That is a fascinating idea. Potentially the Z-depth could be extracted if
you can algorythmically reliably tell the similar pixels in the two images.
The problem is with a anaglyph image you only have one channel of color to
determine similar pixels which is even less data than you get with other
stereo photogrammetry methods. But it's a fascinating idea.
On Monday, February 11, 2013 1:59:05 PM UTC-7, Howard Messing wrote:
> Anyone know how to take a red/green jpg anaglyph and produce an STL file
> out of it for printing? In particular, an photo of a coin?