Discriminative Classification vs Modeling Methods in CBIR

Philippe-Henri Gosselin (1) , Micheline Najjar (2) , Matthieu Cord (3) , Christophe Ambroise (4)

1 MIDI - Multimedia Indexation and Data Integration
2 Heudiasyc - Heuristique et Diagnostic des Systèmes Complexes [Compiègne]
3 ETIS - UMR 8051 - Equipes Traitement de l'Information et Systèmes
4 SG - Laboratoire Statistique et Génome

Statistical learning methods are currently considered with an increasing interest in the content-based image retrieval (CBIR) community. We compare in this article two leader techniques for classification tasks. The first method uses one-class and two-class SVM to discriminate data. The second approach is based on Gaussian Mixture to model classes. To deal with the specificity of the CBIR classifica- tion task, adaptations have been proposed. Experimental tests on a generalist database have been carried out. Ad- vantages and drawbacks are discussed for each method.

Domains: Machine Learning [stat.ML]

Philippe-Henri Gosselin, Micheline Najjar, Matthieu Cord, Christophe Ambroise. Discriminative Classification vs Modeling Methods in CBIR. IEEE International Conference on Advanced Concepts for Intelligent Vision Systems, Sep 2004, Belgium. pp.1. ⟨hal-00520316⟩