FaceAnonyMixer: Cancelable Faces via Identity Consistent Latent Space Mixing

2025-08-07

AI Summary

Existing face anonymization methods struggle to simultaneously achieve identity obfuscation and biometric template protection—namely, revocability, unlinkability, and irreversibility. To address this, we propose a revocable face generation framework grounded in pre-trained generative models: it performs irreversible mixing of real-face latent encodings with key-derived synthetic codes in the latent space, enabling privacy-preserving yet controllable face reconstruction. Our method employs multi-objective loss optimization to ensure generated images maintain semantic fidelity while strictly satisfying biometric template security requirements. Experiments on standard benchmarks demonstrate that our approach reduces identity recognition accuracy on commercial APIs by over 11% compared to state-of-the-art methods, while rigorously guaranteeing revocability, unlinkability, and irreversibility. To the best of our knowledge, this is the first work to systematically realize all three biometric template security principles—revocability, unlinkability, and irreversibility—within a generative face anonymization framework.

Technology Category

• Computer Vision: Biometrics, Face, Gesture & Pose
• Machine Learning: Privacy
• Natural Language Processing: Generation

Application Category

• Security and Privacy: Large-scale security measurements
• User Modeling, Personalization and Recommendation: User privacy protection in personalized systems
• Economics, Online Markets and Human Computation: Economic ramifications for generative AI infrastructure and applications

Abstract

Advancements in face recognition (FR) technologies have amplified privacy concerns, necessitating methods that protect identity while maintaining recognition utility. Existing face anonymization methods typically focus on obscuring identity but fail to meet the requirements of biometric template protection, including revocability, unlinkability, and irreversibility. We propose FaceAnonyMixer, a cancelable face generation framework that leverages the latent space of a pre-trained generative model to synthesize privacy-preserving face images. The core idea of FaceAnonyMixer is to irreversibly mix the latent code of a real face image with a synthetic code derived from a revocable key. The mixed latent code is further refined through a carefully designed multi-objective loss to satisfy all cancelable biometric requirements. FaceAnonyMixer is capable of generating high-quality cancelable faces that can be directly matched using existing FR systems without requiring any modifications. Extensive experiments on benchmark datasets demonstrate that FaceAnonyMixer delivers superior recognition accuracy while providing significantly stronger privacy protection, achieving over an 11% gain on commercial API compared to recent cancelable biometric methods. Code is available at: https://github.com/talha-alam/faceanonymixer.

Problem

• Protect face identity while maintaining recognition utility
• Meet biometric template protection requirements like revocability
• Generate high-quality cancelable faces for existing FR systems

Innovation

• Mixes latent codes for cancelable faces
• Uses multi-objective loss for biometric requirements
• Generates high-quality faces for FR systems