View Single Post
Old 20th December 2018, 03:09   #15  |  Link
abolibibelot
Registered User
 
Join Date: Jun 2015
Location: France
Posts: 46
Quote:
Be aware that MvTools interpolation is quite bad at interpolating where there are even small variation in overall brightness,
dont know how one might prefilter for that, using any kind of temporal brightness stabilising on prefilter clip would screw up interpolaton, I think.
You mean, for FrameSurgeon interpolation ?

I found some interesting insight in this 10 years old thread :
https://forum.doom9.org/showthread.php?t=136025
“The 'trick' with [DePanStabilize] is: use a different (cropped) clip for DepanEstimate. What to crop? It depends on the scene... Focus on something that should not move... A house or something. I use tweak(bright=-100,cont=2.0) on this clip too... Somehow this helps, too.
Also, I nearly always set cutoff to 0.5 and trust between 1 and 1.5.
The same values for dxmax and dymax both in DepanEstimate() and in DepanStablize helps... (30 is a good value)
Sometimes I get better stabilizing with Depan then with Deshaker.. It realy depends on the source.”

If I understand this correctly, the idea is to enhance the contrast on the source clip for DePanEstimate, then run DePanStabilize on the original video.
The script proposed by “g-force” is quite effective at suppressing small vibrations, but alas it's confused by those damn blurry frames : if frame n is blurry, meaning that the edges are doubled, it will align frame n+1 with either the upper edge or the lower edge of n, instead of ignoring it and aligning it with n-1 instead.
This statement reflects my experience thus far :
“I've never been able to dial in the settings for DePanStabilize to be of any use. Doesn't fix quick jitter, and tracks slower pans too well.”
I haven't tried the cropping trick, I'm not sure where I should crop to improve the efficiency, in this particular case, most of the shots are, or should be static, so I don't see how cropping could improve the efficiency, while a few shots have motion everywhere, so cropping around a particular spot on a given frame would be pointless since 5 frames later the picture's content is completely different.
The last suggestion works only with method=0, which in my tests gives very poor results.

And actually, I get those artifacts with the combination of DePanStabilize + FrameSurgeon, not DePanStabilize alone, and not FrameSurgeon alone, it's really weird. Normally, stabilization should improve interpolation, right ?
Here are details cut at exactly the same spot of three untouched consecutive frames :
/ /
Then the same details cut from the same frames stabilized with DePanStabilize (this is one example where it fails to align the neat frames adjacent to a blurry frame) :
/ /
Then here is a detail from the middle frame interpolated by FrameSurgeon after DePanStabilize, with the kind of artifact I'm talking about :

And finally a detail from the middle frame interpolated by FrameSurgeon without the DePanStabilize processing, which is as flawless as can be :

The problem is that in other cases, interpolation by FrameSurgeon without prior stabilization produces an ugly result... So I really don't know what to do... é_è

EDIT : One possibility would be to run the interpolation script without any stabilization, and stabilize in Magix Video Deluxe (which proved to be efficient and reliable as I said earlier), but then I would have to check every interpolated frame again, and do a second interpolation pass for the bad and the ugly ones, at the encoding stage. For this to be practical, I would need to be able to review only the interpolated frames, rather than parsing the whole footage, which is excruciatingly tedious. So is there a convenient way to do that in Avisynth, with my lists of FrameSurgeon commands ?

Last edited by abolibibelot; 20th December 2018 at 03:22.
abolibibelot is offline   Reply With Quote