Additive Margin Softmax for Face Verification

In this letter, we propose a conceptually simple and intuitive learning objective function, i.e., additive margin softmax, for face verification. In general, face verification tasks can be viewed as metric learning problems, even though lots of face verification models are trained in classification...

Full description

Saved in:
Bibliographic Details
Published inIEEE signal processing letters Vol. 25; no. 7; pp. 926 - 930
Main Authors Wang, Feng, Cheng, Jian, Liu, Weiyang, Liu, Haijun
Format Journal Article
LanguageEnglish
Published IEEE 01.07.2018
Subjects
Additives
Annealing
Deep learning
Face
face verification
Measurement
metric learning
Task analysis
Training
Visualization
Online AccessGet full text

Cover

More Information
Summary:In this letter, we propose a conceptually simple and intuitive learning objective function, i.e., additive margin softmax, for face verification. In general, face verification tasks can be viewed as metric learning problems, even though lots of face verification models are trained in classification schemes. It is possible when a large-margin strategy is introduced into the classification model to encourage intraclass variance minimization. As one alternative, angular softmax has been proposed to incorporate the margin. In this letter, we introduce another kind of margin to the softmax loss function, which is more intuitive and interpretable. Experiments on LFW and MegaFace show that our algorithm performs better when the evaluation criteria are designed for very low false alarm rate.
ISSN:1070-9908
1558-2361
DOI:10.1109/LSP.2018.2822810