Recherche avancée

Sur d’autres sites (5457)

• opusdsp : adjust and optimize C function to match assembly

  15 août 2019, par Lynne

  opusdsp : adjust and optimize C function to match assembly
  The C and asm versions behaved differently _outside_ of the codec.
  The C version returned pre-multiplied 'state' for the next execution
  
  to use right away, while the assembly version outputted non-multiplied
  
  'state' for the next execution to multiply to save instructions.
  
  Since the initial state when initialized or seeking is always 0,
  
  and since C and asm versions were never mixed, there was no issue.
  However, comparing outputs directly in checkasm doesn't work without
  
  dividing the initial state by CELT_EMPH_COEFF and multiplying the
  
  returned state by CELT_EMPH_COEFF for the assembly function.
  Since its actually faster to do this in C as well, copy the behavior the
  
  asm versions use. As a reminder, the initial state 0 is divided by
  
  CELT_EMPH_COEFF on seek and init (just in case in the future this is
  
  changed, its technically more correct to init with CELT_EMPH_COEFF than 0,
  
  however when seeking this will result in more audiable pops, unlike with 0
  
  where the output gets in sync over a few samples).
  

  • [DH] libavcodec/opus_celt.c
  • [DH] libavcodec/opusdsp.c

• Extracting video frame image with libavformat and libswscale gives flipped image

  21 novembre 2015, par ivan.ukr

  I am using following code to extract image frames outputted by avcodec_decode_video2() into BMP file :

  // Convert the image from its native format to RGB
  
  int height = sws_scale(state.sws_ctx, 
  
               (uint8_t const * const *)state.frame->data,
  
               state.frame->linesize, 
  
               0, 
  
               state.video_codec_ctx->height,
  
               state.picture.data, 
  
               state.picture.linesize);
  
  
  
  // ..... here BMP header initialization goes ...
  
  
  
  // Extract pixels
  
  {
  
      int y = 0;
  
      size_t frame_line_size = state.picture.linesize[0];
  
      uint8_t* bmp_data = &bmp->data[0];
  
      uint8_t* picture_data = state.picture.data[0];
  
      while(y < height)
  
      {
  
          memcpy(bmp_data, picture_data, image_line_size);
  
          bmp_data += image_line_size_with_padding;
  
          picture_data += frame_line_size;
  
          ++y;
  
      }
  
  }

  But it results flipped image :

  

  Can anyone please help me to understand, what I am doing wrong here ?

  Video stream information :

  Input #0, mov,mp4,m4a,3gp,3g2,mj2, from 'testdata/video3.mp4':
  
    Metadata:
  
      major_brand     : isom
  
      minor_version   : 512
  
      compatible_brands: isomiso2avc1mp41
  
      creation_time   : 1970-01-01 00:00:00
  
      title           : video3.mp4
  
      encoder         : Lavf52.78.3
  
    Duration: 00:00:07.93, start: 0.000000, bitrate: 1467 kb/s
  
      Stream #0.0(und): Video: h264 (High), yuv420p, 640x320 [PAR 1:1 DAR 2:1], 1331 kb/s, 29.73 fps, 29.73 tbr, 29734 tbn, 59.47 tbc
  
      Metadata:
  
        creation_time   : 1970-01-01 00:00:00
  
      Stream #0.1(und): Audio: aac, 48000 Hz, stereo, fltp, 127 kb/s
  
      Metadata:
  
        creation_time   : 1970-01-01 00:00:00
  
  Video file 'testdata/video3.mp4' has 236 frames.

  I have alternate code that uses libopencv, and it gives correct image from the same video file, but I need to reach the correct effect using directly ffmpeg libs.

  

• Prevent libavformat (FFmpeg) from adding "ENCODER" tag to output / help stripping the tags

  9 juillet 2021, par Wyatt Ward

  I made a bash script that transfers audio or video metadata from one file to another, regardless of media container format, via FFMpeg. My problem is that FFMpeg consistently adds a 'ENCODER' tag.

  



  Example :

  



  Before running through FFMpeg :

  



   File : track01.cdda.flac

  



    Metadata:
    ALBUM           : Wish You Were Here
    ARTIST          : Pink Floyd
    COPYRIGHT       : 1975 Harvest Records
    DATE            : 1975
    GENRE           : Experimental Rock
    TITLE           : Shine On You Crazy Diamond
    track           : 1


  



  After running through FFMpeg :

  



   File : track01-iphone-alac.m4a

  



    Metadata:
    ALBUM=Wish You Were Here
    ARTIST=Pink Floyd
    COPYRIGHT=1975 Harvest Records
    DATE=1975
    GENRE=Experimental Rock
    TITLE=Shine On You Crazy Diamond
    track=1
    ENCODER=Lavf55.12.100


  



  So really, I want to either force FFMpeg to not add the 'ENCODER' tag, or I want to strip that tag off afterwards. Is there a way to do this ? I really don't want to spend hours trying to compile FFMpeg again on my Pentium 4 HT - the only working computer I have at the moment. I'd prefer not to have to use another program unless it's related enough to FFMpeg or MPlayer/Mencoder that I won't have to install anything new if I have those installed.