Detecting moving objects in dynamic scenes from sequences of lidar scans isan important task in object tracking, mapping, localization, and navigation.Many works focus on changes detection in previously observed scenes, while avery limited amount of literature addresses moving objects detection. Thestate-of-the-art method exploits Dempster-Shafer Theory to evaluate theoccupancy of a lidar scan and to discriminate points belonging to the staticscene from moving ones. In this paper we improve both speed and accuracy ofthis method by discretizing the occupancy representation, and by removing falsepositives through visual cues. Many false positives lying on the ground planeare also removed thanks to a novel ground plane removal algorithm. Efficiencyis improved through an octree indexing strategy. Experimental evaluationagainst the KITTI public dataset shows the effectiveness of our approach, bothqualitatively and quantitatively with respect to the state- of-the-art.
展开▼
