T3D Re: Cardboarding

[Dylan The Hippy Wabbit <spacey@xxxxxxxxxxxxx>]

Hi All,

William J. Carter, Ph.D. wrote:

>However, if all I want is to test my thesis; that with dual lens
>systems, and in the absence of other depth cues, the brain only
>recognizes edge boundaries,

Psychologists have known for over two decades that the brain calculates
stereo on a far finer-grained basis than edge detection.  Random Dot
Stereograms (both dual image and the single image "Magic Eye" types) rely
on that fact to work at all.

Of course if you don't allow any other cues to be available then that is
what will be used.  Am I missing something?

>This contrasting separation from the background results in card boarding
>when other depth cue aren't present.

In the real world other cues are always present, but limited resolution
can blur them.  That would cause cardboarding due to a relative lack of
cues within an object.

>I also believe that:
>Single lens systems provide the brain with depth information in the form
>of the various sizes and orientations of the circles of confusion; not by
>triangulation based on the centroids. But, I don't know if that belongs
>in this discussion.

I'm not experienced with these split lens systems, but I'm not convinced
by the explanation on your web site.  I still think that a split lens is
best viewed as a closely coupled pair.  A mathematical model based on
trigonometry may well fail though, due to the assymetrical nature of the
half lenses and the considerable prism effect that will be present.  Some
form of integral of the lens area may be needed, but my calculus is *way*
too rusty to even attempt it.

>Here's the question(s) I'd like to answer:
>1. Will I get card boarding with a subject, in sharp focus, which has no
>discernable surface features, regardless of it's having a z-axis
>dimension? I expect the answer is yes, but it seems this needs to be
>proved.

What would need to be answered first is whether you would get cardboarding
viewing such a scene with the unaided eyes.  If you put it in a box to
disallow movement parallax I think so.

>2. With the above subject, will the brain be able to fuse "out-of-focus"
>information and effect depth from a dual lens, if the stereo pair is more
>shallowly focused? I expect so, dunno really. The fact that there is no
>"center information" in a dual lens pair suggests to me that ...
>something's going to be missing!?

I think you are neglecting the viewing aspect of this.  We see
stereoscopically through paired lenses, and the only "center information"
is interpolated by the brain.  Whatever means you use to take the
photograph that is fixed.

BTW, what subject is the Ph.D in?  To put it another way, if I've been
trying to teach psychology to a psychologist I'd better apologise now.
;-)

Dave Spacey

______________________________________________________________________________

Walk tall, walk straight,
Spit the world right in the eye.

-Kevin Godley and Lol Creme


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