Abstract
Graph and hypergraph matching are fundamental problems in computer vision, which have been receiving a steadily growing attention in the last two decades due to the recent modern algorithms and models that are both fast and accurate. Graph and hypergraph matching are successful in tackling many real-world tasks that require 2D and 3D feature matching and alignment. Matching is a general problem in vision and graphs have always been quintessential for modeling information that has both local and longer range connections. Therefore, graph matching will continue to influence computer vision and robotics research, crossing over many different mathematical and computational models, including Bayesian networks, conditional and Markov random fields, spectral graph models, and, more recently, deep neural networks. While graph matching is limited to using pairwise relationships, hypergraph matching permits the use of relationships between sets of features of any order. Consequently, it carries the promise to make matching more robust to changes in scale, deformations, and outliers. In this chapter, we present several key ideas, computational models, and optimization methods that introduced in the literature the possibility to learn graph and hypergraph matching in a semi-supervised or even completely unsupervised fashion. We start by describing two methods for solving graph and hypergraph matching based on spectral matching and integer projected fixed point methods, introduced in Chap. 1. Then, based on the same intuition (also discussed in Chap. 1) that correct assignments form a strong cluster of agreements, we follow up by introducing the main approach to learning as it applies to graph and hypergraph matching, as well as the more general Maximum A Posteriori (MAP) inference problem in probabilistic graphical models (MRFs and CRFs). In the last part of the chapter, we demonstrate experimentally the effectiveness of the algorithms presented and, whenever possible, provide extensive comparisons to other top published methods. Moreover, we show that our learning approach is general and can significantly boost the performance of different state-of-the-art matching algorithms in the literature.
The material presented in this chapter is based in large part on the following papers:
Leordeanu, Marius, and Martial Hebert. “A spectral technique for correspondence problems using pairwise constraints.” IEEE International Conference on Computer Vision (ICCV), 2005.
Leordeanu, Marius, Martial Hebert, and Rahul Sukthankar. “An integer projected fixed point method for graph matching and map inference.” In Advances in neural information processing systems (NIPS) 2009.
Leordeanu, Marius, Rahul Sukthankar, and Martial Hebert. “Unsupervised learning for graph matching.” International journal of computer vision 96, no. 1 (2012): 28–45.
Leordeanu, Marius, Andrei Zanfir, and Cristian Sminchisescu. “Semi-supervised learning and optimization for hypergraph matching.” IEEE International Conference on Computer Vision, 2011.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
References
Gold S, Rangarajan A (1996) A graduated assignment algorithm for graph matching. PAMI 18(4):377–388
Berg A, Berg T, Malik J (2005) Shape matching and object recognition using low distortion correspondences. In: CVPR
Cour T, Srinivasan P, Shi J (2006) Balanced graph matching. In: NIPS
Schellewald C, Schnorr C (2005) Probabilistic subgraph matching based on convex relaxation. In: International conference on energy minimization methods in computer vision and pattern recognition
Torr P (2003) Solving Markov random fields using semidefinite programming. In: International conference on artificial intelligence and statistics
Leordeanu M, Hebert M, Sukthankar R (2009) An integer projected fixed point method for graph matching and map inference. In: NIPS
Cho M, Lee J, Lee KM (2010) Reweighted random walks for graph matching. In: ECCV
Egozi A, Keller Y, Guterman H (2013) A probabilistic approach to spectral graph matching. Pattern Anal Mach Intell 35(1)
Zhou F, la Torre FD (2012) Factorized graph matching. In: Computer vision and pattern recognition
Cho M, Lee K (2012) Progressive graph matching: making a move of graphs via probabilistic voting. In: Computer vision and pattern recognition
Tolias G, Avrithis Y (2011) Speeded-up, relaxed spatial matching. In: ICCV
