Skip to Main content Skip to Navigation
Conference papers

Coordinated Local Metric Learning

Shreyas Saxena 1 Jakob Verbeek 1
1 LEAR - Learning and recognition in vision
Inria Grenoble - Rhône-Alpes, LJK - Laboratoire Jean Kuntzmann, Grenoble INP - Institut polytechnique de Grenoble - Grenoble Institute of Technology
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.
Complete list of metadata

Cited literature [41 references]  Display  Hide  Download
Contributor : Shreyas Saxena Connect in order to contact the contributor
Submitted on : Tuesday, October 13, 2015 - 5:54:16 PM
Last modification on : Thursday, January 20, 2022 - 5:30:18 PM
Long-term archiving on: : Thursday, April 27, 2017 - 4:11:20 AM


Files produced by the author(s)




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⟩



Les métriques sont temporairement indisponibles