Repository navigation
The timecode required by the imported EDL file in DaVinci Resolve's scene cut detection function is 1 frame different from the timecode exported by PySceneDetect. #516
Description
Description:
Slice of EDL format content exported by PYSCENEDETECT:
001 AX V C 00:00:00:00 00:00:01:16 00:00:00:00 00:00:01:16
002 AX V C 00:00:01:17 00:00:03:14 00:00:01:17 00:00:03:14
003 AX V C 00:00:03:15 00:00:05:05 00:00:03:15 00:00:05:05
DaVinci Resolve
DaVinci Resolve's scene cut detection will directly use 00:00:01:16 as the start of 002 instead of the correct 00:00:01:17
However, if I change the data to the following, it imports correctly:
001 AX V C 00:00:00:00 00:00:01:17 00:00:00:00 00:00:01:17
002 AX V C 00:00:01:17 00:00:03:15 00:00:01:17 00:00:03:15
003 AX V C 00:00:03:15 00:00:05:06 00:00:03:15 00:00:05:06
Because the current detection is only a rough detection, DaVinci Resolve's scene cut detection can be used to correct and adjust the clips to the correct size before importing them into the software.
In addition, we hope that PySceneDetect can support reading EDL files for video cutting and saving pictures in the future，thank
Activity
Breakthrough commentedon Aug 21, 2025
Could you provide some more information about the file you're processing (e.g. framerate) and how you're generating the EDL file? I tested locally with quite a few inputs and I never had any off-by-one issues. If you can include some debug logs too that may be useful.
There was a report of this in #502 but that was due to an inaccuracy in the framerate specified by the user. Does this happen with all inputs or just a specific one?
This is possible, you are welcome to file a feature request for this. I assume we would add a
load-edlcommand similar toload-scenes.Breakthrough commentedon Aug 23, 2025
Sorry I misread the output, I see the issue now:
I was not aware of this, but I can certainly add a flag or a fix for this.
[save-edl] Remove PTS correction in end timestamp calculation #516
[save-edl] Remove PTS correction in end timestamp calculation #516
Breakthrough commentedon Aug 25, 2025
@tanglangxia I am going to release v0.6.7 very shortly. It includes a fix for this. Thank you for the bug report.
tanglangxia commentedon Sep 23, 2025
Thanks for the fix