I'm barely able to keep up with my own school work at the moment, but [Bad username or site: @ livejournal.com] did tell me a lot about the work that [Bad username or site: @ livejournal.com] is doing. Binocular vision is another piece of the great AI puzzle, so of course I have some interest in it since my primary interest is in bringing all of the pieces together into a thinking being. Although a lot of the hardware implementation is further beyond what I'm immediately looking at, the vision part is integral in how we classify our environment and therefore plays a large role in the cognitive science area in which I do have an interest. I want the robots to "start from scratch" with vision just as a baby does. A baby is unable to distinguish depth perception and takes a (short) while to distinguish shapes. But outlines and movement are built in. We also have to learn orientation patterns, such as vertical, horizontal, etc. Believe it or not, neuroscience and cognitive psychology tell us that although we have to learn orientations, human faces have a special place in our brain to which we orient naturally. Of course, I'm still not sure if the area is set to orient to the infamous "caretaker" or if it's really set for "human". Still have more research to do and that's not my direct field either.
(no subject)
Date: 2005-03-09 07:00 am (UTC)