Populating huge virtual worlds could be an issue for some game developers, and that’s why NVIDIA GET3D was created. This AI model can populate virtual worlds with a diverse array of 3D buildings, vehicles, characters and more after being trained using only 2D images. The model then generates 3D shapes with high-fidelity textures and complex geometric details.
Even though the AI model is trained with 2D images, these 3D objects are created in the same format used by popular graphics software applications, enabling users to immediately import their shapes into renders and game engines for additional editing. It’s capable of generating a near unlimited number of 3D shapes based on the data it’s trained on. Like an artist who turns a lump of clay into a detailed sculpture, the model transforms numbers into complex 3D shapes.Thankfully, this AI model isn’t as creepy as Facebook’s reverse pass-through VR headset.
No products found.
The researchers note that a future version of GET3D could use camera pose estimation techniques to allow developers to train the model on real-world data instead of synthetic datasets. It could also be improved to support universal generation — meaning developers could train GET3D on all kinds of 3D shapes at once, rather than needing to train it on one object category at a time,” said the team.
