lukasHoel/text2room
Text2Room generates textured 3D meshes from text prompts using 2D text-to-image diffusion models.
Not currently ranked — collecting fresh signals.
star history
This project creates navigable textured 3D room meshes from natural language prompts by iteratively projecting and blending text-to-image outputs into a 3D scene using monocular depth estimation. It combines Stable Diffusion for inpainting with depth maps to establish geometric structure, then fuses multi-view renderings into cohesive textured meshes.
Frequently asked
- What is lukasHoel/text2room?
- Text2Room generates textured 3D meshes from text prompts using 2D text-to-image diffusion models.
- Is text2room open source?
- Yes — lukasHoel/text2room is open source, released under the MIT license.
- What language is text2room written in?
- lukasHoel/text2room is primarily written in Python.
- How popular is text2room?
- lukasHoel/text2room has 1.1k stars on GitHub.
- Where can I find text2room?
- lukasHoel/text2room is on GitHub at https://github.com/lukasHoel/text2room.