Negative evidences and co-occurrences in image retrieval: the benefit of PCA and whitening

Hervé Jégou 1 Ondrej Chum 2
1 TEXMEX - Multimedia content-based indexing
IRISA - Institut de Recherche en Informatique et Systèmes Aléatoires, Inria Rennes – Bretagne Atlantique
Abstract : The paper addresses large scale image retrieval with short vector representations. We study dimensionality reduction by Principal Component Analysis (PCA) and propose improvements to its different phases. We show and explicitly exploit relations between i) mean subtraction and the negative evidence, i.e., a visual word that is mutually missing in two descriptions being compared, and ii) the axis de-correlation and the co-occurrences phenomenon. Finally, we propose an effective way to alleviate the quantization artifacts through a joint dimensionality reduction of multiple vocabularies. The proposed techniques are simple, yet significantly and consistently improve over the state of the art on compact image representations. Complementary experiments in image classification show that the methods are generally applicable.
Type de document :
Communication dans un congrès
ECCV - European Conference on Computer Vision, Oct 2012, Firenze, Italy. 2012
Liste complète des métadonnées



https://hal.inria.fr/hal-00722622
Contributeur : Hervé Jégou <>
Soumis le : jeudi 2 août 2012 - 16:39:09
Dernière modification le : vendredi 13 janvier 2017 - 14:20:20
Document(s) archivé(s) le : vendredi 16 décembre 2016 - 04:44:01

Fichiers

jegou_chum_eccv2012.pdf
Fichiers produits par l'(les) auteur(s)

Identifiants

  • HAL Id : hal-00722622, version 2

Collections

Citation

Hervé Jégou, Ondrej Chum. Negative evidences and co-occurrences in image retrieval: the benefit of PCA and whitening. ECCV - European Conference on Computer Vision, Oct 2012, Firenze, Italy. 2012. <hal-00722622v2>

Partager

Métriques

Consultations de
la notice

846

Téléchargements du document

1483