Drawbacks of using FFMPEG through OpenCV vs directly using FFMPEG libraries

asked 2018-09-05 08:20:45 -0600

844G gravatar image

I have been exploring the option providing Video support through OpenCV. I would like to use FFMPEG as the backend. One of the things that I observed with OpenCV there is not much control on the video encoding settings that I could pass to FFMPEG. Is my understanding correct? What are the advantages and drawbacks of using FFMPEG though OpenCV vs directly calling FFMPEG apart from the programming ease?

edit retag flag offensive close merge delete

Comments

opencv is a computer-vision library.

to edit / encode video files, indeed you're better off using (lib)ffmpeg / gstreamer directly.

berak gravatar imageberak ( 2018-09-05 09:50:00 -0600 )edit

I understand that OpenCV is CV library. We already have OpenCV support. Hence we were looking at leveraging OpenCV itself for the video support. I see that we can pass FPS(frames per second), quality( 0 to 100) parameters to OpenCV video APIs. I would like to understand if these are the only parameters that we can use to control the video encoding. If there are any other possibilities.

844G gravatar image844G ( 2018-09-06 00:25:02 -0600 )edit

"I would like to understand if these are the only parameters that we can use to control the video encoding" -- to my knowledge, yes.

berak gravatar imageberak ( 2018-09-06 01:07:26 -0600 )edit