Ready to be amazed and afraid for a game dev’s job at the same time? Watch this above NVIDIA demo to see the first interactive AI rendered virtual world!
In fewer buzzwords, what you’re seeing is computer graphics created by AI that’s been trained on videos from the real world, no artist required.
“Twenty-five years ago NVIDIA transformed the computer graphics industry by building the first GPU, the modern day tool of the da Vincis and Einsteins of our time. Now, a new deep learning-based model developed by NVIDIA researchers aims to catapult the graphics industry into the AI chapter,” begins NVIDIA in its newest developer post.
Through this above demo, game development costs could fall abruptly, as content creation for making a game has traditionally been a painstaking, time-consuming work. By using AI that learns from the real world, modelers and animators working on a racing game, for example, will have their job made so much easier.
The NVIDIA demo, to be shown at the NeurIPS 2018 Montreal Expo, uses NVIDIA’s Tensor Core GPUs. It lets players drive through a virtual environment where “content is rendered interactively using a neural network that transforms sketches of a 3D world produced by a traditional graphics engine into video.”
For the technical details of this breakthrough, check out NVIDIA’s paper, you’ll get even more ideas about its application.