Learning deep face representation with long-tail data: An aggregate-and-disperse approach

doi:10.1016/j.patrec.2020.02.007

CORC > 计算技术研究所 > 中国科学院计算技术研究所 > 中国科学院计算技术研究所期刊论文 > 英文

	Learning deep face representation with long-tail data: An aggregate-and-disperse approach
	Ma, Yuhao 1,2; Kan, Meina 1,2; Shan, Shiguang 1,2,3; Chen, Xilin 1,2
刊名	PATTERN RECOGNITION LETTERS
	2020-05-01
卷号	133 页码:48-54
关键词	Face recognition Deep representation learning Long-tail distribution Aggregate-and-disperse
ISSN号	0167-8655
DOI	10.1016/j.patrec.2020.02.007
英文摘要	In this work, we study the problem of deep representation learning on a large face dataset with long-tail distribution. Training convolutional neural networks on such dataset with conventional strategy suffers from imbalance problem which results in biased classification boundary, and the few-shot classes lying in tail parts further make the model prone to overfitting. Aiming to learn more discriminative CNN model from long-tail data, we propose a novel aggregate-and-disperse training schema. Firstly, our proposed method aggregates similar classes in tail part to avoid imbalance problem. Based on the aggregated super classes and those original head classes, a model is pre-trained to capture accurate discrimination in head classes as well as coarse discrinimation in tail classes. Secondly, we selectively disperses those aggregated super classes to learn precise inter-class variations and refine the representation for better generalization. We perform extensive experiments on MS-Celeb-1M, BLUFR and MegaFace. Compared with baselines and existing methods, our method achieves better performance of face recognition, demonstrating its effectiveness of handling long-tail distribution. (C) 2020 Published by Elsevier B.V
资助项目	National Key R&D Program of China[2017YFA070 080 0] ; Natural Science Foundation of China[61772496] ; Natural Science Foundation of China[61732004] ; Natural Science Foundation of China[61532018]
WOS研究方向	Computer Science
语种	英语
出版者	ELSEVIER
WOS记录号	WOS:000537129300007
内容类型	期刊论文
源URL	[http://119.78.100.204/handle/2XEOYT63/15247]
专题	中国科学院计算技术研究所期刊论文_英文
通讯作者	Kan, Meina
作者单位	1.Chinese Acad Sci, Inst Comp Technol, CAS, Key Lab Intelligent Informat Proc, Beijing 100190, Peoples R China 2.Univ Chinese Acad Sci, Beijing 100049, Peoples R China 3.CAS Ctr Excellence Brain Sci & Intelligence Techn, Shanghai 200031, Peoples R China
推荐引用方式 GB/T 7714	Ma, Yuhao,Kan, Meina,Shan, Shiguang,et al. Learning deep face representation with long-tail data: An aggregate-and-disperse approach[J]. PATTERN RECOGNITION LETTERS,2020,133:48-54.
APA	Ma, Yuhao,Kan, Meina,Shan, Shiguang,&Chen, Xilin.(2020).Learning deep face representation with long-tail data: An aggregate-and-disperse approach.PATTERN RECOGNITION LETTERS,133,48-54.
MLA	Ma, Yuhao,et al."Learning deep face representation with long-tail data: An aggregate-and-disperse approach".PATTERN RECOGNITION LETTERS 133(2020):48-54.