A novel algorithm for achieving a light-weight tracking system

Abstract

Object tracking is fundamental to automated video surveillance, activity analysis and event recognition. Only a small percentage of the system resources can be allocated for tracking in real time applications, the rest being required for high-level tasks such as recognition, trajectory interpretation, and reasoning. There is a desperate need to carefully optimize the tracking algorithm to keep the computational complexity of a tracker as low as possible yet maintaining its robustness and accuracy. This paper proposes a novel algorithm which attempts to attain a light weight tracking system by reducing undesirable and redundant computations. The frames of the video are passed through a preprocessing stage which transmits only motion detected blocks to the tracking algorithm. Further frames containing little motion in the search area of the target object are detected in preprocessing stage itself and are blocked from further processing. Our experimental results demonstrate that the throughput of the new proposed tracking system is exceptionally higher than the traditional one. © 2010 Springer-Verlag Berlin Heidelberg.