Leading chipmaker Nvidia’s researchers recently demonstrated Nvidia Edify, a multimodal AI architecture, by creating a detailed 3D desert landscape in minutes.
The team showed how, with the help of an AI agent, they could construct and edit the landscape within five minutes, highlighting generative AI’s ability to assist artists by speeding up ideation and generating custom assets.
This technology dramatically reduces ideation time, and could be used to empower 3D artists to be more productive and creative.
Traditionally, creating a full 3D scene is time-consuming, requiring background objects and environment maps.
AI agents enable teams to quickly bring concepts to life and continue iterating for the desired look.
In the demo, an AI agent instructed an Nvidia Edify-powered model to generate 3D assets like cacti and a bull’s skull in seconds. The agent also created potential backgrounds and layouts, adapting to last-minute changes, and rendered the scene as a photorealistic image in Nvidia Omniverse USD Composer.
Nvidia Edify accelerates environment generation, allowing creators to focus on key assets while AI handles background objects.
Nvidia has also announced collaborations with Shutterstock and Getty Images to enhance productivity using generative AI tools powered by Nvidia Edify.
The 3D objects and environments generated are compatible with USD, allowing seamless import into Omniverse USD Composer for further modification.
