中国科学院机构知识库网格
Chinese Academy of Sciences Institutional Repositories Grid
首页 机构 成果 学者
  
  1. CAS IR Grid
  2. 自动化研究所
  3. 中国科学院自动化研究所
  4. 多模态人工智能系统全国重点实验室
  5. 生物识别与安全技术研究中心
Meta-Teacher For Face Anti-Spoofing

文献类型：期刊论文

作者Qin, Yunxiao1,8; Yu, Zitong7; Yan, Longbin1; Wang, Zezheng6; Zhao, Chenxu5; Lei, Zhen2,3,4
刊名IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE
出版日期2022-10-01
卷号44期号:10页码:6311-6326
ISSN号0162-8828
关键词Detectors Face recognition Training Faces Feature extraction Training data Optimization Face anti-spoofing meta-teacher pixel-wise supervision deep-learning
DOI10.1109/TPAMI.2021.3091167
通讯作者Lei, Zhen(zlei@nlpr.ia.ac.cn)
英文摘要Face anti-spoofing (FAS) secures face recognition from presentation attacks (PAs). Existing FAS methods usually supervise PA detectors with handcrafted binary or pixel-wise labels. However, handcrafted labels may are not the most adequate way to supervise PA detectors learning sufficient and intrinsic spoofing cues. Instead of using the handcrafted labels, we propose a novel Meta-Teacher FAS (MT-FAS) method to train a meta-teacher for supervising PA detectors more effectively. The meta-teacher is trained in a bi-level optimization manner to learn the ability to supervise the PA detectors learning rich spoofing cues. The bi-level optimization contains two key components: 1) a lower-level training in which the meta-teacher supervises the detector's learning process on the training set; and 2) a higher-level training in which the meta-teacher's teaching performance is optimized by minimizing the detector's validation loss. Our meta-teacher differs significantly from existing teacher-student models because the meta-teacher is explicitly trained for better teaching the detector (student), whereas existing teachers are trained for outstanding accuracy neglecting teaching ability. Extensive experiments on five FAS benchmarks show that with the proposed MT-FAS, the trained meta-teacher 1) provides better-suited supervision than both handcrafted labels and existing teacher-student models; and 2) significantly improves the performances of PA detectors.
资助项目National Key Research and Development Program of China[2020AAA0140002] ; National Natural Science Foundation of China[61876178] ; National Natural Science Foundation of China[61976229]
WOS研究方向Computer Science ; Engineering
语种英语
出版者IEEE COMPUTER SOC
WOS记录号WOS:000853875300035
资助机构National Key Research and Development Program of China ; National Natural Science Foundation of China
源URL[http://ir.ia.ac.cn/handle/173211/50137]  
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Lei, Zhen
作者单位1.Northwestern Polytech Univ, Xian 710072, Peoples R China
2.Chinese Acad Sci, Ctr Artificial Intelligence & Robot, Hong Kong Inst Sci & Innovat, Hong Kong, Peoples R China
3.Univ Chinese Acad Sci UCAS, Sch Artificial Intelligence, Beijing 100049, Peoples R China
4.Chinese Acad Sci CASIA, Ctr Biometr & Secur Res CBSR, Inst Automat, Natl Lab Pattern Recognit NLPR, Beijing 100190, Peoples R China
5.MiningLamp Technol, Beijing 100000, Peoples R China
6.Beijing Kwai Technol Co Ltd, Beijing 102600, Peoples R China
7.Univ Oulu, Ctr Machine Vis & Signal Anal, Oulu 90014, Finland
8.Commun Univ China, Neurosci & Intelligent Media Inst NIMI, Beijing 100024, Peoples R China
推荐引用方式
GB/T 7714		 Qin, Yunxiao,Yu, Zitong,Yan, Longbin,et al. Meta-Teacher For Face Anti-Spoofing[J]. IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE,2022,44(10):6311-6326.
APA Qin, Yunxiao,Yu, Zitong,Yan, Longbin,Wang, Zezheng,Zhao, Chenxu,&Lei, Zhen.(2022).Meta-Teacher For Face Anti-Spoofing.IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE,44(10),6311-6326.
MLA Qin, Yunxiao,et al."Meta-Teacher For Face Anti-Spoofing".IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE 44.10(2022):6311-6326.

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。