Graph commute times for image representation

Abstract : We introduce a new image representation that encompasses both the general layout of groups of quantized local invariant descriptors as well as their relative frequency. A graph of interest points clusters is constructed and we use the matrix of commute times between the different nodes of the graph to obtain a description of their relative arrangement that is robust to large intra class variation. The obtained high dimensional representation is then embedded in a space of lower dimension by exploiting the spectral properties of the graph made of the different images. Classification tasks can be performed in this embedding space. We expose classification and labelling results obtained on three different datasets, including the challenging PASCAL VOC2007 dataset. The performances of our approach compare favorably with the standard bag of features, which is a particular case of our representation.
Document type :
Conference papers
Liste complète des métadonnées
Contributor : Enzo Ferrante <>
Submitted on : Saturday, December 14, 2013 - 11:49:34 AM
Last modification on : Thursday, April 11, 2019 - 2:34:03 PM




Régis Behmo, Nikos Paragios, Veronique Prinet. Graph commute times for image representation. CVPR 2008 - Computer Vision and Pattern Recognition, Jun 2008, Anchorage, AK, United States. pp.1-8, ⟨10.1109/CVPR.2008.4587840⟩. ⟨hal-00918707⟩



Record views