Title :
Robust model based motion segmentation
Author :
Bab-Hadiashar, Alireza ; Gheissari, Niloofar ; Suter, David
Author_Institution :
Sch. of Eng. & Sci., Swinburne Univ. of Technol., Hawthorn, Vic., Australia
Abstract :
This paper presents a new algorithm for the motion segmentation task. The proposed algorithm addresses the important issue of the interconnectivity between data segmentation, model selection and noise scale estimation. The algorithm is tested on motion segmentation of multiple objects undergoing different types of motion. The results of applying our algorithm to range data segmentation are also included.
Keywords :
image motion analysis; image segmentation; noise; data segmentation; interconnectivity; model selection; multiple objects; noise scale estimation; range data segmentation; robust model based motion segmentation; Australia; Computer vision; Context modeling; Image segmentation; Motion segmentation; Robots; Robustness; Systems engineering and theory; Testing; Tracking;
Conference_Titel :
Pattern Recognition, 2002. Proceedings. 16th International Conference on
Print_ISBN :
0-7695-1695-X
DOI :
10.1109/ICPR.2002.1048412
