Views
No views yet
| Image-to-Scene Results | Text-to-Scene Results |
|---|---|
 |  |
| Model | Download |
|---|---|
| SpatialGen-1.0 | 🤗 HuggingFace |
| FLUX.1-Wireframe-dev-lora | 🤗 HuggingFace |
1# clone the repository
2git clone https://github.com/manycore-research/SpatialGen.git
3cd SpatialGen
4
5python -m venv .venv
6source .venv/bin/activate
7
8pip install -r requirements.txt
9# Optional: fix the [flux inference bug](https://github.com/vllm-project/vllm/issues/4392)
10pip install nvidia-cublas-cu12==12.4.5.81# Single image-to-3D Scene
2bash scripts/infer_spatialgen_i2s.sh
3
4# Text-to-image-to-3D Scene
5# in captions/spatialgen_testset_captions.jsonl, we provide text prompts of different styles for each room,
6# choose a pair of scene_id and prompt to run the text2scene experiment
7bash scripts/infer_spatialgen_t2s.sh1@inproceedings{SpatialGen,
2 title = {SpatialGen: Layout-guided 3D Indoor Scene Generation},
3 author = {Fang, Chuan and Li, Heng and Liang, Yixu and Zheng, Jia and Mao, Yongsen and Liu, Yuan and Tang, Rui and Zhou, Zihan and Tan, Ping},
4 booktitle = {International Conference on 3D Vision},
5 year = {2026}
6}