Aggregating local descriptors into a compact image representation - Archive ouverte HAL Access content directly
Conference Papers Year : 2010

Aggregating local descriptors into a compact image representation

(1) , (2) , (2) , (3)
1
2
3
Hervé Jégou
  • Function : Author
  • PersonId : 833473
Matthijs Douze
  • Function : Author
  • PersonId : 843109
Cordelia Schmid
  • Function : Author
  • PersonId : 831154
Patrick Pérez
  • Function : Author
  • PersonId : 1022281

Abstract

We address the problem of image search on a very large scale, where three constraints have to be considered jointly: the accuracy of the search, its efficiency, and the memory usage of the representation. We first propose a simple yet efficient way of aggregating local image descriptors into a vector of limited dimension, which can be viewed as a simplification of the Fisher kernel representation. We then show how to jointly optimize the dimension reduction and the indexing algorithm, so that it best preserves the quality of vector comparison. The evaluation shows that our approach significantly outperforms the state of the art: the search accuracy is comparable to the bag-of-features approach for an image representation that fits in 20 bytes. Searching a 10 million image dataset takes about 50ms.
Vignette du fichier
fig2.jpg (15.64 Ko) Télécharger le fichier Fichier principal
Vignette du fichier
jegou_compactimagerepresentation.pdf (1.64 Mo) Télécharger le fichier
Vignette du fichier
jegou_compactimagerepresentation_slides.pdf (730.62 Ko) Télécharger le fichier
Format : Figure, Image
Origin : Files produced by the author(s)
Format : Other

Dates and versions

inria-00548637 , version 1 (20-12-2010)

Identifiers

Cite

Hervé Jégou, Matthijs Douze, Cordelia Schmid, Patrick Pérez. Aggregating local descriptors into a compact image representation. CVPR 2010 - 23rd IEEE Conference on Computer Vision & Pattern Recognition, Jun 2010, San Francisco, United States. pp.3304-3311, ⟨10.1109/CVPR.2010.5540039⟩. ⟨inria-00548637⟩
1789 View
8147 Download

Altmetric

Share

Gmail Facebook Twitter LinkedIn More