[论文解读] Privacy-Protection Drone Patrol System based on Face Anonymization

Harim Lee, Myeung Un Kim+3 more|arXiv (Cornell University)|May 29, 2020

Face recognition and analysis参考文献 1被引用 4

一句话总结

本文提出了一种隐私保护型无人机巡逻系统，采用改进的CycleGAN与GauGAN结合的生成对抗网络（GAN）框架，在无人机的配套计算机上实时对人脸进行匿名化处理。该系统将原始人脸转换为逼真但身份不同的新面孔，同时保留对机器人感知至关重要的视觉特征，实现了有效的隐私保护，且未降低SLAM性能。

ABSTRACT

The robot market has been growing significantly and is expected to become 1.5 times larger in 2024 than what it was in 2019. Robots have attracted attention of security companies thanks to their mobility. These days, for security robots, unmanned aerial vehicles (UAVs) have quickly emerged by highlighting their advantage: they can even go to any hazardous place that humans cannot access. For UAVs, Drone has been a representative model and has several merits to consist of various sensors such as high-resolution cameras. Therefore, Drone is the most suitable as a mobile surveillance robot. These attractive advantages such as high-resolution cameras and mobility can be a double-edged sword, i.e., privacy infringement. Surveillance drones take videos with high-resolution to fulfill their role, however, those contain a lot of privacy sensitive information. The indiscriminate shooting is a critical issue for those who are very reluctant to be exposed. To tackle the privacy infringement, this work proposes face-anonymizing drone patrol system. In this system, one person's face in a video is transformed into a different face with facial components maintained. To construct our privacy-preserving system, we have adopted the latest generative adversarial networks frameworks and have some modifications on losses of those frameworks. Our face-anonymzing approach is evaluated with various public face-image and video dataset. Moreover, our system is evaluated with a customized drone consisting of a high-resolution camera, a companion computer, and a drone control computer. Finally, we confirm that our system can protect privacy sensitive information with our face-anonymzing algorithm while preserving the performance of robot perception, i.e., simultaneous localization and mapping.

研究动机与目标

• 为应对配备高分辨率摄像头的监控无人机日益增长的隐私侵犯担忧。
• 开发一种实时人脸匿名化系统，在不损害无人机感知能力的前提下保护个人身份。
• 将隐私保护型视觉处理流程集成到功能完整的无人机巡逻系统中，利用机载深度学习推理实现。
• 在公开数据集和真实世界无人机视频记录上评估系统的有效性。
• 确保人脸匿名化不会降低基于视觉的机器人感知性能，例如ORB-SLAM2的性能。

提出的方法

• 该系统结合CycleGAN进行域迁移和GauGAN进行语义图像合成，构建了用于人脸匿名化的两阶段GAN框架。
• 在生成器中应用改进的损失函数，以在替换身份的同时保留面部结构，确保逼真度与隐私保护。
• 匿名化处理流程部署于安装在定制无人机上的Jetson Xavier配套计算机，并配备ZED立体摄像头。
• 使用ROS框架实时处理视频帧，实现感知与隐私保护的端到端集成。
• 系统对每一帧进行处理，将检测到的人脸替换为新的、无法识别的人脸，同时保持空间与几何一致性。
• 采用ORB-SLAM2评估匿名化是否影响视觉里程计与特征提取，确保与标准机器人感知堆栈的兼容性。

实验结果

研究问题

• RQ1 基于GAN的人脸匿名化系统能否在实时无人机视频流中有效保护个人身份？
• RQ2 人脸匿名化在多大程度上影响基于视觉的机器人感知系统（如ORB-SLAM2）的性能？
• RQ3 结合CycleGAN与GauGAN的混合GAN架构能否实现高质量、保留结构特征的人脸匿名化？
• RQ4 该系统在多样化的面部数据集和真实世界无人机视频场景中的泛化能力如何？
• RQ5 在Jetson Xavier等嵌入式平台部署此类隐私保护系统是否具备实时运行的可行性？

主要发现

• 所提出的匿名化系统成功将原始人脸转换为新的、无法识别的人脸，同时保留了面部组件与结构特征。
• 在CelebA-HQ、Helen和FaceScrub等多个数据集上的评估表明，系统在不同面部外观下均保持一致且高质量的匿名化效果。
• ORB-SLAM2在原始视频帧与匿名化视频帧中的特征提取性能几乎完全一致，检测到的特征点偏差极小。
• ZED摄像头与检测到的特征（如人脸区域）之间的距离测量值保持在可接受范围内，证实深度估计未出现性能下降。
• 系统在Jetson Xavier配套计算机上实现了实时处理，证明了其在移动无人机上部署的可行性。
• 匿名化处理流程与ROS的集成实现了无人机上摄像头、处理单元与感知系统之间的无缝通信。