Recherche avancée

Sur d’autres sites (13817)

• "fixing" RTL texts from logical to visual, before embedding in video as subtitles with ffmpeg

  13 mai 2024, par Berry Tsakala

  I'm searching for the correct way to pre-process my subtitles files before hard-coding them into video clips.

  


  Currently, ffmpeg does not process RTL (right-to-left) languges properly ; I have detailed the problem here :
https://superuser.com/questions/1679536/how-to-embed-rtl-subtitles-in-a-video-hebrew-arabic-with-the-correct-lan

  


  However, there could be 2 programmatic solutions :

  


  

  1. adding certain unicode control characters can fix (or partially fix) the text, which is then fed into ffmpeg, giving good results.

  


  


  

  • character 0x200F at the end of a hebrew clause, after punctuation

  


  • character 0x202B, I haven't yet learned its usage.

  


  


  

  2. I can edit the text so that it will produce the correct results on ffmpeg.
But that requires smart BiDi algorithm.

  


  


  Do you know how to preprocess such text ?

  


  (this is NOT an encoding question. It is about RTL/LTR algorithm to use.)

  


  Thank you

  


• How can I add a background image while combining *.png files using ffmpeg

  12 février 2023, par Walt Howard

  I'd like to add a background image while combining *.png files to make an mp4 file, using ffmpeg.

  


  Currently, I have to first create an animated gif using imagemagick convert, like this :

  


  convert -delay ${DELAY} -dispose Background *.png -coalesce null: ${BACKGROUND} -compose dstOver -layers composite -layers optimize -loop 0 planet.gif


  


  Then I process that animated gif using ffmpeg to optimize it :

  


  ffmpeg -i planet.gif -movflags faststart -pix_fmt yuv420p -vf "scale=trunc(iw/2)*2:trunc(ih/2)*2" planet.mp4


  


  Which sort of works, but is slow and has a problem. I craps out if I go above about 100 input files, and creates animated gifs that are basically just white screens.

  


  ffmpeg alone works great, and 100 times faster stitching the *.png files together

  


  ffmpeg -framerate 10 -pattern_type glob -i '*.png' -movflags faststart -pix_fmt yuv420p -vf "scale=trunc(iw/2)*2:trunc(ih/2)*2" planet.mp4

  


  But, no background there. I haven't found anyway to add that background. Does ffmpeg even support such a thing ?

  


• Pipe streaming video from FFMPEG into OpenCV

  8 juillet 2014, par wunjo

  I am using a Logitech C920 webcam with the beaglebone black running debian. I have successfully written an OpenCV python script that allows me get video from the camera and track objects. Unfortunately, the resolution, FPS, and overall video quality are poor using this method. I have found the proper code that allows me to utilize the webcam’s H264 feature outside of OpenCV.

  I’ve been trying to figure out how to pipe the output from ffmpeg INTO openCV so that I can utilize the camera’s H264 capabilities as my video input for my OpenCV script. My commands look like this :

  ./capture -F -o -c0|avconv -re -i - -vcodec copy -f rtp rtp ://192.168.7.1:1234/|python camera.py

  The first part of the pipe is using the linux v4l2 program that allows me to utilize my camera’s H264 capabilities. The output from that is then fed to ffmpeg and then streamed to RTP. I want to retrieve the RTP stream in my OpenCV script but I haven’t had success in getting it to work. Would I initialize something like the following in my script :

  capture = VideoCapture("rtp ://192.168.7.1:1234/")

  Please let me know if I am even going about this the right way. Is there a way to capture the stream directly from ffmpeg instead of streaming to RTP ? Any suggestions are greatly appreciated. Thank you !