Towards a computational theory of model based vision and perception
Haim Shvaytser · 2002
When given partial data, a model based approach requires that pictures consistent with the model are chosen as plausible interpretations of the data. The author presents a computational theory that relates degrees of freedom in such models to the number of pixels that carry useful information. It is shown that in models with a finite number of degrees of freedom it is always possible to find a consistent interpretation from a finite number of pixels.>