Advanced search

On other websites (4558)

• How can low frame rate video be made to look more smooth?

  26 October 2017, by d3pd

  I am trying to clean up a video that was recorded in 2003 in low-light conditions on what was possibly a cameraphone. The video has been cleaned up somewhat (cropped, logos removed and stabilized), but it remains quite jerky, due in large part to its low frame rate. What are some tricks that might clean up the video in this regard? I feel that I am asking for something a bit like tweening in flash animations, but for pixels, whereby additional frames are generated using nearby frames of the video. Does such a trick exist? Is there another way to approach this problem?

  To reproduce the video processing so far, take the following steps:

  # get video
  
  wget http://www.anwarweb.net/saddamdown.wmv
  
  # crop
  
  ffmpeg -i saddamdown.wmv -filter:v "crop=292:221:14:10" -c:a copy saddamdown_crop.wmv
  
  # remove logo 1
  
  ffmpeg -i saddamdown_crop.wmv -vf delogo=x=17:y=77:w=8:h=54 -c:a copy saddamdown_crop_delogo_1.wmv
  
  # remove logo 2
  
  ffmpeg -i saddamdown_crop_delogo_1.wmv -vf delogo=x=190:y=174:w=54:h=8 -c:a copy saddamdown_crop_delogo_1_delogo_2.wmv
  
  # stabilize
  
  ffmpeg -i saddamdown_crop_delogo_1_delogo_2.wmv -vf deshake saddamdown_crop_delogo_1_delogo_2_deshake.wmv

  Note: The video is of the Saddam Hussein execution.

• avformat/mpegtsenc: do not include adaptation field in teletext TS packets

  21 October 2021, by Alex Shumsky

  avformat/mpegtsenc: do not include adaptation field in teletext TS packets
  From ETSI EN 300 472 V1.3.1 (2003-05) Specification for conveying ITU-R System
  
  B Teletext in DVB bitstreams:
  4.1 Transport Stream (TS) packet format
  
  The standard TS packet syntax and semantics are followed, noting the following
  
  constraint:
  
   adaptation_field_control only the values "01" and "10" are permitted.
  Some set top boxes (Motorola, Arris, Zyxel) refuse non-conforming packets.
  Signed-off-by: Alex Shumsky <alexthreed@gmail.com>
  
  Signed-off-by: Marton Balint <cus@passwd.hu>
  

  • [DH] libavformat/mpegtsenc.c

• Why does ffplay read both video and keyboard input from stdin?

  27 January 2016, by cxrodgers

  I’m trying to compress a video feed from a webcam while simultaneously displaying it, using ffmpeg and ffplay. I do actually have this working, but I want to disable the ffplay window from interpreting keyboard presses.

  It took me a while to figure this out but here’s what I’m using:

  ffmpeg -f video4linux2 -i /dev/video0 -vcodec mpeg4 -f rawvideo - \
  
    | tee output.mkv \
  
    | ffplay -fflags nobuffer -

  (Actually I am doing all of this from a Python script using the subprocess module. Here I have represented it as a straightforward terminal command because the result is the same.)

  So this actually works and does everything I want. The only thing is that if the ffplay window is active, it interprets keypresses (like "F" for fullscreen). Instead I want it to completely ignore all keypresses.

  My questions:

  1. How is this even possible? I thought I was redirecting video input to stdin, and then telling ffplay to read video from stdin. How can keypresses be multiplexed on the same pipe?
  2. How can I disable this behavior? I tried "-nostdin" but it doesn’t work with my version.

  # ffplay -nostdin output.mkv

  ffplay version N-77455-g4707497 Copyright
  (c) 2003-2015 the FFmpeg developers built with gcc 4.8 (Ubuntu
  4.8.4-2ubuntu1 14.04)

  ...

  Failed to set value ’output.mkv’ for option ’nostdin’: Option not
  found