arXiv cs.AI· Dawid Wolkiewicz, Anastasiya Pechko, Przemys{\l}aw Spurek, Piotr Syga·· 11 小时前AI 评分34
AEGIS：用对抗扰动保护 3D 人脸虚拟形象隐私
AEGIS: Preserving Privacy of 3D Facial Avatars with Adversarial Perturbations
AI 导读
AEGIS 是首个针对 3D Gaussian 虚拟形象的隐私保护身份遮蔽框架，通过对高斯原语的基色（DC）系数施加对抗扰动，在多个预训练人脸验证网络引导下将保护嵌入 3D 表示，跨视角与动画生效且无需重训或改动几何。面对黑盒识别器，它将人脸检索与验证准确率降至接近零，同时保持高感知保真度并保留年龄、性别、情绪等面部属性。该工作还提出跨视角泄漏度量，论文已被 WACV 2027 接收。
正文 · 原文
Abstract:The growing adoption of photorealistic 3D facial avatars, particularly those utilizing efficient 3D Gaussian Splatting representations, introduces new risks of online identity theft, especially in systems that rely on biometric authentication. While effective adversarial masking methods have been developed for 2D images, a significant gap remains in achieving robust, viewpoint-consistent identity protection for dynamic 3D avatars. To address this, we present AEGIS, the first privacy-preserving identity masking framework for 3D Gaussian avatars, concealing identity from automated face recognition while preserving the avatar's perceptual realism and animation usability. AEGIS applies adversarial perturbations to the base (DC) color coefficients of the Gaussian primitives, guided by an ensemble of pre-trained face verification networks, embedding the protection directly in the 3D representation so that it applies across viewpoints and animations, without retraining or modifying the avatar's geometry. Evaluated against black-box recognizers, AEGIS reduces face retrieval and verification accuracy to near-zero while maintaining high perceptual fidelity. It also preserves key facial attributes such as age, gender, and emotion, demonstrating strong privacy protection with minimal visual distortion. We further introduce a viewpoint-robust leakage metric that quantifies residual identity leakage across an avatar's rendered viewpoints, which single-view evaluation cannot capture.
| Comments: | Accepted to WACV 2027. Source code at: this https URL |
| Subjects: | Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI) |
| Cite as: | arXiv:2511.17747 [cs.CV] |
| (or arXiv:2511.17747v2 [cs.CV] for this version) | |
| https://doi.org/10.48550/arXiv.2511.17747 arXiv-issued DOI via DataCite |
Submission history
From: Dawid Wolkiewicz [view email]
[v1]
Fri, 21 Nov 2025 19:57:28 UTC (31,299 KB)
[v2]
Wed, 4 Nov 2026 16:37:45 UTC (15,775 KB)
来源：arXiv cs.AI · arxiv.org