'Something Previously Impossible': New AI Makes 3D Worlds Out of a Single Image The new technology allows viewers to explore two-dimensional images in 3D.
By Sherin Shibu Edited by Melissa Malamut
Key Takeaways
- World Labs is an AI startup worth over $1 billion.
- On Monday, the company showcased an early preview of AI that can create an interactive 3D world from just an image.
- In one demo, an 82-year-old painting becomes a 3D world that viewers can move around in.
World Labs, a unicorn AI startup with a valuation of over $1 billion, unveiled a new AI on Monday that can transform any image into a 3D world.
World Labs wrote in a thread on X that its new AI system works by taking an image, approximating 3D geometry, and filling in the scene so that there's more to see. The AI adapts to different art styles and scenes.
The output is fully interactive, so viewers can view the 3D scenes as if they were inside them with a movable camera. They can zoom in and out and move around the space.
Related: Testers Leak OpenAI's New AI Video Generator, Claiming They Were Used as 'Unpaid Labor'
"Most GenAI tools make 2D content like images or videos," World Labs wrote in a blog post. "Generating in 3D instead improves control and consistency. This will change how we make movies, games, simulators, and other digital manifestations of our physical world."
We've been busy building an AI system to generate 3D worlds from a single image. Check out some early results on our site, where you can interact with our scenes directly in the browser!https://t.co/ASD6ZHMwxI
— World Labs (@theworldlabs) December 2, 2024
1/n pic.twitter.com/tuvGXHmepP
While World Labs isn't the first to explore this area, it is the first to make its 3D worlds explorable, per TechCrunch. Viewers can interact with the AI-created scenes on the World Labs website from within a web browser.
In one demonstration, World Labs brings viewers into the 1942 painting Nighthawks by Edward Hopper. With a few clicks, viewers can freely move within the world created by the painting.
We also had some fun peeking into the worlds behind a few creative masterpieces, like the neighborhood surrounding the diner in Edward Hopper's iconic painting Nighthawks.
— World Labs (@theworldlabs) December 2, 2024
6/n pic.twitter.com/QSTDnZfkwb
The painting Nighthawks by Edward Hopper. Photo by VCG Wilson/Corbis via Getty Images
In another demo, content creator and animator Eric Solorio showcases how he uses the technology to quickly generate 3D environments and scenes.
"The process was very fast and easy," Solorio stated. "Something previously impossible, with this level of precision."
@8bit_e shows how our models fill a gap in his creative workflow, making it easy to stage characters within scenes and direct precise camera movements.
— World Labs (@theworldlabs) December 2, 2024
8/n pic.twitter.com/AIbhFtXU1B
World Labs says that AI-generated images can be used as the foundation for 3D spaces, too, and the entire process of creating a 3D world, from the image to the world itself, can be AI-generated.
Although these demos are a "first early preview" of what its technology can do, according to the company, the team is working on making the generated worlds bigger and more true to form. The company hopes to have its first product ready sometime next year.
Related: I Tried 3 AI Headshot Generators and There Was a Clear Winner
