Relations explicites entre différentes représentations d'image dans un modèle de graphe visuel

This paper presents a novel approach, the first to our knowledge, that exploits a complete extension of the language modeling approach from information retrieval to the problem of graph-based image retrieval and categorization. Since photographic images are 2D data, we first use image regions and local interest points (mapped to automatically induced concepts) and then relationships between these regions to build a complete graph representation of images. The results obtained on categorizing of RobotVision collection from ImageCLEF 2009 (containing of 5 rooms in an indoor environment) show that (a) the procedure to automatically induce concepts from an image is effective, and (b) the use of spatial relationships, in addition to concepts, for representing an image content helps improve the classifier accuracy. MOTS-CLÉS : Représentation de graphes, recherche d’image, catégorisation d’image