Recherche avancée

Sur d’autres sites (3571)

• How to force usage of specific codec with ffmpeg ?

  4 septembre 2013, par Srv19

  I am trying to read a certain video file using ffmpeg, and have run into a bit of trouble.

  This is the code that starts to read it :

  int _tmain(int argc, _TCHAR* argv[])
  
  {
  
      if (argc < 2)
  
      {
  
          fprintf( stderr, "Filename is needed as an argument\n" );  
  
          return 1;  
  
      }
  
      /* register all formats and codecs */
  
      av_register_all();
  
      AVFormatContext* fmt_ctx = NULL;
  
      /* open input file, and allocate format context */
  
      const char *src_filename = argv[1];
  
      if (avformat_open_input(&fmt_ctx, src_filename, NULL, NULL) < 0) {
  
          fprintf(stderr, "Could not open source file %s\n", src_filename);
  
          abort();
  
      }
  
      /* retrieve stream information */
  
      AVDictionary *      options;
  
      int res = avformat_find_stream_info(fmt_ctx, &options);
  
      if (res < 0) {
  
          fprintf(stderr, "Could not find stream information\n");
  
          abort();
  
      }
  
      ...
  
  }

  I am consistently getting the following message :

  [avi @ 005f2fe0] Could not find codec parameters for stream 0 (Video :
  none (GREY / 0x59455247), 1280x1024) : unknown codec

  Consider increasing the value for the 'analyzeduration' and 'probesize' options

  I get the same message when i run ffmpeg tool on the same file.

  However, i know what is in the video stream - raw video data with YUV8 format. In fact, when i pass that to ffmpeg via -c:v rawvideo option, there is no problem. Recoding, transforming etc - ffmpeg does it like the magic tool it is.

  Now, the question is : when using ffmpeg api, what is the equivalent of that option ? I desperately need access to the frames of file.

• Decode H.264 stream from live555 using ffmpeg

  3 septembre 2013, par Straticiuc Vicu

  I try to decode H.264 frames for an IP Camera that uses onvif. For this i am using example testRTSPClient from live555. In FAQ is said that function afterGettingFrame() is called every time a new packet came in buffer and that packet is an NAL unit. On live555 part everything looks good. Now for decoding i am using ffmpeg and the frame i pass to the decoder are formed like discribed Here. Now my problem is that the ffmpeg is unable to decode my frames and i am getting errors like :

  top block unavailable for requested intra4x4 mode -1 at 18 0
  
  error while decoding MB 18 0
  
  no picture 
  
  P sub_mb_type 13 out of range at 2 3

  What am i doing wrong with ffmpeg ?

  PS. example of I-Frame passed to ffmpeg Here

   with:
  
   red   is startcode
  
   green is SPS info
  
   black is PPS info 
  
   blue  is I-Frame delivered by live555

• ffmpeg : jpeg-images with different resolution -> video sequence ; using padding instead of scaling [on hold]

  3 septembre 2013, par user2742503

  I am currently struggling with the following problem using ffmpeg : I have a bunch of jpg-files with different resolution (say, one is 320x240, one is 800x600 etc) and want to create an avi-file using mjpeg as a video codec withn a output-video resolution of 800x600. The images itself should not be scaled in any way, that means : What I want to achieve is to just place the image in the center and fill the remaining space with white or black instead of scaling up the smaller images to the target resolution.

  So far I have used

  ffmpeg -f image2 -r 1 -i image_%04d.jpg -vcodec mjpeg -s 1920x1080 -qscale 2 -r 25 output2.avi

  or

  ffmpeg -f image2 -r 1 -i image_%04d.jpg -vcodec mjpeg -vf scale=1920:1080 -qscale 2 -r 25 output2.avi

  in case of scaling to HD resolution.

  Would be very thankful for any support on that !

  Thank you in advance !