Recherche avancée

Sur d’autres sites (13324)

• Error while trying to stream higer resolution using ffmpeg and jsmpeg

  11 avril 2017, par trojek

  I try to do a prof of concept of streaming webcam in browser using jsmpeg. It works perfect when I use code from the documentation which is as follows :

  ffmpeg \
  
      -f v4l2 \
  
          -framerate 25 -video_size 640x480 -i /dev/video0 \
  
      -f mpegts \
  
          -codec:v mpeg1video -s 640x480 -b:v 1000k -bf 0 \
  
      http://localhost:8081/supersecret

  My webcam supports resolutions, framerates and formats as in code below :

  [video4linux2,v4l2 @ 0x2655360] Compressed:       mjpeg :          Motion-JPEG : 1920x1080 1280x720 1024x768 640x480 800x600 1280x1024 320x240
  
  [video4linux2,v4l2 @ 0x2655360] Raw       :     yuyv422 :           YUYV 4:2:2 : 1920x1080 1280x720 1024x768 640x480 800x600 1280x1024 320x240

  While I change in above code the resolution from 640x480 to e.g. 1024x768, I get an error :

  [mpeg1video @ 0x110f4e0] MPEG1/2 does not support 10/1 fps
  
  Error while opening encoder for output stream #0:0 - maybe incorrect parameters such as bit_rate, rate, width or height

  I assume that camera driver change number of fps and MPEG1/2 doesn’t support that numer of fps.
  Probably I get from camera raw data (yuyv422) and it can’t play it with higer framerate when e.f. 10 for 1024x768.
  How Can I modify ffmped execution code in order to stream video in Full HD resulution ?

• configure : arm : Don't add -march= to the compiler if no preference was passed

  20 septembre 2021, par Martin Storsjö

  configure : arm : Don't add -march= to the compiler if no preference was passed
  If no —cpu= option was passed to configure, we detect what the
  
  compiler defaults to. This detected value was then fed back to the
  
  rest of the configure logic, as if it was an explicit choice.
  This breaks on Ubuntu 21.10 with GCC 11.1.
  Since GCC 8, it's possible to add configure extra features via the
  
   march option, like e.g. -march=armv7-a+neon. If the -mfpu= option
  
  is configured to default to 'auto', the fpu setting gets taken
  
  from the -march option.
  GCC 11.1 in Ubuntu seems to be configured to use -mfpu=auto. This
  
  has the effect of breaking any compilation command that specifies
  
   march=armv7-a, because the driver implicitly also adds -mfloat-abi=hard,
  
  and that combination results in this error :
      cc1 : error : ‘-mfloat-abi=hard’ : selected processor lacks an FPU
  One can compile successfully by passing e.g. -march=armv7-a+fp.
  Therefore, restructure configure. If no specific preference was set
  
  (and the 'cpu' configure variable was set as the output of
  
  probe_arm_arch), the value we tried to set via -march= was the same
  
  value that we just tried to detect as the compiler default.
  So instead, just try to detect what the compiler defaults to, with
  
  to allow setting other configure settings (such as 'fast_unaligned'),
  
  but don't try to spell out the compiler's default via the -march flag.
  Signed-off-by : Martin Storsjö <martin@martin.st>
  

  • [DH] configure

• How to convert short video clips to TS without sound "gaps" between the segments ?

  10 novembre 2022, par Zvika

  I am trying to convert a sequence of short video files from MP4 to TS using ffmpeg.
I get valid TS files, but when playing them in any HLS player, there is a noticeable short gap in the sound between segment to segment.

  


  If I first stitch all the short video files to a single video file, and convert this file to TS while slicing it to segments, it plays perfectly fine.

  


  To the gory details :
My software creates short video clips that should be concateenated to an output video and streamed as HLS.
Each short clip is an H.264 video file and WAV audio file (I can create other formats if needed).
I then convert each such pair of H.264+WAV to a TS file using ffmpeg :
ffmpeg -y -i seg_0.mp4 -i seg_0.wav -c:a libvo_aacenc -c:v copy -bsf:v h264_mp4toannexb seg_0.ts ffmpeg -y -i seg_1.mp4 -i seg_1.wav -c:a libvo_aacenc -c:v copy -bsf:v h264_mp4toannexb -output_ts_offset 2.01 seg_1.ts ffmpeg -y -i seg_2.mp4 -i seg_2.wav -c:a libvo_aacenc -c:v copy -bsf:v h264_mp4toannexb -output_ts_offset 4.02 seg_2.ts
etc.

  


  and I create an appropriate M3U8 file to play all the short clips as a sequence.
The result is not satisfying, as I have audio gaps between each segment and segment, as you can hear here :
https://rnd3-temp-public.s3.amazonaws.com/HLS_4/out_seg2.m3u8

  


  However, if I concat all the pairs together, and convert the concatenated sequence to TS, while requesting ffmpeg to slice them again to segments, using a command like :
ffmpeg -y -f concat -i mp4_list.txt -f concat -i wav_list.txt -c:a libvo_aacenc -c:v copy -bsf:v h264_mp4toannexb -flags +cgop -g 30 -hls_time 2 out2.m3u8
it plays perfectly OK, as you can hear here :
https://rnd3-temp-public.s3.amazonaws.com/HLS/out2.m3u8

  


  How can I get a clear audio output by still encoding each segment separately ? (It's crucial for my workflow)

  


  Thanks !