News Categories

Convert simple sketches into photo-realistic masterpieces!

By Vijay Anand - on 19 Mar 2019, 8:01am

Convert simple sketches into photo-realistic masterpieces!

If you ever wished your simple sketches can be quickly translated into a more visual masterpiece to covey your intent a new deep learning model developed by NVIDIA Research can do just that. Leveraging on generative adversarial networks (GANs), it converts simple segmentation maps into photorealistic images.

This fast tracks prototyping and brainstorming sessions into a far more realistic output that can speak to your working group a lot quicker than ever before.

Christened as GauGAN, it is a powerful tool to help content creators, designers and anyone needing to communicate with others visually by creating virtual worlds with simple strokes and boundaries. This will benefit anyone from architects and urban planners to landscape designers and game developers.

Perhaps later in the year, this could make its way into NVIDIA's AI Playground and you can experience GauGAN firsthand!

Join HWZ's Telegram channel here and catch all the latest tech news!
Our articles may contain affiliate links. If you buy through these links, we may earn a small commission.