Alexander Unnervik

Alexander Unnervik is an active researcher in computer vision and biometrics with a focus on face recognition systems, synthetic data generation, and security of deep learning models. His work spans multiple prestigious conferences including CVPR, WACV, BIOSIG, and ECCV workshops.

His research interests center around critical challenges in biometric systems, particularly addressing security vulnerabilities in face recognition systems through anomaly detection approaches for backdoored neural networks. He has made significant contributions to synthetic data generation for improving face recognition systems, organizing and analyzing multiple FRCSyn challenges that have become important benchmarks in the field. His work bridges theoretical security concerns with practical applications in biometric systems.

Analysis of his publication trends reveals a strong progression from automotive applications (driver monitoring systems around 2019-2020) to specialized biometric security research (2022 onward). His recent work demonstrates increasing leadership in organizing major challenges in synthetic data for face recognition, with numerous high-impact publications in 2024 across multiple venues. The research shows a clear trajectory toward addressing security concerns in biometric AI systems while developing robust evaluation frameworks.

Unnervik has established a strong collaborative network with leading researchers in biometrics, particularly working closely with Sébastien Marcel and other members of the international biometrics research community. His work appears in top-tier computer vision and biometrics venues, indicating recognition by the research community.

His laboratory and team affiliations appear connected to major biometrics research initiatives, particularly those focused on synthetic data generation and security evaluation for face recognition systems. He has been instrumental in organizing the FRCSyn challenges, which have brought together researchers from multiple institutions worldwide to address critical problems in biometric security.