Optic flow based on multi-scale anchor point movement and discontinuity-preserving regularization

摘要

We introduce a new method to determine the flow field of an image sequence using multi-scale anchor points. These anchor points manifest themselves in the scale-space representation of an image. The novelty of our method lies largely in the fact that the relation between the scale-space anchor points and the flow field is formulated in terms of soft constraints in a variational method. This leads to an algorithm for the computation of the flow field that differs fundamentally from previously proposed ones based on hard constraints. We show a significant performance increase when our method is applied to the Yosemite image sequence, a standard and well-established benchmark sequence in optic flow research. Also, it is shown that this performance is not sensitive to slight changes in the two parameters used and that, with the same parameter values, our method yields very good results in the Rubber Whale image sequence as well.