FaceMINT : a library for gaining insights into biometric face recognition via mechanistic interpretability

Rot, Peter (avtor), Jutreša, Robert (avtor), Peer, Peter (avtor), Štruc, Vitomir (avtor), Scheirer, Walter J. (avtor), Grm, Klemen (avtor)

Deep-learning models, including those used in biometric recognition, have achieved remarkable performance on benchmark datasets as well as real-world recognition tasks. However, a major drawback of these models is their lack of transparency in decision-making. Mechanistic interpretability has emerged as a promising research field intended to help us gain insights into such models, but its application to biometric data remains limited. In this work, we bridge this gap by introducing the FaceMINT library, a publicly available Python library (build on top of Pytorch) that enables biometric researchers to inspect their models through mechanistic interpretability. It provides a plug-and-play solution that allows researchers to seamlessly switch between the analyzed biometric models, evaluate state-of-the-art sparse autoencoders, select from various image parametrizations, and fine-tune hyperparameters. Using a large scale Glint360K dataset, we demonstrate the usability of FaceMINT by applying its functionality to two state-of-the-art (deep-learning) face recognition models: AdaFace, based on Convolutional Neural Networks (CNN), and SwinFace, based on transformers. The proposed library implements various sparse auto-encoders (SAEs), including vanilla SAE, Gated SAE, JumpReLU SAE, and TopK SAE, which have achieved state-of-the-art results in the mechanistic interpretability of large language models. Our study highlights the promise of mechanistic interpretability in the biometric field, providing new avenues for researchers to explore model transparency and refine biometric recognition systems. The library is publicly available at www.gitlab.com/peterrot/facemint.

• Jezik: Angleški jezik
• Ključne besede: face recognition, biometrics, mechanistic interpretability, sparse autoencoder, library
• Vrsta gradiva: Članek v reviji
• Tipologija: 1.01 - Izvirni znanstveni članek
• Organizacija: FE - Fakulteta za elektrotehniko
• Status publikacije: Objavljeno
• Različica publikacije: Objavljena publikacija
• Leto izida: 2026
• Št. strani: 20 str.
• Številčenje: Vol. 165, art. 105804
• PID: 20.500.12556/RUL-175964
• UDK: 004.93'1
• ISSN pri članku: 0262-8856
• DOI: 10.1016/j.imavis.2025.105804
• COBISS.SI-ID: 257243651
• Datum objave v RUL: 14.11.2025

Gradivo je del revije

• Naslov: Image and vision computing
• Skrajšan naslov: Image vis. comput.
• Založnik: Butterworth Scientific
• ISSN: 0262-8856
• COBISS.SI-ID: 25590016

Licence

• Licenca: CC BY 4.0, Creative Commons Priznanje avtorstva 4.0 Mednarodna
• Povezava: http://creativecommons.org/licenses/by/4.0/deed.sl
• Opis: To je standardna licenca Creative Commons, ki daje uporabnikom največ možnosti za nadaljnjo uporabo dela, pri čemer morajo navesti avtorja.

Sekundarni jezik

• Jezik: Slovenski jezik
• Ključne besede: razpoznavanje obrazov, biometrija, interpretacija mehanizmov, redki samokodirniki, knjižnica

Projekti

• Financer: ARIS - Javna agencija za znanstvenoraziskovalno in inovacijsko dejavnost Republike Slovenije
• Številka projekta: J2-50069
• Naslov: Interpretacija mehanizmov za razložljivo biometrično umetno inteligenco (MIXBAI)

• Financer: ARIS - Javna agencija za znanstvenoraziskovalno in inovacijsko dejavnost Republike Slovenije
• Številka projekta: P2-0250
• Naslov: Metrologija in biometrični sistemi