Geodesic-like features for point matching

doi:10.1016/j.neucom.2016.08.092

	Geodesic-like features for point matching
	Qian, Deheng 1; Chen, Tianshi 2,3; Qiao, Hong 3,4
	2016-12-19
发表期刊	NEUROCOMPUTING
ISSN	0925-2312
卷号	218 页码:401-410
摘要	Point matching problem seeks the optimal correspondences between two sets of points via minimizing the dissimilarities of the corresponded features. The features are widely represented by a graph model consisting of nodes and edges, where each node represents one key point and each edge describes the pair-wise relations between its end nodes. The edges are typically measured depending on the Euclidian distances between their end nodes, which is, however, not suitable for objects with non-rigid deformations. In this paper, we notice that all the key points are spanning on a manifold which is the surface of the target object. The distance measurement on a manifold, geodesic distance, is robust under non-rigid deformations. Hence, we first estimate the manifold depending on the key points and concisely represent the estimation by a graph model called the Geodesic Graph Model (GGM). Then, we calculate the distance measurement on GGM, which is called the geodesic-like distance, to approximate the geodesic distance. The geodesic-like distance can better tackle non-rigid deformations. To further improve the robustness of the geodesic-like distance, a weight setting process and a discretization process are proposed. The discretization process produces the geodesic-like features for the point matching problem. We conduct multiple experiments over widely used datasets and demonstrate the effectiveness of our method. (C) 2016 Elsevier B.V. All rights reserved.
关键词	Point matching Non-rigid deformation Geodesic distance
DOI	10.1016/j.neucom.2016.08.092
收录类别	SCI
语种	英语
资助项目	Strategic Priority Research Program, Chinese Academy of Sciences[XDB02080003] ; BMST[D16110400140000] ; BMST[D161100001416001] ; National Natrual Science Fundation of China[61522211] ; National Natrual Science Fundation of China[61473275]
WOS研究方向	Computer Science
WOS类目	Computer Science, Artificial Intelligence
WOS记录号	WOS:000388053700044
出版者	ELSEVIER SCIENCE BV
引用统计
文献类型	期刊论文
条目标识符	http://119.78.100.204/handle/2XEOYT63/7898
专题	中国科学院计算技术研究所期刊论文_英文
通讯作者	Qiao, Hong
作者单位	1.Samsung Res Inst China Beijing SRC B, Beijing 100028, Peoples R China 2.Chinese Acad Sci, Inst Comp Technol, State Key Lab Comp Architecture, Beijing 100190, Peoples R China 3.Chinese Acad Sci, Ctr Excellence Brain Sci & Intelligence Technol, Shanghai 200031, Peoples R China 4.Chinese Acad Sci, Inst Automat, State Key Lab Management & Control Complex Syst, Beijing 100190, Peoples R China
推荐引用方式 GB/T 7714	Qian, Deheng,Chen, Tianshi,Qiao, Hong. Geodesic-like features for point matching[J]. NEUROCOMPUTING,2016,218:401-410.
APA	Qian, Deheng,Chen, Tianshi,&Qiao, Hong.(2016).Geodesic-like features for point matching.NEUROCOMPUTING,218,401-410.
MLA	Qian, Deheng,et al."Geodesic-like features for point matching".NEUROCOMPUTING 218(2016):401-410.