University of Oulu

Qin, Y., Zhao, C., Zhu, X., Wang, Z., Yu, Z., Fu, T., Zhou, F., Shi, J., & Lei, Z. (2020). Learning Meta Model for Zero- and Few-Shot Face Anti-Spoofing. Proceedings of the AAAI Conference on Artificial Intelligence, 34(07), 11916-11923.

Learning meta model for zero- and few-shot face anti-spoofing

Saved in:
Author: Qin, Yunxiao1,2; Zhao, Chenxu2; Zhu, Xiangyu3;
Organizations: 1Northwestern Polytechnical University, Xian, China
2AIBEE, Beijing, China
3National Laboratory of Pattern Recognition, Institute of Automation, Chinese Academy of Science, Beijing, China
4CMVS, University of Oulu, Oulu, Finland
5Winsense Technology Ltd, Beijing, China
Format: article
Version: accepted version
Access: open
Online Access: PDF Full Text (PDF, 1.6 MB)
Persistent link:
Language: English
Published: Association for the Advancement of Artificial Intelligence, 2021
Publish Date: 2021-09-10


Face anti-spoofing is crucial to the security of face recognition systems. Most previous methods formulate face anti-spoofing as a supervised learning problem to detect various predefined presentation attacks, which need large scale training data to cover as many attacks as possible. However, the trained model is easy to overfit several common attacks and is still vulnerable to unseen attacks. To overcome this challenge, the detector should: 1) learn discriminative features that can generalize to unseen spoofing types from predefined presentation attacks; 2) quickly adapt to new spoofing types by learning from both the predefined attacks and a few examples of the new spoofing types. Therefore, we define face anti-spoofing as a zero- and few-shot learning problem. In this paper, we propose a novel Adaptive Inner-update Meta Face Anti-Spoofing (AIM-FAS) method to tackle this problem through meta-learning. Specifically, AIM-FAS trains a meta-learner focusing on the task of detecting unseen spoofing types by learning from predefined living and spoofing faces and a few examples of new attacks. To assess the proposed approach, we propose several benchmarks for zero- and few-shot FAS. Experiments show its superior performances on the presented benchmarks to existing methods in existing zero-shot FAS protocols.

see all

Series: Proceedings of the AAAI Conference on Artificial Intelligence
ISSN: 2159-5399
ISSN-E: 2374-3468
ISSN-L: 2159-5399
ISBN Print: 978-1-57735-835-0
Pages: 11916 - 11923
DOI: 10.1609/aaai.v34i07.6866
Host publication: 34th AAAI Conference on Artificial Intelligence, AAAI 2020 - AAAI Technical Track: Vision
Conference: Conference on Artificial Intelligence
Type of Publication: A4 Article in conference proceedings
Field of Science: 113 Computer and information sciences
Copyright information: c 2020, Association for the Advancement of Artificial Intelligence ( All rights reserved. This is the authors accepted manuscript version.