T3D Re: S3D 8x12 Brandi phantogram + other anaglyphs

[boris@xxxxxxxxxxxx (Boris Starosta)]

>From: John Bercovitz <bercov@xxxxxxxxxxx>
...
>What is a phantogram?  Is it some variation on a stereogram?  Sorry

I get tech-3d in digest mode, so some of this may be redundant by now.
Last post I saw from John, he figured out that there are no "viewer optics"
for phantograms.

I am collaborating with Paul Talbot, and despite my having successfully
created some phantograms (anaglyphic stereoscopic anamorphs), I am not so
much the mathematician that I can predict the viewing tolerance for his
project (or even for my own, apparently).  The question is, with respect to
taking and or viewing phantograms, what variables are most sensitive, most
likely to contribute to apparent distortions.  In other words, given that a
phantogram stereo pair is shot with only one print size ultimately possible
(for undistorted viewing), when you change that print size, what variable
contributes the most to the  apparent distortions that are introduced?  Is
this making any sense.

Since John appears to have been out of the Phantogram loop the last two
months or so, I'll repost how I did the work.

********

The difficult part in doing the phantogram was the distortion.  It has to
be a perfect perspective distortion, and cannot be done just using a 2-d
perspective tool, such as the one in photoshop.

I shot the model with my 35mm twin rig, in a very standard fashion.  I
scanned the slides (drum scanner) to a high resolution, and took the scans
into "3-d" software - Bryce.  There I used the scans as "gels" within light
sources that projected the image back down onto the "floor." This projected
image I rendered as the print. The geometry was very important, and I had
taken careful notes and measurements during the session.

In essence, I used the computer to reverse the flow of light, from the film
plane, back out through the lenses, past the 3-d image (the subject), and
onto the print.  The print thus became the "shadow" of my subject, with the
cameras functioning as lights.

I had to correct for the falloff in brightness in the distorted image that
occurred on the rendered print, because far away from the light, the image
was fainter.  That correction was easy, and I shall leave the solution as
an exercise for the student.

This description is obviously a bit simplified.  There were a number of
technical details that needed to be solved, with respect specifically to
the use of Bryce, removal of keystone distortion, registration of the two
viewpoints, color control at the print stage (my thanks to Ray Zone for
invaluable advice!) etc.  Lots of small fires to put out over the span of a
week.

Anyway, I got the print two days ago (Dec. 2, 1999), and it turned out
GREAT.  The illusion is even better than I imagined it would be.  It is a
"windowless" stereoscopic experience, about as close to virtual reality as
I've seen (in some ways much better than looking at a slide).  It really
makes for an ultimate anaglyph.

Boris





"The blessed in Heaven will often walk to the battlements,
and look down, and delight in the justice of God being
properly carried out in Hell."

- Unidentified spoken sample from the recording
    "The Serpentine Arcade" by Richard Bone

Boris Starosta            boris@xxxxxxxxxxxx
                          http://www.starosta.com
usa 804 979 3930          http://www.starosta.com/3dshowcase



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

End of TECH-3D Digest 442
*************************