Motion Estimation via Belief Propagation
Giuseppe Boccignone, Angelo Marcelli, Paolo Napoletano, Mario Ferraro · 2007
We present a probabilistic model for motion estimation in which motion characteristics are inferred on the basis of a finite mixture of motion models. The model is graphically represented in the form of a pairwise Markov Random Field network upon which a Loopy Belief Propagation algorithm is exploited to perform inference. Experiments on different video clips are presented and discussed.