Escolano F, Hancock E, Lozano M (2011) Graph matching through entropic manifold alignment. In: Computer vision and pattern recognition
Torresani L, Kolmogorov V, Rother C (2013) A dual decomposition approach to feature correspondence. Pattern Anal Mach Intell 35(2)
Liu Z, Qiao H, Yang X, Hoi S (2014) Graph matching by simplified convex-concave relaxation procedure. Int J Comput Vis 1–18
Zhou F, la Torre FD (2013) Deformable graph matching. In: Computer vision and pattern recognition, pp 2922–2929
Leordeanu M, Hebert M (2005) A spectral technique for correspondence problems using pairwise constraints. In: ICCV
Caetano T, Cheng L, Le Q, Smola AJ (2007) Learning graph matching. In: ICCV
Leordeanu M, Hebert M (2008) Smoothing-based optimization. In: CVPR
Leordeanu M, Sukthankar R, Hebert M (2009) Unsupervised learning for graph matching. IJCV 96(1)
Cho M, Alahari K, Ponce J (2013) Learning graphs to match. In: International conference on computer vision, pp 25–32
Zanfir A, Sminchisescu C (2018) Deep learning of graph matching. In: Proceedings of the IEEE conference on computer vision and pattern recognition, pp 2684–2693
Duchenne O, Bach F, Kweon I, Ponce J (2009) A tensor-based algorithm for high-order graph matching. In: CVPR
Chertok M, Keller Y (2010) Efficient high order matching. IEEE Trans Pattern Anal Mach Intell 32:2205–2215
Zass R, Shashua A (2008) Probabilistic graph and hypergraph matching. In: CVPR
Çeliktutan O, Wolf C, Sankur B, Lombardi E (2014) Fast exact hyper-graph matching with dynamic programming for spatio-temporal data. J Math Imaging Vis 1–21
Wang J, Chen S, Fuh C (2014) Attributed hypergraph matching on a Riemannian manifold. Mach Vis Appl 25(4)
Yan J, Zhang C, Zha H, Liu W, Yang X, Chu SM (2015) Discrete hyper-graph matching. In: Proceedings of the IEEE conference on computer vision and pattern recognition, pp 1520–1528
Nie WZ, Liu AA, Gao Z, Su YT (2015) Clique-graph matching by preserving global and local structure. In: Proceedings of the IEEE conference on computer vision and pattern recognition, pp 4503–4510
Liu AA, Nie WZ, Gao Y, Su YT (2016) Multi-modal clique-graph matching for view-based 3d model retrieval. IEEE Trans Image Process 25(5):2103–2116
Adamczewski K, Suh Y, Mu Lee K (2015) Discrete Tabu search for graph matching. In: Proceedings of the IEEE international conference on computer vision, pp 109–117
Jiang B, Tang J, Ding C, Luo B (2017) Binary constraint preserving graph matching. In: Proceedings of the IEEE conference on computer vision and pattern recognition, pp 4402–4409
Shi X, Ling H, Hu W, Xing J, Zhang Y (2016) Tensor power iteration for multi-graph matching. In: Proceedings of the IEEE conference on computer vision and pattern recognition, pp 5062–5070
Yan J, Cho M, Zha H, Yang X, Chu SM (2015) Multi-graph matching via affinity optimization with graduated consistency regularization. IEEE Trans Pattern Anal Mach Intell 38(6):1228–1242
Leordeanu MD (2010) Spectral graph matching, learning, and inference for computer vision. PhD Thesis, Carnegie Mellon University
Leordeanu M, Zanfir A, Sminchisescu C (2011) Semi-supervised learning and optimization for hypergraph matching. In: IEEE international conference on computer vision
Leordeanu M, Zanfir A, Sminchisescu C (2011) Semi-supervised learning and optimization for hypergraph matching. In: ICCV
Leordeanu M, Hebert M (2006) Efficient map approximation for dense energy functions. In: International conference on machine learning
Cour T, Shi J (2007) Solving Markov random fields with spectral relaxation. In: International conference on artificial intelligence and statistics
Ravikumar P, Lafferty J (2006) Quadratic programming relaxations for metric labeling and Markov random field map estimation. In: International conference on machine learning
Yan P, Khan SM, Shah M (2008) Learning 4D action feature models for arbitrary view action recognition. In: CVPR
Gaur U, Zhu Y, Song B, Roy-Chowdhury A (2011) A string of feature graphs model for recognition of complex activities in natural videos. In: International conference on computer vision, pp 2595–2602
Liu J, Luo J, Shah M (2009) Recognizing realistic actions from videos “in the wild”. In: CVPR
Nayak N, Sethi R, Song B, Roy-Chowdhury A (2011) Motion pattern analysis for modeling and recognition of complex human activities. In: Guide to video analysis of humans: looking at people
