manycore-research/SpatialGen

[3DV 2026] SpatialGen: Layout-guided 3D Indoor Scene Generation

47
/ 100
Emerging

Leverages multi-view multi-modal diffusion to generate coherent 3D scenes from either reference images or text descriptions, conditioned on semantic layout constraints. Employs a two-stage architecture combining SCM-VAE for layout encoding with Gaussian splatting-based 3D reconstruction via Sparse-RaDeGS. Integrates FLUX.1 ControlNet for conditional image generation and builds on Stable Diffusion v2.1 backbone.

360 stars.

No Package No Dependents
Maintenance 10 / 25
Adoption 10 / 25
Maturity 15 / 25
Community 12 / 25

How are scores calculated?

Stars

360

Forks

19

Language

Python

License

MIT

Last pushed

Jan 27, 2026

Commits (30d)

0

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/diffusion/manycore-research/SpatialGen"

Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.