13.3.9 Graph Matching, Continuous Relaxation, Constraint Satisfaction

Barnard, S.T., and Thompson, W.B.,
Disparity Analysis of Images,
PAMI(2), No. 4, July 1980, pp. 333-340. BibRef 8007
Earlier: TR-79-1, CSD, Univ. of MinnesotaJanuary 1979. Relaxation, Results. Matching, Points. Matching for motion. This program finds corresponding pairs of points in a motion analysis system using the similarity of motion with neighboring points. Feature points (such as corners) in both views are used rather than the single view used in Moravec, and a relaxation procedure finds the final global match between the two sets of feature points. The initial assignments of possible matches for the set of feature points is simply all the features with a similar (nearby) position in the second image. Thus, small motions are assumed. An iterative (relaxation based) procedure uses the disparities of the nearby points to eliminate the unlikely assignments from the set of possible assignments. These include points with disparities different from the others in the neighborhood. The formulation of the algorithm is very simple and thus it works for any kind of disparity (such as from observer motion, multiple object motions, or stereo) and it does not require any detailed camera models. This provides a basic matching method to find disparity for a moderate number of points (the feature points) that are generally consistent with the other nearby points (i.e. smooth surfaces), but allowing for edges or changes in the disparity field.
See also Lower-level Estimates and Interpretation of Visual Motion. BibRef

Barnard, S.T.,
The Image Correspondence Problem,
Ph.D.Thesis (CS), U Minn, 1979. The thesis version of his work. BibRef 7900

Kitchen, L.,
Relaxation Applied to Matching Quantitative Relational Structures,
SMC(10), February 1980, pp. 96-101. Fuzzy Logic. Introduction of a new operator defined in terms of fuzzy logic with some examples on synthetic structures. Experiments with the operator on more general problems indicate that there may be problems which are not indicated by the synthetic problems. BibRef 8002

Yamamoto, H.,
Some Experiments on LANDSAT Pixel Classification Using Relaxation Operators,
CGIP(13), No. 1, May 1980, pp. 31-45.
Elsevier DOI BibRef 8005

Kirby, R.L.,
A Product Rule Relaxation Method,
CGIP(13), No. 2, June 1980, pp. 158-189.
Elsevier DOI BibRef 8006

Hwang, J.J., and Hall, E.L.,
Matching of Featured Objects Using Relational Tables from Stereo Images,
CGIP(20), No. 1, September 1982, pp. 22-42.
Elsevier DOI Matching, Regions. Features include regions, lines and vertices. The example is a complex block-like UT. The structure is simply adjacencies. The arrays are used to simplify the search for the matching subset. They use precise knowledge of the camera locations to get search lines in the second image. BibRef 8209

Hwang, J.J., and Hall, E.L.,
Scene Representation Using the Adjacency Matrix and Sampled Shapes of Regions,
PRIP78(250-261). BibRef 7800

Faugeras, O.D., and Price, K.E.,
Semantic Description of Aerial Images Using Stochastic Labeling,
PAMI(3), No. 6, November 1981, pp. 633-642. BibRef 8111 USC Computer Vision BibRef
And: ICPR80(352-357). BibRef
And: DARPA80(89-94). Matching, Regions. Relaxation, Results. The use of an optimization based relaxation method with structural descriptions. This work uses a relaxation approach very similar to that of (
See also Improving Consistency and Reducing Ambiguity in Stochastic Labeling: An Optimization Approach. ) for finding corresponding regions in two images of the same scene and finding regions in the image corresponding to elements in a model of the scene. The relaxation matching procedure has two major steps: finding initial potential matches and computing the updated match rating based on the matches for the neighboring regions. These steps are combined by: (1) Compute the match rating for each region in the model with all regions in the image. Order these and keep only the best (15) matches. (2)Compute the compatibility for each of these possible matches with the current most likely match for all the neighboring (related in the network) regions. (3) Update the match ratings so that compatible matches improve and incompatible ones decrease. (4) If some match is very likely, make the assignment permanent, and continue with the initialization step. Otherwise continue with the compatibility computation step. This procedure works by finding the most obvious match (e.g. largest regions, and all other features match) and building around this one by making assignments to regions related to the obvious match. This matching system makes few assumptions about the types of scenes, though assumptions can be used to improve the efficiency of the match, and is applicable to a variety of tasks.
See also Symbolic Image Registration and Change Detection. BibRef

Price, K.E.,
Hierarchial Matching Using Relaxation,
CVGIP(34), No. 1, April 1986, pp. 66-75.
Elsevier DOI BibRef 8604 USC Computer VisionDiscussion of the use of group level descriptions to aid relaxation. BibRef

