Optimal Transport vs Many-to-many assignment for Graph Matching - Archive ouverte HAL Accéder directement au contenu
Communication Dans Un Congrès Année : 2019

Optimal Transport vs Many-to-many assignment for Graph Matching

Résumé

Graph matching for shape comparison or network analysis is a challenging issue in machine learning and computer vision. Gener-ally, this problem is formulated as an assignment task, where we seek the optimal matching between the vertices that minimizes the differencebetween the graphs. We compare a standard approach to perform graph matching, to a slightly-adapted version of regularized optimal transport,initially conceived to obtain the Gromov-Wassersein distance between structured objects (e.g. graphs) with probability masses associated to thenodes. We adapt the latter formulation to undirected and unlabeled graphs of different dimensions, by adding dummy vertices to cast the probleminto an assignment framework. The experiments are performed on randomly generated graphs onto which different spatial transformations areapplied. The results are compared with respect to the matching cost and execution time, showcasing the different limitations and/or advantagesof using these techniques for the comparison of graph networks.
Fichier principal
Vignette du fichier
GRETSI_2019.pdf (319 Ko) Télécharger le fichier
Origine : Fichiers produits par l'(les) auteur(s)
Loading...

Dates et versions

hal-02279634 , version 1 (05-09-2019)

Identifiants

  • HAL Id : hal-02279634 , version 1

Citer

Anca-Ioana Grapa, Laure Blanc-Féraud, Ellen van Obberghen-Schilling, Xavier Descombes. Optimal Transport vs Many-to-many assignment for Graph Matching. GRETSI 2019 - XXVIIème Colloque francophone de traitement du signal et des images, Aug 2019, Lille, France. ⟨hal-02279634⟩
336 Consultations
877 Téléchargements

Partager

Gmail Facebook X LinkedIn More