🚀 go-pugleaf

In article <4352.2816b547@iccgcc.decnet.ab.com> fleischer@iccgcc.decnet.ab.com writes:
>I saw the most *amazing* "3D" black and white photocopy of an 
>advertisement by PENTICA.  It looked like a random patern of squiggles in
>a box with 2 large black dots above.  When you shift the focus of your eyes
>to make the 2 dots overlap, this flat XEROX'd picture turns 3D.  It's
>like you're looking through a window to a picture about 1 foot below the
>paper with letters and numbers at various heights above the picture.
>
>Does anyone know how this was done?  Is there PC software that will let 
>you build these and print them on a laser printer?  Is there a collection of
>these?  Who did the theory and development?  Does it require more resolution
>than exists on a VGA display?  IS there a dynamic/moving version?

This is a fairly standard procedure called random-dot stereograms,
described in texts on visual information processing. I read about it
in David Marr's book on vision. The idea behind stereograms is that
the pixels in one of the pictures are shifted by a small amount, and
that produces the effect of being in stereo when each image is viewed
by a different eye. The Marr book (and others) has a number of
stereograms in it. I seem to recall that the original work was not
done by Marr, but I can't remember who did it. I have not heard of a
dynamic/moving version, and would theorize that it would be difficult
because you have to keep your focus on the dots, but the movement
would tend to take your attention away, so you would loose the
'focus'. 

What is new in the PENTICA thing is the two dots to help you focus. I
hadn't heard about that before. You normally view the stereograms
through special glasses. The prof I took the vision course from (Kent
Stevens, a student of Marr's) could get to view the stereograms
without the help of spectacles by just crossing his eyes, which is
what the PENTICA approach achieves. I tried and couldn't. The PENTICA
approach would make this much easier.

Hope this helps,

kartik

-- 
Anant Kartik Mithal                                     akm@cs.uoregon.edu
Research Assistant, 					(503)346-4408 (msgs)
Department of Computer Science,                         (503)346-3989 (direct)
University of Oregon, Eugene, OR 97403-1202