Nvidia’s SIGGRAPH Keynote Showcases New AI 3D Background Feature Developed With Shutterstock

Nvidia and Shutterstock have used Nvidia Picasso to develop a foundation model that can generate photorealistic, 8K, 360-degree high-dynamic-range imaging (HDRi) environment maps.

What is Nvidia Picasso? Nvidia Picasso is a cloud service for building generative AI–powered visual applications. Users can run inference on their models, train Nvidia Edify foundation models on proprietary data, or start from pre-trained models to generate image, video, and 3d models from text and image prompts.

The latest: Revealed today at SIGGRAPH, the new feature developed in partnership with Shutterstock will allow users to enhance and light 3d scenes using text or image prompts. Importantly, it will do so using AI models built entirely on licensed, rights-reserved data.

How does it work? The user provides a text or image prompt and the model will quickly generate a custom 360-degree, 8K-resolution HDRi environment map which can be used to set a background and light scene. During the keynote, it was also explained that 3d software and tool provider Autodesk will integrate generative AI services, developed using foundation models in Picasso, into its popular 3d software Maya.

How might this benefit artists? There’s a very clear application here for 3d production and according to Nvidia, by using the new background-generating tools artists will have more time to work on hero 3d assets in a scene. In the past, productions often needed expensive 360-degree cameras to create realistic backgrounds and environment maps. Now, Nvidia says users need only a prompt entered into the 360 HDRi toolset to receive full panoramic images.

What are they saying? Said Dade Orgeron, vice president of 3d innovation at Shutterstock:

We’re committed to hyper-enabling 3d artists and collaborators — helping them build the immersive environments they envision faster than ever before and streamlining their content-creation workflows using Nvidia Picasso.