Title :
The model selection criteria as merging criteria
Author :
Gheissari, Niloofar ; Bab-Hadiashar, Alireza
Author_Institution :
Swinburne Univ. of Technol., Melbourne, Vic., Australia
Abstract :
A comparative study of a wide range of different model selection criteria for merging 3D range data is presented. First, we measure and compared the success rate of every criterion in detecting step and crease discontinuities in 3D range data followed by measuring the performance in merging non-existent discontinuities. As a conclusion, we have recommended a small group of criteria that are suitable for all cases.
Keywords :
computer vision; data models; merging; 3D range data; merging criteria; merging nonexistent discontinuities; model selection criteria; Application software; Australia; Computer vision; Libraries; Merging; Parameter estimation; Relays;
Conference_Titel :
Pattern Recognition, 2004. ICPR 2004. Proceedings of the 17th International Conference on
Print_ISBN :
0-7695-2128-2
DOI :
10.1109/ICPR.2004.1334149
