Recherche avancée

Sur d’autres sites (10570)

• Artifacts in ffmpeg fifo, low fps, stream ends

  11 août 2020, par Ben Gardner

  I'm using a Raspberry Pi 3B and 4 for this, neither works.

  


  I'm trying to both pass my capture card's input (/dev/video0) through a fifo file so I can play it on the screen via omxplayer (1080p/30fps), and also grab frames of /dev/video0 out to a series of jpgs (1920x1080 right now, but I'd like it to be 640x480) so I can do analysis on it while it's being played. The input to the capture card is television via HDMI.

  


  This is the command I use to make the stream go to the fifo and jpgs.

  


  ffmpeg -y -f v4l2 -input_format mjpeg -framerate 30 -video_size 1920x1080 \
-thread_queue_size 16384 -i /dev/video0 -f alsa -ac 1 \
-thread_queue_size 16384 -i hw:CARD=U0x534d0x2109,DEV=0 \
-c:v copy -b:v 32000k -preset faster -x264opts keyint=50 \
-g 25 -pix_fmt yuvj422p -c:a aac -b:a 128k -codec:v copy -f tee \
-map 0:v -map 1:a "fifo.mkv|output_%3d.jpg"


  


  Here is my output, which gives me 30fps originally, sometimes dipping into 29-28 fps, and then skipping (both audio and video) and artifacts in the video after around 5-10 minutes with the severity eventually increasing until it stops :

  


  [mjpeg @ 0x1504490] EOI missing, emulating
Input #0, video4linux2,v4l2, from '/dev/video0':
  Duration: N/A, start: 27151.039849, bitrate: N/A
    Stream #0:0: Video: mjpeg, yuvj422p(pc, bt470bg/unknown/unknown), 1920x1080, 30 fps, 30 tbr, 1000k tbn, 1000k tbc
Guessed Channel Layout for Input Stream #1.0 : mono
Input #1, alsa, from 'hw:CARD=U0x534d0x2109,DEV=0':
  Duration: N/A, start: 1596773777.825328, bitrate: 1536 kb/s
    Stream #1:0: Audio: pcm_s16le, 96000 Hz, mono, s16, 1536 kb/s
Stream mapping:
  Stream #0:0 -> #0:0 (copy)
  Stream #1:0 -> #0:1 (pcm_s16le (native) -> aac (native))
Press [q] to stop, [?] for help
Output #0, tee, to 'fifo.mkv|output_%3d.jpg':
  Metadata:
    encoder         : Lavf58.20.100
    Stream #0:0: Video: mjpeg, yuvj422p(pc, bt470bg/unknown/unknown), 1920x1080, q=2-31, 32000 kb/s, 30 fps, 30 tbr, 1000k tbn, 1000k tbc
    Stream #0:1: Audio: aac (LC), 96000 Hz, mono, fltp, 128 kb/s
    Metadata:
      encoder         : Lavc58.35.100 aac
[alsa @ 0x1507300] Thread message queue blocking; consider raising the thread_queue_size option (current value: 16384)
[alsa @ 0x1507300] ALSA buffer xrun.time=00:13:55.89 bitrate=N/A speed=0.972x
[alsa @ 0x1507300] ALSA buffer xrun.time=00:14:16.02 bitrate=N/A speed=0.974x
[alsa @ 0x1507300] ALSA buffer xrun.time=00:14:27.25 bitrate=N/A speed=0.972x
[alsa @ 0x1507300] ALSA buffer xrun.time=00:14:33.35 bitrate=N/A speed=0.97x
[alsa @ 0x1507300] ALSA buffer xrun.time=00:14:44.71 bitrate=N/A speed=0.969x
[alsa @ 0x1507300] ALSA buffer xrun.time=00:14:49.71 bitrate=N/A speed=0.97x
[alsa @ 0x1507300] ALSA buffer xrun.time=00:15:01.51 bitrate=N/A speed=0.967x
[alsa @ 0x1507300] ALSA buffer xrun.time=00:15:07.78 bitrate=N/A speed=0.969x
[alsa @ 0x1507300] ALSA buffer xrun.time=00:15:14.27 bitrate=N/A speed=0.962x
[alsa @ 0x1507300] ALSA buffer xrun.time=00:15:26.44 bitrate=N/A speed=0.96x
[alsa @ 0x1507300] ALSA buffer xrun.time=00:15:32.40 bitrate=N/A speed=0.96x
[alsa @ 0x1507300] ALSA buffer xrun.time=00:15:38.63 bitrate=N/A speed=0.963x
[alsa @ 0x1507300] ALSA buffer xrun.time=00:15:45.60 bitrate=N/A speed=0.959x
[alsa @ 0x1507300] ALSA buffer xrun.time=00:15:50.93 bitrate=N/A speed=0.957x
    Last message repeated 1 times
[alsa @ 0x1507300] ALSA buffer xrun.time=00:16:00.79 bitrate=N/A speed=0.951x
    Last message repeated 1 times
[alsa @ 0x1507300] ALSA buffer xrun.time=00:16:05.29 bitrate=N/A speed=0.949x
[alsa @ 0x1507300] ALSA buffer xrun.time=00:16:07.19 bitrate=N/A speed=0.949x
[alsa @ 0x1507300] ALSA buffer xrun.time=00:16:11.72 bitrate=N/A speed=0.945x
[alsa @ 0x1507300] ALSA buffer xrun.time=00:16:16.02 bitrate=N/A speed=0.944x
[alsa @ 0x1507300] ALSA buffer xrun.time=00:16:26.39 bitrate=N/A speed=0.953x
[alsa @ 0x1507300] ALSA buffer xrun.time=00:16:30.22 bitrate=N/A speed=0.938x
[alsa @ 0x1507300] ALSA buffer xrun.time=00:16:30.98 bitrate=N/A speed=0.937x
[alsa @ 0x1507300] ALSA buffer xrun.time=00:16:37.66 bitrate=N/A speed=0.941x
[alsa @ 0x1507300] ALSA buffer xrun.time=00:16:41.28 bitrate=N/A speed=0.935x
[alsa @ 0x1507300] ALSA buffer xrun.time=00:16:42.95 bitrate=N/A speed=0.934x
[alsa @ 0x1507300] ALSA buffer xrun.time=00:16:45.82 bitrate=N/A speed=0.933x
[alsa @ 0x1507300] ALSA buffer xrun.time=00:16:46.76 bitrate=N/A speed=0.932x
[alsa @ 0x1507300] ALSA buffer xrun.time=00:16:47.05 bitrate=N/A speed=0.931x
[alsa @ 0x1507300] ALSA buffer xrun.time=00:16:50.74 bitrate=N/A speed=0.927x
[alsa @ 0x1507300] ALSA buffer xrun.time=00:16:57.05 bitrate=N/A speed=0.927x
[alsa @ 0x1507300] ALSA buffer xrun.time=00:16:58.91 bitrate=N/A speed=0.927x
[alsa @ 0x1507300] ALSA buffer xrun.time=00:17:09.41 bitrate=N/A speed=0.92x
[alsa @ 0x1507300] ALSA buffer xrun.time=00:17:12.04 bitrate=N/A speed=0.917x
[alsa @ 0x1507300] ALSA buffer xrun.time=00:17:12.61 bitrate=N/A speed=0.916x
Killed31579 fps= 28 q=-1.0 size=N/A time=00:17:32.56 bitrate=N/A speed=0.919x


