3D generation models create 3D objects, such as meshes or point clouds, from a text prompt or an image.
In plain words
Making a 3D asset by hand takes skill and time. These models do it automatically. Labelled 3D data is scarce, so some methods borrow a 2D image model: they keep adjusting a 3D shape until pictures of it from every angle look like the prompt. Others first draw one image, then turn it into 3D. The result can be viewed from any side and placed in a scene.
A quick example
Imagine typing 'a small wooden chair'. The model produces a 3D chair you can turn around, light from any direction and drop into a 3D scene, instead of an artist modelling it by hand.
Where you'll meet it
- Creating 3D objects to place into scenes and 3D environments.
Related
Sources
- PaperDreamFusion: Text-to-3D using 2D Diffusion, Poole et al. (arXiv) · read 29 Sept 2026
- PaperShap-E: Generating Conditional 3D Implicit Functions, Jun and Nichol (arXiv) · read 29 Sept 2026
- PaperPoint-E: A System for Generating 3D Point Clouds from Complex Prompts, Nichol et al. (arXiv) · read 29 Sept 2026
Checked by WhatIsAI against the sources above · 29 Sept 2026. Spotted an error? Report it.