Coordinated Local Metric Learning - Archive ouverte HAL Access content directly
Conference Papers Year :

Coordinated Local Metric Learning

(1) , (1)
1
Shreyas Saxena
  • Function : Author
  • PersonId : 778901
  • IdRef : 223376027
Jakob Verbeek

Abstract

Mahalanobis metric learning amounts to learning a linear data projection, after which the L2 metric is used to compute distances. To allow more flexible metrics, not restricted to linear projections, local metric learning techniques have been developed. Most of these methods partition the data space using clustering, and for each cluster a separate metric is learned. Using local metrics, however, it is not clear how to measure distances between data points assigned to different clusters. In this paper we propose to embed the local metrics in a global low-dimensional representation, in which the L2 metric can be used. With each cluster we associate a linear mapping that projects the data to the global representation. This global representation directly allows computing distances between points regardless to which local cluster they belong. Moreover, it also enables data visualization in a single view, and the use of L2 based efficient retrieval methods. Experiments on the Labeled Faces in the Wild dataset show that our approach improves over previous global and local metric learning approaches.
Fichier principal
Vignette du fichier
paper.pdf (2.47 Mo) Télécharger le fichier
Vignette du fichier
thumbnail.jpg (25.17 Ko) Télécharger le fichier
Origin : Files produced by the author(s)
Format : Figure, Image
Origin : Files produced by the author(s)
Loading...

Dates and versions

hal-01215272 , version 1 (13-10-2015)

Identifiers

Cite

Shreyas Saxena, Jakob Verbeek. Coordinated Local Metric Learning. ICCV ChaLearn Looking at People workshop, Dec 2015, Santiago, Chile. pp.369-377, ⟨10.1109/ICCVW.2015.56⟩. ⟨hal-01215272⟩
538 View
1178 Download

Altmetric

Share

Gmail Facebook Twitter LinkedIn More