Triangulation embedding and democratic aggregation for image search - Inria - Institut national de recherche en sciences et technologies du numérique Access content directly
Conference Papers Year : 2014

Triangulation embedding and democratic aggregation for image search

Hervé Jégou
  • Function : Author
  • PersonId : 833473
Andrew Zisserman
  • Function : Author
  • PersonId : 878447

Abstract

We consider the design of a single vector representation for an image that embeds and aggregates a set of local patch descriptors such as SIFT. More specifically we aim to construct a dense representation, like the Fisher Vector or VLAD, though of small or intermediate size. We make two contributions, both aimed at regularizing the individual contributions of the local descriptors in the final representation. The first is a novel embedding method that avoids the dependency on absolute distances by encoding directions. The second contribution is a ''democratization" strategy that further limits the interaction of unrelated descriptors in the aggregation stage. These methods are complementary and give a substantial performance boost over the state of the art in image search with short or mid-size vectors, as demonstrated by our experiments on standard public image retrieval benchmarks.
Fichier principal
Vignette du fichier
cvpr2014_democratic_appendices.pdf (1.56 Mo) Télécharger le fichier
Vignette du fichier
thumb_democratic.png (12.88 Ko) Télécharger le fichier
Origin : Files produced by the author(s)
Format : Figure, Image
Loading...

Dates and versions

hal-00977321 , version 1 (11-04-2014)
hal-00977321 , version 2 (14-04-2014)

Identifiers

  • HAL Id : hal-00977321 , version 2

Cite

Hervé Jégou, Andrew Zisserman. Triangulation embedding and democratic aggregation for image search. CVPR - International Conference on Computer Vision and Pattern Recognition, Jun 2014, Columbus, United States. ⟨hal-00977321v2⟩
8300 View
3307 Download

Share

Gmail Facebook X LinkedIn More