Re: weighting depth clues

[T3D Jim Roberts <xjim@xxxxxxx>]

John Bercovitz wonders:

>I wonder how depth clues are weighted.  For my talk at NSA

I don't have any answers, but perhaps this perspective can
elicit some smart folks to do some thinking . . .

A fascinating experiment with depth cues can be done with
Ken Dunkley's pinhole arrangement. (The 3D effect from 2D
shots, if you remember.)   I tried his setup and was not
not too impressed - I achieved the effect, or so I thought
but was just barely whelmed.  After I sent it back to
him, though, I got curious and I tried again, this time with
a similar device I made with smaller pinholes in aluminum foil.

The results can be absolutely striking!  I found that choice
of subject makes all the difference in the world, and that
the chromatic effects were not nearly as important in the
effect as I thought.  It seems to me that the various effects
of the pinhole arrangement (chromatic effect, sharpness of
focus, isolation, and other stuff I haven't thought of) trigger
the processing of other depth cues to much stronger than
normal levels.  I checked out a nice mountain lake landscape
slide in a Sawyer's Bi-Lens with pinholes over the eyepieces
and got fantastic realism, including what I am calling 'perceived
stereopsis.'  Since there was little in the way of true stereo
difference, and that little was induced by the pinholes, it was
amazing that the depth was so strong.

I, too, am curious about depth clues, and am much more interested
than I was before in Ken Dunkley's device and how it works.

Jim Roberts


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