Replies: 1 comment 3 replies
-
I built it that way initially and moved away from it. Opencv only supports a limited number of ffmpeg options compared to the command line. Using ffmpeg directly allows much more flexibility for use of hardware acceleration and keeping the libraries up to date. It also has the benefit of running in a separate process and not blocking the GIL while being easier to test external to frigate. |
Beta Was this translation helpful? Give feedback.
3 replies
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
-
OpenCV has the ability to read the video feed from a live stream using the videocapture function. This supports multiple streams, like RTSP, HTTP, etc...
If we could change the video.py to support this, most compatibility issues will be solved.
So my question for discussion here is, why did you choose the ffmpeg method instead of the OpenCV?
Is there a reason behind this approach?
I will be happy to help do the transition if you agree with the change.
Beta Was this translation helpful? Give feedback.
All reactions