P3D Re: Alternatives to "insufficient depth"

[roberts@xxxxxxxxxxxxxxxxx (John W Roberts)]

>Date: Fri, 4 Sep 1998 02:50:49 -0600
>From: "Greg Wageman" <gjw@xxxxxxxxxx>
>Subject: P3D Re: Alternatives to "insufficient depth"

>From: Gabriel Jacob <jacob@xxxxxxxxxxxxxx>
>>I agree BUT the problem is.. EVEN a properly exposed picture that
>>is sharp doesn't capture reality. This is because of the limitations
>>of film. Alot of distant scenic shots that supposedly look like
>>backdrops in real life don't look like backdrops to me except on film!
>>Now it can be claimed this is because of my limitations as an amateur
>>photographer BUT it seems everyone claims those distant mountains
>>(or whatever) don't "look" 3-D in real life, which I claim otherwise
>>(unless of course it's hazy).

>Motion parallax?  Power of suggestion, perhaps?  (You know the scene is
>real because you are "there", therefore it "must" have depth, and so you
>"see" depth?)
>Or, maybe without the MTF of a pair of triplet lenses, film chips and
>viewer lenses in the way, there is sufficient optical resolution to
>produce some stereopsis when you're there "live"?

There are many depth cues (including visual depth cues) other than stereopsis.
For example, differing amounts of scattering / color shift can provide an
indication of relative distances of distant objects. Another example might be
the relative angles of shadows.

Some of these cues may be very subtle, and perhaps not consciously noticed
by the observer without a deliberate effort, but they contribute to the brain's
effort to make a 3D model of the world around it.

Generally there is some loss of information in the photographic process.
It is entirely possible that some of these subtle depth cues are not
represented as accurately in a 2D photograph as they are in real life,
and thus a human observer may feel that the 2D photograph doesn't convey
a sense of depth as well as actually seeing the distant scene, without
necessarily being able to specifically state what the difference is.

There may be cases in which two separate photos are slightly better than one at
conveying subtle non-stereopsis depth cues, as the brain averages together
the characteristics of the two photos.

Another possibility to consider: most of us have seen a sufficient number
of 3D photographs to become accustomed to certain common characteristics,
for example the fact that the grain pattern is different in the two
images. It is possible that there could be some degree of conditioning -
if we see two separate photos of the same distant scene, and from subtle
cues are able to detect that it is two separate photos, there may be some
expectation of seeing depth, which could relate to the power of suggestion
as you mentioned. [Analogous situation: have you ever noticed in a movie
that there's sometimes an attempt to "fake" a motionless scene by stringing
together many copies of the same still image? Somehow there are subtle
differences that make it pretty easy to spot.]

John R

ps: Yet another possibility: as most amateur astronomers know,
atmospheric distortion can affect "seeing", generally to a greater extent
as the diameter of the light collection area increases, due to the nature
of atmospheric turbulence, convection patterns, etc. (Under some conditions,
a small aperture telescope will get a better image than a large aperture
telescope.) In addition to motion cues, for stereo photographs of distant 
objects, relative distortions of the two images may provide a cue on the 
separation of the two camera lenses. So a stereo photo of a distant 
mountain, taken with a Realist, may actually look slightly different from 
two photos taken with less separation than the Realist. (I believe 
astronomers speak of "convection cells" in the atmosphere, with sizes 
(at least at night, and looking up) sometimes in the range of inches.)


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