Recherche avancée

Sur d’autres sites (4338)

• FFMPEG Encoding to prores 4444 with alpha doesn't retain original alpha [closed]

  19 mai 2023, par Nick Weeden

  I'm trying to convert an rgba png to prores 4444 with an alpha. When I run the following command it will create prores file with an alpha. But, the alpha is full white, it doesn't retain the values I put in.

  


  ffmpeg -start_number 1001 -r 24.000 - "pngWithAlpha_%06d.png" -c:v prores_ks -profile:v 4 -alpha_bits 16 -y "proresWithAlpha.mov"


  


  I've tried this with both FFMPEG 4.2.2 and 6.0.0 on a linux system, I'm not sure if I'm doing it wrong or if it is a bug.

  


  Looking at this post it seems it should be possible (they're complaining about quality not a fully missing transfer of the alpha).

  


  I would expect it to pass the alpha straight through but it doesn't. I used alpha_extract to copy the alpha in to the rgb to confirm I'm able to read the alpha from the png, which worked. But the alpha was still full white.

  


• ffmpeg zoompan multiple zoomin/zoomout on different areas of big image [closed]

  10 janvier 2024, par LinuxAdmin

  I need to create a video, with zooming in and out different areas of one big square image.
I tried different zoompan examples, but can't do what I want. What I resolved is keeping aspect ratio of square image, setting setsar=ratio='(1/1)' before zoompan and setdar=ratio='(1/1) after.

  


  Video begins with full image (keeping aspect ratio), then zooms in to a sector of image (1sec), stay 1 second, and zoomout back to full size (1sec). I have 40 sectors, all same size. I can iterate every with bash, adding zoompan line for every sector, setting right x/y coordinates.

  


  Please help me evaluate zoompan expression for getting such result.

  


• avutil/threadmessage : split the pthread condition in two

  1er décembre 2015, par Clément Bœsch

  avutil/threadmessage : split the pthread condition in two
  Fix a dead lock under certain conditions. Let’s assume we have a queue of 1
  
  message max, 2 senders, and 1 receiver.
  Scenario (real record obtained with debug added) :
  
      [...]
  
      SENDER #0 : acquired lock
  
      SENDER #0 : queue is full, wait
  
      SENDER #1 : acquired lock
  
      SENDER #1 : queue is full, wait
  
      RECEIVER : acquired lock
  
      RECEIVER : reading a msg from the queue
  
      RECEIVER : signal the cond
  
      RECEIVER : acquired lock
  
      RECEIVER : queue is empty, wait
  
      SENDER #0 : writing a msg the queue
  
      SENDER #0 : signal the cond
  
      SENDER #0 : acquired lock
  
      SENDER #0 : queue is full, wait
  
      SENDER #1 : queue is full, wait
  Translated :
  
   - initially the queue contains 1/1 message with 2 senders blocking on
  
     it, waiting to push another message.
  
   - Meanwhile the receiver is obtaining the lock, read the message,
  
     signal & release the lock. For some reason it is able to acquire the
  
     lock again before the signal wakes up one of the sender. Since it
  
     just emptied the queue, the reader waits for the queue to fill up
  
     again.
  
   - The signal finally reaches one of the sender, which writes a message
  
     and then signal the condition. Unfortunately, instead of waking up
  
     the reader, it actually wakes up the other worker (signal = notify
  
     the condition just for 1 waiter), who can’t push another message in
  
     the queue because it’s full.
  
   - Meanwhile, the receiver is still waiting. Deadlock.
  This scenario can be triggered with for example :
  
      tests/api/api-threadmessage-test 1 2 100 100 1 1000 1000
  One working solution is to make av_thread_message_queue_send,recv()
  
  call pthread_cond_broadcast() instead of pthread_cond_signal() so both
  
  senders and receivers are unlocked when work is done (be it reading or
  
  writing).
  This second solution replaces the condition with two : one to notify the
  
  senders, and one to notify the receivers. This prevents senders from
  
  notifying other senders instead of a reader, and the other way around.
  
  It also avoid broadcasting to everyone like the first solution, and is,
  
  as a result in theory more optimized.
  

  • [DH] libavutil/threadmessage.c