/unique3d-diffusion

Primary LanguagePython

Unique3D Diffusion Models

This project contains the image-to-multiview diffusion and image-to-normal diffusion models from Unique3D. demo

Install

pip install git+https://github.com/TingtingLiao/unique3d-diffusion.git

or

git clone https://github.com/TingtingLiao/unique3d-diffuser.git 
cd unique3d-diffuser
pip install -e .

Usage

Download unique3d models from huggingface.

# single image 
python3 demo.py --ckpt_dir ./ckpt --img data/disney/belle.jpeg 

# processing img_dir  
python3 demo.py --ckpt_dir ./ckpt --img_dir data/disney

or

from unique3d_diffusion import Unique3dDiffuser

seed = 0 
ckpt_dir = "./ckpt"
save_dir = "./output"
image_path = "data/disney/belle.jpeg" 
model = Unique3dDiffuser(ckpt_dir, seed, save_dir)
images, normals = model(image_path, save=True)

Acknowledgement

The code is adapted from unique3d. Please consiter cite:

@misc{wu2024unique3d,
      title={Unique3D: High-Quality and Efficient 3D Mesh Generation from a Single Image}, 
      author={Kailu Wu and Fangfu Liu and Zhihan Cai and Runjie Yan and Hanyang Wang and Yating Hu and Yueqi Duan and Kaisheng Ma},
      year={2024},
      eprint={2405.20343},
      archivePrefix={arXiv},
      primaryClass={cs.CV}
}