Recherche avancée

Sur d’autres sites (12434)

• FFmpeg convert video to images with complex logic

  18 juillet 2020, par Udi

  I'm trying to use FFMPEG in order to solve some complex logic on my videos.

  


  The business logic is the following :
I get videos from the formats : avi, mp4, mov.

  


  I don't know what is the content in the video. It can be from 1M to 5G.

  


  I want to output a list of images from this video with the higher quality I can get. Capturing only frames that have big changes from their previous frame. (new person, a new angle, big movement, etc)

  


  In addition, I want to limit the number of frames per second that even if the video is dramatically fast and has changed all the time it will not produce more frames per second than this parameter.

  


  I'm using now the following command :

  


  ./ffmpeg -i "/tmp/input/fast_movies/3_1.mp4" -vf fps=3,mpdecimate=hi=14720:lo=7040:frac=0.5 -vsync 0 -s hd720 "/tmp/output/fast_movies/(#%04d).png"


  


  According to my understanding it doing the following :
fps=3 - first cut the video to 3 frames per second (So that is the limit I talked about)

  


  mpdecimate - filter frames that doesn't have greater changes than the thresholds I set.

  


  -vsync 0 - sync video timestamp - I'm not sure why but without it - it makes hundereds of duplicate frames ignoring the fps and mpdecimate command. Can someone explain ?

  


  -s hd720 - set video size to

  


  It works pretty well but I'm not so happy with the quality. Do you think I miss something ? Is there any parameter in FFMPEG that I better use it instead of these ones ?

  


• Any other better ways to extract frames from a large size video according to the timecode given ?

  6 avril 2020, par Yong En

  Given a video from youtube with at least 600 MB. The video is annotated with labels that happens in multiple timecodes. The timecodes are in millisecond (SSSS.ss). I am trying to get the frames that fall within a time period (2 timecodes). There are TWO approaches that I used with different tools, one is using openCV in python and FFmpeg in bash script :

  



  I would stick with few variables here,

  



  

  • fps = 25

  


  • timecode (after convert into second) = 333.44 to 334.00 take note that I am dealing with time period that might less than a second.

  


  



  openCV

  



  

  1. Using openCV in python, I read the video as frames into a numpy array. Using the fps from video, 25. I can estimate where do the frames that fall under this time period by diving total video duration in second with length of the numpy array.

  


  2. Problem here is I will miss out some frames here due to the video fps is not really 25 as given by the video meta info, it could be 24.xx. Any solutions ?

  


  



  ffmpeg

  



  

  1. What I did, every time I want to get the frames, I run the script.

  


  2. Problem here is I need to read the video 100 times if I have 100 time periods. Any ways to overcome this ?

  


  



  Thanks for reading it.

  


• aarch64 : vp9 : loop_filter : fix typo in skip flatout8 check

  14 novembre 2016, par Janne Grunau

  aarch64 : vp9 : loop_filter : fix typo in skip flatout8 check
  The 16_16 loop filter functions could miss an early exit before
  
  flatout8.
  Signed-off-by : Martin Storsjö <martin@martin.st>
  

  • [DBH] libavcodec/aarch64/vp9lpf_neon.S