Support Vector Guided Softmax Loss for Face Recognition

Abstract

Face recognition has witnessed significant progresses due to the advances of deep convolutional neural networks (CNNs), the central challenge of which, is feature discrimination. To address it, one group tries to exploit mining-based strategies (\textit{e.g.}, hard example mining and focal loss) to focus on the informative examples. The other group devotes to designing margin-based loss functions (\textit{e.g.}, angular, additive and additive angular margins) to increase the feature margin from the perspective of ground truth class. Both of them have been well-verified to learn discriminative features. However, they suffer from either the ambiguity of hard examples or the lack of discriminative power of other classes. In this paper, we design a novel loss function, namely support vector guided softmax loss (SV-Softmax), which adaptively emphasizes the mis-classified points (support vectors) to guide the discriminative features learning. So the developed SV-Softmax loss is able to eliminate the ambiguity of hard examples as well as absorb the discriminative power of other classes, and thus results in more discrimiantive features. To the best of our knowledge, this is the first attempt to inherit the advantages of mining-based and margin-based losses into one framework. Experimental results on several benchmarks have demonstrated the effectiveness of our approach over state-of-the-arts.

References in corpus (6)

• Very Deep Convolutional Networks for Large-Scale Image Recognition
• Deep Learning Face Representation by Joint Identification-Verification
• Residual Attention Network for Image Classification
• CosFace: Large Margin Cosine Loss for Deep Face Recognition
• Learning Deep Features via Congenerous Cosine Loss for Person Recognition
• Ensemble Soft-Margin Softmax Loss for Image Classification

Cited by in corpus (13)

• MagFace: A Universal Representation for Face Recognition and Quality Assessment
• PoseFace: Pose-Invariant Features and Pose-Adaptive Loss for Face Recognition
• A Dataset and Benchmark for Large-scale Multi-modal Face Anti-spoofing
• A Dataset and Benchmark Towards Multi-Modal Face Anti-Spoofing Under Surveillance Scenarios
• CASIA-SURF: A Large-scale Multi-modal Benchmark for Face Anti-spoofing
• Mis-classified Vector Guided Softmax Loss for Face Recognition
• Learning Meta Face Recognition in Unseen Domains
• Improving Face Recognition from Hard Samples via Distribution Distillation Loss
• An Efficient Training Approach for Very Large Scale Face Recognition
• Inter-class Discrepancy Alignment for Face Recognition
• More Information Supervised Probabilistic Deep Face Embedding Learning
• DSAM: A Distance Shrinking with Angular Marginalizing Loss for High Performance Vehicle Re-identificatio
• Recent Advances in Large Margin Learning