Fast and Stable Human Detection Using Multiple Classifiers Based on Subtraction Stereo with HOG Features

摘要：

In this paper, we propose a fast and stable human detection based on “subtraction stereo which can measure distance information of foreground regions. Scanning an input image by detection windows is controlled in their window sizes and number using the distance information obtained from subtraction stereo. This control can skip a large number of detection windows and leads to reduce the computational time and false detection for fast and stable human detection. Additionally, we propose two-step boosting as a new training way of classifier with whole and upper human body models. Experimental results show that the proposal is faster and less false detection than the method described in the reference 1.

作者: Makoto Arie Alessandro Moro Yuma Hoshikawa Toru Ubukata Kenji Terabayashi Kazunori Umeda

作者单位: Course of Precision Engineering,School of Science and Engineering,Chuo University/CREST,JST,1-13-27 Department of Industrial and Information Engineering,Univ. of Trieste/CREST,JST,P. le Europe 1,34127 Department of Precision Mechanics,Faculty of Science and Engineering,Chuo University/CREST,JST,1-13-

会议类型: 国际会议

会议名称: 2011 IEEE International Conference on Robotics and Automation(2011年IEEE世界机器人与自动化大会 ICRA 2011)

会议地点: 上海

会议语种:英文

页码: 868-873

在线出版日期: 2011-05-09（万方平台首次上网日期，不代表论文的发表时间）