This is about all I have at the moment:

Basically there are 2 inputs (1 would be a stream from a ca,era and the other a reference image (for difference detection etc.)

There is a main output which is the image that would have the crosshairs for target tracking or whatever. Then there are two other windows for output like histograms and colour breakdown and all that shit Michael did in his thesis.

Links:

http://cs-people.bu.edu/aaron/turret/turret.htm

http://www.techeblog.com/index.php/tech-gadget/video-real-world-motion-tracking

Q and A

is this thing going to be having its code in java or something, because i think it would be cool if we could write a few indpendently and do comparisons or something? -if you dont mind me horning in on the action? I don't really mind what it is written in, I suppose if we can get a good codebase from somewhere we can use whatever. Also, the more people in the action the better.