Negative evidences and co-occurrences in image retrieval: the benefit of PCA and whitening - Inria - Institut national de recherche en sciences et technologies du numérique Access content directly
Conference Papers Year : 2012

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

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.
Fichier principal
Vignette du fichier
jegou_chum_eccv2012.pdf (290.07 Ko) Télécharger le fichier
Vignette du fichier
thumb.png (11.93 Ko) Télécharger le fichier
Origin : Files produced by the author(s)
Format : Figure, Image
Loading...

Dates and versions

hal-00722622 , version 1 (02-08-2012)
hal-00722622 , version 2 (02-08-2012)

Identifiers

  • HAL Id : hal-00722622 , version 2

Cite

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. ⟨hal-00722622v2⟩
1444 View
3027 Download

Share

Gmail Facebook X LinkedIn More