Motion Compensation Based Detecting and Tracking Targets in Dynamic Scene

In most circumstances of target tracking in video sequences, the scene is dynamic due to the mobile cameras. So developing an approach to handle this problem is of great importance. Aiming at this problem, we propose a motion compensation based target detection and tracking algorithm, in which, motion compensation is generated by pyramidal optical flow. In target detection phase, motion compensation and background update algorithms are developed, while in the tracking phase, motion compensation and RJ-MCMC particle filtering algorithms are used. Experiment shows that the proposed approach can handle detection and tracking problems well in dynamic scene.