Graph matching using spectral embedding and alignment

被引：14

作者：

Bai, X ^{[1
]}

Yu, H ^{[1
]}

Hancock, ER ^{[1
]}

机构：

[1] Univ York, Dept Comp Sci, York YO10 5DD, N Yorkshire, England

来源：

PROCEEDINGS OF THE 17TH INTERNATIONAL CONFERENCE ON PATTERN RECOGNITION, VOL 3 | 2004年

关键词：

D O I：

10.1109/ICPR.2004.1334550

中图分类号：

TP18 [人工智能理论];

学科分类号：

081104 ; 0812 ; 0835 ; 1405 ;

摘要：

This paper describes how graph-spectral methods can be used to transform the node correspondence problem into one of point-set alignment. We commence by using the ISOMAP algorithm to embed the nodes of a graph in a low-dimensional Euclidean space. With the nodes in the graph transformed to points in a metric space, we can recast the problem of graph-matching into that of aligning the points. Here we use a variant of the Scott and Longuet-Higgins algorithm to find point correspondences. We experiment with the resulting algorithm on a number of real-world problems.

引用

页码：398 / 401

页数：4

共 18 条

[1]

[Anonymous], INT C PATT REC

[2]

BELKIN M, 2000, NEURAL COMPUTATION, V15

[3]

Busemann H., 1955, GEOMETRY GEODESICS

[4]

CHRISTMAS W, 1995, IEEE PAMI, V17

[5]

GOLD S, 1996, IEEE PAMI, V18

[6] Properties of embedding methods for similarity searching in metric spaces [J].

Hjaltason, GR ;

Samet, H .

IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE, 2003, 25 (05) :530-549

[7]

KOSINOV S, 2002, LNCS, V2396, P133

[8] GEOMETRY OF GRAPHS AND SOME OF ITS ALGORITHMIC APPLICATIONS [J].

LINIAL, N ;

LONDON, E ;

RABINOVICH, Y .

COMBINATORICA, 1995, 15 (02) :215-245

[9]

Luo B, 2002, INT C PATT RECOG, P164, DOI 10.1109/ICPR.2002.1048263

[10] Structural graph matching using the EM algorithm and singular value decomposition [J].

Luo, B ;

Hancock, ER .

IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE, 2001, 23 (10) :1120-1136

← 1 2 →