Zhu Y, Nayak N, Gaur U, Song B, Roy-Chowdhury A (2013) Modeling multi-object interactions using string of feature graphs. Comput Vis Image Underst 117(10)
Kim G, Faloutsos C, Hebert M (2008) Unsupervised modeling of object categories using link analysis techniques. In: CVPR
Kim G, Faloutsos C, Hebert M (2008) Unsupervised modeling and recognition of object categories with combination of visual contents and geometric similarity links. In: ACM international conference on multimedia information retrieval
Leordeanu M, Collins R, Hebert M (2005) Unsupervised learning of object features from video sequences. In: CVPR
Parikh D, Chen T (2007) Unsupervised learning of hierarchical semantics of objects (hSOs). In: CVPR
Parikh D, Chen T (2007) Unsupervised identification of multiple objects of interest from multiple images: discover. In: Asian conference on computer vision
Leordeanu M, Hebert M, Sukthankar R (2007) Beyond local appearance: category recognition from pairwise interactions of simple features. In: CVPR
de Aguiar E, Stoll C, Theobalt C, Ahmed N, Seidel HP, Thrun S (2008) Performance capture from sparse multi-view video. In: SIGGRAPH
Huhle B, Jenke P, Straer W (2008) On-the-fly scene acquisition with a handy multi-sensor system. Int J Intell Syst Technol Appl 5(3–4):255–263
Ren X (2007) Learning and matching line aspects for articulated objects. In: CVPR
Chertok M, Keller Y (2010) Spectral symmetry analysis. PAMI 32(7):1227–1238
Hays J, Leordeanu M, Efros A, Liu Y (2006) Discovering texture regularity as a higher-order correspondence problem. In: ECCV
Yi L, Kim VG, Ceylan D, Shen I, Yan M, Su H, Lu C, Huang Q, Sheffer A, Guibas L et al (2016) A scalable active framework for region annotation in 3d shape collections. ACM Trans Graph (TOG) 35(6):210
Im K, Raschle NM, Smith SA, Ellen Grant P, Gaab N (2015) Atypical sulcal pattern in children with developmental dyslexia and at-risk kindergarteners. Cereb Cortex 26(3):1138–1148
Im K, Pienaar R, Lee JM, Seong JK, Choi YY, Lee KH, Grant PE (2011) Quantitative comparison and analysis of sulcal patterns using sulcal graph matching: a twin study. Neuroimage 57(3):1077–1086
Im K, Guimaraes A, Kim Y, Cottrill E, Gagoski B, Rollins C, Ortinau C, Yang E, Grant PE (2017) Quantitative folding pattern analysis of early primary sulci in human fetuses with brain abnormalities. Am J Neuroradiol 38(7):1449–1455
Sotiras A, Davatzikos C, Paragios N (2013) Deformable medical image registration: a survey. IEEE Trans Med Imaging 32(7):1153
Singh R, Xu J, Berger B (2008) Global alignment of multiple protein interaction networks with application to functional orthology detection. Proc Natl Acad Sci 105(35):12,763–12,768
Jiang ZP, Xi W, Li X, Tang S, Zhao JZ, Han JS, Zhao K, Wang Z, Xiao B (2014) Communicating is crowdsourcing: Wi-Fi indoor localization with CSI-based speed estimation. J Comput Sci Technol 29(4):589–604
Hu Y, Liao S, Lei Z, Yi D, Li S (2013) Exploring structural information and fusing multiple features for person re-identification. In: Proceedings of the IEEE conference on computer vision and pattern recognition workshops, pp 794–799
Zhang S, Zhu Y, Roy-Chowdhury AK (2016) Context-aware surveillance video summarization. IEEE Trans Image Process 25(11):5469–5478
Mirzaalian H, Lee TK, Hamarneh G (2016) Skin lesion tracking using structured graphical models. Med Image Anal 27:84–92
Kang U, Hebert M, Park S (2013) Fast and scalable approximate spectral graph matching for correspondence problems. Inf Sci 220:306–318
Park S, Park SK, Hebert M (2013) Fast and scalable approximate spectral matching for higher order graph matching. IEEE Trans Pattern Anal Mach Intell 36(3):479–492
Yang X, Liu CK, Liu ZY, Qiao H, Wang BF, Wang ZD (2014) A probabilistic spectral graph matching algorithm for robust correspondence between lunar surface images. In: Proceeding of the 11th world congress on intelligent control and automation. IEEE, pp 385–390
Brendel W, Todorovic S (2010) Segmentation as maximum-weight independent set. In: NIPS
Jain A, Gupta A, Rodriguez M, Davis L (2013) Representing videos using mid-level discriminative patches. In: Computer vision and pattern recognition, pp 2571–2578
Semenovich D (2010) Tensor power method for efficient map inference in higher-order MRFs. In: ICPR
