Lights, Camera, Matching: The Role of Image Illumination in Fair Face Recognition

2025-01-15

AI Summary

This study addresses algorithmic unfairness in facial recognition, specifically the accuracy gap between White and Black women attributable to image brightness disparities. We propose a systematic illumination calibration method that first decouples median skin-region brightness from its overall distributional characteristics; skin regions are precisely localized using facial landmarks, and illumination is normalized via median brightness adjustment and histogram matching. A novel contribution is the use of the discriminability index *d′* to quantify fairness improvement across demographic groups. Experiments demonstrate that distribution-level brightness calibration significantly outperforms median-only normalization: *d′* decreases by up to 57.6%, while mean match scores for White and Black women improve by 5.9% and 3.7%, respectively—effectively narrowing the performance gap between groups.

Technology Category

• Computer Vision: Bias, Fairness & Privacy
• Machine Learning: Ethics, Bias, and Fairness
• Natural Language Processing: Ethics — Bias, Fairness, Transparency & Privacy

Application Category

• User Modeling, Personalization and Recommendation: Fairness-aware retrieval and ranking
• Responsible Web: Algorithmic accountability and transparency on the web
• Social Networks and Social Media: Fairness and bias in social network and social media analysis

Abstract

Facial brightness is a key image quality factor impacting face recognition accuracy differentials across demographic groups. In this work, we aim to decrease the accuracy gap between the similarity score distributions for Caucasian and African American female mated image pairs, as measured by d' between distributions. To balance brightness across demographic groups, we conduct three experiments, interpreting brightness in the face skin region either as median pixel value or as the distribution of pixel values. Balancing based on median brightness alone yields up to a 46.8% decrease in d', while balancing based on brightness distribution yields up to a 57.6% decrease. In all three cases, the similarity scores of the individual distributions improve, with mean scores maximally improving 5.9% for Caucasian females and 3.7% for African American females.

Problem

• Facial Recognition
• Accuracy Disparity
• Illumination Variation

Innovation

• Brightness Adjustment
• Facial Recognition Fairness
• Algorithmic Bias Reduction