Robust object tracking with active context learning

摘要

This paper proposes a method to deal with long-term robust object tracking in unconstrained environment. The approach exploits both target and background information on the fly automatically. It builds the structural constraint using active context learning to enhance the adaptability for variation of the target and stability of tracking. An optical-flow-based motion region extraction method is integrated into the context learning framework to address the problem of fast target motion or abrupt camera motion. Experimental results on challenging real-world video sequences demonstrate the effectiveness and robustness of our approach. Comparisons with several state-of-the-art methods are provided.