期刊名称：International Journal of Signal Processing, Image Processing and Pattern Recognition
印刷版ISSN：2005-4254
出版年度：2015
卷号：8
期号：10
页码：233-240
DOI：10.14257/ijsip.2015.8.10.25
出版社：SERSC
摘要：Most of the state-of-the-art tracking algorithms rely on either intensity or color information. Therefore, seeking an effective image descriptor to construct a robust tracker is still a challenging problem. This paper investigates the contribution of saliency map for object tracking, and proposes a saliency detection method which is combined with the location information. Our method formulates the tracking framework based on a Bayesian framework, which models the statistical correlation between the target and the surrounding background with saliency values and location information. Additionally, our results show that the proposed method provides superior performance in terms of accuracy, robustness, and speed
