Towards Real-Time Privacy-Preserving Minutiae Matching

While biometric data, such as fingerprints, are increasingly used for identification and authentication, their inability to be revoked once compromised raises privacy concerns. To address these concerns, this work explores the use of Multiparty Computation (MPC), which enables secure computations on encrypted data for privacy-preserving fingerprint matching. Despite MPC's well-known drawback of slowing down computation, recent advances have made it a viable option for real-world applications. Our research focuses on implementing and optimizing a minutiae-based fingerprint matching algorithm with MPC that addresses the challenge of preserving privacy while ensuring reasonable computation times. Furthermore, this work presents a privacy-preserving implementation of SourceAFIS using the MP-SPDZ framework. In addition, we demonstrate the adaptability of the MPC implementation by integrating different minutiae extractors, leading to significant enhancements in error rates. These findings underscore the feasibility and effectiveness of using modern MPC techniques for privacy-preserving fingerprint matching and paves the way for further optimizations.