Abstract
A new formulation of metric learning is introduced by assimilating the kernel ridge regression (KRR) and weighted side-information linear discriminant analysis (WSILD) to enjoy the best of both worlds for unconstrained face verification task. To be specific, we formulate a doublet constrained metric learning problem by means of a second degree polynomial kernel function. The said metric learning problem can be solved analytically for Mahalanobis distance metric due to simplistic nature of KRR in which we named KRRML. In addition, the WSILD further enhances the learned Mahalanobis distance metric by leveraging the within-class and between-class scatter matrix of doublets. We evaluate the proposed method with Labeled Faces in the Wild database, a large benchmark dataset targeted for unconstrained face verification. The promising result attests the robustness and feasibility of the proposed method.
Original language | English |
---|---|
Title of host publication | Neural Information Processing - 23rd International Conference, ICONIP 2016, Proceedings |
Editors | Seiichi Ozawa, Kazushi Ikeda, Derong Liu, Akira Hirose, Kenji Doya, Minho Lee |
Publisher | Springer Verlag |
Pages | 401-410 |
Number of pages | 10 |
ISBN (Print) | 9783319466712 |
DOIs | |
Publication status | Published - 2016 |
Event | 23rd International Conference on Neural Information Processing, ICONIP 2016 - Kyoto, Japan Duration: 2016 Oct 16 → 2016 Oct 21 |
Publication series
Name | Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) |
---|---|
Volume | 9948 LNCS |
ISSN (Print) | 0302-9743 |
ISSN (Electronic) | 1611-3349 |
Other
Other | 23rd International Conference on Neural Information Processing, ICONIP 2016 |
---|---|
Country/Territory | Japan |
City | Kyoto |
Period | 16/10/16 → 16/10/21 |
Bibliographical note
Funding Information:The authors would like to thank Malaysia’s Fundamental Research Grant Scheme for supporting the research under grants MMUE/140026.
Publisher Copyright:
© Springer International Publishing AG 2016.
All Science Journal Classification (ASJC) codes
- Theoretical Computer Science
- Computer Science(all)