P3D SIRDS from photos

["H a r o l d B a i z e" <baize@xxxxxxxxxxxxx>]

Michael Henderson wrote:

>Ok, here's a dumb question...  I've been out of the loop in SIRDS
technology
>for a few years now, although I was one of the first people to use
Stare-E-O
>Workshop.  Many of the modern SIRDS today seem to start out as images
>similar to the POVray images I've seen; just stereo pairs cut into strips
of
>left & right images for divergence viewing.  My question is this:  If a
>computer-developed stereo pair can be converted to SIRDS, why can't a
>photograph?

Michael... there are no dumb questions! Here's the scoop. Most SIRDS
programs use dark-light to map depth. If you simply input a b&w
photograph then a person's eyebrows would be behind his/her face!
The computer rendered images use a special rendering technique, where
depth (z-axis) is rendered as a gradient from black to white.
It should be possible to use a stereo pair to calculate depth and
apply a gradient to depth, but I have not seen that yet. It is way
beyond my graphics programming capabilities.

Another reason it hasn't been (or is not) done, is the relatively
LOW resolution of SIRDS. All programs I've found use only up to
256 levels of depth. A few years ago I wanted to use SIRDS to
test a hypothesis about depth perception, but I gave up on the
project when I realized that I couldn't get any existing SIRDS
programs to deliver the necessary resolution, and I can't spare
the time to learn graphics programming to write the program.

Harolddd



------------------------------