P3D Re: Cardboarding

[Larry Berlin <lberlin@xxxxxxxxx>]

>Date: Sat, 24 Jan 1998 
>From: Michael Kersenbrock  writes:
>...............
>
>Note that McKay did *not* say that it *became* ortho at any point, in 
>fact later on down in the quote it says something explicitely to that
>effect.  Even above he talks about approaching an approximation and then
>goes on to explain the details of what he means by approximation.
>

****  Remember too, that McKay did not have the benefit of digital image
processing. Using today's tools, though some of them are not generally
available to the public yet, one could in fact create an accurate Ortho
image using long lenses, a wider stereo base and a collection of images
taken from vantage points in a circle of a diameter the same as the chosen
stereo base. With digital processes it would be possible to exactly
reconfigure the actual vantage point of an ortho image taken at a
proportionately closer distance, and significantly increase the total
resolution at the same time.

I have not done this process since I don't have access to some of the
advanced processing software that is used in certain other industries for
other purposes. However an ortho result is possible by this method because
it would record all the critical elements, both vertical and horizontal,
needed for reconstructing the Ortho requirements.

When the stereo base is extended all by itself, it provides horizontal
parallax equivalent to a closer vantage point, useful to a limited degree.
It DOES NOT provide the vertical component of the closer vantage point,
contributing to the distortions observed with base-only exaggeration. That's
what pictures from this circle would accomplish, a combination of horizontal
and vertical components. The simplest form of this circle would be
represented by four cameras. Two for the wide stereo base and two for the
top to bottom vertical component.

Since most discussions on this topic do not involve the vertical component
or a solution to the vertical problem, there is a perceived limit due to the
limited method (base only). If one can collect all the needed information
and process it accordingly, ortho results are possible from many
circumstances, and the limit can be seen for what it really is, conditional
and relative instead of an insurmountable concrete wall. 

Larry Berlin

Email: lberlin@xxxxxxxxx
http://www.sonic.net/~lberlin/
http://3dzine.simplenet.com/


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