Re: [photo-3d] Re: Re: do you believe it

[John W Roberts <roberts@xxxxxxxxxxxxxxxxx>]

From: Jim Crowell <jcrowell@xxxxxxxxxxxxxxxxxxxxxxxxxxxxx>
Date: Mon, 16 Oct 2000 11:13:42 -0400

[Bob Maxey]
> I went to the site, still not convinced. What you are seeing is some
> visual trick, not stereoscopic vision. Again, Stereoscopic images require
> a left and right eye view. 

More formally, stereoscopic images require left and right eye information.
There are several ways this information can be encoded into the images.
It might not be obvious that SL3D, for example, uses separate "views".
I would be interested in knowing whether there is a term to specify
stereoscopic images designed in an attempt to recreate a generic and coherent
3-dimensional structure, in contrast to those that may cause the viewer
to perceive depth, but with no coherent structure to the depth (for example,
chromostereoptic viewing of a random 2-D color photo).

[Jim C]
The 3-DVG has been discussed to death several times over the past few years.
It does introduce binocular disparities between parts of the images with
different colors (among other effects), but they generally aren't consistent
with the 3-D structure of the scene in the photograph.  Only if e.g. all the
warm-colored objects in the scene are nearer or farther away than all the
cool-colored objects.  It's called chromostereopsis, results from the pupil
(or in this case, the pinhole) being offset from the center of the lens of
the eye.

I don't remember whether it's been discussed on this list, but I've seen some
very clever paintings based on chromostereopsis. I think they're produced
by a particular company, and always depict underwater scenes. The clever
feature is that the dispersion characteristics of water (warmer colors
for nearby objects, cooler colors for distant objects) approximates the
coloring needed for the chromostereoptic depth cues, so the coloring
appears plausible, whether viewed in 2D or in 3D.

John Roberts