T3D orthoscopic afocal perspectives

[John Toeppen <toeppen@xxxxxxxx>]

Two lenses separated by the sum of their focal lengths will produce an
afocal image relay.  Keystone distortion in stereo will still exist
unless you translate the camera without changing its pointing(slide on a
bar).  Ariel photographic mapping methods scale.  While this may be
great for macrophotogrammetric mapping, it may not be good stereo.

Two 5" projection lenses could be placed with their screen sides
pointing towards each other separated by about 10".  The film plane is
placed 10" behind one of the lenses producing a working distance of
10".  This would produce an image size of 1" for a 1" object.  If the
object were moved to 13" away it would be 1" if you refocused to 8" from
the lens.  Anyway, there is no perspective foreshortening with such a
system.

A more practical approach for a similar effect is to use a long lens
with an extension tube on a slide bar.  This approach avoids the
keystone effect due to convergance and produces a near othoscopic view.
A 35mm twin rig with 135mm fls on extenders could do macro very well.

Distortion due to curvature of field is usually due to not viewing a
display from the correct distance.  A fisheye view photo has correct
perspective if viewed from, say, six inches. For this to work you need a
magnifying lens with a focal length equal to the lens used to take the
photo.  If the angles of the image exposure are preserved in the
playback there is no distortion.  So a slide taken with a 28mm lens
should be viewed with a 28mm lens, or it would looky "fishey" in a 50mm
fl red button viewer.  But that may be good stereo!


John Toeppen
http://members.home.net/toeppen/welcome.htm


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

End of TECH-3D Digest 484
*************************