Originally posted by: Jaiman Tuckuh
The P&S & Widescreen, taken together, would contain information that falls "between" the pixels scanned for either.
The P&S & Widescreen, taken together, would contain information that falls "between" the pixels scanned for either.
Yes, matching the two images is difficult. It has to take film wobble, lost details and artifacts into account.
When the match is done, the letterbox image would be upscaled using the fullscreen image as 'hint', color correction could be done automatically and the fullscreen "window" could leave high-res details behind when it scans left and right. These extra functions would contribute to mask the borders of the fullscreen image, but they all depend on the initial match having been done right.
If the match was done badly, then there would be visible artifacts, especially when the camera (or even just the fullscreen image) is panning ... and those are only the problems that I have thought of so far.
It is possible that this automatic approach would be a total waste of time, and not produce a better result than manual compositing. I don't know much about manual compositing. That is why I ask.