Regularized estimation of occluded displacement vector fields

Occluded regions and motion boundaries introduce displacement vector field (DVF) discontinuities that must be reconciled to accurately estimate image flow. In this work, the robust regularized estimation of the DVF is considered in the presence of these discontinuities. A robust convex estimation criterion is presented that preserves motion boundaries and allows for a globally optimal estimate of the DVF. A new class of robust convex measures is introduced for edge preserving regularization and an occlusion weighted gradient is proposed as mechanism for managing DVF discontinuities due to occlusion. Results using synthetic image sequences are presented.