Recherche avancée

Sur d’autres sites (7137)

• Can AWS Transcoder merge an audio file and a video file ?

  7 juillet 2017, par Daniel Bernstein

  I’m interested in capturing video via html5. The problem is that the video is captured separately from the audio. So to produce a video with audio, one must merge them together. It appears that ffmpeg will do the trick.

  Does anyone know if AWS Transcoder will perform this operation as well ?

• Is seeking to specific frames of an MJPEG video faster than doing the same for an h.264 file ?

  9 février 2018, par David Parks

  I am recording video and will need to seek to a specific frame in the video quickly and randomly.

  I can record the video in MJPEG or h.264 compression standards.

  I understand that MJPEG produces individual jpegs and produce larger file sizes than h.264, and h.264 compresses across multiple frames of video.

  Does this difference mean that I will get faster seek times when
  seeking to a random location in the file using ffmpeg or gstreamer
  programmatically ?

  Will the MJPEG allow the frame-seek operation to reduce the IO requirements when reading just 1 frame from a video file ?

• FFMPEG : Mapping Decoded frame buffer to FFMPEG's output frame buffer

  9 décembre 2013, par Zax

  I have integrated my custom decoder into FFMPEG's multimedia framework.
  I have the AVCodec structure inside my video decoder as shown below :

  AVCodec ff_hello_decoder = {
  
  
  
      .name           = "myDec",
  
      .type           = AVMEDIA_TYPE_VIDEO,
  
      .id             = AV_CODEC_ID_HELLO,
  
      .init           = myDec_decode_init,
  
      .close          = myDec_decode_close,
  
      .decode         = myDec_decode_frame,
  
      .pix_fmts       = (const enum AVPixelFormat[]) {AV_PIX_FMT_YUV420P},
  
  };

  In function myDec_decode_init() i perform all initializations, memory allocation etc. In function myDec_decode_close(), all the allocated memory is deallocated. In function myDec_decode_frame(), i read the input stream, and decode a frame as shown below :

  static int myDec_decode_frame(AVCodecContext *avctx, void *data,int *got_frame_ptr, AVPacket *avpkt)
  
  {
  
  
  
       AVFrame *frame=data;
  
       unsigned char *YUVptr;
  
  
  
       //Here i read the input stream and pass it to the decoder.
  
  
  
       //Decoder gives me a address that points to the YUV data, i collect the address in YUVptr as shown below
  
  
  
       YUVptr=myDecodeFrame(...);
  
  
  
       //Inorder to map this pointer to the output frame pointer provided by ffmpeg, i perform the below operation
  
       frame->buf=YUVptr;
  
  
  
       //give indication to ffmpeg framework that the data has been mapped to frame buffer
  
       *got_frame_ptr=1;
  
  
  
       return 0;
  
  
  
  }

  When i execute the ffmpeg command from terminal i get the following message :

  ./ffmpeg -vcodec myDec -i input.bin output.yuv
  
  
  
  Codec 0xFF1081 is not in the full list
  
  
  
  Output file #0 does not contain any stream

  My Question is :

  1) Why am i getting the first message i.e. Codec ... is not in the full list ? What has to be done to get this solved

  2) Why am i getting the message Output file #0 does not contain any stream, althought in the definition of AVCodec i have specified that pix_fmts are of type AV_PIX_FMT_YUV420P ? How can this be resolved ?

  3) Lastly, i still doubt, that i'm actually not mapping the yuv buffer data properly with the ffmpeg's output buffer pointer i.e. void* data parameter. Am i doing something wrong here ?

  Any information regarding the same shall be really helpful to me. Please do provide you valuable suggestions. Thanks in advance.