Re: Computer 3-goD

[P3D Gregory J. Wageman <gjw@xxxxxxxxxx>]

John Roberts replies:

>>Translating a single 2D image into a 3D image requires the computer to
>>recognize the *content* of the image 

>Not necessarily - that's how a human would do it, but machines aren't
>constrained to use the same techniques humans do. A photograph contains
>information that a human viewer doesn't necessarily use.

I disagree.  The human viewer has a HUGE repository of knowlege and
processing capacity including shape recognition and the ability to
"fill in" holes in the information.  You can cut off Aunt Tilly's
head at the neck, and you would still recognize that it's Aunt Tilly.
Traditional machine-based pattern matching would have a difficult
time dealing with that.

For example, we have a very strong visual recognition system for faces.
It is so strong in fact, that we see faces when they aren't even really
there (like that famous 'face' on Mars, and the various "Old Man in the
Mountain" formations around the world).  Our visual cortex has processed
the image so heavily before we consciously "see" it and we aren't even
aware of how much it is doing.  This automatic processing is what makes
us subject to all of those optical illusions that make straight lines
seem bent, and so on.

Humans are capable of recognizing objects that are out-of-focus and
obscured.  Ever watch a badly snowy television picture and still make
sense out of it?  Try that with a computer!

>It would be more productive to try to avoid phrases like "a computer knows
>this" or "a computer does that" - though I'm sure you meant it as a shorthand
>expression, it can be confusing to people who don't know much about computers.
>In general, from the viewpoint of the outside world, a computer doesn't
>*know* anything, and unless it's a specialized device, left to itself it
>usually just sits there passively. It's the *program* (I'm including stored
>data bases in this) that embodies the "knowledge" and implements the actions
>of the computer. Most computers today can implement the model of the
>"Turing machine", in the sense that anything that can be fully and explicitly
>described, and doesn't require too much memory or too many computations,
>can be done by them in a "reasonable" period of time. Fully and explicitly
>stating what is to be done is what countless thousands of computer programmers
>are being paid to do.

Thanks for the lesson in computers.  I've only been paid to program them
for the last 17 years.  I didn't use the phrases you ascribe to me-- I
said that the *human* recognizes which shapes are part of one object
and not part of another object, even when they overlap, which a computer
cannot successfully do without the same sort of reasoning process occurring.
It must *recognize* that an object is a person, say-- and that people
generally do not have benches growing out of their sides, and that therefore
the bench must be a separate object.  Edge detection alone doesn't cut
it, it is too easily fooled under normal photographic conditions.
Practical machine vision implementations usually require very special
lighting conditions and a carefully controlled environment for success.

>While the research continues, I am not
>aware of any convincing evidence presently available that human thought and
>perception are *inherently* indescribable.

Since we humans cannot even yet explain the process by which we "think", and
as far as I know no one has yet proven whether or not the "mind" is simply
a product of brain function, as opposed to something more metaphysical,
I don't see how you are anyone can confidently predict that we will ever
be able to make a machine think like we do.  As you yourself pointed out,
a process must be described to the computer unambiguously and in the
smallest detail.

>That could be true. I also recently saw a web page at a university, in which
>it was stated that if you consider the body of a Stradivarius violin as an
>analog computer to convert the energy provided by the bow into specifically
>configured acoustic energy, the digital equivalent of its computational
>power would be on the order of gigaflops.

Yamaha, Korg and other musical instrument manufacturers have just begun
shipping a new generation of instruments based on just such a physical
modelling algorithm, using a host of special-purpose Digital Signal
Processors.  They algorithmically model the string, the resonant chamber
of the body, and the excitation of the string by the bow.  They can also
do things like model a trumpet mouthpiece exciting the resonant chamber
of the violin body, producing heretofore unheard of sounds.  The computing
power to do this has only very recently become practical, althought the
concept was demonstrated in university labs some time ago.

(If you wish to continue this, we should take it to e-mail, we are WAY
off topic.)

        -Greg W.
        Member, Technical Staff
        Taos Mountain Software
        http://www.taos.com


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

End of PHOTO-3D Digest 1660
***************************
***************************
 Trouble? Send e-mail to 
 wier@xxxxxxxxxxxxxxx 
 To unsubscribe select one of the following,
 place it in the BODY of a message and send it to:
 listserv@xxxxxxxxxxxxxxx 
   unsubscribe photo-3d
   unsubscribe mc68hc11
   unsubscribe overland-trails
   unsubscribe icom
 ***************************