Semantic video object segmentation and tracking using mathematical morphology and perspective motion model

This paper introduces a novel semantic video object extraction system using mathematical morphology and a perspective motion model. Inspired by the results from a study of the human visual system, we solve the semantic video object extraction problem in two steps: semi-automatic I-frame segmentation and automatic P-frame tracking. First, the precise semantic video object boundary can be found with the combined power of human assistance and a morphological segmentation tool. Second, the semantic video object in the remaining frames can be obtained using a global perspective motion estimation and compensation of the previous semantic video object plus boundary refinement as used for frames