Biometric authentication often relies on noisy, high-dimensional data, making privacy-preserving matching a challenging problem. This internship explores new cryptographic protocols that enable biometric verification without revealing the underlying biometric information.

The focus is on building efficient cryptographic techniques and protocols for the key building blocks of a private biometric pipeline. Optional: integration with advanced authentication protocols such as PAKE, MPC, 2PC, or related secure computing techniques.

The goal is to develop practical methods which form part of a bigger platform for secure, noise-tolerant biometric verification.

Duration: flexible, to be agreed (typically 3-5 months), starting time flexible

• PhD student in Computer Science/Engineering.

• Experience with MPC, FHE or Zero-Knowledge Proofs.

• Strong programming skills (Rust, C++, …).

• Interest in privacy-preserving authentication, secure computation, or cryptographic protocol design.

• Optional: knowledge of PAKE, 2PC, or related protocols

• You are expected to familiarize yourself with a range of different cryptographic techniques: MPC, FHE & ZKP, and understand their limits and tradeoffs.

• Given your interest, you can optionally familiar yourself with privacy-preserving fuzzy protocols used in biometric authentication, such as fuzzy PAKE/PSI/extractors.

• Explore the most efficient and effective protocol and tools for the problems at hand and ultimately selecting a few promising combinations for further development.

• You will develop functional prototypes of the selective data filtering, privacy-preserving feature extraction and thoroughly evaluate the results.

• Consolidate your research outcome in a scientific publication.