Path: utzoo!utgpu!jarvis.csri.toronto.edu!mailrus!husc6!m2c!wpi!lfoard
From: lfoard@wpi.wpi.edu (Lawrence C Foard)
Newsgroups: comp.graphics
Subject: Re: 3-D perceptual abilities
Message-ID: <1200@wpi.wpi.edu>
Date: 5 Mar 89 17:31:53 GMT
References: <2492@masada.cs.swarthmore.edu> <3241@ficc.uu.net> <2860@osiris.UUCP>
Reply-To: lfoard@wpi.wpi.edu (Lawrence C Foard)
Organization: Worcester Polytechnic Institute, Worcester, MA. USA
Lines: 13

It is interesting how many ways there are of fooling the human visual system!
This brings up a problem with artificial vision, if you look at something that
doesn't make complete sence to you brain by itself (strange wallpaper etc) you
can still recognize it by comparing it with other things you have seeing, once
you associate it with something familiar you now see what it is (or atleast
what you think it is). The problem this brings up for artificial vision is
that even with very good visual abilitys we are unable to form a unique 3D
model of an object when it is not clearly visible, instead we guess what it is
and check if it matchs something we have seen before. Has this method ever been
used for artificial vision? 
-- 
Disclaimer: My school does not share my views about FORTRAN.
            FORTRAN does not share my views about my school.