Learning a Distance Metric from Relative Comparisons between Quadruplets of Images - Sorbonne Université Access content directly
Journal Articles International Journal of Computer Vision Year : 2016

Learning a Distance Metric from Relative Comparisons between Quadruplets of Images

Marc T. Law
  • Function : Correspondent author
Nicolas Thome
Matthieu Cord

Abstract

This paper is concerned with the problem of learning a distance metric by considering meaningful and dis-criminative distance constraints in some contexts where rich information between data is provided. Classic metric learning approaches focus on constraints that involve pairs or triplets of images. We propose a general Mahalanobis-like distance metric learning framework that exploits distance constraints over up to four different images. We show how the integration of such constraints can lead to unsupervised or semi-supervised learning tasks in some applications. We also show the benefit on recognition performance of this type of constraints, in rich contexts such as relative attributes, class taxonomies and temporal webpage analysis.
Fichier principal
Vignette du fichier
Law_2016_Learning_a_Distance.pdf (9.37 Mo) Télécharger le fichier
Origin : Files produced by the author(s)
Loading...

Dates and versions

hal-01346190 , version 1 (18-07-2016)

Identifiers

Cite

Marc T. Law, Nicolas Thome, Matthieu Cord. Learning a Distance Metric from Relative Comparisons between Quadruplets of Images. International Journal of Computer Vision, 2016, pp.1-30. ⟨10.1007/s11263-016-0923-4⟩. ⟨hal-01346190⟩
235 View
334 Download

Altmetric

Share

Gmail Facebook X LinkedIn More