Recherche avancée

Sur d’autres sites (11512)

• using ffmpeg to process webcam data and save it in multiple files

  21 septembre 2017, par Deepak Nellurvalappil

  I am trying to process streaming video data from a usb webcam using ffmpeg. the process involves encoding the raw data into hevc format. Upto this point I am able to do.
  But now, I want to slice the processed data in chunks of 10 seconds and save it in a separate file. This should repeat until I kill/interrupt the process manually.

  ffmpeg -i /dev/video1 -f segment -segment_times 10 -c:v hevc cam_1_%02d.mp4

  the above code does create multiple files but only the first file is readable using vlc ; other files look corrupt.

  And while I was running the command, I could see the following messages :-

  Stream mapping:  Stream #0:0 -> #0:0 (rawvideo (native) -> h264 (libx264)) Press [q] to stop, [?] for help Past duration 0.601555 too largeN/A time=00:00:14.63 bitrate=N/A dup=11 drop=0      Past duration
  
  0.601463 too large Past duration 0.601555 too largeN/A time=00:00:15.16 bitrate=N/A dup=11 drop=0     Past duration 0.601585 too large
  
      Last message repeated 1 times Past duration 0.601646 too large Past duration 0.601677 too large Past duration 0.601707 too largeN/A time=00:00:15.66 bitrate=N/A dup=11 drop=0

  What am I missing here ?

• save row ffmepg data packets as a video

  7 août 2019, par MattéoLuci

  I have modified the muxing.c ffmpeg example program to be able to save a video from the PC or on an RTSP stream, and it works quite good.
  I would like to make the video saving as fast and efficient as possible by saving directly the packets without changing them into AVFrames (I don’t need to print them).

  I tried to save the AVPacket coming from av_read_frame(InFmtCtx, &packet) with av_write_frame(OutFmtCtx, &packet).

  AVPacket packet;
  
  static int count = 0;
  
  while (av_read_frame(InFmtCtx, &packet)>=0) {
  
     if(packet.stream_index == videoStream) {
  
  
  
         packet.dts = count;
  
         packet.pts = count++;
  
  
  
         av_packet_rescale_ts(&packet, ost->enc->time_base, 
  
                                           ost->st->time_base);
  
         packet.stream_index = ost->st->index;
  
         if (av_write_frame(OutFmtCtx, &packet) < 0) {
  
            fprintf(stderr, "Error writing video frame: \n");
  
            exit(1);
  
         }
  
         ost->enc->frame_number++;
  
     }
  
  }

  The program runs correctly and the video file has the right size and the right duration.
  When I run it with VLC or ffmpeg it shows no error message, the time runs correctly, but no image is printed

• lavf/webvttdec : save cue id and settings as side data

  1er juin 2013, par Matthew Heaney

  lavf/webvttdec : save cue id and settings as side data
  Currently the WebVTT demuxer parses the cues but throws away
  
  the cue id (the optional first line of the cue) and cue
  
  settings (the optional rendering instructions that follow
  
  the timestamp).
  However, in order to write inband text tracks (to WebM
  
  files), the entire cue payload from the WebVTT source must
  
  be preserved.
  This commit makes no change to the data part of the output
  
  buffer packet (where the actual cue text is stored), but
  
  does add the cue id and settings as a side data items, if
  
  they’re present in the cue. Existing code that cares only
  
  about the data part of the packet can continue to ignore the
  
  side data.
  There are two new packet data type flags,
  
  AV_PKT_DATA_WEBVTT_IDENTIFIER and
  
  AV_PKT_DATA_WEBVTT_SETTINGS.
  

  • [DH] libavcodec/avcodec.h
  • [DH] libavcodec/version.h
  • [DH] libavformat/webvttdec.c