Price, K.E.,
Relaxation Matching Techniques: A Comparison,
PAMI(7), No. 5, September 1985, pp. 617-623. BibRef 8509 USC Computer Vision BibRef
And: ICPR84(987-989). Relaxation, Evaluation. Comparison of several relaxation methods, for accuracy and time. BibRef

Price, K.E.,
Symbolic Matching of Images and Scene Models,
DARPA82(299-308). BibRef 8200 USC Computer Vision BibRef
And: CVWS82(105-112). Several discussions on relaxation techniques in one paper. The
See also Relaxation Matching Techniques: A Comparison. and
See also Hierarchial Matching Using Relaxation. supersede this one. BibRef

Price, K.E.,
Relaxation Matching Applied to Aerial Images,
DARPA81(22-25). BibRef 8100 USC Computer VisionDiscussion of more recent results. Not much else. BibRef

Price, K.E.,
Symbolic Matching and Analysis with Substantial Changes in Orientation,
DARPA78(93-99). BibRef 7800 USC Computer Vision BibRef
And: PRAI-78(19-21). BibRef

Hummel, R.A.[Robert A.],
A Design Method for Relaxation Labeling Applications,
AAAI-83(168-171). BibRef 8300
Earlier: NYUCS Dept., TR 68, March 1983. A discussion of how to set up a relaxation labeling system. BibRef

Ogawa, H.,
A Fuzzy Relaxation Technique For Partial Shape-Matching,
PRL(15), No. 4, April 1994, pp. 349-355. BibRef 9404

Qin, C., Luh, J.Y.S.,
Ambiguity Reduction by Relaxation Labeling,
PR(27), No. 1, January 1994, pp. 165-180.
Elsevier DOI BibRef 9401

Ranganath, H.S.[Heggere S.], Chipman, L.J.[Laure J.],
Fuzzy Relaxation Approach for Inexact Scene Matching,
IVC(10), No. 9, November 1992, pp. 631-640.
Elsevier DOI Matching, Regions. BibRef 9211

Cooper, P.R.[Paul R.], Swain, M.J.[Michael J.],
Arc Consistency: Parallelism and Domain Dependence,
AI(58), No. 1-3, 1992, pp. 207-23.5
Elsevier DOI BibRef 9200

Cooper, P.R.[Paul R.], Swain, M.J.[Michael J.],
Domain Dependence in Parallel Constraint Satisfaction,
IJCAI89(54-59). BibRef 8900

Swain, M.J.[Michael J.], Cooper, P.R.[Paul R.],
Parallel Hardware for Constraint Satisfaction,
AAAI-88(682-686). BibRef 8800

Gold, S.[Steven], Rangarajan, A.[Anand],
A Graduated Assignment Algorithm for Graph Matching,
PAMI(18), No. 4, April 1996, pp. 377-388.
IEEE DOI BibRef 9604 YaleDCS/RR-1062, January 1995. BibRef
And:
Graph Matching by Graduated Assignment,
CVPR96(239-244).
IEEE DOI Matching O(lm). Similar to relaxation (annealing) approach. (But not quite). Uses hand labeled features in the image for matching (multiple features on an object). They note that relaxation labeling does poorly on pure subgraph isomorphism (no attributed nodes), and does poorly when noise is high for attributed graph matching. (Though the comparison is with the most basic relaxation methodology.) 9605
BibRef

Gold, S.[Steven],
Matching and Learning Structural and Spatial Representation with Neural Networks,
Ph.D.Thesis, Yale, 1995. BibRef 9500

Gold, S.[Steven], Rangarajan, A.[Anand], and Mjolsness, E.,
Learning with Preknowledge: Clustering with Point and Graph Matching Distance Measures,
NeurComp(8), 1966, pp. 787-804. BibRef 6600

Sitaraman, R.[Ramesh], Rosenfeld, A.[Azriel],
Probabilistic Analysis of Two Stage Matching,
PR(22), No. 3, 1989, pp. 331-343.
Elsevier DOI BibRef 8900

Finch, A.M.[Andrew M.], Wilson, R.C., Hancock, E.R.[Edwin R.],
Matching Delaunay Graphs,
PR(30), No. 1, January 1997, pp. 123-140.
Elsevier DOI 9702
BibRef
Earlier: A1, A3 only: CIAP95(56-61).
Springer DOI 9509
BibRef

Finch, A.M.[Andrew M.], Wilson, R.C.[Richard C.], Hancock, E.R.[Edwin R.],
Matching delaunay triangulations by probabilistic relaxation,
CAIP95(350-358).
Springer DOI 9509
BibRef

Finch, A.M., Hancock, E.R.,
Matching Deformed Delaunay Triangulations,
SCV95(31-36).
IEEE DOI Univ. of York. Relaxation applied to matching graphs composed of triangles. BibRef 9500

