计算机科学与探索 ›› 2022, Vol. 16 ›› Issue (11): 2471-2486.DOI: 10.3778/j.issn.1673-9418.2203082

• 综述·探索 • 上一篇    下一篇

领域外人脸活体检测综述

史屹琛1, 封筠1,+(), 肖立轩1, 贺晶晶1, 胡晶晶2   

  1. 1.石家庄铁道大学 信息科学与技术学院,石家庄 050043
    2.北京理工大学 计算机科学与技术学院,北京 100081
  • 收稿日期:2022-03-21 修回日期:2022-05-16 出版日期:2022-11-01 发布日期:2022-11-16
  • 通讯作者: + E-mail: fengjun@stdu.edu.cn
  • 作者简介:史屹琛(1998—),男,山西太原人,硕士研究生,主要研究方向为人脸活体检测、迁移学习。
    封筠(1971—),女,河北石家庄人,博士,教授,主要研究方向为计算机视觉、机器学习等。
    肖立轩(1999—),男,河北定州人,硕士研究生,主要研究方向为人脸活体检测。
    贺晶晶(2000—),女,湖南衡阳人,硕士研究生,主要研究方向为人脸活体检测。
    胡晶晶(1978—),女,江苏徐州人,博士,副教授,主要研究方向为Web智能、信息安全等。
  • 基金资助:
    国家自然科学基金(61772070);国家自然科学基金(61972267);河北省高等学校科学技术研究重点项目(ZD2021333)

Out of Domain Face Anti-spoofing: A Survey

SHI Yichen1, FENG Jun1,+(), XIAO Lixuan1, HE Jingjing1, HU Jingjing2   

  1. 1. School of Information Science and Technology, Shijiazhuang Tiedao University, Shijiazhuang 050043, China
    2. School of Computer Science and Technology, Beijing Institute of Technology, Beijing 100081, China
  • Received:2022-03-21 Revised:2022-05-16 Online:2022-11-01 Published:2022-11-16
  • About author:SHI Yichen, born in 1998, M.S. candidate. His research interests include face anti-spoofing and transfer learning.
    FENG Jun, born in 1971, Ph.D., professor. Her research interests include computer vision, machine learning, etc.
    XIAO Lixuan, born in 1999, M.S. candidate. His research interest is face anti-spoofing.
    HE Jingjing, born in 2000, M.S. candidate. Her research interest is face anti-spoofing.
    HU Jingjing, born in 1978, Ph.D., associate professor. Her research interests include Web intelligence, information security, etc.
  • Supported by:
    National Natural Science Foundation of China(61772070);National Natural Science Foundation of China(61972267);Key Projects of Science and Technology Research in Colleges and Universities of Hebei Province(ZD2021333)

摘要:

人脸活体检测(FAS)作为保护人脸识别模型的重要手段,能够确保系统在面对各种呈现攻击时仍然安全、可靠。当前基于深度学习的人脸活体检测模型在测试数据与训练数据服从同一分布时结果令人满意,但当训练好的模型在领域外场景进行推理时,如遇到跨域迁移、分布外场景时,模型的准确性会出现较大的下降。主要阐述了静默型人脸活体检测模型在真实场景中会遇到的问题,即模型遇到未知环境和未知攻击方式。将相应的解决方案分为四类:基于领域自适应的方法、基于领域泛化的方法、基于零样本/小样本学习的方法以及基于异常检测的方法。对各解决方案及其包含的深度学习模型方法进行总结、比较,归纳了主要方法的机制、模型结构、优势、局限性以及适用场景。介绍了领域外场景下人脸活体检测常用的公共数据集、评价指标、测评协议以及在部分测评协议下当前先进方法的测试结果。最后讨论了人脸活体检测在实际应用中存在的难点与挑战,总结了未来的研究方向。

关键词: 人脸活体检测(FAS), 领域自适应, 领域泛化, 零样本/小样本学习, 异常检测, 深度学习

Abstract:

Face anti-spoofing (FAS), as an important means to protect face recognition models, can ensure that the system remains secure and reliable in the face of various presentation attacks. The current deep learning-based face anti-spoofing model has satisfactory results when the test data and training data obey the same distribution, but the accuracy of the model decreases considerably when the trained model infers in the scene outside the domain, such as cross-domain transfer and out-of-distribution scenarios. The problems that silent face anti-spoofing models will encounter in real scenarios, i.e., the models encounter unknown environments and unknown attack methods, are mainly described. The corresponding solutions are classified into four categories: methods based on domain adaptation, methods based on domain generalization, methods based on zero shot or few shot learning, and methods based on anomaly detection. Each solution and its deep learning model methods are summarized and compared. The mechanism, network structure, advantages, limitations and application scenarios of some major methods are summarized. After that, common public datasets, evaluation metrics, measurement protocols commonly used for face anti-spoofing in out of domain scenarios and test results of state-of-the-art methods under some protocols are introduced. Finally, the difficulties and challenges of face anti-spoofing in practical applications are discussed, and future research directions are summarized.

Key words: face anti-spoofing (FAS), domain adaptation, domain generalization, zero shot/few shot learning, anomaly detection, deep learning

中图分类号: