Florent Perronnin, Chris Dance, Gabriela Csurka, Marco Bressan
ECCV, 9th European Conference on Computer Vision, Graz, Austria, May 7-13, 2006.
Many state of the art Generic visual categorization (GVC) systems are built around a vocabulary of visual terms and characterize images with one histogram of visual word counts. We propose a novel approach to GVC based on a universal vocabulary, which describes the content of all the considered classes of images, and class vocabularies obtained through the adaptation of the universal vocabulary using class-specific data. An image is characterized by a set of histograms - one per class - where each histogram describes whether the image content is best modeled by the universal vocabulary or the corresponding class vocabulary. It is found experimentally that this novel representation greatly improves the performance compared to an approach based on a single vocabulary at the cost of a modest increase in the computation
Report number: