NVIDIA Explained How ChatGPT Can Be Used for 3D Content Generation

The company created an AI Room Generator extension in NVIDIA Omniverse.

NVIDIA has published a blog post where it explains how generative AI tools, including ChatGPT, can be used for 3D content creation to reduce manual tasks and help creators focus on the enjoyable aspects of their work.

"By combining GPT-4 with  NVIDIA Omniverse DeepSearch, a smart AI librarian that’s able to search through massive databases of untagged 3D assets, we were able to quickly develop a custom extension that quickly retrieves 3D objects with simple, text-based prompts and automatically add them to a 3D scene."

The company created an AI Room Generator extension in NVIDIA Omniverse, which allows users to enter text prompts to automatically generate and place high-fidelity objects. Objects generated from the extension are based on USD SimReady assets – physically-accurate 3D objects that can be used in any simulation and behave as they would in the real world. 

Users can circle an area using the Pencil tool in Omniverse, type in the kind of room/environment they want to generate, and that area will be created.

The ChatGPT prompt comes in four pieces: system input, user input example, assistant output example, and user prompt. This prompt is then sent to the AI and Omniverse DeepSearch API, which finds suitable objects and lets you add them to the scene.

If you don't like the result from DeepSearch, NVIDIA has a small companion extension so you can browse for similar objects and swap them in with just a click. It takes an object generated via DeepSearch as an argument and offers two buttons to get the next or previous object from the related DeepSearch query.

NVIDIA also notes that the new GPT-4 model is much better at solving complex tasks and comprehending complex instructions, which allows for more accurate prompt results.

"GPT-4 not only places objects within the right boundaries of the room, but also places objects logically: a bedside table will actually show up on the side of a bed, a coffee table will be placed in between two sofas, and so on."

The company believes AI will open doors to a wide range of tools beyond scene building. Developers will be able to build AI-powered extensions in Omniverse for lighting, cameras, animations, character dialog, and other elements that optimize creator workflows. Meanwhile, NVIDIA is working on making experimental generative AI examples available to Omniverse creators and developers soon.

Read more about it here and don't forget to join our 80 Level Talent platform and our Telegram channel, follow us on Instagram and Twitter, where we share breakdowns, the latest news, awesome artworks, and more.

Join discussion

Comments 2

  • Dubois Peter

    Nothing good has come into the world from OpenAI. This kind of "helper" aims to replace artists in the long run. If the company had a real professional awareness of how to really develop tools for artists and not against them, it could have a better image.

    1

    Dubois Peter

    ·a year ago·
  • Garcia Justin

    To Dubois point,
    To be clear, this is NVIDIA. While this specific application utilizes ChatGPT, they do have others which are homemade and specific to their technology such as their upcoming Picasso project.
    With that said, ai is a tool to be utilized. If mediocre workers are displaced because they fail to grasp the mechanics, that is their own fault. Should even a fairly new artist or developer pair themselves WITH the ai systems rather than against them, they'd find their productivity levels skyrocketing and the quality of their work getting better as well.
    Times change. Technology and tools change as well. It's best to get onboard now.

    Cheers

    0

    Garcia Justin

    ·a year ago·

You might also like

We need your consent

We use cookies on this website to make your browsing experience better. By using the site you agree to our use of cookies.Learn more