MFDTs: mean field dynamic trees

Abstract

Tree structured belief networks are attractive for image segmentation tasks. However, networks with fixed architectures are not very suitable as they lead to blocky artefacts, and led to the introduction of dynamic trees (DTs). The Dynamic trees architecture provide a prior distribution over tree structures, and simulated annealing (SA) was used to search for structures with high posterior probability. In this paper we introduce a mean field approach to inference in DTs. We find that the mean field method captures the posterior better than just using the maximum a posteriori solution found by SA.