Title :
Empirical Bayesian motion segmentation
Author :
Vasconcelos, Nuno ; Lippman, Andrew
Author_Institution :
Compaq Comput. Corp., Cambridge, MA, USA
fDate :
2/1/2001 12:00:00 AM
Abstract :
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 qualitative description of the prior, avoiding the requirement for a quantitative specification of its parameters. This eliminates the need for trial-and-error strategies for the determination of these parameters and leads to better segmentations
Keywords :
Bayes methods; Markov processes; image motion analysis; image segmentation; parameter estimation; Markov random field prior; empirical Bayesian motion segmentation; hyperparameter estimation; observed motion field; parameter estimation; prior beliefs; Bayesian methods; Computer vision; Image segmentation; Layout; Markov random fields; Motion estimation; Motion segmentation; Random variables; Shape control; Statistical learning;
Journal_Title :
Pattern Analysis and Machine Intelligence, IEEE Transactions on
