A visual-SLAM for first person vision and mobile robots

Abstract

SLAM(Simultaneous Localization and Mapping) is one of the core subjects in computer vision and robotics. In order to avoid the effects of noise, SLAM systems need devises to remove the moving object such as human beings and cars in real-world environment. In this paper, we propose a method which excludes dynamic features and generate a map in crowded environment, called ICGM2.5. Experiments were conducted in indoor and outdoor crowded real environments. Experimental results show that our approach has superior performance compared to conventional approaches in terms of accuracy.