基于中心偏差估计和自适应间隔的人脸识别算法

何志浩; 王浩; 曹文明; 何志权

doi:10.12263/DZXB.20230533

您当前的位置：

首页 >

文章列表页 >

基于中心偏差估计和自适应间隔的人脸识别算法

学术论文 | 更新时间：2025-12-08

- 基于中心偏差估计和自适应间隔的人脸识别算法
- Face Recognition Based on Center Bias Estimation and Adaptive Margin
- 电子学报 2024年52卷第8期页码：2866-2877
- 作者机构：
  
  1.深圳大学广东省多媒体信息服务工程技术研究中心，广东深圳 518060
  2.广东省智能信息处理重点实验室，广东深圳 518060
  3.深圳大学射频异质异构集成全国重点实验室，广东深圳 518060
- 作者简介：
  
  [ "何志浩男，2002年1月出生，广东佛山人.深圳大学电子与信息工程学院本科生，主要研究方向为人工智能算法." ]
  [ "王浩男，1990年1月出生，河南鹤壁人.深圳大学电子与信息工程学院助理教授.主要研究方向为机器学习、深度学习、大数据挖掘 Email:haowang@szu.edu.cn" ]
  [ "曹文明男，1965年11月出生，江苏南京人.深圳大学广东省多媒体信息服务工程中心主任，教授，博士生导师，俄罗斯自然科学院外籍院士.主要研究方向为几何代数在传感器网络、多光谱图像、目标检测与评估 Email:wmcao@szu.edu.cn" ]
  [ "何志权男，1978年2月出生，湖南邵阳人.深圳大学电子与信息工程学院副教授.主要研究方向为机器学习、深度学习、图像处理. Email:zhiquan@szu.edu.cn" ]
- 基金信息：
  
  国家自然科学基金(62206178);深圳市稳定支持A类项目(20200826104014001)
- DOI：10.12263/DZXB.20230533
  中图分类号： TP391.4;TN911.73
- 收稿：2023-06-13，
  
  修回：2023-11-27，
  
  纸质出版：2024-08-25
- 稿件说明：
移动端阅览
何志浩, 王浩, 曹文明, 等. 基于中心偏差估计和自适应间隔的人脸识别算法[J]. 电子学报, 2024, 52(08): 2866-2877.

HE Zhi-hao, WANG Hao, CAO Wen-ming, et al. Face Recognition Based on Center Bias Estimation and Adaptive Margin[J]. Acta Electronica Sinica, 2024, 52(08): 2866-2877.
何志浩, 王浩, 曹文明, 等. 基于中心偏差估计和自适应间隔的人脸识别算法[J]. 电子学报, 2024, 52(08): 2866-2877. DOI：10.12263/DZXB.20230533

HE Zhi-hao, WANG Hao, CAO Wen-ming, et al. Face Recognition Based on Center Bias Estimation and Adaptive Margin[J]. Acta Electronica Sinica, 2024, 52(08): 2866-2877. DOI：10.12263/DZXB.20230533

摘要

损失函数的设计在深度人脸识别中至关重要.常见做法是给所有类别添加固定的间隔项，以修改类别间的决策边界，压缩类内特征间距，提高模型分离不同类别特征的能力.然而，为所有类别添加相同的间隔项可能会忽略人脸识别数据集内类别间的不一致性.为进一步提升模型效果，模型应依据类别的学习难易程度，对不同类别样本特征给予不同程度的关注.文中设计了基于类均值中心与类权重中心之间的偏差挖掘难类的方法，称之为中心偏差估计.本文提出的方法会根据中心偏差估计的程度，为不同类别自适应分配不同大小的间隔项.同时，为解决训练前期中心偏差计算不稳定问题，提出了动态变化的收敛参数，调整中心偏差估计的可信度，开展相关实验验证收敛参数的有效性.在人脸验证基准数据集中，本文提出的方法比基准方法的平均准确率提高了0.26%，达到96.62%.在2个大型人脸验证测试数据集上，在FPR等于0.01%时，提出方法的TPR分数分别提高了0.58%和0.22%，获得88.47%和92.29%的实验结果，且多组实验结果表明提出的方法优于一般现有算法.实现代码参见

https：//github.com/TCCofWANG/FR-Centers-Bias

https://github.com/TCCofWANG/FR-Centers-Bias

Abstract

The design of the loss function is crucial in deep face recognition. A common practice is to add a fixed margin term to all classes to modify the decision boundary between classes

