You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Right now it is difficult as avcut relies on the number of decoded frames as a last resort to match packets and frames in some cases. I had an idea for another approach but this needs more research.
When you have a long video, it takes minutes to run, even when the dropped segment is very short.
Is it possible to jump to the timestamp and scan backwards/forward until it finds a keyframe instead of scanning the whole file?
The text was updated successfully, but these errors were encountered: