Auto-scaled ISL tracking for region based rate control infrastructure and applications in video surveillance

Object tracking and video rate control are usually challenge tasks for the infrastructure and applications of standard video surveillance systems. In this paper, we proposed a novel auto-seated incremental subspace learning (ISL) to track the salient distortion areas continually in the video and employ this tracking mechanism to serve the purpose of region based rate control of video surveillance applications. Compared to other tracking/locating methods, the auto-scaled ISL tracking can track the salient distortion areas more robustly and accurately, and specifically in real-time. In addition, for the case that there exists the overlap/occlusion between different salient distortion areas, the proposed method can also obtain accurate results, which could make the region based rate control and bit allocation to reach higher efficiency in many applications. The experiment results of the proposed approach demonstrate the subject visual quality of the video has been improved greatly.