Bhattacharya, P.,
Some Remarks on Fuzzy Graphs,
PRL(6), 1987, pp. 297-302. BibRef 8700

Pelillo, M., Fanelli, A.M.,
Autoassociative Learning in Relaxation Labeling Networks,
PRL(18), No. 1, January 1997, pp. 3-12. 9704
BibRef
Earlier: ICPR96(IV: 105-110).
IEEE DOI 9608
(Univ. Ca Foscari Venezia, I) BibRef

Doa, K.H.[Kyeong-Hoon], Kima, Y.S.[Yong-Suk], Uama, T.U.[Tae-Uk], Ha, Y.H.[Yeong-Ho],
Iterative Relaxational Stereo Matching Based on Adaptive Support Between Disparities,
PR(31), No. 8, August 1998, pp. 1049-1059.
Elsevier DOI 9807
Stereo, Matching. BibRef

Skomorowski, M.[Marek],
Use of random graph parsing for scene labelling by probabilistic relaxation,
PRL(20), No. 8, August 1999, pp. 949-956. BibRef 9908

Torsello, A.[Andrea], Pelillo, M.[Marcello],
Continuous-time relaxation labeling processes,
PR(33), No. 11, November 2000, pp. 1897-1908.
Elsevier DOI 0011
BibRef

Medasani, S., Krishnapuram, R., Choi, Y.S.,
Graph Matching by Relaxation of Fuzzy Assignments,
Fuzzy(9), No. 1, 2001, pp. 173-182. BibRef 0100

Bengoetxea, E.[Endika], Larrañaga, P.[Pedro], Bloch, I.[Isabelle], Perchant, A.[Aymeric], Boeres, C.[Claudia],
Inexact graph matching by means of estimation of distribution algorithms,
PR(35), No. 12, December 2002, pp. 2867-2880.
Elsevier DOI 0209
BibRef
Earlier: A1, A2, A3, A4, Only:
Estimation of Distribution Algorithms: A New Evolutionary Computation Approach for Graph Matching Problems,
EMMCVPR01(454-469).
Springer DOI 0205
BibRef

Perchant, A., Bloch, I.,
Graph Fuzzy Homomorphism Interpreted as Fuzzy Association Graphs,
ICPR00(Vol II: 1042-1045).
IEEE DOI 0009
BibRef

Aldea, E.[Emanuel], Fouquier, G.[Geoffroy], Atif, J.[Jamal], Bloch, I.[Isabelle],
Kernel Fusion for Image Classification Using Fuzzy Structural Information,
ISVC07(II: 307-317).
Springer DOI 0711
BibRef

Aldea, E.[Emanuel], Atif, J.[Jamal], Bloch, I.[Isabelle],
Image Classification Using Marginalized Kernels for Graphs,
GbRPR07(103-113).
Springer DOI 0706
BibRef

Fouquier, G.[Geoffroy], Atif, J.[Jamal], Bloch, I.[Isabelle],
Local Reasoning in Fuzzy Attribute Graphs for Optimizing Sequential Segmentation,
GbRPR07(138-147).
Springer DOI 0706
BibRef

Atif, J.[Jamal], Hudelot, C., Bloch, I.[Isabelle],
Explanatory Reasoning for Image Understanding Using Formal Concept Analysis and Description Logics,
SMCS(44), No. 5, May 2014, pp. 552-570.
IEEE DOI 1405
algebra Algebraic erosion over the concept lattice of a background theory.
See also Mathematical morphology on hypergraphs, application to similarity and positive kernel. BibRef

Hudelot, C.[Céline], Atif, J.[Jamal], Bloch, I.[Isabelle],
ALC(F): A New Description Logic for Spatial Reasoning in Images,
CVONT14(370-384).
Springer DOI 1504
BibRef

Cesar, Jr., R.M.[Roberto M.], Bengoetxea, E.[Endika], Bloch, I.[Isabelle], Larrañaga, P.[Pedro],
Inexact graph matching for model-based recognition: Evaluation and comparison of optimization algorithms,
PR(38), No. 11, November 2005, pp. 2099-2113.
Elsevier DOI 0509
BibRef
Earlier: A1, A2, A3, Only:
Inexact graph matching using stochastic optimization techniques for facial feature recognition,
ICPR02(II: 465-468).
IEEE DOI 0211
BibRef

Sminchisescu, C.[Cristian], Triggs, B.[Bill],
Building Roadmaps of Minima and Transitions in Visual Models,
IJCV(61), No. 1, January 2005, pp. 81-101.
DOI Link 0410
BibRef
Earlier:
Building Roadmaps of Local Minima of Visual Models,
ECCV02(I: 566 ff.).
Springer DOI 0205
Avoiding local minima in searching techniques. BibRef

