Motion field estimation using segmentation and consistency constraint

This paper presents an approach to realistic motion field estimation. With this approach, an image is first segmented into homogeneous regions using a new multiscale gradient algorithm followed by watershed transformation. The multiscale gradient algorithm efficiently solves the oversegmentation problem of watershed transformation, increases segmentation accuracy and reduces the computational cost. The motion field is then estimated using block matching with a consistency constraint. The consistency constraint function is defined by neighbouring motion vectors and the segmentation map. Simulation results show that motion fields generated by block matching with the consistency constraint are very smooth within each object, approaching realistic motion fields, even when a small block size is used.