/AerialDiffusion

Codebase for the paper Aerial Diffusion: Text Guided Ground-to-Aerial View Translation from a Single Image using Diffusion Models

Primary LanguagePython

Paper: Aerial Diffusion: Text Guided Ground-to-Aerial View Translation from a Single Image using Diffusion Models (SIGGRAPH Asia 2023, Conference Proceedings)

Please cite our paper if you find it useful. 

@article{kothandaraman2023aerial, 
  title={Aerial Diffusion: Text Guided Ground-to-Aerial View Translation from a Single Image using Diffusion Models}, 
  author={Kothandaraman, Divya and Zhou, Tianyi and Lin, Ming and Manocha, Dinesh}, 
  journal={arXiv preprint arXiv:2303.11444}, 
  year={2023} 
}

Using the code

Modify the text prompt (line 29) and the path to the ground-view image (line 31) in run_aerialDiffusion.py.

Execute

python run_aerialDiffusion.py

Dependencies

torch
cv2
diffusers
numpy
scipy
accelerate
packaging
transformers

Method

Acknowledgements

This codebase is heavily borrowed from https://github.com/huggingface/diffusers/blob/main/examples/community/imagic_stable_diffusion.py.