we are pleased to announce the IJCB 2026 Competition on Adapting Foundation Models for Face Recognition, organized as part of the International Joint Conference on Biometrics (IJCB 2026).

This competition aims to foster research on adaptation strategies for foundation models in face recognition (FR), particularly in challenging scenarios with limited training data or where privacy constraints restrict access to real biometric datasets.

Competition Tracks:

Track 1 – Full Data Track: Participants can use a large-scale provided training dataset to adapt foundation models.

Track 2 – Limited Data Track: Participants are restricted to a small subset of the training data, encouraging data-efficient adaptation methods.

The competition is based on fully synthetic face datasets generated using state-of-the-art generative techniques. Participants are required to exclusively use the provided datasets for model adaptation.

We warmly invite researchers, students, and industry practitioners from computer vision, biometrics, and machine learning to participate.

Competition website and registration: https://sites.google.com/view/ijcb-afmfr-2026/home

The competition is co-organized with Tahar Chettaoui, Shiqi Yu, Vitomir Štruc, and Fadi Boutros, in affiliation to ATHENE Center, Fraunhofer IGD, TU Darmstadt, University of Ljubljana, and Southern University of Science and Technology.

We look forward to your participation!

Best regards,

Naser Damer

Fraunhofer IGD and TU Darmstadt