Richards, J.A.[John A.], Jia, X.P.[Xiu-Ping],
A Dempster-Shafer Relaxation Approach to Context Classification,
GeoRS(45), No. 5, May 2007, pp. 1422-1431.
IEEE DOI 0704
BibRef

Schellewald, C.[Christian], Roth, S.[Stefan], Schnorr, C.[Christoph],
Evaluation of a convex relaxation to a quadratic assignment matching approach for relational object views,
IVC(25), No. 8, 1 August 2007, pp. 1301-1314.
Elsevier DOI 0706
Quadratic assignment; Weighted graph matching; Combinatorial optimization; Convex programming; Object recognition BibRef

Schellewald, C.[Christian],
Conves Mathematical Programs for Relational Matching of Object Views,
Ph.D.Thesis, Univ. of Mannhein, 2004. 0905
BibRef

Werner, T.[Tomas],
A Linear Programming Approach to Max-Sum Problem: A Review,
PAMI(29), No. 7, July 2007, pp. 1165-1179.
IEEE DOI 0706
Constraint Satisfaction. Maximization of a sum of binary functions. Explore a formulation from early Russian paper. BibRef

Werner, T.[Tomas],
Revisiting the Linear Programming Relaxation Approach to Gibbs Energy Minimization and Weighted Constraint Satisfaction,
PAMI(32), No. 8, August 2010, pp. 1474-1488.
IEEE DOI 1007
E.g. Gibbs energy minimization, link to constraint programming. BibRef

Werner, T.[Tomas],
High-arity interactions, polyhedral relaxations, and cutting plane algorithm for soft constraint optimisation (MAP-MRF),
CVPR08(1-8).
IEEE DOI 0806
BibRef

Werner, T.[Tomas],
Combinatorial constraints on multiple projections of a set of points,
ICCV03(1011-1016).
IEEE DOI 0311
BibRef

Potetz, B.[Brian], Lee, T.S.[Tai Sing],
Efficient belief propagation for higher-order cliques using linear constraint nodes,
CVIU(112), No. 1, October 2008, pp. 39-54.
Elsevier DOI 0810
BibRef
Earlier: A1, Only:
Efficient Belief Propagation for Vision Using Linear Constraint Nodes,
CVPR07(1-8).
IEEE DOI 0706
Belief propagation; Higher-order cliques; Non-pairwise cliques; Factor graphs; Continuous Markov random fields BibRef

Choi, Y.H.[Young-Hun], Jun, C.H.[Chi-Hyuck],
A causal discovery algorithm using multiple regressions,
PRL(31), No. 13, 1 October 2010, pp. 1924-1934.
Elsevier DOI 1003
Causal discovery; Conditional independence test; Markov blanket; Multiple regression BibRef

Bui, A.T.[Anh Tuan], Jun, C.H.[Chi-Hyuck],
Learning Bayesian network structure using Markov blanket decomposition,
PRL(33), No. 16, 1 December 2012, pp. 2134-2140.
Elsevier DOI 1210
Causal structure learning; Conditional independence test; Directed acyclic graph; Directed global Markov property; Moral graph; V structure BibRef

Pock, T.[Thomas], Cremers, D.[Daniel], Bischof, H.[Horst], Chambolle, A.[Antonin],
Global Solutions Of Variational Models With Convex Regularization,
SIIMS(3), No. 4, 2010, pp. 1122-1145.
WWW Link.
DOI Link BibRef 1000
Earlier: A1, A4, A2, A3:
A convex relaxation approach for computing minimal partitions,
CVPR09(810-817).
IEEE DOI 0906
variational methods; calibrations; total variation; convex optimization BibRef

Chambolle, A.[Antonin], Cremers, D.[Daniel], Pock, T.[Thomas],
A Convex Approach to Minimal Partitions,
SIIMS(5), No. 4, 2012, pp. 1113-1158.
DOI Link 1211
BibRef

Pock, T.[Thomas], Chambolle, A.[Antonin],
Diagonal preconditioning for first order primal-dual algorithms in convex optimization,
ICCV11(1762-1769).
IEEE DOI 1201
BibRef

Goldluecke, B.[Bastian], Cremers, D.[Daniel],
Introducing total curvature for image processing,
ICCV11(1267-1274).
IEEE DOI 1201
Menger-Melnikov curvature of the Radon measure. For regularizer. BibRef

Goldluecke, B.[Bastian], Cremers, D.[Daniel],
Convex Relaxation for Multilabel Problems with Product Label Spaces,
ECCV10(V: 225-238).
Springer DOI 1009
BibRef

