Graph matching with a dual-step EM algorithm

被引:146
作者
Cross, ADJ [1 ]
Hancock, ER [1 ]
机构
[1] Univ York, Dept Comp Sci, York YO1 5DD, N Yorkshire, England
关键词
EM algorithm; graph-matching; affine geometry; perspective geometry; relational constraints; Delaunay graph; discrete relaxation;
D O I
10.1109/34.730557
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 a new approach to matching geometric structure in 2D point-sets. The novel feature is to unify the tasks of estimating transformation geometry and identifying point-correspondence matches. Unification is realized by constructing a mixture model over the bipartite graph representing the correspondence match and by affecting optimization using the EM algorithm. According to our EM framework, the probabilities of structural correspondence gate contributions to the expected likelihood function used to estimate maximum likelihood transformation parameters. These gating probabilities measure the consistency of the matched neighborhoods in the graphs. The recovery of transformational geometry and hard correspondence matches are interleaved and are realized by applying coupled update operations to the expected log-likelihood function. In this way, the two processes bootstrap one another. This provides a means of rejecting structural outliers. We evaluate the technique on two real-world problems. The first involves the matching of different perspective views of 3.5-inch floppy discs. The second example is furnished by the matching of a digital map against aerial images that are subject to severe barrel distortion due to a line-scan sampling process. We complement these experiments with a sensitivity study based on synthetic data.
引用
收藏
页码:1236 / 1253
页数:18
相关论文
共 45 条
[1]   IMAGE REPRESENTATION USING VORONOI TESSELLATION [J].
AHUJA, N ;
AN, B ;
SCHACHTER, B .
COMPUTER VISION GRAPHICS AND IMAGE PROCESSING, 1985, 29 (03) :286-295
[2]   DOT PATTERN PROCESSING USING VORONOI NEIGHBORHOODS [J].
AHUJA, N .
IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE, 1982, 4 (03) :336-343
[3]   EXTRACTION OF EARLY PERCEPTUAL STRUCTURE IN DOT PATTERNS - INTEGRATING REGION, BOUNDARY, AND COMPONENT GESTALT [J].
AHUJA, N ;
TUCERYAN, M .
COMPUTER VISION GRAPHICS AND IMAGE PROCESSING, 1989, 48 (03) :304-356
[4]   3-D POSE FROM 3 POINTS USING WEAK-PERSPECTIVE [J].
ALTER, TD .
IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE, 1994, 16 (08) :802-808
[5]   Graphical templates for model registration [J].
Amit, Y ;
Kong, A .
IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE, 1996, 18 (03) :225-236
[6]  
BEARDSLEY PA, 1994, P 3 EUR C COMP VIS, P85
[7]   GEOMETRIC STRUCTURES FOR 3-DIMENSIONAL SHAPE REPRESENTATION [J].
BOISSONNAT, JD .
ACM TRANSACTIONS ON GRAPHICS, 1984, 3 (04) :266-286
[8]   COMBINING POINT DISTRIBUTION MODELS WITH SHAPE MODELS BASED ON FINITE-ELEMENT ANALYSIS [J].
COOTES, TF ;
TAYLOR, CJ .
IMAGE AND VISION COMPUTING, 1995, 13 (05) :403-409
[9]   ACTIVE SHAPE MODELS - THEIR TRAINING AND APPLICATION [J].
COOTES, TF ;
TAYLOR, CJ ;
COOPER, DH ;
GRAHAM, J .
COMPUTER VISION AND IMAGE UNDERSTANDING, 1995, 61 (01) :38-59
[10]  
Cross ADJ, 1998, ADV NEUR IN, V10, P780