基于近邻关系聚合的人脸聚类方法

计算机与现代化 ›› 2022, Vol. 0 ›› Issue (12): 81-87.

基于近邻关系聚合的人脸聚类方法

（1.河北农业大学信息科学与技术学院,河北保定071001;2.河北省农业大数据重点实验室,河北保定071001）

出版日期:2023-01-04 发布日期:2023-01-04
作者简介:文紫鑫(1996—),女,河北任丘人,硕士研究生,研究方向:计算机视觉,机器视觉,E-mail: zixin_wen@163.com; 李少英(1997—),男,河北秦皇岛人,硕士研究生,研究方向：计算机视觉,机器视觉,E-mail: 20202060087@pgs.hebau.edu.cn; 王斌成(1998—),男,河北廊坊人,硕士研究生,研究方向:计算机视觉,机器视觉，E-mail： 20212060108@pgs.hebau.edu.cn; 通信作者:刘博(1981—),男,河北保定人,副教授,博士生导师,研究方向:计算机视觉,机器学习,E-mail: boliu@hebau.edu.cn。
基金资助:
国家自然科学基金资助项目(61972132); 河北农业大学自主培养人才科研专项(PY201810)

Face Clustering Method Based on Nearest Neighborhood Aggregation

(1. College of Information Science and Technology, Hebei Agricultural University, Baoding 071001, China;
2.Hebei Key Laboratory of Agricultural Big Data, Baoding 071001, China)

Online:2023-01-04 Published:2023-01-04

摘要/Abstract

摘要： 人脸聚类是人脸标注、人脸识别等工作的预处理过程。其主要对人脸图像进行分组，用来为人脸识别模型提供高质量的标注信息，从而有效降低人工标注的成本。人脸聚类的关键在于如何学习大规模人脸数据中整体及局部的结构关系，并把其迁移至待标注数据集。针对这一问题，本文提出一种基于近邻关系聚合的人脸聚类方法（Nearest Neighborhood Aggregation Clustering, NNAC）。该方法把局部结构的学习建模为一个近邻关系预测问题，通过堆叠多个改进的基于残差-全连接模块（Residual Fully-Connected Block, ResFCB）以提取多尺度的邻接关系特征。实验结果表明，相比主流人脸聚类方法，该方法在基准数据集上能够有效提升聚类精度。

关键词: 人脸聚类, 近邻关系, 连接预测, 残差模块

Abstract: Face clustering is a pre-processing process for face annotation, face recognition and other tasks. It can reduce the labelling burden and provide high-quality annotation for face recognition models by grouping face images. The challenge of face clustering is to extract the global and local structural knowledge in large-scale face datasets and transfer it to the unlabelled ones. To address the issue, a face clustering method based on nearest neighbor aggregation is proposed. The method formulates local structure learning as a link prediction problem. It extracts multi-scale neighborhood characteristics by multiple improved residual Fully-Connected block. The experimental results show that the proposed method can effectively improve the clustering accuracy on the benchmark compared with the mainstream face clustering methods.

Key words: face clustering, nearest neighborhood, link prediction; , residual block

文紫鑫, 李少英, 王斌成, 刘博, . 基于近邻关系聚合的人脸聚类方法[J]. 计算机与现代化, 2022, 0(12): 81-87.

WEN Zi-xin, LI Shao-ying, WANG Bin-cheng, LIU Bo, . Face Clustering Method Based on Nearest Neighborhood Aggregation[J]. Computer and Modernization, 2022, 0(12): 81-87.

