Probability distributions of optical flow

Gradient methods are widely used in the computation of optical ow. We discuss extensions of these methods which compute probability distributions of optical ow. The use of distributions allows representation of the uncertainties inherent in the optical ow computation, facilitating the combination with information from other sources. We compute distributed optical ow for a synthetic image sequence and demonstrate that the probabilistic model accounts for the errors in the ow estimates. We also compute distributed optical ow for a real image sequence.