Recherche avancée

Sur d’autres sites (5640)

• How to force avcodec to use unaligned frame data planes ?

  26 février 2015, par user3244284

  I have been searching high and low for an option to force avcodec to use unaligned memory for its AVFrame data.

  Depending on the pixel format, the horizontal planes of an AVFrame->data may be padded with extra data to be aligned to memory for performance.

  eg : a 1920 * 1080 video with 4 bytes per pixel will have 1920 * 4 = 7680 bytes per plane.

  With avcodec if you are decoding this video it will create 7808 bytes per plane.

  This adds 7808 - 7680 = 128 bytes of extra padding.

  For my purposes I would like to force avcodec to use unaligned data so I can copy an entire continuous chunk of frame data instead of copying and formatting smaller pieces one at a time to a continuous chunk.

  The following flag found in the headers :

  /* encoding support
  
     These flags can be passed in AVCodecContext.flags before initialization.
  
     Note: Not everything is supported yet.
  
  */
  
  
  
  /**
  
   * Allow decoders to produce frames with data planes that are not aligned
  
   * to CPU requirements (e.g. due to cropping).
  
   */
  
  #define CODEC_FLAG_UNALIGNED 0x0001

  Setting this AVCodecContext.flags to be CODEC_FLAG_UNALIGNED, the assumption is that the AVFrame->data is now unaligned, this is not the case.

  I’m not sure if I am looking at the right place or using this flag correctly.

  Regards,

  Curious George

• QAudioSink play AVFrame data decoded by ffmpeg

  26 mai 2022, par liuyulvv

  I want to using QAudioSink to play PCM data decoded by ffmpeg.

  


  I first try to play a pcm file and I made it :

  


  Here is the override function readData of QIODevice

  


  qint64 PCMDevice::readData(char *data, qint64 maxlen)
{
    int len = m_inputFile.size();
    len = len < maxlen ? len : maxlen;

    m_inputFile.read(data, len);
    return len;
}


  


  And I get the right result, the file is playing.

  


  But I want to play AVFrame data from the memory :

  


  qint64 PCMDevice::readData(char *data, qint64 maxlen)
{
    AVFrame *res = getFrame();
    if (res == nullptr)
        return 0;

    // Something wrong here
    auto ret = res->nb_samples;
    memcpy(data, res->data[0], ret);

    delete res;
    res = nullptr;
    return ret;
}


  


  How to make AVFrame's data to the data that QAudioSink need ?

  


  I'm pretty sure I'm using the correct sample_fmt, sample_rate etc.

  


  Where is the data of the stereo AVFrame and how to copy it to the data correctly ?

  


• I want to separate MPEG2ts data into images and metadata (klv)

  4 août 2020, par 夏目たかし

  please. Help me.
  
I would like to separate the streamed MPEG2ts data into images and metadata (klv).
  

  


  Delivery
  
ffmpeg -fflags +genpts -re -i NightFlightIR.mpg -f mpegts -c copy -map 0:v -map 0:d udp ://127.0.0.1:5004
  

  
reception
  
ffmpeg -i 'udp ://127.0.0.1:5004' -vsync 0 -q:v 2 -f image2 -update 1 img.jpg -c copy -map d:0 -f data klv.bin
  

  
The image is output, but the klv.bin is empty.
  

  
This command outputs data to standard output, but cannot output to a file.
  
ffmpeg -i 'udp ://127.0.0.1:5004' -vsync 0 -q:v 2 -f image2 -update 1 img.jpg -c copy -map d:0 -f data -
  

  
On ffmpeg2.8.6, the klv.bin contains data, but on 3.x and 4.x, it's empty.
  
Is it possible to write klv data to klv.bin on 4.x ?