ChildGAN: Large Scale Synthetic Child Facial Data Using Domain Adaptation in StyleGAN

TL;DR

This paper introduces ChildGAN, a novel GAN-based method for generating large-scale, high-quality synthetic child facial data with diverse features, validated through multiple computer vision applications, offering a cost-effective alternative to real data collection.

Contribution

ChildGAN is a new GAN framework that produces realistic, diverse synthetic child facial images using domain transfer and StyleGAN2, creating a large dataset for research and applications.

Findings

Generated dataset exceeds 300,000 samples with diverse features.

Synthetic data performs well in face recognition and detection tasks.

High-quality synthetic data can replace real child data in various applications.

Abstract

In this research work, we proposed a novel ChildGAN, a pair of GAN networks for generating synthetic boys and girls facial data derived from StyleGAN2. ChildGAN is built by performing smooth domain transfer using transfer learning. It provides photo-realistic, high-quality data samples. A large-scale dataset is rendered with a variety of smart facial transformations: facial expressions, age progression, eye blink effects, head pose, skin and hair color variations, and variable lighting conditions. The dataset comprises more than 300k distinct data samples. Further, the uniqueness and characteristics of the rendered facial features are validated by running different computer vision application tests which include CNN-based child gender classifier, face localization and facial landmarks detection test, identity similarity evaluation using ArcFace, and lastly running eye detection and eye aspect ratio tests. The results demonstrate that synthetic child facial data of high quality offers an alternative to the cost and complexity of collecting a large-scale dataset from real children.