Longitudinal Study of Child Face Recognition

TL;DR

This study evaluates how well face recognition systems perform on children over several years, highlighting challenges and improvements in identifying children for safety and law enforcement.

Contribution

It provides a comprehensive longitudinal analysis of child face recognition performance and fine-tunes open-source models for better accuracy in this domain.

Findings

Face recognition accuracy decreases over time in children.

Fine-tuning improves open-source face matcher performance.

Analysis supports use in child identification and safety applications.

Abstract

We present a longitudinal study of face recognition performance on Children Longitudinal Face (CLF) dataset containing 3,682 face images of 919 subjects, in the age group [2, 18] years. Each subject has at least four face images acquired over a time span of up to six years. Face comparison scores are obtained from (i) a state-of-the-art COTS matcher (COTS-A), (ii) an open-source matcher (FaceNet), and (iii) a simple sum fusion of scores obtained from COTS-A and FaceNet matchers. To improve the performance of the open-source FaceNet matcher for child face recognition, we were able to fine-tune it on an independent training set of 3,294 face images of 1,119 children in the age group [3, 18] years. Multilevel statistical models are fit to genuine comparison scores from the CLF dataset to determine the decrease in face recognition accuracy over time. Additionally, we analyze both the verification and open-set identification accuracies in order to evaluate state-of-the-art face recognition technology for tracing and identifying children lost at a young age as victims of child trafficking or abduction.