Mean-Shift Blob Tracking With Adaptive Feature Selection And Scale Adaptation

Dawei Liang,Qingming Huang,Shuqiang Jiang,Hongxun Yao,Wen Gao

DOI: https://doi.org/10.1109/ICIP.2007.4379323

2007-01-01

Abstract:When the appearances of die tracked object and surrounding background change during tracking, fixed feature space tends to cause tracking failure. To address this problem, we propose a method to embed adaptive feature selection into mean shift tracking framework. From a feature set, the most discriminative features are selected after ranking these features based on their Bayes error rates, which are estimated from object and background samples. For the selected features, a criterion is proposed to evaluate their stability for tracking and to guide feature reselection. The selected features are used to generate a weight image, in which mean shift is employed to locate the object. Moreover, a simple yet effective scale adaptation method is proposed to deal with object changing in size. Experiments on several video sequences show the effectiveness of the proposed method.