TY - JOUR
T1 - Minimum margin loss for deep face recognition
AU - Wei, Xin
AU - Wang, H.
AU - Scotney, Bryan
AU - Wan, Huan
PY - 2020/1/31
Y1 - 2020/1/31
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://pure.ulster.ac.uk/en/publications/minimum-margin-loss-for-deep-face-recognition
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
SN - 0031-3203
VL - 97
JO - Pattern Recognition
JF - Pattern Recognition
M1 - 107012
ER -