lukasHoel/text2room

Health v2 · maintenance only

Flags: no_releases

Adoption not part of the score

1089 stars · 75 forks observed · 2026-08-28

What it is AI-extracted, prompt v1, taxonomy v1, 2026-08-30, confidence not recorded

Text2Room is a research codebase that generates room-scale textured 3D meshes from a text prompt by leveraging pre-trained 2D text-to-image diffusion models, monocular depth estimation, and text-conditioned inpainting. It is the official implementation of the ICCV 2023 paper and outputs fused PLY meshes plus rendered scene images.

Use cases

• generate a textured 3D room mesh from a text prompt
• create 3D scenes from text descriptions
• research text-to-3D scene generation with diffusion models
• produce room-scale 3D geometry for graphics experiments
• reproduce ICCV 2023 Text2Room paper results
• explore lifting 2D text-to-image outputs into 3D meshes

When to choose

• you need text-to-3D room-scale mesh generation with explicit geometry
• you want a research baseline for lifting 2D diffusion outputs into 3D scenes
• you can run GPU-heavy Stable Diffusion and PyTorch3D pipelines

When to avoid

• you need production-ready, fast, or interactive 3D generation
• you want real-time or game-ready asset pipelines without heavy GPU setup
• you need actively maintained software with frequent updates

Facets

library · maturity maintenance

machine-learning image-processing graphics deep-learning computer-vision graphics artificial-intelligence deep-learning python 3d-generation text-to-3d diffusion-models mesh-generation stable-diffusion depth-estimation research-code iccv-2023 linux gpu

Member repositories

For agents

markdown · JSON · MCP: product_card(name="lukasHoel/text2room")