Generative Adversarial Networks (GANs) are one of the hottest topics in computer science in recent times. They are a clever way of training a generative model (unsupervised learning) by framing the problem as a supervised learning problem. The main idea is that two different models are trained simultaneously by an adversarial process. Generative adversarial networks are based on a game theoretic scenario in which the generator network must compete against an adversary. The generator network directly produces samples.

Taken from: A Style-Based Generator Architecture for Generative Adversarial Networks. We can review each of these changes in more detail. The StyleGAN generator and discriminator models are trained using the progressive growing GAN training method. This means that both models start with small images, in this case, 4 4 images. The models are fit until stable, then both discriminator and generator are expanded to double the width and height (quadruple the area), e.g. 8 8. A new block is added to each model to support the larger image size, which is faded in slowly over training. Once faded-in, the models are again trained until reasonably stable and the process is repeated with ever-larger image sizes until the desired target image size is met, such as 1024 1024.

Authors: Tero Karras (NVIDIA) Samuli Laine (NVIDIA) Timo Aila (NVIDIA) Abstract: We propose an alternative generator architecture for generative adversarial networks, borrowing from style transfer literature. The new architecture leads to an automatically learned, unsupervised separation of high-level attributes (e.g., pose and identity when trained on human faces) and stochastic variation in the generated images (e.g., freckles, hair), and it enables intuitive, scale-specific control of the synthesis. The new generator improves the state-of-the-art in terms of traditional distribution quality metrics, leads to demonstrably better interpolation properties, and also better disentangles the latent factors of variation. To quantify interpolation quality and disentanglement, we propose two new, automated methods that are applicable to any generator architecture. Finally, we introduce a new, highly varied and high-quality dataset of human faces.