AdaCos: Adaptively Scaling Cosine Logits for Effectively Learning Deep Face Representations

Abstract

The cosine-based softmax losses and their variants achieve great success in deep learning based face recognition. However, hyperparameter settings in these losses have significant influences on the optimization path as well as the final recognition performance. Manually tuning those hyperparameters heavily relies on user experience and requires many training tricks. In this paper, we investigate in depth the effects of two important hyperparameters of cosine-based softmax losses, the scale parameter and angular margin parameter, by analyzing how they modulate the predicted classification probability. Based on these analysis, we propose a novel cosine-based softmax loss, AdaCos, which is hyperparameter-free and leverages an adaptive scale parameter to automatically strengthen the training supervisions during the training process. We apply the proposed AdaCos loss to large-scale face verification and identification datasets, including LFW, MegaFace, and IJB-C 1:1 Verification. Our results show that training deep neural networks with the AdaCos loss is stable and able to achieve high face recognition accuracy. Our method outperforms state-of-the-art softmax losses on all the three datasets.

arXiv commentCVPR 2019 Oral

References in corpus (10)

• Practical Bayesian Optimization of Machine Learning Algorithms
• MXNet: A Flexible and Efficient Machine Learning Library for Heterogeneous Distributed Systems
• Deep Learning Face Representation by Joint Identification-Verification
• Learning Face Representation from Scratch
• YFCC100M: The New Data in Multimedia Research
• DeepID3: Face Recognition with Very Deep Neural Networks
• CosFace: Large Margin Cosine Loss for Deep Face Recognition
• Rethinking Feature Discrimination and Polymerization for Large-scale Recognition
• Learning Deep Features via Congenerous Cosine Loss for Person Recognition
• Crystal Loss and Quality Pooling for Unconstrained Face Verification and Recognition

Cited by in corpus (5)

• A Survey of Deep Learning-based Object Detection
• Circle Loss: A Unified Perspective of Pair Similarity Optimization
• FaceX-Zoo: A PyTorch Toolbox for Face Recognition
• GB-CosFace: Rethinking Softmax-based Face Recognition from the Perspective of Open Set Classification
• More Information Supervised Probabilistic Deep Face Embedding Learning