compress the distance between intra-class features

and improv

e the ability of the model to separate features of different classes. However

adding the same margin term for all classes may ignore the inconsistency between classes in the face recognition dataset. In order to further improve the effectiveness of the model

we argue that the model should pay different attention to the samples of different classes according to the learning difficulty of the class. In this paper

we introduce a method for hard class mining based on the bias between the center of the class mean and the center of the class weight

called center bias estimation. The method proposed in this paper adaptively assigns margin terms of different sizes to different classes according to the value of center bias estimation. At the same time

to solve the problem of unstable calculation of center bias estimation in the early stage of training

we propose an adaptively changing convergence parameter to adjust the credibility of center bias estimation and design relevant experiments to prove the effectiveness of the convergence parameters. In the face verification baseline dataset

the proposed method in this paper is improved by 0.26% on average accuracy compared with the baseline method

reaching 96.62%. In two large face verification test datasets

when FPR is equal to 0.01%

the TPR scores of our method is improved by 0.58% and 0.22%

respectively

and the experimental results of 88.47% and 92.29% are obtained

and multiple experimental results show that our method is better than the general existing algorithms. The implementation code is published on

https://github.com/TCCofWANG/FR-Centers-Bias

关键词

Keywords

references

SUN Y , CHEN Y H , WANG X G , et al . Deep learning face representation by joint identification-verification [C ] // Proceedings of the 27th International Conference on Neural Information Processing Systems . New York : ACM , 2014 : 1988 - 1996 .

SCHROFF F , KALENICHENKO D , PHILBIN J . FaceNet: A unified embedding for face recognition and clustering [C ] // 2015 IEEE Conference on Computer Vision and Pattern Recognition (CVPR) . Piscataway : IEEE , 2015 : 815 - 823 .

WEN Y , ZHANG K , LI Z , et al . A discriminative feature learning approach for deep face recognition [C ] // 2016 14th Computer Vision–ECCV . Amsterdam : Springer , 2016 , 499 - 515 .

LIU W Y , WEN Y D , YU Z D , et al . SphereFace: Deep hypersphere embedding for face recognition [C ] // 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR) . Piscataway : IEEE , 2017 : 6738 - 6746 .

WANG H , WANG Y T , ZHOU Z , et al . CosFace: Large margin cosine loss for deep face recognition [C ] // 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition . Piscataway : IEEE , 2018 : 5265 - 5274 .

DENG J K , GUO J , XUE N N , et al . ArcFace: Additive angular margin loss for deep face recognition [C ] // 2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) . Piscataway : IEEE , 2019 : 4685 - 4694 .

RIPPEL O , PALURI M , DOLLAR P , et al . Metric learning with adaptive density discrimination [J ] . ( 2015-11-18 )[ 2023-06-08 ] . http://arXiv.org/pdf/1511.05939 http://arXiv.org/pdf/1511.05939 .

BOUTROS F , DAMER N , KIRCHBUCHNER F , et al . ElasticFace: Elastic margin loss for deep face recognition [C ] // 2022 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW) . Piscataway : IEEE , 2022 : 1577 - 1586 .

MENG Q , ZHAO S C , HUANG Z D , et al . MagFace: A universal representation for face recognition and quality assessment [C ] // 2021 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) . Piscataway : IEEE , 2021 : 14220 - 14229 .

TERHÖRST P , IHLEFELD M , HUBER M , et al . QMagFace: Simple and accurate quality-aware face recognition [C ] // 2023 IEEE/CVF Winter Conference on Applications of Computer Vision (WACV) . Piscataway : IEEE , 2023 : 3473 - 3483 .

KIM M , JAIN A K , LIU X M . AdaFace: Quality adaptive margin for face recognition [C ] // 2022 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) . Piscataway : IEEE , 2022 : 18729 - 18738 .

SHRIVASTAVA A , GUPTA A , GIRSHICK R . Training region-based object detectors with online hard example mining [C ] // 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR) . Piscataway : IEEE , 2016 : 761 - 769 .

WANG X B , ZHANG S F , WANG S , et al . Mis-classified vector guided softmax loss for face recognition [J ] . Proceedings of the AAAI Conference on Artificial Intelligence , 2020 , 34 ( 7 ): 12241 - 12248 .