Yang, Y.[Yang], Huang, Z.[Zi], Yang, Y.[Yi], Liu, J.J.[Jia-Jun], Shen, H.T.[Heng Tao], Luo, J.B.[Jie-Bo],
Local image tagging via graph regularized joint group sparsity,
PR(46), No. 5, May 2013, pp. 1358-1368.
Elsevier DOI 1302
Local image tagging; Group sparse coding; Graph regularization; Tag propagation BibRef

Ortiz-Bayliss, J.C.[José Carlos], Terashima-Marín, H.[Hugo], Conant-Pablos, S.E.[Santiago Enrique],
Learning vector quantization for variable ordering in constraint satisfaction problems,
PRL(34), No. 4, 1 March 2013, pp. 423-432.
Elsevier DOI 1302
Constraint satisfaction; Hyper-heuristics; Learning vector quantization; Variable and value ordering BibRef

Zach, C.[Christopher], Hane, C.[Christian], Pollefeys, M.[Marc],
What Is Optimized in Convex Relaxations for Multilabel Problems: Connecting Discrete and Continuously Inspired MAP Inference,
PAMI(36), No. 1, 2014, pp. 157-170.
IEEE DOI 1312
BibRef
Earlier:
What is optimized in tight convex relaxations for multi-label problems?,
CVPR12(1664-1671).
IEEE DOI 1208
Markov random fields BibRef

Liu, Z.Y.[Zhi-Yong], Qiao, H.[Hong], Yang, X.[Xu], Hoi, S.C.H.[Steven C. H.],
Graph Matching by Simplified Convex-Concave Relaxation Procedure,
IJCV(109), No. 3, September 2014, pp. 169-186.
Springer DOI 1408
BibRef

Yang, X.[Xu], Qiao, H.[Hong], Liu, Z.Y.[Zhi-Yong],
Feature correspondence based on directed structural model matching,
IVC(33), No. 1, 2015, pp. 57-67.
Elsevier DOI 1412
Feature correspondence BibRef

Yang, X.[Xu], Qiao, H.[Hong], Liu, Z.Y.[Zhi-Yong],
Outlier robust point correspondence based on GNCCP,
PRL(55), No. 1, 2015, pp. 8-14.
Elsevier DOI 1503
Feature correspondence BibRef

Yang, X.[Xu], Liu, Z.Y.[Zhi-Yong], Qiao, H.[Hong],
A Continuation Method for Graph Matching Based Feature Correspondence,
PAMI(42), No. 8, August 2020, pp. 1809-1822.
IEEE DOI 2007
Optimization, Task analysis, Linear programming, Image processing, Pattern matching, Smoothing methods, combinatorial optimization BibRef

Yang, X.[Xu], Liu, Z.Y.[Zhi-Yong],
Adaptive Graph Matching,
Cyber(48), No. 5, May 2018, pp. 1432-1445.
IEEE DOI 1804
Adaptation models, Control systems, Cybernetics, Linear programming, Manganese, Optimization, regularization method BibRef

Yan, J., Wang, J., Zha, H., Yang, X., Chu, S.,
Consistency-Driven Alternating Optimization for Multigraph Matching: A Unified Approach,
IP(24), No. 3, March 2015, pp. 994-1009.
IEEE DOI 1502
BibRef

Åström, F.[Freddie], Petra, S.[Stefania], Schmitzer, B.[Bernhard], Schnörr, C.[Christoph],
Image Labeling by Assignment,
JMIV(58), No. 2, June 2017, pp. 211-238.
Springer DOI 1704
BibRef
Earlier:
A Geometric Approach to Image Labeling,
ECCV16(V: 139-154).
Springer DOI 1611
BibRef
And:
The Assignment Manifold: A Smooth Model for Image Labeling,
DIFF-CV16(963-971)
IEEE DOI 1612
BibRef

Zern, A.[Artjom], Zisler, M.[Matthias], Petra, S.[Stefania], Schnörr, C.[Christoph],
Unsupervised Assignment Flow: Label Learning on Feature Manifolds by Spatially Regularized Geometric Assignment,
JMIV(62), No. 6-7, July 2020, pp. 982-1006.
Springer DOI 2007
BibRef
Earlier: A2, A1, A3, A4:
Unsupervised Labeling by Geometric and Spatially Regularized Self-assignment,
SSVM19(432-444).
Springer DOI 1909
BibRef

Zisler, M.[Matthias], Zern, A.[Artjom], Petra, S.[Stefania], Schnörr, C.[Christoph],
Self-Assignment Flows for Unsupervised Data Labeling on Graphs,
SIIMS(13), No. 3, 2020, pp. 1113-1156.
DOI Link 2010
BibRef

Zern, A.[Artjom], Zisler, M.[Matthias], Åström, F.[Freddie], Petra, S.[Stefania], Schnörr, C.[Christoph],
Unsupervised Label Learning on Manifolds by Spatially Regularized Geometric Assignment,
GCPR18(698-713).
Springer DOI 1905
BibRef

