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

05/01/2019
by   Xiao Zhang, et al.
0

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.

READ FULL TEXT

page 1

page 2

page 3

page 4

research
05/07/2019

P2SGrad: Refined Gradients for Optimizing Deep Face Models

Cosine-based softmax losses significantly improve the performance of dee...
research
05/25/2019

Hyperparameter-Free Out-of-Distribution Detection Using Softmax of Scaled Cosine Similarity

The ability of detecting out-of-distribution (OOD) samples is important ...
research
04/03/2018

Crystal Loss and Quality Pooling for Unconstrained Face Verification and Recognition

In recent years, the performance of face verification and recognition sy...
research
07/27/2020

Contraction Mapping of Feature Norms for Classifier Learning on the Data with Different Quality

The popular softmax loss and its recent extensions have achieved great s...
research
03/28/2017

L2-constrained Softmax Loss for Discriminative Face Verification

In recent years, the performance of face verification systems has signif...
research
01/23/2022

Basket-based Softmax

Softmax-based losses have achieved state-of-the-art performances on vari...
research
07/25/2023

An Explainable Model-Agnostic Algorithm for CNN-based Biometrics Verification

This paper describes an adaptation of the Local Interpretable Model-Agno...

Please sign up or login with your details

Forgot password? Click here to reset