Multilayer matching of metric structures using hierarchically well-separated trees

A novel method to match metric distributions is proposed.Method utilizes both spatial distribution of nodes and individual features.Method allows matching to be done in multiple layers of detail.A primal-dual approximation algorithm is provided for improving the performance of the method.Application of the method to the image matching problem is demonstrated. Display Omitted We consider the matching problem between two metric distributions where establishing a one-to-one matching of features may not always be possible. Although many-to-many graph matching techniques achieve the desired multi map between features, they ignore the spatial structure of the nodes. We propose a novel technique, multilayer matching, for solving the matching problem which utilizes both the individual node features and the clustering information of nodes. Our method uses the hierarchically well-separated trees (HSTs) to represent the metric distribution such that non-leaf nodes in the tree representation corresponds to a constellation of features in the original structure. By using HSTs in a linear programming setup, we obtain a matching between features through finding a mapping between non-leaf nodes among the two HSTs. We further provide a primal-dual approximation algorithm for the multilayer matching which runs several order of magnitudes faster while achieving comparable success rates. Application of the method to the image matching problem is also presented in the paper. Empirical evaluation of the method and its primal-dual extension on a set of recognition tests show the robustness and efficiency of the overall approach.

[1]  M. Fatih Demirci,et al.  Object Recognition as Many-to-Many Feature Matching , 2006, International Journal of Computer Vision.

[2]  M. Fatih Demirci,et al.  Many-to-many graph matching via metric embedding , 2003, 2003 IEEE Computer Society Conference on Computer Vision and Pattern Recognition, 2003. Proceedings..

[3]  David P. Williamson,et al.  Primal-Dual Approximation Algorithms for Integral Flow and Multicut in Trees, with Applications to Matching and Set Cover , 1993, ICALP.

[4]  Terry Caelli,et al.  An eigenspace projection clustering method for inexact graph matching , 2004, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[5]  David G. Lowe,et al.  Distinctive Image Features from Scale-Invariant Keypoints , 2004, International Journal of Computer Vision.

[6]  Daniel P. Huttenlocher,et al.  Comparing Images Using the Hausdorff Distance , 1993, IEEE Trans. Pattern Anal. Mach. Intell..

[7]  Salih O. Duffuaa,et al.  A Linear Programming Approach for the Weighted Graph Matching Problem , 1993, IEEE Trans. Pattern Anal. Mach. Intell..

[8]  Piotr Indyk,et al.  Algorithmic applications of low-distortion geometric embeddings , 2001, Proceedings 2001 IEEE International Conference on Cluster Computing.

[9]  M. Fatih Demirci,et al.  Efficient many-to-many feature matching under the l1 norm , 2011, Comput. Vis. Image Underst..

[10]  Horst Bunke,et al.  Graph Edit Distance with Node Splitting and Merging, and Its Application to Diatom Idenfication , 2003, GbRPR.

[11]  Ali Shokoufandeh,et al.  Delay minimizing user association in cellular networks via hierarchically well-separated trees , 2015, 2015 IEEE International Conference on Communications (ICC).

[12]  Alberto Del Bimbo,et al.  A Graph Edit Distance Based on Node Merging , 2004, CIVR.

[13]  LingHaibin,et al.  An Efficient Earth Mover's Distance Algorithm for Robust Histogram Comparison , 2007 .

[14]  J. van Leeuwen,et al.  Graph Based Representations in Pattern Recognition , 2003, Lecture Notes in Computer Science.

[15]  George B. Dantzig,et al.  A PRIMAL--DUAL ALGORITHM , 1956 .

[16]  Robert M. Haralick,et al.  Structural Descriptions and Inexact Matching , 1981, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[17]  Satish Rao,et al.  A tight bound on approximating arbitrary metrics by tree metrics , 2003, STOC '03.

[18]  Stephen A. Cook,et al.  The complexity of theorem-proving procedures , 1971, STOC.

[19]  Ali Shokoufandeh,et al.  Multi-layer Tree Matching Using HSTs , 2015, GbRPR.

[20]  G LoweDavid,et al.  Distinctive Image Features from Scale-Invariant Keypoints , 2004 .

[21]  Sameer A. Nene,et al.  Columbia Object Image Library (COIL100) , 1996 .

[22]  Francis R. Bach,et al.  Many-to-Many Graph Matching: a Continuous Relaxation Approach , 2010, ECML/PKDD.

[23]  Yair Bartal,et al.  Probabilistic approximation of metric spaces and its algorithmic applications , 1996, Proceedings of 37th Conference on Foundations of Computer Science.

[24]  Haibin Ling,et al.  An Efficient Earth Mover's Distance Algorithm for Robust Histogram Comparison , 2007, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[25]  Ali Shokoufandeh,et al.  A primal-dual approach to delay minimizing user association in cellular networks , 2015, 2015 53rd Annual Allerton Conference on Communication, Control, and Computing (Allerton).

[26]  Éva Tardos,et al.  Approximation algorithms for classification problems with pairwise relationships: metric labeling and Markov random fields , 2002, JACM.

[27]  Yll Haxhimusa,et al.  Approximative graph pyramid solution of the E-TSP , 2009, Image Vis. Comput..

[28]  Edwin R. Hancock,et al.  Deterministic search for relational graph matching , 1999, Pattern Recognit..

[29]  Horst Bunke,et al.  Error Correcting Graph Matching: On the Influence of the Underlying Cost Function , 1999, IEEE Trans. Pattern Anal. Mach. Intell..