Recherche avancée

Sur d’autres sites (10172)

• FFMPEG - Converting TS frame to PNG creating invalid images

  20 juillet 2015, par spelley

  I am attempting to snag a single frame from a TS live stream and generate a PNG image. The command I’ve been using is :

  ffmpeg -y -i "filenamehere" -f image2 -vframes 1 -s 250x250 ./snapshot-250x250.png

  When doing this on my local Windows machine using the FFMPEG executable, it generates the image as expected. When using Ubuntu, after doing an apt-get install ffmpeg and using the same command, I am generating grey squares with sporadic noise even while targeting the same file with the exact same command.

  Here is the (anonymized) output :

  [h264 @ 0xa256a0] no frame!
  
  [h264 @ 0xa256a0] non-existing SPS 0 referenced in buffering period
  
  [h264 @ 0xa256a0] non-existing PPS referenced
  
  [h264 @ 0xa256a0] non-existing SPS 0 referenced in buffering period
  
  [h264 @ 0xa256a0] non-existing PPS 0 referenced
  
  [h264 @ 0xa256a0] decode_slice_header error
  
  [h264 @ 0xa256a0] non-existing PPS 0 referenced
  
  [h264 @ 0xa256a0] decode_slice_header error
  
  [h264 @ 0xa256a0] non-existing PPS 0 referenced
  
  [h264 @ 0xa256a0] decode_slice_header error
  
  [h264 @ 0xa256a0] non-existing PPS 0 referenced
  
  [h264 @ 0xa256a0] decode_slice_header error
  
  [h264 @ 0xa256a0] no frame!
  
  [mpegts @ 0xa1d1e0] max_analyze_duration reached
  
  [NULL @ 0xa28080] start time is not set in estimate_timings_from_pts
  
  
  
  Seems stream 0 codec frame rate differs from container frame rate: 59.94 (60000/1001) -> 1000.00 (1000/1)
  
  Input #0, mpegts, from '<file here="here">':
  
    Duration: 00:00:09.07, start: 35968.174711, bitrate: 697 kb/s
  
    Program 1
  
      Stream #0.0[0x100]: Video: h264 (High), yuv420p, 1280x720 [PAR 1:1 DAR 16:9], 44.52 fps, 1k tbr, 90k tbn, 59.94 tbc
  
      Stream #0.1[0x101]: Audio: aac, 48000 Hz, stereo, s16, 96 kb/s
  
      Stream #0.2[0x102]: Data: [21][0][0][0] / 0x0015
  
  Incompatible pixel format 'yuv420p' for codec 'png', auto-selecting format 'rgb24'
  
  [buffer @ 0xa43480] w:1280 h:720 pixfmt:yuv420p
  
  [avsink @ 0xa28780] auto-inserting filter 'auto-inserted scaler 0' between the filter 'src' and the filter 'out'
  
  [scale @ 0xa26520] w:1280 h:720 fmt:yuv420p -> w:1280 h:720 fmt:rgb24 flags:0x4
  
  Output #0, image2, to './snapshot-250x250.png':
  
    Metadata:
  
      encoder         : Lavf53.21.1
  
      Stream #0.0: Video: png, rgb24, 1280x720 [PAR 1:1 DAR 16:9], q=2-31, 200 kb/s, 90k tbn, 1k tbc
  
  Stream mapping:
  
    Stream #0.0 -> #0.0
  
  Press ctrl-c to stop encoding
  
  frame=    1 fps=  0 q=0.0 Lsize=      -0kB time=0.01 bitrate= -17.6kbits/s dup=31 drop=0
  
  video:8kB audio:0kB global headers:0kB muxing overhead -100.261749%
  
  </file>

  The initial "no frame !" I believe is just it seeking to the first valid keyframe (And thus no cause for concern) however the only difference from my Windows output is the inclusion of this line :

  Incompatible pixel format 'yuv420p' for codec 'png', auto-selecting format 'rgb24'

  Windows has an encoder of Lavc56.46.100 instead of Lavf53.21.1 and the stream mapping line on Windows has more information :

  Stream #0:0 -> #0:0 **(h264 (native) -> png (native))**

  I’ve attempted to manually seek to frames, change the output format from PNG -> GIF, JPEG and just cant seem to find the issue which has no problems on my local Windows machine. Any advice on where to look would be greatly appreciated ! Below is an example of the garbled output.

  

• avcodec/mpeg12enc : Disallow using MPEG-2 intra VLC table for mpeg1video

  24 novembre 2020, par Andreas Rheinhardt

  avcodec/mpeg12enc : Disallow using MPEG-2 intra VLC table for mpeg1video
  Using MPEG-2 intra VLC tables is spec-incompliant for MPEG-1 and given
  
  that an MPEG-1 bitstream can't signal whether MPEG-2 intra VLC tables
  
  have been used the output is broken. Therefore this option is removed
  
  immediately without any deprecation period.
  Reviewed-by : James Almer <jamrial@gmail.com>
  
  Reviewed-by : Marton Balint <cus@passwd.hu>
  
  Reviewed-by : Anton Khirnov <anton@khirnov.net>
  
  Signed-off-by : Andreas Rheinhardt <andreas.rheinhardt@gmail.com>
  

  • [DH] libavcodec/mpeg12enc.c

• Using ffmpeg to encode web dash and audio drift

  1er octobre 2016, par John Joske

  I’m trying to use ffmpeg to create webm-dash encoded files from a usb web cam. Adapted from this example http://wiki.webmproject.org/adaptive-streaming/instructions-to-do-webm-live-streaming-via-dash

  I have managed to get something nearly works with the following command :

  ffmpeg -f video4linux2 -framerate 30 -s 640x360 -i /dev/video0
  -thread_queue_size 512 -f alsa -ar 44100 -ac 2 -i hw:2 -map 0:0 -pix_fmt yuv420p -c:v libvpx-vp9 -s 640x360 -keyint_min 60 -g 60 -speed 6 -tile-columns 4 -frame-parallel
  1 -threads 8 -static-thresh 0 -max-intra-rate 300 -deadline realtime
  -lag-in-frames 0 -error-resilient 1 -f webm_chunk -header video_360.hdr" -chunk_start_index 1 video_360_%d.chk -map 1:0 -c:a libvorbis -b:a 128k -ar 44100
  -f webm_chunk -copytb 1 -audio_chunk_duration 2000 -header video_171.hdr -chunk_start_index 1 video_171_%d.chk

  However over a period of time the audio chunks slowly get behind the video chunks, can anyone suggest a way to ensure they stay the same.