2020/02/27 by Martins Bruveris, Jochem Gietema, Bruveris, Martins +5
Computer Science · #Advanced Image and Video Retrieval Techniques #Computer Vision and Pattern Recognition (cs.CV) #FOS: Computer and information sciences #Face and Expression Recognition #Face recognition and analysis #cs.CV
paper · pdf · doi:10.48550/arxiv.2002.12093
Demographic Variation in the Performance of Biometric Systems workshop at WACV 2020
arxiv created 2020/02/27 · openalex publication_date 2020/02/27 · arxiv updated 2020/02/28 · openalex created_date 2025/10/10 · openalex updated_date 2026/07/28
As face recognition algorithms become more accurate and get deployed more widely, it becomes increasingly important to ensure that the algorithms work equally well for everyone. We study the geographic performance differentials-differences in false acceptance and false rejection rates across different countries-when comparing selfies against photos from ID documents. We show how to mitigate geographic performance differentials using sampling strategies despite large imbalances in the dataset. Using vanilla domain adaptation strategies to fine-tune a face recognition CNN on domain-specific doc-selfie data improves the performance of the model on such data, but, in the presence of imbalanced training data, also significantly increases the demographic bias. We then show how to mitigate this effect by employing sampling strategies to balance the training procedure.