Fei-Fei Li's World Labs Launches 3D Scene Generation from a Single Image

Fei-Fei Li's World Labs launches groundbreaking 3D scene generation from a single image or sentence, marking a major leap in spatial intelligence and immersive AI experiences.

This is the first step toward spatial intelligence.

"No matter how theoretical we get, it’s still difficult to describe the experience of interacting with 3D scenes generated from photos or sentences using words," said Fei-Fei Li.

Today, renowned AI scholar and Stanford University professor Fei-Fei Li unveiled her first entrepreneurial project—a magical initiative that generates 3D worlds from a single image or a sentence.

Generative AI has now entered a new era of 3D and fully immersive experiences.

How detailed can a 3D world generated from just one image be? World Labs claims the results look like this:

Here are more examples of the effects:

Previously, we’ve witnessed a series of generative AI tools capable of creating 2D content, which can output images or videos.

In contrast, methods for generating 3D content are less common. Directly generating 3D content improves control and consistency, which will revolutionize the way we create movies, games, simulators, and other digital representations of the physical world.