  


  I also occasionally get this error :

  


  [tee @ 0x17001c0] Non-monotonous DTS in output stream 0:1; previous: 99251754, current: 99247503; changing to 99251755. This may result in incorrect timestamps in the output file.


  


  I'm assuming this has something to do with the audio getting routed to the jpg. I've tried [select=\'v\'] in front of the jpg, which doesn't change the behavior as well as [map=\'1\:a\'] in front of the mkv, which says [matroska @ 0xe446f0] Unknown option 'map'.

  


  I should also disclaim that I don't have much of an idea of what this command is doing compression-wise, I basically just copy/pasted that part.

  


  What edits do I need to make to get this into a fifo and series of jpgs at the same time ?

  


• mpjpeg : probe should require same constraints as packet reader - both proper content...

  13 septembre 2015, par Alex Agranovsky

  mpjpeg : probe should require same constraints as packet reader - both proper content-type and content-size must be present
  return AVPROBE_SCORE_MAX, rather than random positive number on success
  Signed-off-by : Michael Niedermayer <michael@niedermayer.cc>
  

  • [DH] libavformat/mpjpegdec.c

• shorten : Use a checked bytestream reader for the wave header

  11 septembre 2013, par Martin Storsjö

  shorten : Use a checked bytestream reader for the wave header
  Reported-by : Mateusz "j00ru" Jurczyk and Gynvael Coldwind
  
  CC : libav-stable@libav.org
  
  Signed-off-by : Martin Storsjö <martin@martin.st>
  

  • [DBH] libavcodec/shorten.c