Direct point-based registration for precise non-rigid surface matching using Student’s-t mixture model

详细信息    查看全文

• 作者：Zhiyong Zhou ; Baotong Tong ; Chen Geng ; Jisu Hu ; Jian Zheng ; Yakang Dai ; ^{daiyk@sibet.ac.cn}
• 关键词：Surface matching ; Point set registration ; Student&rsquo ; s-t mixture model ; Dirichlet distribution
• 刊名：Biomedical Signal Processing and Control
• 出版年：2017
• 出版时间：March 2017
• 年：2017
• 卷：33
• 期：Complete
• 页码：10-18
• 全文大小：3798 K
• 卷排序：33

文摘

A direct point set registration method based on Student’s-t mixture model for precise non-rigid surface matching is proposed. Dirichlet distribution is introduced into the Student’s-t mixture model for formulating the prior probabilities. Local spatial relationships between neighboring points are modeled by representing their posterior probabilities in a linear smoothing filter. Directional springs that are defined as the orientation changes of the undirected edges are introduced to preserve structures of surfaces.