中国科学院机构知识库网格
Chinese Academy of Sciences Institutional Repositories Grid
Age estimation via attribute-region association

文献类型:期刊论文

作者Chen, Yiliang1; He, Shengfeng1; Tan, Zichang2; Han, Chu3; Han, Guoqiang1; Qin, Jing4
刊名NEUROCOMPUTING
出版日期2019-11-20
卷号367页码:346-356
ISSN号0925-2312
关键词Age estimation Multi-task learning Attribute-region association
DOI10.1016/j.neucom.2019.08.034
通讯作者He, Shengfeng(hesfe@scut.edu.cn)
英文摘要Human age has been treated as an important biometric trait in many practical applications. In this paper, we propose an Attribute-Region Association Network (ARAN) to tackle the challenging age estimation problem. Instead of performing prediction from a global perspective, we delve into the relationship between face attributes and regions. First, the proposed network is guided by the auxiliary demographic information, as different demographic information (e.g., gender and ethnicity) intrinsically correlates to human age. Second, different face components are separately handled and then involved in the proposed ensemble network, as these components vary differently along with human age. To explore both global and local information, the proposed network consists of several sub-network, each of them takes the global face and a face sub-region as input. Each sub-network leverages the intrinsic correlation between different face attributes (i.e., age, gender, and ethnicity), and it is trained in a multi-task manner. These attribute-region sub-networks are associated to yield the final predictions. To properly train and coordinate such a complex network, a new hierarchical-scheduling training method is proposed to balance the learning complexity in the multi-task learning. In this way, the performance of the most difficult task (i.e., age estimation) can be significantly improved. Extensive experiments on the MORPH Album II and FG-NET show that the proposed method outperforms the state-of-the-art age estimation methods by a significant margin. In particular, for the challenging age estimation, the Mean Absolute Errors (MAE) are decreased to 2.51 years compared to the state-of-the-arts on the MORPH Album II dataset. (C) 2019 Elsevier B.V. All rights reserved.
WOS关键词FRAMEWORK ; GENDER ; IMAGE
资助项目National Natural Science Foundation of China[61472145] ; National Natural Science Foundation of China[61972162] ; National Natural Science Foundation of China[61702194] ; Innovation and Technology Fund of Hong Kong[ITS/319/17] ; Special Fund of Science and Technology Research and Development on Application From Guangdong Province (SFSTRDA-GD)[2016B010127003] ; Guangzhou Key Industrial Technology Research fund[201802010036] ; Guangdong Natural Science Foundation[2017A030312008] ; CCFTencent Openfund
WOS研究方向Computer Science
语种英语
出版者ELSEVIER
WOS记录号WOS:000489017500033
资助机构National Natural Science Foundation of China ; Innovation and Technology Fund of Hong Kong ; Special Fund of Science and Technology Research and Development on Application From Guangdong Province (SFSTRDA-GD) ; Guangzhou Key Industrial Technology Research fund ; Guangdong Natural Science Foundation ; CCFTencent Openfund
源URL[http://ir.ia.ac.cn/handle/173211/26429]  
专题自动化研究所_模式识别国家重点实验室_生物识别与安全技术研究中心
通讯作者He, Shengfeng
作者单位1.South China Univ Technol, Sch Comp Sci & Engn, Guangzhou, Guangdong, Peoples R China
2.Chinese Acad Sci, Inst Automat, Beijing, Peoples R China
3.Chinese Univ Hong Kong, Dept Comp Sci & Engn, Hong Kong, Peoples R China
4.Hong Kong Polytech Univ, Dept Nursing, Hong Kong, Peoples R China
推荐引用方式
GB/T 7714
Chen, Yiliang,He, Shengfeng,Tan, Zichang,et al. Age estimation via attribute-region association[J]. NEUROCOMPUTING,2019,367:346-356.
APA Chen, Yiliang,He, Shengfeng,Tan, Zichang,Han, Chu,Han, Guoqiang,&Qin, Jing.(2019).Age estimation via attribute-region association.NEUROCOMPUTING,367,346-356.
MLA Chen, Yiliang,et al."Age estimation via attribute-region association".NEUROCOMPUTING 367(2019):346-356.

入库方式: OAI收割

来源:自动化研究所

浏览0
下载0
收藏0
其他版本

除非特别说明,本系统中所有内容都受版权保护,并保留所有权利。