Abstract

A central theme of computational vision research has been the realization that reliable estimation of local scene properties requires propagating measurements across the image. Many authors have therefore suggested solving vision problems using architectures of locally connected units updating their activity in parallel. Unfortunately, the convergence of traditional relaxation methods on such architectures has proven to be excruciatingly slow and in general they do not guarantee that the stable point will be a global minimum. In this paper we show that an architecture in which Bayesian Beliefs about image properties are propagated between neighboring units yields convergence times which are several orders of magnitude faster than traditional methods and avoids local minima. In particular our architecture is non-iterative in the sense of Marr [5]: at every time step, the local estimates at a given location are optimal given the information which has already been propagated to that loc...

Citations

5662	Probabilistic reasoning in intelligent systems: networks of plausible inference - Pearl - 1988
1180	Fundamentals of speech recognition - Rabiner - 1993
307	Markov random field modeling in computer vision - Li - 2001
288	Scene labeling by relaxation operations - Rosenfeld, Hummel, et al. - 1976
135	The measurement of visual motion - Hildreth - 1983
93	Efficient multiscale regularization with applications to the computation of optical flow - Luettgen, Karl, et al. - 1992
33	Introduction to Applied Mathematics, WellesleyCambridge - Strang - 1993
22	Ecient multiscale regularization with applications to the computation of optical - Luettgen, Karl, et al. - 1994
19	Intermediate-level visual representations and the construction of surface perception - Sajda, Finkel - 1995
18	Rabiner and Biing-Hwang Juang, Fundamentals of Speech Recognition - Lawrence - 1993