Re: Converging fields of view

[P3D Gregory J. Wageman <gjw@xxxxxxxxxx>]

Larry Berlin writes:

>It is true that the fields overlap but it is NOT true that they converge.
>The fields spread out fan like and OVERLAP as they proceed in a generally
>parallel arrangement. The center of each field is OFFSET from the other and
>these centers are supposed to be parallel for the least amount of
>distortion. It is the viewer's eyes that converge on one pair of details or
>another within the overlapping areas of these fields of view. If there were
>no offset and overlapping areas you would not have stereo information.

Here we go again. :-)

The problem is we keep talking about different portions of the camera
*system*, ignoring others to make our points.

We all agree on the following points, I think.

1) The lens axes are parallel.

2) The film planes are co-planar and perpendicular to the lens axes.

3) The the centers of the apertures are offset outward from the centers
   of their corresponding lenses to set the window.

This next is the problem point.

The fields of view of the *captured images* do, in fact, converge.
Not the fields of view of the *lenses*, which we all agree are parallel,
neither converging nor diverging with respect to each other.

Draw lines from the corners of the images through the centers of
their lens out into the scene.  Push a plane perpendicular to the lens
axes down these lines, to form a rectangle.  We see that the boxes
converge as we move away from the lenses until they are co-incident at
the window distance, and then diverge beyond the window so that the right
image's box contains more of the left side of the scene, and left image's
box contains more of the right side of the scene once you're beyond the
window.

The disagreements arose because one of us always concentrated on one
of these points to the virtual exclusion of the other.  I believe the
above accurately describes the entire camera system.

	-Greg


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