Recherche avancée

Sur d’autres sites (13733)

• AWS Lambda. Error=20 (Not a directory) when moving FFmpeg to /tmp

  18 novembre 2017, par Omar Riaz

  I am using the ffmpeg-cli-wrapper to run FFmpeg from an application I upload onto AWS lambda. I was initially getting the

  error=13 Permission Denied

  when trying to call it via the wrappers ProcessBuilder. My inital solution was to chmod 755 ffmpeg before uploading, but it didn’t help.

  For information, my FFmpeg and ffprobe files are located in the following classpath : static/ffmpeg/ffmpeg and static/ffmpeg/ffprobe. They are also statically linked.

  AWS Lambda permission denied when trying to use ffmpeg

  I’ve tried to follow the instructions given in the example above, but when I try to perform either mv or cp command :Runtime.exec("mv " + pathToFFmpeg + " /tmp"), I get the

  error=20, Not a directory

  error.

  I know that I have the correct path for FFmpeg because the following command mv *pathToFFmpeg* *an arbitrary name* runs without error, meaning that the file is there and so the mv command just renames it as it’s supposed to do.

• log : allow color highlighting in Cygwin’s mintty

  4 avril 2014, par James Darnley

  log : allow color highlighting in Cygwin’s mintty
  Configure will detect the availability of the Windows’ console functions and set
  
  HAVE_SETCONSOLETEXTATTRIBUTE.  Meaning av_log will use those functions to
  
  control colours.  When ffmpeg is run in Cygwin’s mintty terminal emulator it
  
  will not use colour highlighting in this case.
  Mintty responds to the usual escape code colours (it even supports 256 colours).
  
  Windows’ cmd.exe does not.  Fortunately it seems that Cygwin’s emulation layer
  
  now translates the basic 16 colours into Windows’ Console command functions.
  That means that we can have av_log use the standard colour commands and let
  
  ffmpeg print colours in both mintty and cmd.
  Signed-off-by : Michael Niedermayer <michaelni@gmx.at>
  

  • [DH] libavutil/log.c

• Is there a way to extract a video frame, encode a sentence behind it, put it back, and then decode it ?

  29 juillet 2021, par Jesse Hix

  I am trying to do stenography on an MP4 video where I am pulling a frame out at a certain time, 5.52 sec, saving that frame as a jpg or jpeg using any image stenography tool, in this case stegosuite, to encode a sentence behind the image and then reinsert that image back into the video at the same time and then extract it again with the same command and using the same steno tool read the message.

  


  Up until decoding the message everything runs without issue.
I know that it is not working because the videos are not the same in size meaning that the frankenstein video is not the same as the original and thus messing up the frame extraction.

  


  Question :
  Is there a way to extract a frame encode a sentence behind it put it back and then decode it ?

  


  What I tried :
  
Used to extract the frame both times :

  


  ffmpeg -ss 5.52 -i original.mp4 -vframes 1 frames_%d.jpg


  


  Used to combine the image back into the video :

  


  ffmpeg -i original.mp4 -i frames_1_embed.jpg -filter_complex "[1]setpts=5.52/TB[im];[0][im]overlay=eof_action=pass" -c:a copy out_1.mp4