P3D Re: Real world test? - Caution

[roberts@xxxxxxxxxxxxxxxxx (John W Roberts)]

>Date: Tue, 16 Dec 1997 11:49:00 -0700
>From: fj834@xxxxxxxxxxxxxxxxxxxxx (Dr. George A. Themelis)
>Subject: P3D Re: Real world test? - Caution

>>Of course, caution should
>>be exercised to minimize unwanted cues such as size, shadows, etc.

>I get the feeling that any effort to minimize these unwanted cues
>(including motion parallax!) would like very much like an artificial
>contraption as if looking through a viewer.  (The beauty of artificial
>testing is that they can be made without any extraneous cue.)

Artificial testing is fine if I'm completely unaware that it's artificial
testing.

>Which brings the question:  Why would looking at an artificial scene with
>only built-in parallax would be any different than looking at a real scene
>with only stereoscopic depth information?

The difference (which *might* be significant) is that my brain knows that
it's looking at an artificial scene through a viewer and not a real scene,
so the interpretation process might be different. (Analogy: if I'm conducting
an ESP test, and I tell you that I'm going to try to mentally project an
image of the Mona Lisa, so see if you can pick it up, I'm not going to get
much useful information from that test. :-)

Here are some examples of things I believe I interpret differently depending
on whether I'm looking at the real world or at an artificial view through
a viewer:
 - Most stereo pairs are mounted to the stereo window rather than to infinity.
   If I look at a stereo photo and the moon appears to be 100 feet away,
   I don't think of it as tiny and close - I decouple the convergence cue
   to the point where I'm willing to believe that it's an image of the real
   moon, at infinity. If I were to look up in the sky and see the moon as
   being 100 feet away, I would be very worried.
 - If I look at an extreme close-up in real life, I see a tremendous amount
   of keystone distortion - but I expect it, and I'm not surprised. If I
   see that much distortion in a close-up stereo photograph, I consider it
   a terrible photograph.
 - When I went to my first stereo club meeting, back in November, I watched
   a few projected slides, then noticed brief intervals when the depth
   appeared to "wobble" in and out. I was somewhat worried: "Here I just
   looked at a few projected slides, and it's blown out the stabilizer in
   my depth perception!" After a few more slides, I figured out that it was
   just the projection crew adjusting the alignment of the slides. When
   I went to the December meeting, some of the slides were similarly
   adjusted while being projected, but I no longer perceived it as the
   depth moving in and out - I was able to shut off that interpretation.
   But I would definitely notice a similar effect in the real world.

When we view stereo photos, we are confronted with a number of conflicting
visual cues - we learn to discount some of those cues, and concentrate on
the ones that provide enjoyable stereo viewing. So when we know we're looking
at a stereo photo or similar image, we're willing to make allowances in
our interpretation of the image. I believe that this process is at least
partially under conscious control. So I think it is at least possible
that in the case of an artificial depth perception test, prior experience
with viewing stereo photos might cause a person to perceive depth differently
when they know they're looking at an artificial scene than when they think
they're looking at the real world.

John R


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

End of PHOTO-3D Digest 2464
***************************