Biometric systems leak enough feedback to reconstruct user templates
How Much Feedback Is Needed to Break Template Protection? The Geometric Vulnerability of Biometric Authentication
Cryptography and Security
Summary
Biometric systems like face recognition protect stored templates to keep them private, but the way they respond to login attempts can reveal hidden information. The authors studied how even just a simple yes-or-no answer to an authentication attempt (accept or reject) can be used to figure out the original biometric data. They developed new attack methods that need fewer tries and use knowledge about typical face shapes to guess the template more accurately. Their findings show that biometric security has a fundamental weakness because the authentication process itself can leak detailed geometric information about users.
What this means in practice
- •For biometric system developers: Evaluate and strengthen authentication systems against partial feedback attacks exploiting geometric leakage to enhance template privacy.
- •For security auditors: Conduct more realistic penetration tests on biometric systems by using limited feedback attacks that reveal vulnerabilities in template protection.
Authors
Eliron Rahimi, Margarita Osadchy, Orr Dunkelman
Abstract
Biometric authentication systems increasingly use privacy-preserving mechanisms to protect sensitive templates. However, the authentication interface itself may still leak information that enables reconstruction attacks. Prior attacks exploit released similarity scores, whereas more advanced systems reveal only the binary accept/reject decision. This binary feedback provides substantially less information, making template reconstruction far more challenging. In this work, we introduce query-efficient reconstruction attacks under extremely limited feedback, addressing two key limitations of prior attacks. First, we introduce the first attack that reconstructs protected biometric templates using only authentication success/failure feedback by exploiting the geometric constraints of biometric authentication. Second, we introduce distribution-aware query-efficient reconstruction methods for score-releasing systems, achieving a more realistic threat model by reducing the number of required queries by a factor of $5\times$ compared to recent prior score-based attacks. Finally, we combine these two insights by incorporating a pretrained facial diffusion prior into our geometric binary attack, reducing the number of required queries by a factor of $5\times$. Our findings reveal a fundamental security limitation of biometric authentication: independently of the template protection mechanism, the authentication process itself exposes geometric information that can enable perfect template reconstruction up to numerical precision. Moreover, the distribution of biometric templates can be leveraged to make significantly more effective reconstruction attacks.