Object tracking using particle filter and template matching

被引：0

作者：

Lin, Shinfeng D. ^{[1
]}

Chen, Ting-Yi ^{[1
]}

Lin, Jia-Jen ^{[1
]}

机构：

[1] Department of Computer Science and Information Engineering, National Dong Hwa University, No. 1, Sec. 2, Da Hsueh Road, Shoufeng, Hualien, Taiwan

来源：

ICIC Express Letters | 2015年 / 9卷 / 11期

关键词：

Monte Carlo methods - Decision trees - Tracking (position) - Bandpass filters - Color matching;

D O I：

暂无

中图分类号：

学科分类号：

摘要：

In recent years, existing trackers have made success under some situations. However, it is still a very challenging problem such as large appearance changes, illumination changes, and occlusion. In this paper, we propose an object tracking using particle filter and template matching. This method contains three major parts: feature extraction, template matching and particles weighting. The decision tree is presented for the tracker result of template matching, which considers color, spatial distance, and motion vector. To compensate for template matching, particle filter with Speeded Up Robust Features (SURF) is used in the failed tracking. Experimental results with challenging video sequences are presented to demonstrate the effectiveness and robustness of the proposed method. © 2015 ICIC International.

引用

页码：3093 / 3099