P3D Re: Cardboarding and telephoto lenses

[Larry Berlin <lberlin@xxxxxxxxx>]

>Date: Sat, 24 Jan 1998
>From: "Dr. George A. Themelis" writes:
>................
>I love McKay and this book too, but this statement is questionable:
>
>>If we maintain a proportional increase between base and magnification, 
>>the type of reproduction closely approximates that of orthostereo. 
>
>The question is "how closely"?

****  To answer that you have to ask how close do you want to approximate
it? In theory, one could be extremely precise about the arrangement and
create an image that is indistinguishable from Ortho under some
circumstances. In practice one might be less precise, so use the term
*closely approximate*. Even an Orhto image is itself a close approximation
to a collective average experience. It comes down to which method of *close*
do you want or is most appropriate to the circumstances.

>
>In this situation there is a distortion that can be noticeable:  The
>objects further away appear larger.  
>For example, if you take a portrait
>with 135 mm lenses and view it with the normal 45-50 mm viewing lenses
>then the ears will appear to be disproportionate large, compared to the 
>nose (or visa versa).  This makes sense if you think about it.

****  You don't really mean *larger*, more like proportionately less
diminished by distance than a shorter focal length camera lens provides.
Each set of lenses provides it's own proportionate scale to what is observable.

>
>McKay was a practical man and he believed that if you don't see it, it
>is not there.  I was not seeing this distortion untily it was pointed 
>to me by Dr. Milligan and then I saw it and found it disturbing.

*****  Then one has to take into account the wide variety of experiences
common to different individuals. Beginners sometimes don't see some of the
smaller parallax factors through sheer lack of familiarity with the stereo
image process. Does that mean it doesn't exist? No. We can train our
eyes/mind  to better observe details thus gaining the ability to see that
which we formerly thought was absent or simply didn't notice. Once we see an
effect it becomes either something we avoid, or can use to advantage. Stereo
imaging is richer in this sort of working range than 2D and is part of the
attractiveness of stereo for me.

Any effect that you don't like can be fine tuned in other ways. First one
has to notice the effect and decide they don't like it. Understanding where
it comes from is instrumental to finding a fix. In the use of stereobase to
compensate for cardboarding, one might not go so far as to create some other
effect that is undesirable. In the telephoto example with maybe the subject
of a Lion, perhaps it's safer to use telephoto and live with a distance
factor that doesn't diminsh as rapidly as normal and thereby prevent your
own life signs from diminishing! There is still the option to use the
telephoto lion image composited into a more *normal* scene, thereby getting
the best of two methods. 

Larry Berlin

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


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