+- Selur's Little Message Board (https://forum.selur.net)
+-- Forum: Hybrid - Support (https://forum.selur.net/forum-1.html)
+--- Forum: Problems & Questions (https://forum.selur.net/forum-3.html)
+--- Thread: [HELP] Find And (Optionally) Fix Bad Frames (/thread-2920.html)
Pages:
1
2
Find And (Optionally) Fix Bad Frames - Miranda - 19.11.2022
Is this script included in Hybrid's Avisynth filter options?
Find And (Optionally) Fix Bad Frames
RE: Find And (Optionally) Fix Bad Frames - Selur - 19.11.2022
No, and I won't add that script because it's ugly written (not generic, but requires manual editing of the script whenever used).
Cu Selur
RE: Find And (Optionally) Fix Bad Frames - Miranda - 19.11.2022
Is there another way of detecting digital glitched frames?
RE: Find And (Optionally) Fix Bad Frames - Selur - 19.11.2022
Quote:Is there another way of detecting digital glitched frames?That script will not detect digital glitched frames, at least not when the glitch is caused by transfer issues or similar.
Quote:#It works very well for finding both blank frames and also "flash" frames (like those causedIt will replace any frames based on luma differences, and will replace a frame if it's luma difference is high in comparison to the frame before and after it.
#by a photographer's flash). It will also find single frames which have lots of
#static or pixels. It can also find a frame with large x or y displacement from adjacent frames, like
#a film frame that wasn't properly registered in the film gate, or an analog
#video frame that lost vertical sync.
=> It does not detect digital glitched frames.
Cu Selur
RE: Find And (Optionally) Fix Bad Frames - Miranda - 19.11.2022
OK, thanks. I guess I will need to search for glitched frames manually by sight.
RE: Find And (Optionally) Fix Bad Frames - zspeciman - 27.06.2024
Has there been a new tool developed since then?
I have a damaged vhs-c camcoder tape where in certain frames from the past are are relocated to the future. Hard to describe but its something like this
for example frame 20 22 23 24 25 26 27 28 21 30 31 32 33. The glitch is visible when 21 appears out of sequence, where it is missing not as much if at all.
It is easy to find the bad frame when it is out of sequence, such as between 28 and 30, but harder to find where it belonged originally.
What would be an approach to fix this? Manually delete the bad frames?
RE: Find And (Optionally) Fix Bad Frames - Selur - 27.06.2024
Assuming the file is sync (and not vfr) I would probably replace the bad frames with interpolated frames of the neighboring frames.
In the Avisynth script linked in the first post this is done through mvtools, personally I would rather use svp or rife.
Cu Selur
RE: Find And (Optionally) Fix Bad Frames - zspeciman - 27.06.2024
Thank you for the suggestion, ReplaceSingle looks like a solution. Is there a limit how many frames I can enter in that field? It might be as high as 500 frames.
RE: Find And (Optionally) Fix Bad Frames - Selur - 27.06.2024
iirc. the limit for a QString is 32767 characters, so 500 frames should be no problem, if you got more of those clips spending time to come up with a way to identify those frames automatically might not be a bad idea.
RE: Find And (Optionally) Fix Bad Frames - zspeciman - 27.06.2024
auto detection would be great. Is there such a tool, that lists those frames?
I've attached a sample of the video at its worst part
Another question. Is it better to replace the frames before or after deinterlacing or it makes no difference?