NVIDIANvidia has launched software that will turn MS Paint-style images into landscapes that look incredibly realistic.
GauGAN was created by the company that makes GPUs to make photorealistic images in a matter of seconds. Using just a paint bucket, pen and pencil, to create a series of objects. If you wanted to create water in your picture, select the water option and you can draw where you want the river to go.
Nvidia/YouTubeIt converts amateur images into beautiful scenery super fast. It’s so accurate that even if two people tried to draw the same scene, the project’s software would conjure up different results.
To see how this amazing technology works for yourself, check out the video from Nvidia’s YouTube channel below:
Nvidia demonstrated this software on an RDX Titan GPU platform, which allowed it to produce results in real-time. If you were to draw in a pond, nearby elements like trees and rocks will appear as reflections in the water. Swap a segment label from “grass” to “snow” and the entire image will change to a winter scene.
In order to train the software, Nvidia used 1 million images on Flickr to teach the neural network. The company used them to program GauGAN to look at hundreds of thousands of objects and their relation to other objects in the real world. Doing this has helped to make the pictures look so realistic.
Bryan Catanzaro, VP of Applied Deep Learning Research, explained the groundbreaking technology in a press release:
It’s like a coloring book picture that describes where a tree is, where the sun is, where the sky is.
And then the neural network is able to fill in all of the detail and texture, and the reflections, shadows and colors, based on what it has learned about real images.
This technology is not just stitching together pieces of other images, or cutting and pasting textures… it’s actually synthesizing new images, very similar to how an artist would draw something.
The program could benefit video game designers, architects and map developers in the future to really bring ideas and concepts to life. While there doesn’t appear to be any plans to release the software commercially, Nvidia may release it on a public trial. Doing this would make GauGAN available for anyone to use.
If you have a story you want to tell send it to UNILAD via [email protected]
Matt Weston is a lover of electric cars, artificial intelligence and space. From Cornwall, he’s a UCLan graduate that still dreams of being a Formula One driver in the very near future. Previously work includes reporting for regional newspapers and freelance video for the International Business Times.