Kernel on Graphs based on Dictionary of Paths for Image Retrieval.

Jean-Emmanuel Haugeard (1) , Sylvie Philipp-Foliguet (1) , Philippe-Henri Gosselin (2)

1 ETIS - UMR 8051 - Equipes Traitement de l'Information et Systèmes
2 MIDI - Multimedia Indexation and Data Integration

Recent approaches of graph comparison consider graphs as sets of paths. Kernels on graphs are then computed from kernels on paths. A common strat- egy for graph retrieval is to perform pairwise compar- isons. In this paper, we propose to follow a different strategy, where we collect a set of paths into a dictio- nary, and then project each graph to this dictionary. Then, graphs can be classified using powerful classifi- cation methods, such as SVM. Furthermore, we collect the paths through interaction with a user. This strategy is ten times faster than a straight comparisons of paths. Experiments have been carried out on a database of city windows.

Domaines: Machine Learning [stat.ML]

Jean-Emmanuel Haugeard, Sylvie Philipp-Foliguet, Philippe-Henri Gosselin. Kernel on Graphs based on Dictionary of Paths for Image Retrieval.. International Conference on Pattern Recognition, Aug 2010, Turkey. pp.1. ⟨hal-00520295⟩