NAS-FAS: Static-Dynamic Central Difference Network Search for Face Anti-Spoofing | |
Yu, Zitong1; Wan, Jun2,3; Qin, Yunxiao4; Li, Xiaobai1; Li, Stan Z.5; Zhao, Guoying1 | |
刊名 | IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE |
2021-09-01 | |
卷号 | 43期号:9页码:3005-3023 |
关键词 | Task analysis Face recognition Convolution Testing Computer architecture Protocols Search problems Face anti-spoofing neural architecture search convolution pooling static-dynamic CASIA-SURF 3DMask |
ISSN号 | 0162-8828 |
DOI | 10.1109/TPAMI.2020.3036338 |
通讯作者 | Wan, Jun(jun.wan@ia.ac.cn) ; Zhao, Guoying(guoying.zhao@oulu.fi) |
英文摘要 | Face anti-spoofing (FAS) plays a vital role in securing face recognition systems. Existing methods heavily rely on the expert-designed networks, which may lead to a sub-optimal solution for FAS task. Here we propose the first FAS method based on neural architecture search (NAS), called NAS-FAS, to discover the well-suited task-aware networks. Unlike previous NAS works mainly focus on developing efficient search strategies in generic object classification, we pay more attention to study the search spaces for FAS task. The challenges of utilizing NAS for FAS are in two folds: the networks searched on 1) a specific acquisition condition might perform poorly in unseen conditions, and 2) particular spoofing attacks might generalize badly for unseen attacks. To overcome these two issues, we develop a novel search space consisting of central difference convolution and pooling operators. Moreover, an efficient static-dynamic representation is exploited for fully mining the FAS-aware spatio-temporal discrepancy. Besides, we propose Domain/Type-aware Meta-NAS, which leverages cross-domain/type knowledge for robust searching. Finally, in order to evaluate the NAS transferability for cross datasets and unknown attack types, we release a large-scale 3D mask dataset, namely CASIA-SURF 3DMask, for supporting the new 'cross-dataset cross-type' testing protocol. Experiments demonstrate that the proposed NAS-FAS achieves state-of-the-art performance on nine FAS benchmark datasets with four testing protocols. |
资助项目 | Academy of Finland[316765] ; Infotech Oulu ; Chinese National Natural Science Foundation[61961160704] ; Chinese National Natural Science Foundation[61876179] ; Science and Technology Development Fund of Macau[0025/2019/A1] ; ICT 2023 project[328115] |
WOS研究方向 | Computer Science ; Engineering |
语种 | 英语 |
出版者 | IEEE COMPUTER SOC |
WOS记录号 | WOS:000681124300013 |
资助机构 | Academy of Finland ; Infotech Oulu ; Chinese National Natural Science Foundation ; Science and Technology Development Fund of Macau ; ICT 2023 project |
内容类型 | 期刊论文 |
源URL | [http://ir.ia.ac.cn/handle/173211/45661] |
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 | Wan, Jun; Zhao, Guoying |
作者单位 | 1.Univ Oulu, Ctr Machine Vis & Signal Anal, Oulu 90014, Finland 2.Chinese Acad Sci, Inst Automat, Natl Lab Pattern Recognit, Beijing, Peoples R China 3.Univ Chinese Acad Sci, Beijing 100190, Peoples R China 4.Northwestern Polytech Univ, Xian 710072, Peoples R China 5.Westlake Univ, Sch Engn, Hangzhou 310012, Zhejiang, Peoples R China |
推荐引用方式 GB/T 7714 | Yu, Zitong,Wan, Jun,Qin, Yunxiao,et al. NAS-FAS: Static-Dynamic Central Difference Network Search for Face Anti-Spoofing[J]. IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE,2021,43(9):3005-3023. |
APA | Yu, Zitong,Wan, Jun,Qin, Yunxiao,Li, Xiaobai,Li, Stan Z.,&Zhao, Guoying.(2021).NAS-FAS: Static-Dynamic Central Difference Network Search for Face Anti-Spoofing.IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE,43(9),3005-3023. |
MLA | Yu, Zitong,et al."NAS-FAS: Static-Dynamic Central Difference Network Search for Face Anti-Spoofing".IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE 43.9(2021):3005-3023. |
个性服务 |
查看访问统计 |
相关权益政策 |
暂无数据 |
收藏/分享 |
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。
修改评论