Combining the Spatial and Temporal Eigen-space for Visual Tracking

摘要：

Visual tracking is an important research topic in computer vision community. Most subspace based tracking algorithms focus on the time correlation between the image observations of the object, but the spatial layout information of the object is ignored. This paper proposes a robust visual tracking algorithm which effectively combines the spatial and temporal eigen-space of the object. In order to captures the variations of object appearance, an incremental updating strategy is developed to update the eigen-space and mean of the object Experimental results demonstrate that, compared with the state-of-the-art subspace based tracking algorithms, the proposed tracking algorithm is more robust and effective.

关键词： object tracking subspace learning incremental learning

作者: Xiaoqin Zhang Qiuyun Cheng Xingchu Shi Weiming Hu Zhenjie Hong

作者单位: College of Mathematics & Information Science, Wenzhou University, Zhejiang, China Zhengzhou Institute of Aeronautical Industry Management, Henan, China National Laboratory of Pattern Recognition, Institute of Automation, Beijing, China

会议类型: 国际会议

会议名称: The 2010 International Conference on Computer Application and System Modeling(2010计算机应用与系统建模国际会议 ICCASM 2010)

会议地点: 太原

会议语种:英文

页码: 152-155

在线出版日期: 2010-10-22（万方平台首次上网日期，不代表论文的发表时间）