Monroy A, Bell P, Ommer B (2014) Morphological analysis for investigating artistic images. Image Vis Comput 32(6)
He L, Yang X, Liu ZY (2018) Generalizing integer projected graph matching algorithm for outlier problem. In: 2018 IEEE 3rd international conference on image vision and computing (ICIVC). IEEE, pp 50–54
Lu Y, Huang K, Liu CL (2016) A fast projected fixed-point algorithm for large graph matching. Pattern Recognit 60:971–982
Zaslavskiy M, Bach F, Vert JP (2008) A path following algorithm for graph matching. In: ICSP
Zaslavskiy M (2010) Graph matching and its application in computer vision and bioinformatics. PhD thesis, lEcole nationale superieure des mines de Paris
Besag J (1986) On the statistical analysis of dirty pictures. J R Stat Soc 48(5):259–302
Frank M, Wolfe P (1956) An algorithm for quadratic programming. Nav Res Logist Q 3(1–2):95–110
LeBlanc LJ, Helgason RV, Boyce DE (1985) Improved efficiency of the frank-wolfe algorithm for convex network programs. Transp Sci 19(4):445–462
Mirzaalian H, Lee T, Hamarneh G (2012) Uncertainty-based feature learning for skin lesion matching using a high order MRF optimization framework. In: Medical image computing and computer-assisted intervention, pp 98–105
Leordeanu M, Zanfir A, Sminchisescu C (2013) Locally affine sparse-to-dense matching for motion and occlusion estimation. In: ICCV
Semenovich D, Sowmya A (2010) Tensor power method for efficient map inference in higher-order MRFs. In: ICPR
Neudecker Magnus (1999) Matrix differential calculus with applications in statistics and econometrics. Wiley
Cour T, Shi J, Gogin N (2005) Learning spectral graph segmentation. In: International conference on artificial intelligence and statistics
Bach F, Jordan M (2003) Learning spectral clustering. In: NIPS
Baratchart L, Berthod M, Pottier L (1998) Optimization of positive generalized polynomials under lp constraints. J Convex Anal 5(2):133
Kofidis E, Regallia PA (2002) On the best rank-1 approximation of higher-order supersymmetric tensors. SIAM J Matrix Anal Appl 23(3)
Belongie S, Malik J, Puzicha J (2002) Shape matching and object recognition using shape context. PAMI 24(4):509–522
Kumar S (2005) Models for learning spatial interactions in natural images for context-based classification. PhD thesis, The Robotics Institute, Carnegie Mellon University
Leordeanu M, Hebert M (2009) Unsupervised learning for graph matching. In: CVPR
Guo M, Chou E, Huang DA, Song S, Yeung S, Fei-Fei L (2018) Neural graph matching networks for Fewshot 3d action recognition. In: Proceedings of the European conference on computer vision (ECCV), pp 653–669
Li Y, Gu C, Dullien T, Vinyals O, Kohli P (2019) Graph matching networks for learning the similarity of graph structured objects. arXiv:190412787
Martineau M, Raveaux R, Conte D, Venturini G (2018) Learning error-correcting graph matching with a multiclass neural network. Pattern Recognit Lett
Wang R, Yan J, Yang X (2019) Neural graph matching network: learning Lawler’s quadratic assignment problem with extension to hypergraph and multiple-graph matching. arXiv:191111308
Xu K, Wang L, Yu M, Feng Y, Song Y, Wang Z, Yu D (2019) Cross-lingual knowledge graph alignment via graph matching neural network. arXiv:190511605
Nowak A, Villar S, Bandeira AS, Bruna J (2018) Revised note on learning quadratic assignment with graph neural networks. In: 2018 IEEE data science workshop (DSW). IEEE, pp 1–5
Wang W, Song H, Zhao S, Shen J, Zhao S, Hoi SC, Ling H (2019) Learning unsupervised video object segmentation through visual attention. In: Proceedings of the IEEE conference on computer vision and pattern recognition, pp 3064–3074
Weinzaepfel P, Revaud J, Harchaoui Z, Schmid C (2013) DeepFlow: large displacement optical flow with deep matching. In: ICCV
Author information
Authors and Affiliations
Corresponding author
Rights and permissions
Copyright information
© 2020 Springer Nature Switzerland AG
About this chapter
Cite this chapter
Leordeanu, M. (2020). Unsupervised Learning of Graph and Hypergraph Matching. In: Unsupervised Learning in Space and Time. Advances in Computer Vision and Pattern Recognition. Springer, Cham. https://doi.org/10.1007/978-3-030-42128-1_2
Download citation
DOI: https://doi.org/10.1007/978-3-030-42128-1_2
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-030-42127-4
Online ISBN: 978-3-030-42128-1
eBook Packages: Computer ScienceComputer Science (R0)