华东师范大学学报(自然科学版) ›› 2023, Vol. 2023 ›› Issue (6): 49-60.doi: 10.3969/j.issn.1000-5641.2023.06.005

• 计算机科学 • 上一篇    下一篇

面向身份相互关系一致性的人脸去识别化方法

步一凡1, 王晓玲1, 贺珂珂2, 卢兴见1, 王文萱2,*()   

  1. 1. 华东师范大学 计算机科学与技术学院, 上海 200062
    2. 复旦大学 计算机科学技术学院, 上海 200438
  • 收稿日期:2022-06-18 出版日期:2023-11-25 发布日期:2023-11-23
  • 通讯作者: 王文萱 E-mail:wxwang.iris@gmail.com

Towards an identity inter-relationship-consistent face de-identification method

Yifan BU1, Xiaoling WANG1, Keke HE2, Xingjian LU1, Wenxuan WANG2,*()   

  1. 1. School of Computer Science and Technology, East China Normal University, Shanghai 200062, China
    2. School of Computer Science, Fudan University, Shanghai 200438, China
  • Received:2022-06-18 Online:2023-11-25 Published:2023-11-23
  • Contact: Wenxuan WANG E-mail:wxwang.iris@gmail.com

摘要:

智能手机、监控摄像头等智能设备的普及导致了严重的人脸隐私安全问题. 人脸去识别化技术被认为是解决上述问题的有效途径, 该技术旨在将人脸图像中的身份信息去除以有效保护人脸的隐私. 然而, 现有的研究工作大多缺乏对去识别化人脸图像在身份上显式的控制与可控的变化, 导致去识别化后的图像无法应用于人脸验证和检索等与身份相关的任务. 因此, 考虑到去识别化人脸图像在身份相关任务上的可用性, 提出了一种新的面向身份相互关系一致性的人脸去识别化任务, 即任意两个被去识别化的人脸之间的身份相互关系与去识别化前的人脸保持一致. 为解决这一问题, 引入了一种任务驱动的身份相互关系一致性生成对抗网络算法. 该方法设计了一种基于特征向量旋转的去识别化模块, 将原有的身份特征可控修改为具有身份相互关系一致性的去识别化特征. 此外, 还引入了身份控制损失, 以保证去识别化后生成的人脸身份的准确性. 大量的定性和定量结果表明, 提出的方法在人脸的去识别化和去识别人脸的身份相互关系一致性上的表现远优于现有的方法.

关键词: 隐私保护, 人脸去识别化, 生成对抗网络, 身份相互关系一致性

Abstract:

The popularity of intelligent devices such as smartphones and surveillance cameras has led to serious face privacy problems. Face de-identification is considered an effective tool for protecting face privacy by concealing identity information. However, most de-identification methods lack explicit control and controllable changes in identifying de-identified face images, resulting in de-identified images that are inapplicable to face authentication and retrieval and other identity-related tasks. Therefore, this study proposes an identity inter-relationship-consistent face de-identification task in which the identity inter-relationship between two arbitrary de-identified faces maintained the same as before de-identification. To this end, a task-driven identity inter-relationship consistent generative adversarial network is introduced to generate de-identified faces with a consistent identity inter-relationship. A rotation-based de-identifier was designed to modify the original identity features to be de-identified with identity inter-relationship consistency. In addition, identity control loss is introduced to guarantee a precise identity generation using a de-identified generator. Qualitative and quantitative results show that our method achieves improvements compared with exiting methods for de-identifying de-identified faces as well as for maintaining their identity inter-relationship consistent.

Key words: privacy protection, face de-identification, generative adversarial network, identity inter-relationship consistency

中图分类号: