Content based image retrieval using bag-of-regions

Abstract

In this work we introduce the Bag-Of-Regions model, inspired from the Bag-Of-Visual-Words. Instead of clustering local image patches represented by SIFT or related descriptors, low level descriptors are extracted and clustered from image regions, as given by a segmentation algorithm. The Bag-Of-Region model allows to define visual dictionaries that capture extra information with respect to Bag-Of-Visual-Words. Combined description schemes and ad-hoc incremental clustering for visual dictionnaries are proposed. The results on public datasets are promising. © 2012 Springer-Verlag.