Heterogeneous Face Recognition via Face Synthesis With Identity-Attribute Disentanglement
文献类型:期刊论文
作者 | Yang, Ziming1,5; Liang, Jian2,3,4; Fu, Chaoyou2,3,4; Luo, Mandi2,3,4; Zhang, Xiao-Yu1,5 |
刊名 | IEEE Transactions on Information Forensics and Security |
出版日期 | 2022 |
卷号 | 17页码:1344-1358 |
ISSN号 | 1556-6013 |
关键词 | Heterogeneous face recognition cross-domain face augmentation face disentanglement |
英文摘要 | Heterogeneous Face Recognition (HFR) aims to match faces across different domains (e.g., visible to near-infrared images), which has been widely applied in authentication and forensics scenarios. However, HFR is a challenging problem because of the large cross-domain discrepancy, limited heterogeneous data pairs, and large variation of facial attributes. To address these challenges, we propose a new HFR method from the perspective of heterogeneous data augmentation, named Face Synthesis with Identity-Attribute Disentanglement (FSIAD). Firstly, the identity-attribute disentanglement (IAD) decouples face images into identity-related representations and identity-unrelated representations (called attributes), and then decreases the correlation between identities and attributes. Secondly, we devise a face synthesis module (FSM) to generate a large number of images with stochastic combinations of disentangled identities and attributes for enriching the attribute diversity of synthetic images. Both the original images and the synthetic ones are utilized to train the HFR network for tackling the challenges and improving the performance of HFR. Extensive experiments on five HFR databases validate that FSIAD obtains superior performance than previous HFR approaches. Particularly, FSIAD obtains 4.8% improvement over state of the art in terms of VR@FAR=0.01% on LAMP-HQ, the largest HFR database so far. |
语种 | 英语 |
源URL | [http://ir.ia.ac.cn/handle/173211/48274] |
专题 | 自动化研究所_智能感知与计算研究中心 |
通讯作者 | Zhang, Xiao-Yu |
作者单位 | 1.School of Cyber Security, University of Chinese Academy of Sciences 2.Center for Excellence in Brain Science and Intelligence Technology, CAS 3.Center for Research on Intelligent Perception and Computing, CASIA 4.National Laboratory of Pattern Recognition, CASIA 5.Institute of Information Engineering, Chinese Academy of Sciences |
推荐引用方式 GB/T 7714 | Yang, Ziming,Liang, Jian,Fu, Chaoyou,et al. Heterogeneous Face Recognition via Face Synthesis With Identity-Attribute Disentanglement[J]. IEEE Transactions on Information Forensics and Security,2022,17:1344-1358. |
APA | Yang, Ziming,Liang, Jian,Fu, Chaoyou,Luo, Mandi,&Zhang, Xiao-Yu.(2022).Heterogeneous Face Recognition via Face Synthesis With Identity-Attribute Disentanglement.IEEE Transactions on Information Forensics and Security,17,1344-1358. |
MLA | Yang, Ziming,et al."Heterogeneous Face Recognition via Face Synthesis With Identity-Attribute Disentanglement".IEEE Transactions on Information Forensics and Security 17(2022):1344-1358. |
入库方式: OAI收割
来源:自动化研究所
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。