Y. Qin, Z. Yu, L. Yan, Z. Wang, C. Zhao and Z. Lei, "Meta-Teacher For Face Anti-Spoofing," in IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 44, no. 10, pp. 6311-6326, 1 Oct. 2022, doi: 10.1109/TPAMI.2021.3091167
Meta-teacher for face anti-spoofing
|Author:||Qin, Yunxiao1,2; Yu, Zitong3; Yan, Longbin2;|
1Neuroscience and Intelligent Media Institute (NIMI), Communication University of China, Beijing 100024, China
2Northwestern Polytechnical University, Xian 710072, China
3Center for Machine Vision and Signal Analysis, University of Oulu, 90014 Oulu, Finland
4Beijing Kwai Technology Co., Ltd, Beijing 102600, China
5MiningLamp Technology, Beijing 100000, China
6National Laboratory of Pattern Recognition (NLPR), Center for Biometrics and Security Research (CBSR), Institute of Automa- tion, Chinese Academy of Sciences (CASIA), Beijing 100190, China
7School of Artificial Intelligence, University of Chinese Academy of Sciences (UCAS), Beijing 100049, China
8Centre for Artificial Intelligence and Robotics, Hong Kong Institute of Science & Innovation, Chinese Academy of Sciences, Hong Kong
|Online Access:||PDF Full Text (PDF, 2.2 MB)|
|Persistent link:|| http://urn.fi/urn:nbn:fi-fe2022100661277
Institute of Electrical and Electronics Engineers,
|Publish Date:|| 2022-10-06
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.
IEEE transactions on pattern analysis and machine intelligence
|Pages:||6311 - 6326|
|Type of Publication:||
A1 Journal article – refereed
|Field of Science:||
113 Computer and information sciences
This work was supported by the National Key Research and Development Program of China (No. 2020AAA0140002). This work was also supported in part by the National Natural Science Foundation of China (No. 61876178, 61976229).
© 2021 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting/republishing this material for advertising or promotional purposes, creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works.