Recherche avancée

Sur d’autres sites (11067)

• Can't open encoder when use libavcodec

  1er novembre 2013, par liuyanghejerry

  I’m using libavcodec, version 9.7, to write a simple demo, almost exactly like example in official example.

  However, I can’t open encoder. Also, av_opt_set(context->priv_data, "preset", "slow", 0) always leads to crush.

  This is my code :

  // other code...
  
  int ret = 0;
  
  avcodec_register_all();
  
  AVCodec* codec = NULL;
  
  AVCodecContext* context = NULL;
  
  AVFrame* frame = NULL;
  
  uint8_t endcode[] = { 0, 0, 1, 0xb7 };
  
  codec = avcodec_find_encoder(AV_CODEC_ID_H264);
  
  if(!codec){
  
      qDebug()<<"cannot find encoder";
  
      return;
  
  }
  
  qDebug()<<"encoder found";
  
  
  
  context = avcodec_alloc_context3(codec);
  
  if(!context){
  
      qDebug()<<"cannot alloc context";
  
      return;
  
  }
  
  qDebug()<<"context allocted";
  
  
  
  context->bit_rate = 400000;
  
  /* resolution must be a multiple of two */
  
  context->width = 352;
  
  context->height = 288;
  
  /* frames per second */
  
  context->time_base= (AVRational){1,25};
  
  context->gop_size = 10; /* emit one intra frame every ten frames */
  
  context->max_b_frames=1;
  
  context->pix_fmt = AV_PIX_FMT_YUV420P;
  
  qDebug()<<"context init";
  
  
  
  // av_opt_set(context->priv_data, "preset", "slow", 0); // this will crush
  
  AVDictionary *d = NULL;
  
  av_dict_set(&d, "preset", "ultrafast",0); // this won't
  
  
  
  ret = avcodec_open2(context, codec, &d);
  
  if ( ret < 0) {
  
      qDebug()<<"cannot open codec"</ other code...

  This outputs :

  encoder found

  context allocted

  context init

  cannot open codec -22

  [libx264 @ 0340B340] [IMGUTILS @ 0028FC34] Picture size 0x10 is invalid

  [libx264 @ 0340B340] ignoring invalid width/height values

  [libx264 @ 0340B340] Specified pix_fmt is not supported

  I don’t think the width/height is invalid and format there either. I have no idea what’s wrong here.

  Any help. plz ?

• RuntimeError : abort(OOM). Build with -s ASSERTIONS=1 for more info

  26 avril 2021, par KhoPhi

  I'm using ffmpeg.wasm 0.9.7 (at the time of this question).
Browser is Brave (Version 1.23.71 Chromium : 90.0.4430.72 (Official Build) (64-bit))

  


  I get this error running an overlay ffmpeg command in the browser, slapping an png over an hevc mp4 4K video of size, almost 40 Megabyte.

  


  RuntimeError: abort(OOM). Build with -s ASSERTIONS=1 for more info.

  


  Here's my code in Angular. Stripped down for brevity

  


  async generatePreviews(event: any) {

    if (!ffmpeg.isLoaded()) {
      await ffmpeg.load();
    }

    for (let index = 0; index < this.files.length; index++) {

      this.video = this.files[index];
      const file_name = this.video.name.split('.')[0] + '-preview.mp4';

      ffmpeg.FS('writeFile', 'tempLogo.png', await fetchFile(this.watermark));
      ffmpeg.FS('writeFile', 'tempVideo.mp4', await fetchFile(this.video));


      // working command in normal terminal
      // ffmpeg -i tempVideo.mp4 -i tempLogo.png -filter_complex "overlay=(W-w)/2:(H-h)/2" temp.mp4

      await ffmpeg.run(
        '-i',
        'tempVideo.mp4',
        '-i',
        'tempLogo.png',
        '-filter_complex',
        'overlay=(W-w)/2:(H-h)/2',
        'temp.mp4'
      );

      const data = ffmpeg.FS('readFile', 'temp.mp4');

      var blob = new Blob([data.buffer], { type: 'video/mp4' });
      saveAs(blob, file_name); // using filesaver.js to save the blob

    }
  }
}



  


  In chrome, I read up to 2Gb of file is possible to convert. Not sure why the OOM issues. Any settings I need to set or changes I need to do ?

  


  Update (4/26/2021)

  


  This thread offered a solution, by building the ffmpeg wasm with a few tweaks. I am able to build, but using the built files even causes the OOM faster than the npm built from the ffmpeg wasm repo.

  


• Decoder return of av_find_best_stream vs. avcodec_find_decoder

  7 octobre 2016, par Jason C

  The docs for libav’s av_find_best_stream function (libav 11.7, Windows, i686, GPL) specify a parameter that can be used to receive a pointer to an appropriate AVCodec :

  decoder_ret - if non-NULL, returns the decoder for the selected stream

  There is also the avcodec_find_decoder function which can find an AVCodec given an ID.

  However, the official demuxing + decoding example uses av_find_best_stream to find a stream, but chooses to use avcodec_find_decoder to find the codec in lieu of av_find_best_stream’s codec return parameter :

  ret = av_find_best_stream(fmt_ctx, type, -1, -1, NULL, 0);
  
  ...
  
  stream_index = ret;
  
  st = fmt_ctx->streams[stream_index];
  
  ...
  
  /* find decoder for the stream */
  
  dec = avcodec_find_decoder(st->codecpar->codec_id);

  As opposed to something like :

  ret = av_find_best_stream(fmt_ctx, type, -1, -1, &dec, 0);

  My question is pretty straightforward : Is there a difference between using av_find_best_stream’s return parameter vs. using avcodec_find_decoder to find the AVCodec ?

  The reason I ask is because the example chose to use avcodec_find_decoder rather than the seemingly more convenient return parameter, and I can’t tell if the example did that for a specific reason or not. The documentation itself is a little spotty and disjoint, so it’s hard to tell if things like this are done for a specific important reason or not. I can’t tell if the example is implying that it "should" be done that way, or if the example author did it for some more arbitrary personal reason.