Score Normalization for Demographic Fairness in Face Recognition

TL;DR

This paper proposes demographic-aware score normalization techniques to improve fairness in face recognition systems without sacrificing verification accuracy, addressing distribution disparities across demographic groups.

Contribution

It introduces demographic-integrated score normalization methods extending Z/T-norm, enhancing fairness across demographics without retraining models.

Findings

Improved fairness across gender and ethnicity groups

Maintained verification performance while increasing fairness

Effective on multiple face recognition networks

Abstract

Fair biometric algorithms have similar verification performance across different demographic groups given a single decision threshold. Unfortunately, for state-of-the-art face recognition networks, score distributions differ between demographics. Contrary to work that tries to align those distributions by extra training or fine-tuning, we solely focus on score post-processing methods. As proved, well-known sample-centered score normalization techniques, Z-norm and T-norm, do not improve fairness for high-security operating points. Thus, we extend the standard Z/T-norm to integrate demographic information in normalization. Additionally, we investigate several possibilities to incorporate cohort similarities for both genuine and impostor pairs per demographic to improve fairness across different operating points. We run experiments on two datasets with different demographics (gender and ethnicity) and show that our techniques generally improve the overall fairness of five state-of-the-art pre-trained face recognition networks, without downgrading verification performance. We also indicate that an equal contribution of False Match Rate (FMR) and False Non-Match Rate (FNMR) in fairness evaluation is required for the highest gains. Code and protocols are available.