a live camera at a given moment.

  


  However... it takes me about 4 seconds to read the frames, and about 2.5 seconds to open a socket between my computer and the camera's computer.

  


  Is there a way, to have a socket/connection always open between my computer and the camera's computer, to save the 2.5 seconds ?

  


  I read something about fifo_size and overrun_fatal. I thought that maybe I can set fifo_size to be equal to NUM_WANTED_FRAMES, and overrun_fatal to True ? Will this solve my problem ? Or is there a different and simpler/better solution ?

  


  Should I try to record always (no -vframes flag) store the frames in a queue(With max size), and upon a wish to slice the video, read from my queue buffer ? Will it work well with the keyframe ?

  


  Also... What to do when ffmpeg fails ? restart the ffmpeg command ?

  


• Creating periodic offsets with ffmpeg segment muxer for seamless ffmpeg-concat transitions

  30 janvier 2021, par Soren Wray

  Suppose I begin with a 1 minute video with a framerate output of 24 frames/second and then use ffmpeg's segment muxer to split it into 60 equal parts of 1 second each with the timesptamps reset :

  


  ffmpeg -i input.mp4 -map 0 -f segment -segment_time 1 -reset_timestamps 1 -g 24 output_%03d.mp4

  


  I then use ffmpeg-concat to string output_000.mp4-output_059.mp4 together into a single video with a periodic 0.5 second dreamy transition effect between each segment : ffmpeg-concat -t dreamy -d 500 -o dreamy.mp4 output_*.mp4

  


  Unfortunately, the transition effects introduce a negative 0.5 second offset between the segments, resulting in a video that is 32 seconds in duration. I don't want to deform the audio and video in this manner. I want to use transition effects without changing the duration or creating any other syncing issues. How to best achieve these seamless transitions is the question.

  


  One potential solution would be to add an offset to the segments, e.g. 0.25 seconds at the beginning and end of every segment, such that the segment muxer produces 60 segments of 1.5 second duration from a 1 minute video. In this manner, the 0.5 second transition effects should blend seamlessly into the offsets.

  


  I don't know where to begin to achieve this potential solution. I've read the official guide, but I can't decipher the relevant filter options for this particular use case. ffmpeg-concat doesn't seem to provide any native options around this problem either.