Racial Faces in-the-Wild: Reducing Racial Bias by Information Maximization Adaptation Network

TL;DR

This paper introduces the Racial Faces in-the-Wild dataset and proposes a deep unsupervised domain adaptation network to reduce racial bias in face recognition, demonstrating improved cross-race generalization.

Contribution

It provides a new racial face dataset and a novel deep information maximization adaptation network for mitigating racial bias in face recognition.

Findings

IMAN reduces racial bias in face recognition models.

IMAN improves cross-race recognition accuracy.

Extensive experiments validate the effectiveness of IMAN.

Abstract

Racial bias is an important issue in biometric, but has not been thoroughly studied in deep face recognition. In this paper, we first contribute a dedicated dataset called Racial Faces in-the-Wild (RFW) database, on which we firmly validated the racial bias of four commercial APIs and four state-of-the-art (SOTA) algorithms. Then, we further present the solution using deep unsupervised domain adaptation and propose a deep information maximization adaptation network (IMAN) to alleviate this bias by using Caucasian as source domain and other races as target domains. This unsupervised method simultaneously aligns global distribution to decrease race gap at domain-level, and learns the discriminative target representations at cluster level. A novel mutual information loss is proposed to further enhance the discriminative ability of network output without label information. Extensive experiments on RFW, GBU, and IJB-A databases show that IMAN successfully learns features that generalize well across different races and across different databases.