Re:Suitability of telescope oculars for stereoscopes

[T3D John W Roberts <roberts@xxxxxxxxxxxxxxxxx>]

[PJ Homer]
>>I have thought along similar lines myself to what extent does the curved
>>surface of the retina enable the eyes to get away with toing-in without
>>because it curves around to meet the outer rays unlike the flat surface of
>>a film. Or is it the brain again that corrects the distortion

The human visual system does have keystone distortion. If you hold an object
with sharp boundaries (i.e. not a cotton ball or puff of smoke) very close
to your face (and can focus on it), you'll notice that you can't really see
all of it all at once as a single image - there's always something a little
funny about the edges. It seems that the brain is able to discard some of
this extraneous information when interpreting a scene.

[Larry Berlin]
>*******  The mind processes it easily because no matter how distorted the
>actual sensory input, it is identical to all received input from the
>beginnings of sight. We are literally wired to perceive what comes in
>according to our knowledge of the space around us.

Agree.

>That's why I think electronic digital imaging would work better than film.
>The accumulated data which has known distortions, can be reprocessed to
>correspond to the proportions of the image the way we are used to seeing
>things. Essentially how the mind interprets the eyes. What I wonder, is
>whether there is some inherent optical property of such an arrangement that
>would make it's reception inherently more immune to directional placement
>distortions than exists with planar systems? We do in fact converge our eyes
>on objects we see. We know that flat field cameras must be maintained
>parallel to provide undistorted stereo correlation. Perhaps the reason is
>this basic geometrical difference.

>The concave surface would result in the various incident light angles
>behaving differently than on a plane. The angle between incoming rays and a
>perpendicular to the sensory surface with planar systems is constantly
>increasing in one direction at a somewhat steady rate. That same angle under
>some conditions, changes in the opposite direction and at an ever increasing
>rate for a concave surface. (a description matching yours above in different
>terms) I'm not sure how to relate this to distortion sensitivity in a scene
>when the optic sysytem is rotated slightly as opposed to translated. Would
>there be less tendency for the rotation to distort the reception? Or is it
>again just the mind's familiarity with the process that compensates?

An intriguing idea! I know there's some keystoning, but the shape of the eye
may affect the extent of the distortion.

The lower spatial resolution of the peripheral vision probably helps too -
things are probably fairly "flat" in the region of the fovea.

John R


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