Researchers at NVIDIA have developed a deep learning model that can turn rough doodles into photorealistic landscapes using generative adversarial networks, or GANs. The interactive app has been named GauGAN, after the post-Impressionist painter Paul Gauguin.
The deep learning model has been trained on a million images. When you draw the shapes of the objects you want, the neural network is able to fill in the details, texture, color, reflections, and shadows, based on what it has learned from those real images.
GauGAN could be a powerful tool for architects, urban planners, landscape designers, game developers, and more. Using this ‘smart paintbrush’, professionals will be able to develop prototypes, and make rapid changes to a synthetic scene with ease. Check it out below.
Watch the AI in action below
What do you think of NVIDIA’s new technology? Share this post with a designer friend and voice your views in the comments below. All images © NVIDIA.