Recherche avancée

Sur d’autres sites (4292)

• How to overlay a box on a transparent video in FFmpeg

  28 août 2023, par Sidney

  I'm trying to overlay a box and image on a transparent video in FFmpeg and Python, and have it show up in video editors.

  


  I've written the following code :

  


  import ffmpeg

stream = ffmpeg.input("/dev/zero", f='rawvideo', pix_fmt='rgba', s='{}x{}'.format(1920, 1080),
                              r=30)

stream = ffmpeg.overlay(stream, ffmpeg.input("tmp.png"), x=0, y=0)
stream = ffmpeg.drawbox(stream, x=900, y=100, width=500, height=500, color='blue', thickness="fill")
stream = ffmpeg.trim(stream, duration=1)
stream = ffmpeg.output(stream, "out.webm",
                       pix_fmt="yuva420p")
ffmpeg.run(stream)


  


  While the box displays in most video players :
  

  
When I try to overlay it on another image in any video editor like Kdenlive or Shotcut, or open in FireFox, it doesn't appear :
  

  

  


  This seems odd considering the overlayed image displays fine. Is there something wrong with the drawbox call ?

  


  EDIT :
The same issue occurs when exporting as PNGs, where no frames contain the box when it's not over the image :

  


  stream = ffmpeg.output(stream, "out/out%d.png")


  


  

  


• fate/matroska : Add test for mastering display metadata

  17 février 2021, par Andreas Rheinhardt

  fate/matroska : Add test for mastering display metadata
  The FATE suite already contains a file containing mastering display
  
  and content light level metadata : Meridian-Apple_ProResProxy-HDR10.mxf
  
  This file is used to test both the Matroska muxer and demuxer.
  Reviewed-by : Ridley Combs <rcombs@rcombs.me>
  
  Signed-off-by : Andreas Rheinhardt <andreas.rheinhardt@gmail.com>
  

  • [DH] tests/fate/matroska.mak
  • [DH] tests/ref/fate/matroska-mastering-display-metadata

• Audio MP2 encoding and decoding producing half data C++ and FFMPEG

  7 juillet 2016, par hockeyislife

  I have 16 bit PCM data with stereo setup which I grab from a microphone.

  Once I get the data I encode it using the following encoder settings

  AVCodec* audio_codec = avcodec_find_encoder(AV_CODEC_ID_MP2);
  
  
  
  AVCodecContext* audio_codec_ctx = avcodec_alloc_context3(audio_codec);
  
  audio_codec_ctx->bit_rate = 64000;
  
  audio_codec_ctx->channels = 2;
  
  audio_codec_ctx->channel_layout = AV_CH_LAYOUT_STEREO;
  
  audio_codec_ctx->sample_rate = 44100;
  
  audio_codec_ctx->sample_fmt = AV_SAMPLE_FMT_S16;

  When I pass the audio data into the encoder I see that it takes 4608 bytes of data each time and encodes it correctly to MP2 data. PCM data grabbed by the microphone is 88320 bytes and the encoder takes 4608 bytes each time and compresses it.

  If I take each 4608 byte section that has been encoded and pass it through a decoder with the same settings as above but with a decoder.

  AVCodecID audio_codec_id = AV_CODEC_ID_MP2;
  
  AVCodec * audio_decodec = avcodec_find_decoder(audio_codec_id);
  
  
  
  audio_decodecContext = avcodec_alloc_context3(audio_decodec); 
  
  audio_decodecContext->bit_rate = 64000;
  
  audio_decodecContext->channels = 2;
  
  audio_decodecContext->channel_layout = AV_CH_LAYOUT_STEREO;
  
  audio_decodecContext->sample_rate = 44100;
  
  audio_decodecContext->sample_fmt = AV_SAMPLE_FMT_S16;

  The decoding works and is successful but when I look at the data size it is exactly half 2034 of what was encoded. I dont understand why that would be. I would of imagined I would get 4608 considering the encoder and decoder are the same.

  Can anyone shed some light into why this would be happening. Anything I should be setting ?