Nvidia’s new AI model quickly generates 3D objects to incorporate into virtual worlds

MADRID, 23 Sep. (Portaltic/EP) –

Nvidia has introduced a new artificial intelligence model (AI) that generates three-dimensional objects for later use in virtual worlds in sectors such as video games, architecture or robotics.

NVIDIA GET3Dthe name given to this model, has been trained with about one million 2D images of 3D objects captured from different angles, using the Nvidia A100 Tensor Core GPU, which took the researchers around two days.

From such training, the AI ​​model is capable of creating three-dimensional objects “with high-fidelity textures and intricate geometric details“, as the company states in its official blog. And generate them, moreover, quickly, since it can produce about 20 forms per second.

All this means that, for example, from the images captured in 2D of cars, the model can generate a collection of sedans, trucks, race cars and vans; or if it is images of animals, creatures such as foxes, rhinos, horses and bears.

These three-dimensional objects can be imported later for use in game engines, 3D modelers and movie renderers, and edit to apply more realistic lighting effects, for example.

Nvidia has indicated that in the future they hope to incorporate camera pose estimation techniques and universal generation, to expand the capabilities of this tool.

Leave a Comment

Your email address will not be published.