Password-conditioned Anonymization and Deanonymization with Face Identity Transformers
Password-conditioned Anonymization and Deanonymization with Face Identity Transformers
复制标题
DOI：
10.1007/978-3-030-58592-1_43
复制
发表时间：
2019-11
期刊：
影响因子：
--
通讯作者：
Xiuye Gu;Weixin Luo;M. Ryoo;Yong Jae Lee
中科院分区：
文献类型：
--
作者：
Xiuye Gu;Weixin Luo;M. Ryoo;Yong Jae Lee
相机在我们的日常生活中非常普遍，并使许多基于计算机视觉技术的有用系统成为可能，例如智能相机和家用机器人。然而，由于所捕获的图像/视频可能包含隐私敏感信息（例如， 面部识别）。我们提出了一个novelface身份transformerwhich使自动照片逼真的基于密码的匿名化和deanonymization的人脸出现在视觉数据。我们的面部身份Transformer经过训练，可以（1）在匿名化后删除面部身份信息，（2）在给定正确密码时恢复原始面部，以及（3）在给定错误密码时返回错误但照片般逼真的面部。通过我们精心设计的密码方案和多任务学习目标，我们使用同一个网络实现了匿名和去匿名。大量的实验表明，我们的方法，使多模态密码条件匿名和deanonymizations，而不牺牲隐私相比，现有的匿名方法。
Cameras are prevalent in our daily lives, and enable many useful systems built upon computer vision technologies such as smart cameras and home robots for service applications. However, there is also an increasing societal concern as the captured images/videos may contain privacy-sensitive information (e.g., face identity). We propose a novelface identity transformerwhich enables automated photo-realistic password-based anonymization and deanonymization of human faces appearing in visual data. Our face identity transformer is trained to (1) remove face identity information after anonymization, (2) recover the original face when given the correct password, and (3) return a wrong—but photo-realistic—face given a wrong password. With our carefully designed password scheme and multi-task learning objective, we achieve both anonymization and deanonymization using the same single network. Extensive experiments show that our method enables multimodal password conditioned anonymizations and deanonymizations, without sacrificing privacy compared to existing anonymization methods.