Minimum Margin Loss for Deep Face Recognition

Research output: Contribution to journalArticle

Abstract

Face recognition has achieved great success owing to the fast development of deep neural networks in the past few years. Different loss functions can be used in a deep neural network resulting in different performance. Most recently some loss functions have been proposed, which have advanced the state of the art. However, they cannot solve the problem of margin bias which is present in class imbalanced datasets, having the so-called long-tailed distributions. In this paper, we propose to solve the margin bias problem by setting a minimum margin for all pairs of classes. We present a new loss function, Minimum Margin Loss (MML), which is aimed at enlarging the margin of those overclose class centre pairs so as to enhance the discriminative ability of the deep features. MML, together with Softmax Loss and Centre Loss, supervises the training process to balance the margins of all classes irrespective of their class distributions. We implemented MML in Inception-ResNet-v1 and conducted extensive experiments on seven face recognition benchmark datasets, MegaFace, FaceScrub, LFW, SLLFW, YTF, IJB-B and IJB-C. Experimental results show that the proposed MML loss function has led to new state of the art in face recognition, reducing the negative effect of margin bias.
LanguageEnglish
Article number107012
Number of pages9
JournalPattern Recognition
Volume97
Early online date17 Aug 2019
DOIs
Publication statusE-pub ahead of print - 17 Aug 2019

Fingerprint

Face recognition

Keywords

  • Deep learning
  • Convolutional neural networks
  • Face recognition
  • Minimum Margin Loss
  • Minimum margin loss

Cite this

@article{b33269cb7dfd4c8fae8a6ebdf99cfb51,
title = "Minimum Margin Loss for Deep Face Recognition",
abstract = "Face recognition has achieved great success owing to the fast development of deep neural networks in the past few years. Different loss functions can be used in a deep neural network resulting in different performance. Most recently some loss functions have been proposed, which have advanced the state of the art. However, they cannot solve the problem of margin bias which is present in class imbalanced datasets, having the so-called long-tailed distributions. In this paper, we propose to solve the margin bias problem by setting a minimum margin for all pairs of classes. We present a new loss function, Minimum Margin Loss (MML), which is aimed at enlarging the margin of those overclose class centre pairs so as to enhance the discriminative ability of the deep features. MML, together with Softmax Loss and Centre Loss, supervises the training process to balance the margins of all classes irrespective of their class distributions. We implemented MML in Inception-ResNet-v1 and conducted extensive experiments on seven face recognition benchmark datasets, MegaFace, FaceScrub, LFW, SLLFW, YTF, IJB-B and IJB-C. Experimental results show that the proposed MML loss function has led to new state of the art in face recognition, reducing the negative effect of margin bias.",
keywords = "Deep learning, Convolutional neural networks, Face recognition, Minimum Margin Loss, Minimum margin loss",
author = "Xin Wei and H. Wang and Bryan Scotney and Huan Wan",
year = "2019",
month = "8",
day = "17",
doi = "10.1016/j.patcog.2019.107012",
language = "English",
volume = "97",
journal = "Pattern Recognition",
issn = "0031-3203",
publisher = "Elsevier",

}

Minimum Margin Loss for Deep Face Recognition. / Wei, Xin; Wang, H.; Scotney, Bryan; Wan, Huan.

In: Pattern Recognition, Vol. 97, 107012, 31.01.2020.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Minimum Margin Loss for Deep Face Recognition

AU - Wei, Xin

AU - Wang, H.

AU - Scotney, Bryan

AU - Wan, Huan

PY - 2019/8/17

Y1 - 2019/8/17

N2 - Face recognition has achieved great success owing to the fast development of deep neural networks in the past few years. Different loss functions can be used in a deep neural network resulting in different performance. Most recently some loss functions have been proposed, which have advanced the state of the art. However, they cannot solve the problem of margin bias which is present in class imbalanced datasets, having the so-called long-tailed distributions. In this paper, we propose to solve the margin bias problem by setting a minimum margin for all pairs of classes. We present a new loss function, Minimum Margin Loss (MML), which is aimed at enlarging the margin of those overclose class centre pairs so as to enhance the discriminative ability of the deep features. MML, together with Softmax Loss and Centre Loss, supervises the training process to balance the margins of all classes irrespective of their class distributions. We implemented MML in Inception-ResNet-v1 and conducted extensive experiments on seven face recognition benchmark datasets, MegaFace, FaceScrub, LFW, SLLFW, YTF, IJB-B and IJB-C. Experimental results show that the proposed MML loss function has led to new state of the art in face recognition, reducing the negative effect of margin bias.

AB - Face recognition has achieved great success owing to the fast development of deep neural networks in the past few years. Different loss functions can be used in a deep neural network resulting in different performance. Most recently some loss functions have been proposed, which have advanced the state of the art. However, they cannot solve the problem of margin bias which is present in class imbalanced datasets, having the so-called long-tailed distributions. In this paper, we propose to solve the margin bias problem by setting a minimum margin for all pairs of classes. We present a new loss function, Minimum Margin Loss (MML), which is aimed at enlarging the margin of those overclose class centre pairs so as to enhance the discriminative ability of the deep features. MML, together with Softmax Loss and Centre Loss, supervises the training process to balance the margins of all classes irrespective of their class distributions. We implemented MML in Inception-ResNet-v1 and conducted extensive experiments on seven face recognition benchmark datasets, MegaFace, FaceScrub, LFW, SLLFW, YTF, IJB-B and IJB-C. Experimental results show that the proposed MML loss function has led to new state of the art in face recognition, reducing the negative effect of margin bias.

KW - Deep learning

KW - Convolutional neural networks

KW - Face recognition

KW - Minimum Margin Loss

KW - Minimum margin loss

UR - https://www.sciencedirect.com/science/article/abs/pii/S0031320319303152

UR - http://www.scopus.com/inward/record.url?scp=85071110645&partnerID=8YFLogxK

U2 - 10.1016/j.patcog.2019.107012

DO - 10.1016/j.patcog.2019.107012

M3 - Article

VL - 97

JO - Pattern Recognition

T2 - Pattern Recognition

JF - Pattern Recognition

SN - 0031-3203

M1 - 107012

ER -