Recherche avancée

Sur d’autres sites (9315)

• lavc/vvc : Ensure subpictures don't overlap

  22 février, par Frank Plowman

  lavc/vvc : Ensure subpictures don't overlap
  This is essentially a re-implementation of
  
  https://patchwork.ffmpeg.org/project/ffmpeg/patch/20241005223955.54158-1-post@frankplowman.com/
  That patch was not applied last time.  Instead we opted to identify
  
  issues which could be caused by invalid subpicture layouts and remedy
  
  those issues where they manifest, either through error detection or code
  
  hardening.  This was primarily implemented in the set
  
  https://patchwork.ffmpeg.org/project/ffmpeg/list/?series=13381.
  This has worked to some degree, however issues with subpicture layouts
  
  continue to crop up from the fuzzer and I've fixed a number of bugs
  
  related to subpicture layouts since then.  I think it's best to return
  
  to the initial plan and simply check if the subpicture layout is valid
  
  initially.
  This implementation is also lighter than the first time — by doing a
  
  bit more logic in pps_subpic_less_than_one_tile_slice, we are able to
  
  store a tile_in_subpic map rather than a ctu_in_subpic map.  This
  
  reduces the size of the map to the point it becomes possible to allocate
  
  it on the stack.  Similar to 8bd66a8c9587af61c7b46558be3c4ee317c1af5a,
  
  the layout is also validated in the slice map construction code, rather
  
  than in the CBS, which avoids duplicating some logic.
  Signed-off-by : Frank Plowman <post@frankplowman.com>
  

  • [DH] libavcodec/vvc/ps.c

• ffmpeg probesize and analyzeduration flags

  7 mai 2014, par Xerphiel

  Could someone explain to me in simple terms what these flags/options do ?

  Running "ffmpeg -h full | grep probesize/analyzeduration" yields the following documentation :

  -probesize         <int>        .D.... set probing size (from 32 to INT_MAX) (default 5e+06)
  
  -analyzeduration   <int>        .D.... specify how many microseconds are analyzed to probe the input (from 0 to INT_MAX) (default 5e+06)
  
  </int></int>

  This is beyond my understanding, so am hoping for a simpler explanation.

  All the discussions on this subject I’ve found via googling do not describe the basic function of the options.

  Any help is appreciated.

  Thank you.

• Wrong video duration when trying to make screen capture with libav\ffmpeg

  16 juin 2023, par Ларионов Фёдор

  I am performing screen capture using libav/c++ with the gdbigrab InputFormat and saving the transcoded video with h.264 to an .mp4 container. Despite doing all the necessary conversions from pts/dts frames and packets, the final video has an issue with its duration. When playing the video, it stops a few seconds before the end and rewinds back to the end.

  


  To diagnose the problem, I used FFprobe to examine the recorded video stream and found the following issues :

  


  ffprobe my_video.mp4 -show_streams

  


  Provides the following information :

  


  time_base=1/12800
start_pts=28096
start_time=2.195000
duration_ts=38945
duration=3.042578


  


  Although I tracked the frame and burst timestamps while recording, starting at zero and ending at the expected pts for the number of frames recorded, I tried to set the start and end time of the stream manually. I observed this problem with different time_base and different video durations, do some minor and major fixes that does not affected this issue. Interestingly, the start time always has the same offset, and the expected/specified video duration in the code is close to the sum of the duration of the video stream and its start time. What am I missing ?

  


  


  UPDATE :

  


  


  There is the source of problem :

  


  [mov,mp4,m4a,3gp,3g2,mj2 @ 000002a69f15b380] Processing st: 0, edit list 0 - media time: -1, duration: 27059
[mov,mp4,m4a,3gp,3g2,mj2 @ 000002a69f15b380] Processing st: 0, edit list 1 - media time: 0, duration: 38912 


  


  which i get by that command ffmpeg -hide_banner -y -v 9 -loglevel 99 -i sc_test.mp4

  


  The question is where this edits happend.