参考文献［30］

［1］	JING L, TIAN Y. Self-supervised visual feature learning with deep neural networks: Asurvey［J］. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2021,43(11):4037-4058.
［2］	LLOYD S. Least squares quantization in PCM［J］. IEEE Transactions on Information Theory, 1982,28(2):129-137.
［3］	ESTER M, KRIEGEL H, SANDER J, et al. A density-based algorithm for discovering clusters in large spatial databases with noise［C］// Proceedings of the 2nd International Conference on Knowledge Discovery and Data Mining. 1996:226-231.
［4］	WANG L, DING S, JIA H. An improvement of spectral clustering via message passing and density sensitivesimilarity［J］. IEEE Access, 2019,7:101054-101062.
［5］	LIN W A, CHEN J C, CASTILLO C D, et al. Deep density clustering of unconstrained faces［C］// Proceedings of the 2018 IEEE Conference on Computer Vision and Pattern Recognition. 2018:8128-8137.
［6］	LIN W A, CHEN J C, CHELLAPPA R. A proximity-aware hierarchical clustering of faces［C］// 2017 12th IEEE International Conference on Automatic Face & Gesture Recognition (FG 2017). 2017:294-301.
［7］	ZHU C, WEN F, SUN J. A rank-order distance based clustering algorithm for face tagging［C］// Proceedings of the 2011 IEEE Conference on Computer Vision and Pattern Recognition. 2011:481-488.
［8］	ZHANG M, CHEN Y. Link prediction based on graph neural networks［C］// Proceedings of the 32nd International Conference on Neural Information Processing Systems. 2018:5171-5181.
［9］	OTTO C, WANG D, JAIN A K. Clustering millions of faces byidentity［J］. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2018,40(2):289-303.
［10］	ZHAN X, LIU Z, YAN J, et al. Consensus-driven propagation in massive unlabeled data for face recognition［C］// Proceedings of the 2018 European Conference on Computer Vision (ECCV). 2018:568-583.
［11］	WANG Z, ZHENG L, LI Y, et al. Linkagebased face clustering via graph convolution network［C］// 2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR). 2019:1117-1125.
［12］	KIPF T N, WELLING M. Semi-supervised classification with graph convolutional networks［J］. arXiv preprint arXiv:1609.02907, 2016.
［13］	VELI〖XCC1.TIF,JZ〗KOVI〖XCC.TIF;%90%90,JZ〗 P, CUCURULL G, CASANOVA A, et al. Graphattention networks［J］. arXiv preprint arXiv:1710.10903, 2017.
［14］	王文博,罗恒利. 基于图卷积神经网络的完全图人脸聚类［J］. 计算机科学, 2021,48(z2):275-277.
［15］	LI Q, HAN Z, WU X. Deeper insights into graph convolutional networks for semi-supervised learning［C］// Proceedings of the 32nd AAAI Conference on Artificial Intelligence and 30th Innovative Applications of Artificial Intelligence Conference and 8th AAAI Symposium on Educational Advances in Artificial Intelligence. 2018:3538-3545.
［16］	TOLSTIKHIN I, HOULSBY N, KOLESNIKOV A, et al. MLP-mixer: An all-MLP architecture for vision［J］. arXiv preprint arXiv:2105.01601, 2021.
［17］	DING X H, ZHANG X Y, HAN J G, et al. RepMLP: Re-parameterizing convolutions into fully-connected layers for image recognition［J］. arXiv preprint arXiv:2105.01883, 2021.
［18］	TOUVRON H, BOJANOWSKI P, CARON M, et al.ResMLP: Feedforward networks for image classification with data-efficient training［J］. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2022.DOI:10.1109/TPAMI.2022.3206148.
［19］	TOUVRON H, CORD M, DOUZE M, et al. Training data-efficient image transformers & distillation throughattention［J］. arXiv preprint arXiv:2012.12877, 2020.
［20］	RAM P, SINHA K. Revisiting kd-tree for nearest neighbor search［C］// Proceedings of the 25th ACM SIGKDD International Conference on Knowledge Discovery & Data Mining. 2019:1378-1388.
［21］	ANTALAN J R M, CAMPENA F J. A breadth-first search tree construction for multiplicative circulant graphs［J］. European Journal of Pure and Applied Mathematics, 2021,14(1):248-264.
［22］	MCDAID A F, GREENE D, HURLEY N. Normalized Mutual Information to evaluate overlapping community finding algorithms［J］. arXiv preprint arXiv:1110.2515, 2011.
［23］	AMIG E, GONZALO J, ARTILES J, et al. A comparison of extrinsic clustering evaluation metrics based on formalconstraints［J］. Information Retrieval, 2009,12(4):461-486.
［24］	BANERJEE A, KRUMPELMAN C, GHOSH J, et al. Model-based overlapping clustering［C］// Proceedings of the 11th ACM SIGKDD International Conference on Knowledge Discovery in Data Mining. 2005:532-537.
［25］	SHI Y, OTTO C, JAIN A K. Face clustering: Representation and pairwise constraints［J］. IEEE Transactions on Information Forensics and Security, 2018,13(7):1626-1640.
［26］	GUO Y, ZHANG L, HU Y, et al. MS-celeb-1m: A dataset and benchmark for large-scale face recognition［C］// 2016 European Conference on Computer Vision. 2016:87-102.
［27］	DENG J, GUO J, XUE N, et al. ArcFace: Additive angular margin loss for deep face recognition［C］// 2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR). 2019:4685-4694.
［28］	SCULLEY D. Web-scale k-means clustering［C］// Proceedings of the 19th International Conference on World Wide Web. 2010:1177-1178.
［29］	SIBSON R. SLINK: An optimally efficient algorithm for the single-link cluster method［J］. The Computer Journal, 1973,16(1). DOI:10.1093/comjnl/16.1.30.
［30］	HE K M, ZHANG X, REN S Q, et al. Deep residual learning for image recognition［C］// 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR). 2016:770-778.