[photo-3d] Stereoscopic Display

["Steve Berezin" <sbere@xxxxxxxxxxxxx>]

I think my earlier post may have been bounced because of HTML code.  This is
a second try.

This is a continuation of the coverage of the conference in San Jose that I
am working on for the NSA publication stereo world. These are rough
notes…the final article should be a little more refined.

The second day of the conference was mainly concerned with Autostereoscopic
displays (3D without glasses). Most of the early papers concentrated on
improvement of head tracking capabilities in existing autostereoscopic
applications. The team at NYU, the Korean Institute of Science and
Technology, and the Electronic Visualization Lab, at the University of
Illinois at Chicago all went over refinements in this area. Sharp Labs in
the UK went over improvements in their 2D/3D switchable parallax barrier
that has the advantage of disappearing when not in use.

Several Korean University research teams went about the challenge using
different approaches. Kyung-Wan University used a reflective vibrating
scanner array to isolate images and then played them back on a system using
8 separate hard drives. Seoul National University recorded and played back
an integrated picture by first shooting through a lenticular system and then
playing it back through a similar system.

Some of the more novel concepts detailed were volumetric display concepts.
Actuality-Systems is working on a system where they are attempting to create
interactive three-dimensional imagery by projecting a series of 2D images on
a rotating screen. The device, which looks like a spinning disc in a glass
globe, will rotate and images projected on it at 4000 frames per second.
They felt it would be good for applications that do not require
photo-realism.

John Rupkalvis (SCSC and Photo 3D member) gave an enjoyable presentation
entitled "Human Vision Considerations in Stereoscopic Displays". Some of the
concepts he touched on involved proper room lighting for stereoscopic
presentations, flat vs. stadium seating (stadium seating like for Imax tends
to be shot from a higher angle to account for the audience position), and
editing styles. MTV type quick edits are not recommended for stereo video
because it takes people more time to process a stereo image than a flat
image.

John Roberts (NIST, also a Photo 3D member) detailed other issues important
in stereo display usability. He recommended that there be rules for cross
platform stereo applications and that perhaps even stereo theatres should
adapt a certification system to insure consistent quality.

Later a Panel discussion on Standards in Stereoscopic Imaging was given.
Michael Weissman of Karl Storz Imaging (stereo endoscopy) chaired, and
Andrew Woods (Photo 3D Member) and Dave Swift (VREX) gave presentations.
Discussed were standards such as the VESA (Video Electronic Standard
Association) standard for shutter glass connectors and standards for the
video signal for those glasses.

Wendesday will feature actual demonstrations of these products and should
give a clearer picture of the actual performance of the devices.