Hühnerbein, R.[Ruben], Savarino, F.[Fabrizio], Åström, F.[Freddie], Schnörr, C.[Christoph],
Image Labeling Based on Graphical Models Using Wasserstein Messages and Geometric Assignment,
SIIMS(11), No. 2, 2018, pp. 1317-1362.
DOI Link 1807
BibRef

Hühnerbein, R.[Ruben], Savarino, F.[Fabrizio], Petra, S.[Stefania], Schnörr, C.[Christoph],
Learning Adaptive Regularization for Image Labeling Using Geometric Assignment,
SSVM19(393-405).
Springer DOI 1909
BibRef

Åström, F.[Freddie], Hühnerbein, R.[Ruben], Savarino, F.[Fabrizio], Recknagel, J.[Judit], Schnörr, C.[Christoph],
MAP Image Labeling Using Wasserstein Messages and Geometric Assignment,
SSVM17(373-385).
Springer DOI 1706
BibRef

Savarino, F.[Fabrizio], Schnörr, C.[Christoph],
A Variational Perspective on the Assignment Flow,
SSVM19(547-558).
Springer DOI 1909
BibRef

Savarino, F.[Fabrizio], Hühnerbein, R.[Ruben], Åström, F.[Freddie], Recknagel, J.[Judit], Schnörr, C.[Christoph],
Numerical Integration of Riemannian Gradient Flows for Image Labeling,
SSVM17(361-372).
Springer DOI 1706
BibRef

Sitenko, D.[Dmitrij], Boll, B.[Bastian], Schnorr, C.[Christoph],
A Nonlocal Graph-PDE and Higher-Order Geometric Integration for Image Labeling,
SIIMS(16), No. 1, 2023, pp. 501-567.
DOI Link 2305

See also Image Labeling by Assignment. BibRef

Boll, B.[Bastian], Schwarz, J.[Jonathan], Gonzalez-Alvarado, D.[Daniel], Sitenko, D.[Dmitrij], Petra, S.[Stefania], Schnörr, C.[Christoph],
Modeling Large-scale Joint Distributions and Inference by Randomized Assignment,
SSVM23(730-742).
Springer DOI 2307
BibRef

Schwarz, J.[Jonathan], Boll, B.[Bastian], Gonzalez-Alvarado, D.[Daniel], Sitenko, D.[Dmitrij], Gärttner, M.[Martin], Albers, P.[Peter], Schnörr, C.[Christoph],
Quantum State Assignment Flows,
SSVM23(743-756).
Springer DOI 2307
BibRef

Lin, G.F.[Guang-Feng], Liao, K.Y.[Kai-Yang], Sun, B.Y.[Bang-Yong], Chen, Y.J.[Ya-Jun], Zhao, F.[Fan],
Dynamic graph fusion label propagation for semi-supervised multi-modality classification,
PR(68), No. 1, 2017, pp. 14-23.
Elsevier DOI 1704
Dynamic graph fusion BibRef

Pruša, D.[Daniel], Werner, T.[Tomáš],
LP Relaxation of the Potts Labeling Problem Is as Hard as Any Linear Program,
PAMI(39), No. 7, July 2017, pp. 1469-1475.
IEEE DOI 1706
BibRef
Earlier:
How Hard Is the LP Relaxation of the Potts Min-Sum Labeling Problem?,
EMMCVPR15(57-70).
Springer DOI 1504
Approximation algorithms, Computational modeling, Cost function, Graphical models, Labeling, Measurement, Minimization, MAP inference, Markov random field, Potts model, discrete energy minimization, graphical model, linear programming relaxation, uniform metric labeling problem, valued constraint satisfaction. BibRef

Bergmann, R.[Ronny], Fitschen, J.H.[Jan Henrik], Persch, J.[Johannes], Steidl, G.[Gabriele],
Iterative Multiplicative Filters for Data Labeling,
IJCV(123), No. 3, July 2017, pp. 435-453.
Springer DOI 1706
for the supervised partitioning of data Derived from:
See also Image Labeling by Assignment. BibRef

Magri, L.[Luca], Fusiello, A.[Andrea],
Multiple structure recovery via robust preference analysis,
IVC(67), No. 1, 2017, pp. 1-15.
Elsevier DOI 1710
BibRef
Earlier:
Multiple Models Fitting as a Set Coverage Problem,
CVPR16(3318-3326)
IEEE DOI 1612
BibRef
Earlier:
Robust Multiple Model Fitting with Preference Analysis and Low-rank Approximation,
BMVC15(xx-yy).
DOI Link 1601
BibRef
And:
Fitting Multiple Models via Density Analysis in Tanimoto Space,
CIAP15(I:73-84).
Springer DOI 1511
BibRef
And:
Scale Estimation in Multiple Models Fitting via Consensus Clustering,
CAIP15(II:13-25).
Springer DOI 1511
BibRef
And:
T-Linkage: A Continuous Relaxation of J-Linkage for Multi-model Fitting,
CVPR14(3954-3961)
IEEE DOI 1409
Multi-model fitting BibRef

