Study on Moving Object Segmentation Using High Order Statistics

摘要：

Detecting and segmenting moving objects in a static scene is an important computer vision task. In order to improve the efficiency of moving object segmentation, a HOS algorithm is proposed and implemented. The method of searching data table is used to reduce computation complexity. Additionally, using coarse-to-fine technique improves the computation speed greatly. Experiments are provided to show the validation of our algorithm.

关键词： Moving Object Segmentation High order statistics Coarse-to-fine technique

作者: Li Yingchun Liu Qitao Tang Liming

作者单位: Department of Optical and Electrical Equipment,The Academy of Equipment Command & Technology, Beijing, 101416

会议类型: 国际会议

会议名称: 第七届国际测试技术研讨会

会议地点: 北京

会议语种:英文

在线出版日期: 2007-08-05（万方平台首次上网日期，不代表论文的发表时间）