STOCHASTIC-MODELS FOR DIV-CURL OPTICAL-FLOW METHODS

In this letter, we consider Suter's DIV-CURL optical flow methods, whe rein the problem of computing a velocity field from an image sequence is regularized using smoothness conditions based on the divergence and curl of the field. In particular, we develop stochastic formulations of DIV-CURL splines using the linear smoothing theory of Adams, Willsk y, and Levy. Our models are shown to be well posed and thus can be use d in both simulating and estimating velocity fields having known stoch astic properties. As a special case, our stochastic model reduces to t hat developed by Rougee, Levy, and Whisky for the classical Horn and S chunck's optical flow.