Unmasking the Uniqueness: A Glimpse into Age-Invariant Face Recognition of Indigenous African Faces

TL;DR

This work addresses the underrepresentation of Indigenous African faces in age-invariant face recognition (AIFR) by building and evaluating a VGGFace-based AIFR model trained on a newly created Indigenous African dataset, FAGE_v2, and compared to an African-American subset of CACD. FAGE_v2 comprises 5,000 images from 500 individuals across 10 African countries, while CACD provides an African-American benchmark; the model achieves 81.8% accuracy on FAGE_v2 and 91.5% on CACD, revealing significant performance gaps linked to ethnic representation and ageing pattern differences. The study demonstrates that indigenous ageing patterns differ from non-indigenous ones and that pre-trained models biased toward non-indigenous data underperform on Indigenous African faces. The authors advocate expanding Indigenous African datasets and extending cross-ethnicity analyses to develop robust AIFR systems for African populations and other underrepresented groups.

Abstract

The task of recognizing the age-separated faces of an individual, Age-Invariant Face Recognition (AIFR), has received considerable research efforts in Europe, America, and Asia, compared to Africa. Thus, AIFR research efforts have often under-represented/misrepresented the African ethnicity with non-indigenous Africans. This work developed an AIFR system for indigenous African faces to reduce the misrepresentation of African ethnicity in facial image analysis research. We adopted a pre-trained deep learning model (VGGFace) for AIFR on a dataset of 5,000 indigenous African faces (FAGE\_v2) collected for this study. FAGE\_v2 was curated via Internet image searches of 500 individuals evenly distributed across 10 African countries. VGGFace was trained on FAGE\_v2 to obtain the best accuracy of 81.80\%. We also performed experiments on an African-American subset of the CACD dataset and obtained the best accuracy of 91.5\%. The results show a significant difference in the recognition accuracies of indigenous versus non-indigenous Africans.

Figures (2)

• Figure 1: A deep learning-based AIFR model using VGG-FACE
• Figure 2: Sample of the faces in the FAGE_v2 dataset. Each row shows the age-separated images of an individual.