Multilevel Image Coding with Hyperfeatures

Ankur Agarwal 1 William Triggs 2, 3
2 LEAR - Learning and recognition in vision
Inria Grenoble - Rhône-Alpes, LJK - Laboratoire Jean Kuntzmann, INPG - Institut National Polytechnique de Grenoble
3 AI - Artificial Intelligence
LJK - Laboratoire Jean Kuntzmann
Abstract : Histograms of local appearance descriptors are a popular representation for visual recognition. They are highly discriminant with good resistance to local occlusions and to geometric and photometric variations, but they are not able to exploit spatial co-occurrence statistics over scales larger than the local input patches. We present a multilevel visual representation that remedies this. The starting point is the notion that to detect object parts in images, in practice it often suffices to detect co-occurrences of more local object fragments. This can be formalized by coding image patches against a codebook of known fragments or a more general statistical model and locally histogramming the resulting labels to capture their co-occurrence statistics. Local patch descriptors are converted into somewhat less local histograms over label occurrences. The histograms are themselves local descriptor vectors so the process can be iterated to code ever larger assemblies of object parts and increasingly abstract or 'semantic' image properties. We call these higher-level descriptors "hyperfeatures". We formulate the hyperfeature model and study its performance under several different image coding methods including k-means based Vector Quantization, Gaussian Mixtures, and combinations of these with Latent Dirichlet Allocation. We find that the resulting highlevel features provide improved performance in several object image and texture image classification tasks.
Type de document :
Article dans une revue
International Journal of Computer Vision, Springer Verlag, 2008, 78 (1), pp.15-27. <10.1007/s11263-007-0072-x>
Liste complète des métadonnées


https://hal.archives-ouvertes.fr/hal-00192599
Contributeur : William Triggs <>
Soumis le : mercredi 28 novembre 2007 - 17:33:12
Dernière modification le : mercredi 29 juillet 2015 - 01:19:24
Document(s) archivé(s) le : jeudi 27 septembre 2012 - 10:26:11

Fichier

ijcv06.pdf
Fichiers produits par l'(les) auteur(s)

Identifiants

Collections

Citation

Ankur Agarwal, William Triggs. Multilevel Image Coding with Hyperfeatures. International Journal of Computer Vision, Springer Verlag, 2008, 78 (1), pp.15-27. <10.1007/s11263-007-0072-x>. <hal-00192599>

Partager

Métriques

Consultations de
la notice

452

Téléchargements du document

402