P3D Re: Cardboarding

[michaelk@xxxxxxxxxxxxxxxxxx (Michael Kersenbrock)]

> >>I think you are very right that a telephoto lens portrait
> >>could cause the same effect with things at the right distances
> >>and w/o hyper "quasi-compensation".
> >
> >Now you're throwing in another variable, namely stereobase.  Attempting to
> >compensate for cardboarding by extending the stereobase will fail, as was
> >discussed on P3D late last year (I think - certainly within the last six
> >months).
> 
> *****  This statement is without foundation. Of course stereobase is a
> significant part of the situation since it definitely affects the appearance
> of cardboarding by affecting *local parallax* for objects themselves. There

Yes, despite what may have been mentioned, a telescopically taken photo
of a person can be done with the person showing depth (not a cardboard
person) according to Herbert C. McKay.  I quote from his book (below)
explaining things (I love this book).  Not that one can't also mess
things up by doing it wrong.... :-)


............... START OF QUOTE .....................................

        Three Dimentional Photography, principals of stereoscopy
                    (revised) by Herbert C. McKay
                          pp 268, 269

         It must be remembered that the conventional telescope does not, as
so commonly stated, bring objects nearer. It simply increases the
perspective values, the angular value of the field of vision filled by the
object.  Therefore we shall make use of long-focus lenses only when the
object is too small to be satisfactorily recorded from the nearest
accessible camera position.

        If we maintain a proportional increase between base and
magnification, the type of reproduction closely approximates that of
orthostereo. We reproduce the objects at their full natural telescopic size,
and with that distance and degree of relief which is fully natural for that
size. Thus if the lenses are doubled in focal length and the base is also
doubled, objects are shown in the size and relief normal to a distance which
is just half that of their true distance.

        If the object is small and far away, the proportionate increase just
described, that is, the use of parastereoscopic technique, will have much
the same effect as that of using a telephoto lens upon a planar
camera-except that we have stereo relief. That is to say, we are shown the
object as it would appear from a point of view midway between the real
camera position and the object. The same distortions appear as in the planar
telescopic record, but like them, these distortions are extremely difficult
to perceive even when they are pointed out in detail. Any distortion as
subtle as this may be completely disregarded without detriment to the
quality of the stereogram.

        Therefore, in making stereograms of objects at a great distance, we
first determine the base necessary to introduce the desired degree of
relief. The next step is to select the lens which will cause the image to
occupy the desired space upon the film, and if it is desirable that strictly
normal proportions be preserved, the first two steps are compromised so that
the two increases are of the same degree.
..
..
..

        It will be seen that the choice of the base in hyperstereo, or the
choice of the lens-base combination in parastereoscopy, is determined, not
by the traditions of old time stereo, not by some complex mathematical
computation, but solely by the exigencies of existing circumstances.

  Increase the stereo base (hyperstereo) to bring the object nearer.
  Increase the focal length of the lenses to make the object larger.
  The two functions are not at all identical.

        Let it be emphasized that there is no valid reason why the
stereographer should not vary both base and focal length of his lenses if by
so doing be is enabled to obtain a result which is visually superior to that
which would have resulted otherwise.  The freedom of choice extends beyond
that limit. If the stereographer wishes to make a stereo caricature through
the use of exaggerated relief, that is perfectly good stereography. In fact,
if well done it constitutes expert stereography, because understood theory
has been used to exert a wanted control.

............... END OF QUOTE .....................................



> By the example I've used with digital images of small scale, I could propose
> that cardboarding is better defined by the simple rule as stated by Gregory
> Wageman. His rule of thumb works under any of the possible conditions:
> 
> >My working definition of "cardboarding" for some time now has been:
> >Lack of stereopsis *within* an object, while stereopsis is present *between*
> >objects.

I like that definition much better than mine (which it now is equivalent
to :-).


Mike K.


P.S. - I see things like cardboarding in 3D animation being one due to the
       "lack of resolution" reason.  The animation isn't a photo of cardboard
       characters (in my mind) but a low-resolution image of the 3-D characters
       that exist in the artist's mind.  Had the artist implemented a
       higher resolution image ("drawn photograph") it *could* have looked
       non-cardboardy (say, done by a Pixar artist). Yup, just lack of resolution... :-)






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

End of PHOTO-3D Digest 2539
***************************