Adaptive spatio-temporal context learning for visual target tracking

While visual target tracking is one of the noteworthy and the most active research areas in computer vision and machine learning, many challenges are still unresolved. In this paper, an adaptive generic target tracker is proposed that includes the adaptive determination of learning parameters from spatio-temporal context model, analysis of prior targets and confidence map for accurate target localization, and modified scale estimation scheme based on confidence map. According to spatio-temporal context model, the learning parameters are adaptively determined for achieving confidence map and target scale robustly. Moreover, analysis of the confidence map helps our tracker to change context feature set and accurately estimate target location in critical situations such as occlusion, appearance changes, and pose variation. Finally, we propose a modified scale estimation scheme based on confidence map that corrects adopted scale when the confidence map is not reliable. Experimental results on numerous video sequences show that the proposed adaptive generic tracker outperforms both the accuracy and robustness compared to the dense spatio-temporal context learning (STC) tracker effectively.