Advanced search

On other websites (10562)

• avcodec/pngdec: support alpha blending for palette apng

  5 March 2022, by Paul B Mahol

  avcodec/pngdec: support alpha blending for palette apng
  Update clock test, as PAL8 apngs are now decoded as RGBA.
  

  • [DH] libavcodec/pngdec.c
  • [DH] tests/ref/fate/apng-clock

• How to achive exactly 25 images per second. Pygame

  14 September 2021, by Vax

  Each frame I take audio frequencies read them and convert them into a visual way. Then I take a picture.

  


  My problem is that the following approach takes a different amount of images each second. So when I convert images into a video it becomes shorter than the audio that played as input ( 5-10 seconds shorter).

  


  FPS = 25
clock = pygame.time.Clock()

pygame.mixer.music.load(filename)
pygame.mixer.music.play(0)

running = True 
while running and pygame.mixer.music.get_busy():

    print(clock.get_fps()) #output 21 - 25 FPS

    t = pygame.time.get_ticks()
    deltaTime = (t - getTicksLastFrame) / 1000.0
    getTicksLastFrame = t

    for event in pygame.event.get():
        if event.type == pygame.QUIT:
            running = False

    screen.fill((0, 0, 0))

    for b in bars:
        b.update(deltaTime, get_decibel(pygame.mixer.music.get_pos() / 1000.0, b.freq))
        b.render(screen)

    pygame.display.flip()

    pygame.display.update()
    video.make_png(screen)
    clock.tick(FPS)


  


  I was thinking about slow down audio, so I will take only (12+- images per second), but this approach wouldn't work either because the FPS cap isn't stable.
Another way I was thinking about is to stretch somehow the video to the length of the audio with an FFmpeg... However, I believe that there should be a more elegant solution.

  


  Will really appreciate any advice!

  


• opencv vs live555 vs ffmpeg for decoding rtsp stream with added metadata?

  17 August 2021, by Peter Parker

  From an IP camera I must decode a h.264 encoded rtsp transport stream containing a program clock reference (PCR), transmitted in the adaptation field of the TS packet. I am using a Windows machine.

  


  This PCR field is transmitted at least every 100ms.

  


  I need to extract this timestamp data for each frame of the video.

  


  I would also like to use OpenCV to process each frame, whilst acknowledging the current time, based on the PCR field value.

  


  OpenCV is not capable of decoding the live rtsp url AND it discard the ancillary data I need.

  


  Do I need to use live555 along with ffmpeg to decode the stream? Or is it possible to extract this data using ffmpeg and opencv libraries alone?

  


  Suppose I embed a SMPTE/MISB compliant Microsecond Timestamp (KLV metadata packet) in the VANC space after achieving this.

  


  Will the same strategy work?

  


  Thanks in advance for any pointers.