-
Aligning two SEPARATE film scans of the SAME footage?
Greetings,
I have headache brewing on my screen: I need to align two clips that contain the same footage via different film scans. The reason for doing this is to use an older, poorer scan to restore some crushed/clipped shadow and highlight details done by the last person who messed with the newer scan (I do not have access to the source files nor the film footage else this situation wouldn’t exist!). I’m doing this by using the extract effect to leave only the mid range of the higher quality clip.
Since they are scans of the same footage the motion lines up almost perfectly (they both have been stabilized). The tripping point for me is the geometry of the scans; both scans differ slightly in their geometry and the differences aren’t linear, which means I cannot just adjust the corners. I’ve tried to use meshwarp to align various section of the image however it ends up looking too muddled, liquefy is too imprecise, and the warp functions aren’t too helpful as the clips need to be overlapped for the duration, instead of changing from one to the other.
Any advice on how to accomplish this obtuse task?
-
3 Replies
-
just use displacement map set as luminescence/alpha of the other layer that is Gaussian blurred. It is important that both scans are the same size.
-
Both scans are HD however the masking and geometry is different; one scan shows more information on the top and bottom.
I was thinking a displacement map would be a good idea but I’m unsure how I’d generate it for this purpose; given that I’m unsure of how exactly the geometry in the two scans is different, how can I know what to put in to the displacement map? Would it just be a case of trial and error?
-
masking? like black borders? they need to be close to work. I don’t know exactly what you got, but the displacement map
could be created by hand by visually comparing parallax. white is closest and black is furthest, to keying, masking, or
even a precomposed difference matte.I still don’t understand how they scanned it so wrong, even if it were another angle, AE’s Corner Pin effect should have worked no problem. also, the Lens Compensation effect should have fixed the focal length(if there was any). Until I see a screenshot, this one’s got me stumped.
Reply to this Discussion! Login or Sign Up