Data pertaining to Chapter 4 of the PhD dissertation: “Protecting User Privacy in Gaze Estimation Services"

This dataset accompanies the paper “PrivateGaze: Preserving User Privacy in Black-box Mobile Gaze Tracking Services” (Chapter 4 of the PhD dissertation). This research focuses on protecting user privacy in black-box gaze tracking services without compromising gaze estimation performance. Specifically, we proposed a novel framework to train a privacy preserver that converts full-face images into obfuscated counterparts, which are effective for gaze estimation while containing no privacy information. This dataset contains the codes to reproduce the results presented in the paper and the real-time video demo of the proposed method.

本数据集配套于论文《PrivateGaze：在黑盒（black-box）式移动端视线追踪服务中保护用户隐私》（博士学位论文第4章）。本研究聚焦于在不损害视线估计（gaze estimation）性能的前提下，保护黑盒式视线追踪服务中的用户隐私。具体而言，我们提出了一种新颖的框架，用于训练隐私保护器，该保护器可将全脸图像转换为混淆后的图像版本，此类混淆图像既可有效用于视线估计任务，又不包含任何隐私信息。本数据集包含了复现论文中所述实验结果所需的代码，以及所提方法的实时视频演示。