Markov Random Fields with E cient Approximations

Markov Random Fields (MRF's) can be used for a wide variety of vision problems. In this paper we focus on MRF's with two-valued clique potentials, which form a generalized Potts model. We show that the maximum a posteriori estimate of such an MRF can be obtained by solving a multiway minimum cut problem on a graph. We develop eecient algorithms for computing good approximations to the minimum mul-tiway cut. The visual correspondence problem can be formulated as an MRF in our framework; this yields quite promising results on real data with ground truth. We also apply our techniques to MRF's with linear clique potentials.