Longitudinal Study of Child Face Recognition

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 COTSA 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.

1. Introduction

The United Nations Convention on the Rights of the Child defines child as “a human being below the age of 18 years unless under the law applicable to the child, majority is attained earlier” [1]. This definition is ratified by 192 of the 194 countries that are members of the United Nations. According to the United Nations Children's Fund (UNICEF), nearly 2 million children under the age of 20 are subjected to prostitution in the global sex trade. On average, victims range from 11 to 14 years old and are expected to survive only 7 years. The United Nations Office on Drugs and Crime reports the percentage of child trafficking victims has risen about 25% from 2009 to 2012, where the victims are in the age group of 1 to 18 years [2]. For every three child victims, two are girls and one is a boy. According to Kolkata's Child in Need Institute, 1, 628 kidnapped children, in the age group of 4 to 15 years, were retrieved from a single railway station; among these, 134 were girls and the youngest was only four years old [3]. Of course, these are official statistics, and do not necessarily reflect the true numbers of child kidnapping and sex trafficking in a population of around 1.2 billion in India. Figure 1:

Sharbat gula (a) at age 12, photographed in 1984 and (b) at age 30, photographed in 2002 [14]. She was identified based on iris recognition [15].

Figure 2:

Saroo breirley (a) before he went missing at age 5 and (b) after reuniting with his biological mother at age 30 [16].