Pré-Publication, Document De Travail Année : 2025

Fairer analysis and demographically balanced face generation for fairer face verification

Résumé

Face recognition and verification are two computer vision tasks whose performances have advanced with the introduction of deep representations. However, ethical, legal, and technical challenges due to the sensitive nature of face data and biases in real-world training datasets hinder their development. Generative AI addresses privacy by creating fictitious identities, but fairness problems remain. Using the existing DCFace SOTA framework, we introduce a new controlled generation pipeline that improves fairness. Through classical fairness metrics and a proposed in-depth statistical analysis based on logit models and ANOVA, we show that our generation pipeline improves fairness more than other bias mitigation approaches while slightly improving raw performance.
Fichier principal
Vignette du fichier
afm2024fairer_analysis_face.pdf (9.76 Mo) Télécharger le fichier
Origine Fichiers produits par l'(les) auteur(s)

Dates et versions

cea-04910990 , version 1 (24-01-2025)

Licence

Identifiants

  • HAL Id : cea-04910990 , version 1

Citer

Alexandre Fournier-Montgieux, Michael Soumm, Adrian Popescu, Bertrand Luvison, Hervé Le Borgne. Fairer analysis and demographically balanced face generation for fairer face verification. 2025. ⟨cea-04910990⟩
0 Consultations
0 Téléchargements

Partager

More