Quote:
Originally Posted by slavanap
Mug Funky,
And what can you say about creating 3-D film from 2-D or stereopair form single image?
Don't users want to do that on their PCs?
|
They can, if they're willing to manually select areas and assign a depth (or gradient of depths) in every single frame. I can't see a plugin ever doing that. There's just not enough depth information in a video for a computer to do it.
I had an idea for this where I thought maybe the algorithm could look at the relative bluriness of neighboring pixels to get an idea of that pixel's depth (because the optics of a camera progressively blur objects further away and closer to the focal point). However, after reading several technical papers, I've come to the conclusion that it's just not enough. (Truthfully, I didn't understand most of the math, but you don't have to in order to get the point that more is needed to determine depth with any kind of accuracy and precision.)