Issue No. 02 - February (2001 vol. 23)
DOI Bookmark: http://doi.ieeecomputersociety.org/10.1109/34.908972
<p><b>Abstract</b>—We introduce an empirical Bayesian procedure for the simultaneous segmentation of an observed motion field and estimation of the hyperparameters of a Markov random field prior. The new approach exhibits the Bayesian appeal of incorporating prior beliefs, but requires only a <it>qualitative</it> description of the prior, avoiding the requirement for a <it>quantitative</it> specification of its parameters. This eliminates the need for trial-and-error strategies for the determination of these parameters and leads to better segmentations.</p>
Motion segmentation, layered representations, empirical Bayesian procedures, estimation of hyperparameters, statistical learning, expectation-maximization.
A. Lippman and N. Vasconcelos, "Empirical Bayesian Motion Segmentation," in IEEE Transactions on Pattern Analysis & Machine Intelligence, vol. 23, no. , pp. 217-221, 2001.