An object tracking algorithm based on occlusion mesh model

The concept of an occlusion mesh model is introduced. An object tracking algorithm based on an occlusion mesh model is proposed. A modified occlusion detection method is considered to improve the detection accuracy. A mesh nodes motion estimation method based on feature window matching is presented to achieve sub-pixel resolution and overcome block artifacts produced by block matching. Experimental results indicate that the proposed algorithm is practical and feasible, it can be used for object tracking effectively. Also it solves the 2D motion estimation problem existing in occlusion regions and shows better visual performance.