Skip to Main content Skip to Navigation
Journal articles

Pairwise Identity Verification via Linear Concentrative Metric Learning

Lilei Zheng 1 Stefan Duffner 1 Khalid Idrissi 1 Christophe Garcia 1 Atilla Baskurt 1 
1 imagine - Extraction de Caractéristiques et Identification
LIRIS - Laboratoire d'InfoRmatique en Image et Systèmes d'information
Abstract : This paper presents a study of metric learning systems on pairwise identity verification, including pairwise face verification and pairwise speaker verification, respectively. These problems are challenging because the individuals in training and testing are mutually exclusive, and also due to the probable setting of limited training data. For such pairwise verification problems, we present a general framework of metric learning systems and employ the stochastic gradient descent algorithm as the optimization solution. We have studied both similarity metric learning and distance metric learning systems, of either a linear or shallow nonlinear model under both restricted and unrestricted training settings. Extensive experiments demonstrate that with limited training pairs, learning a linear system on similar pairs only is preferable due to its simplicity and superiority, i.e. it generally achieves competitive performance on both the LFW face dataset and the NIST speaker dataset. It is also found that a pre-trained deep nonlinear model helps to improve the face verification results significantly.
Complete list of metadata

Cited literature [73 references]  Display  Hide  Download
Contributor : Stefan Duffner Connect in order to contact the contributor
Submitted on : Friday, January 13, 2017 - 7:26:23 PM
Last modification on : Friday, September 30, 2022 - 11:34:16 AM
Long-term archiving on: : Friday, April 14, 2017 - 7:34:01 PM


Files produced by the author(s)



Lilei Zheng, Stefan Duffner, Khalid Idrissi, Christophe Garcia, Atilla Baskurt. Pairwise Identity Verification via Linear Concentrative Metric Learning. IEEE Transactions on Cybernetics, IEEE, 2018, pp.1 - 12. ⟨10.1109/TCYB.2016.2634011⟩. ⟨hal-01435368⟩



Record views


Files downloads