Re-Training StyleGAN -- A First Step Towards Building Large, Scalable Synthetic Facial Datasets
This work addresses the need for scalable synthetic facial datasets for researchers, but it is incremental as it focuses on retraining an existing method without introducing new techniques.
The paper tackled the problem of generating synthetic facial data by retraining StyleGAN on various public datasets, presenting practical challenges and comparative validation results, but did not include concrete numbers for the outcomes.
StyleGAN is a state-of-art generative adversarial network architecture that generates random 2D high-quality synthetic facial data samples. In this paper, we recap the StyleGAN architecture and training methodology and present our experiences of retraining it on a number of alternative public datasets. Practical issues and challenges arising from the retraining process are discussed. Tests and validation results are presented and a comparative analysis of several different re-trained StyleGAN weightings is provided 1. The role of this tool in building large, scalable datasets of synthetic facial data is also discussed.