Magri, L.[Luca], Fusiello, A.[Andrea],
Multiple structure recovery with maximum coverage,
MVA(29), No. 1, January 2018, pp. 159-173.
WWW Link. 1801
BibRef

Zoidi, O.[Olga], Tefas, A., Nikolaidis, N.[Nikos], Pitas, I.[Ioannis],
Positive and Negative Label Propagations,
CirSysVideo(28), No. 2, February 2018, pp. 342-355.
IEEE DOI 1802
BibRef
Earlier: A1, A3, A4, Only:
Label propagation on data with multiple representations through multi-graph locality preserving projections,
ICIP14(1505-1509)
IEEE DOI 1502
Cost function, Face recognition, Laplace equations, Manifolds, Semisupervised learning, Training, Action recognition, label propagation (LP). Accuracy BibRef

Uzun, A.O.[Arif Orhun], Usta, T.[Tugba], Dündar, E.B.[Enes Burak], Korkmaz, E.E.[Emin Erkan],
A solution to the classification problem with cellular automata,
PRL(116), 2018, pp. 114-120.
Elsevier DOI 1812
Classification, Cellular automata, Heat Transfer, Big Data BibRef

Nassif, R., Vlaski, S., Richard, C., Sayed, A.H.,
A Regularization Framework for Learning Over Multitask Graphs,
SPLetters(26), No. 2, February 2019, pp. 297-301.
IEEE DOI 1902
approximation theory, gradient methods, graph theory, inference mechanisms, learning (artificial intelligence), distributed implementation BibRef

Ji, W.[Wei], Li, X.[Xi], Wei, L.[Lina], Wu, F.[Fei], Zhuang, Y.T.[Yue-Ting],
Context-Aware Graph Label Propagation Network for Saliency Detection,
IP(29), 2020, pp. 8177-8186.
IEEE DOI 2008
saliency detection, superpixel pooling, graph neural network BibRef

Hoang, T., Do, T., Nguyen, T.V., Cheung, N.,
Unsupervised Deep Cross-modality Spectral Hashing,
IP(29), 2020, pp. 8391-8406.
IEEE DOI 2008
Binary codes, Semantics, Optimization, Correlation, Sparse matrices, Task analysis, Training data, Cross-modal retrieval, constraint optimization BibRef

Lim, K.L.[Kart-Leong], Jiang, X.D.[Xu-Dong],
Variational posterior approximation using stochastic gradient ascent with adaptive stepsize,
PR(112), 2021, pp. 107783.
Elsevier DOI 2102
Dirichlet process mixture, Stochastic gradient ascent, Fisher information, Scalable algorithm BibRef

Strecke, M.[Michael], Goldluecke, B.[Bastian],
Sublabel-Accurate Convex Relaxation with Total Generalized Variation Regularization,
GCPR18(263-277).
Springer DOI 1905
BibRef

Lê-Huu, D.K., Paragios, N.,
Continuous Relaxation of MAP Inference: A Nonconvex Perspective,
CVPR18(5533-5541)
IEEE DOI 1812
Convex functions, Markov processes, Optimization, Indexes. BibRef

Dong, J.X.[Jiang-Xin], Liu, R.S.[Ri-Sheng], Tang, K.W.[Ke-Wei], Wang, Y.Y.[Yi-Yang], Zhang, X.D.[Xin-Dong], Su, Z.X.[Zhi-Xun],
Sparse Gradient Pursuit for Robust Visual Analysis,
ACCV16(I: 369-384).
Springer DOI 1704
BibRef

Li, D.[Dong], Hung, W.C.[Wei-Chih], Huang, J.B.[Jia-Bin], Wang, S.J.[Sheng-Jin], Ahuja, N.[Narendra], Yang, M.H.[Ming-Hsuan],
Unsupervised Visual Representation Learning by Graph-Based Consistent Constraints,
ECCV16(IV: 678-694).
Springer DOI 1611
BibRef

Kim, K.I.[Kwang In], Tompkin, J.[James], Pfister, H.[Hanspeter], Theobalt, C.[Christian],
Context-Guided Diffusion for Label Propagation on Graphs,
ICCV15(2776-2784)
IEEE DOI 1602
Anisotropic magnetoresistance BibRef

