NVIDIA Updates Its Generative Adversarial Network to StyleGAN2

NVIDIA StyleGan2

NVIDIA continues to incorporate its latest technologies into real-world applications. The company is constantly evolving AI to power or enhance a plethora of solutions. These can range from use in scientific and medical applications, industrial and manufacturing, to content creation and artistic endeavors. One such use is in GANs or General Adversarial Networks. No, this is not the latest step toward Skynet, nor an attempt to improve those wonky AI opponents in your favorite online combat game.

Definition of GAN

A generative adversarial network (GAN) is a machine learning (ML) model in which two neural networks compete with each other to become more accurate in their predictions. GANs typically run unsupervised and use a cooperative zero-sum game framework to learn.

Source: Techtarget/SearchAI

On the heels of its latest Ampere-based AI workstation, the DGX A100, NVIDIA has updated its GAN application called StyleGAN. StyleGAN2 has now been upgraded to use image modeling that has been trained with cuDNN-accelerated Tensorflows from eight NVIDIA V100 GPUs in a DGX.

This new project called StyleGAN2, presented at CVPR 2020, uses transfer learning to generate a seemingly infinite number of portraits in an infinite variety of painting styles. The work builds on the team’s previously published StyleGAN project. 

Shown in this new demo, the resulting model allows the user to create and fluidly explore portraits. This is done by separately controlling the content, identity, expression, and pose of the subject. Users can also modify the artistic style, color scheme, and appearance of brush strokes. 

An application such as this could have many uses. Game developers could use it to create more unique and individualized characters in games. Those working in CGI could do the same for animated projects. As you can see in the demo, it is not limited to human subjects, either. Wildlife can also be enhanced with StyleGAN2. Someone less inclined toward the professional front could make a stylized image for a family photo. The ability to simulate classical painting is now within keystrokes.

Peter Brosdahl
As a child of the 70’s I was part of the many who became enthralled by the video arcade invasion of the 1980’s. Saving money from various odd jobs I purchased my first computer from a friend of my dad, a used Atari 400, around 1982. Eventually it would end up being a lifelong passion of upgrading and modifying equipment that, of course, led into a career in IT support.

Recent News