HUANG Y G , WANG Y H , TAI Y , et al . CurricularFace: Adaptive curriculum learning loss for deep face recognition [C ] // 2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) . Piscataway : IEEE , 2020 : 5900 - 5909 .

WU S J , GONG X . BoundaryFace: A mining framework with noise label self-correction for face recognition [C ] // European Conference on Computer Vision . Cham : Springer , 2022 : 91 - 106 .

HUANG Y G , WU J X , XU X K , et al . Evaluation-oriented knowledge distillation for deep face recognition [C ] // 2022 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) . Piscataway : IEEE , 2022 : 18719 - 18728 .

CHAWLA N V , BOWYER K W , HALL L O , et al . SMOTE: Synthetic minority over-sampling technique [J ] . Journal of Artificial Intelligence Research , 2002 , 16 : 321 - 357 .

吴洪兴 , 彭宇 , 彭喜元 . 适用于不平衡样本数据处理的支持向量机方法 [J ] . 电子学报 , 2006 , 34 ( S1 ): 2395 - 2398 .

WU H X , PENG Y , PENG X Y . Support vector machine method for unbalanced sample data processing [J ] . Acta Electronica Sinica , 2006 , 34 ( S1 ): 2395 - 2398 . (in Chinese)

KIM B , KO Y , SEO J . Novel regularization method for the class imbalance problem [J ] . Expert Systems with Applications , 2022 , 188 : 115974 .

SUH Y , HAN B , KIM W , et al . Stochastic class-based hard example mining for deep metric learning [C ] // 2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) . Piscataway : IEEE , 2019 : 7244 - 7252 .

LIU B Y , DENG W H , ZHONG Y Y , et al . Fair loss: Margin-aware reinforcement learning for deep face recognition [C ] // 2019 IEEE/CVF International Conference on Computer Vision (ICCV) . Piscataway : IEEE , 2019 : 10051 - 10060 .

LIU H , ZHU X Y , LEI Z , et al . AdaptiveFace: Adaptive margin and sampling for face recognition [C ] // 2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) . Piscataway : IEEE , 2019 : 11939 - 11948 .

AN X , DENG J K , GUO J , et al . Killing two birds with one stone: Efficient and robust training of face recognition CNNs by partial FC [C ] // 2022 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) . Piscataway : IEEE , 2022 : 4032 - 4041 .

WANG F , XIANG X , CHENG J , et al . NormFace: L 2 hypersphere embedding for face verification [C ] // Proceedings of the 25th ACM international conference on Multimedia . New York : ACM , 2017 : 1041 - 1049 .

RANJAN R , CASTILLO C D , CHELLAPPA R . L 2 -constrained softmax loss for discriminative face verification [J ] . ( 2017-03-28 )[ 2023-06-08 ] . http://arXiv.org/pdf/1703.09507 http://arXiv.org/pdf/1703.09507 .

汪宇玲 , 陈立 , 黎明 , 等 . 基于迹变换和旋转增量调制特征的模糊人脸识别 [J ] . 电子学报 , 2021 , 49 ( 12 ): 2437 - 2448 .

WANG Y L , CHEN L , LI M , et al . Rotational delta modulation feature and its application in blurry face recognition based on trace transform [J ] . Acta Sinica Electronica , 2021 , 49 ( 12 ): 2437 - 2448 . (in Chinese)

EBRAHIMI SAADABADI M S , RAHIMI MALAKSHAN S , ZAFARI A , et al . A quality aware sample-to-sample comparison for face recognition [C ] // 2023 IEEE/ VF Winter Conference on Applications of Computer Vision (WACV) . Piscataway : IEEE , 2023 : 6118 - 6127 .

SIMONYAN K , ZISSERMAN A . Very deep convolutional networks for large-scale image recognition [J ] . ( 2014-09-04 )[ 2023-06-08 ] . http://arXiv.org/pdf/1409.1556 http://arXiv.org/pdf/1409.1556 .

HE K M , ZHANG X Y , REN S Q , et al . Deep residual learning for image recognition [C ] // 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR) . Piscataway : IEEE , 2016 : 770 - 778 .

CHEN S , LIU Y , GAO X , et al . MobileFaceNets: Efficient CNNs for accurate real-time face verification on mobile devices [C ] // Chinese Conference on Biometric Recognition . Cham : Springer , 2018 : 428 - 438 .

浏览量

下载量

CSCD

文章被引用时，请邮件提醒。

提交

工具集

关联资源

暂无数据