Souiai, M.[Mohamed], Oswald, M.R.[Martin R.], Keef, Y.[Youngwook], Kim, J.[Junmo], Pollefeys, M.[Marc], Cremers, D.[Daniel],
Entropy Minimization for Convex Relaxation Approaches,
ICCV15(1778-1786)
IEEE DOI 1602
Computer vision BibRef

Zhang, Z.[Zhao], Li, F.Z.[Fan-Zhang], Zhao, M.B.[Ming-Bo],
Transformed Neighborhood Propagation,
ICPR14(3792-3797)
IEEE DOI 1412
Control charts BibRef

Oswald, M.R.[Martin R.], Cremers, D.[Daniel],
Surface Normal Integration for Convex Space-time Multi-view Reconstruction,
BMVC14(xx-yy).
HTML Version. 1410
BibRef
Earlier:
A Convex Relaxation Approach to Space Time Multi-view 3D Reconstruction,
4DMOD13(291-298)
IEEE DOI 1403
image reconstruction BibRef

Stuhmer, J.[Jan], Schroder, P.[Peter], Cremers, D.[Daniel],
Tree Shape Priors with Connectivity Constraints Using Convex Relaxation on General Graphs,
ICCV13(2336-2343)
IEEE DOI 1403
Medical Imaging; Optimization; Segmentation BibRef

Wang, B.[Bo], Tsotsos, J.K.[John K.],
Dynamic Label Propagation for Semi-supervised Multi-class Multi-label Classification,
PR(52), No. 1, 2016, pp. 75-84.
Elsevier DOI 1601
Dynamic label propagation BibRef
Earlier: Add A2: Tu, Z.W.[Zhuo-Wen], ICCV13(425-432)
IEEE DOI 1403
Dynamic Label Propagation; Multi-class; Multi-label BibRef

Ebert, S.[Sandra], Fritz, M.[Mario], Schiele, B.[Bernt],
Pick Your Neighborhood: Improving Labels and Neighborhood Structure for Label Propagation,
DAGM11(152-162).
Springer DOI 1109
propogate labels on graph in learning. BibRef

Kasprzak, W.[Wlodzimierz], Czajka, L.[Lukasz], Wilkowski, A.[Artur],
A Constraint Satisfaction Framework with Bayesian Inference for Model-Based Object Recognition,
ICCVG10(II: 1-8).
Springer DOI 1009
BibRef

Pawan Kumar, M., Torr, P.H.S.,
Fast Memory-Efficient Generalized Belief Propagation,
ECCV06(IV: 451-463).
Springer DOI 0608
BibRef

Coito, F.J.[Fernando J.], Lemos, J.M.[João M.],
Adaptive Optimization with Constraints: Convergence and Oscillatory Behaviour,
IbPRIA05(II:19).
Springer DOI 0509
BibRef

Yuille, A.L.[Alan L.],
A Double-Loop Algorithm to Minimize the Bethe Free Energy,
EMMCVPR01(3-18).
Springer DOI 0205
BibRef
Earlier:
A Double-Loop Algorithm to Minimize the Bethe and Kikuchi Free Energies,
SCTV01(xx-yy). 0106
BibRef

Yedidia, J., Freeman, W.T., Weiss, Y.,
Bethe free energy, Kikuchi approximations, and belief propagation algorithms,
SCTV01(xx-yy). 0106
Stable points of belief propagation algorithms for graphs with loops correspond to extrema of the Bethe free energy. BibRef

Haddon, J., Boyce, J.,
Spatio-Temporal Relaxation Labelling Applied to Segmented Infrared Image Sequences,
ICPR96(II: 171-175).
IEEE DOI 9608
(Defence Res. Agency, UK) BibRef

Horiuchi, T., Yamamoto, K., Yamada, H.,
Robust Relaxation Method for Structural Matching Under Uncertainty,
ICPR96(II: 176-180).
IEEE DOI 9608
(Univ. of Tsukuba, J) BibRef

Shao, Z., Kittler, J.V.,
Fuzzy Non-Iterative ARG Labeling with Multiple Interpretations,
ICPR96(II: 181-185).
IEEE DOI 9608
(Univ. of Surrey, UK) BibRef

Hatef, M., Kittler, J.V.,
Combining symbolic with numeric attributes in multi-class object recognition problems,
ICIP95(III: 364-367).
IEEE DOI 9510
BibRef

Choate, J.A., Gennert, M.A.,
Multiscale relaxation labeling of fractal images,
CVPR93(674-675).
IEEE DOI 0403
BibRef

McLean, C.R., Dyer, C.R.,
An Analog Relaxation Processor,
ICPR80(58-60). BibRef 8000

Chapter on Matching and Recognition Using Volumes, High Level Vision Techniques, Invariants continues in
Continuous Relaxation Theory, Constraint Satisfaction .