Re: Stereo Window and Convergence Examples

[P3D Dr. George A. Themelis <fj834@xxxxxxxxxxxxxxxxxxxxx>]

Gregg Podnar writes:

>Criticisms of the imagery not being realistic and differences in 
>perception of a real scene are certainly valid.  If most of us can see 
>the different effects in these overly-simplified images, then it is not
>hard to believe the effect generalizes to more real-world images.

I think the danger lies in the reverse.  If some of us do not see the
different effects in the ASCII diagrams, does that mean that the effects do
not exist when real-world images are examined?

I only see changes in the size of the stereo window when it is examined in
connection with real-life objects.  I am talking about what Larry called
"perceptual window".  These effects cannot be seen in ASCII drawings.  The
fact that they are not seen in ASCII drawings does not mean that they are
not perceived in more realistic images.

Here is a challenge for Gregg:  Can you make an ASCII drawing of a
hyperstereo?  Will this generate the impression of looking at a model of
the original scene?  This is not possible.  Because to get this perception
you need a REAL scene.  

Perception is a complicated thing, not easily explained with simple
diagrams or geometry.  If you reverse an ASCII drawing, the depth is
reversed right away.  Yet many people are unable to perceive this reversal
of depth when shown photographs of real-world objects (even simple ones).

I am not saying that these ASCII drawings are not useful in demonstrating
some basic concepts.  What I am saying is that there is a danger when
conclusions from these drawings are drawn for complicated human perception
concepts like the "stereo window".

-- George Themelis 


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