Mailinglist Archives:
Infrared
Panorama
Photo-3D
Tech-3D
Sell-3D
MF3D
|
|
Notice |
This mailinglist archive is frozen since May 2001, i.e. it will stay online but will not be updated.
|
|
P3D overlap subjects, not windows
- From: John Toeppen <toeppen@xxxxxxxx>
- Subject: P3D overlap subjects, not windows
- Date: Fri, 3 Mar 2000 00:23:26 -0700
I wrote about the eye's "windows" and Eddie wrote:
"It does if you are looking though a window, and that's the convention
you
force on the viewer anytime you can't fill the field of vision."
The forward field of human vision contains some overlap in the "stereo
zone". This is not our entire right and left fields - peripheral vision
is obviously part of what we see. Our view of the real world does not
have "matching windows"
I notice that the window edges on my right and left chips do not match
in my viewers. And I don't consider window alignment critical in
projection since I align to the image and not the boarders when using my
TDC. If my subject is the distant mountain I will set convergance to
overlap this on the screen. The trees in the foreground are beyond the
edges of the screen - but Halfdome looks great! If my subject is a
flower then I converge that on the screen plane. Yes, it is
uncomfortable to view the out of focus far field beyond the flower, but
boy the flower is great. Yes, I spend some time tweeking to get the
view "perfect" this defined by getting our from behind my projector and
sitting in front of it. Different viewing positions yield different
effects. Only rear projection is "distortion free" (at least you're are
on axis).
Projection requirements do not neccesiate that the frame of the two
images overlap. The objects of primary interest must be comfortable to
view with low ghosting. A clear comfortable view of some objects in the
field may be scarificed to do justice to the subject of the image. If
the image is of low contrast and with low crosstalk then diverging to 2
1/2" for the far field may make sense. Otherwise convergance produces
the most believable image and is not unlike viewing conventional slides.
Consider the questions about the panorama camera and its optimal use for
stereo. Imagine that the cameras are 20 degrees divergant(walleyed),
that one shoots slide film, and has a cylindrical lenticular screen, and
a pair of medium format projectors (wide images over a short thow
distance)with crossed polaroids. Do you think that one could produce a
panorama with a wide central stereo field along with a peripheral right
and left image?
The "window" is an artifact of the recording and playback system. It is
also a tool used to achieve consistant results in and environment where
throughput is important. I certainly know why people do this, but I
don't have the time to mount all of my slides that way.
John Toeppen
|