Arbitrary-steps Image Super-resolution via Diffusion Inversion

Official Gradio demo for Arbitrary-steps Image Super-resolution via Diffuion Inversion.
🔥 InvSR is an image super-resolution method via Diffusion Inversion, supporting arbitrary sampling steps.
Number of steps
Chopping size
Examples
Input: Low Quality Image Number of steps Chopping size Ranom seed

If you've found InvSR useful for your research or projects, please show your support by ⭐ the Github Repo. Thanks!

If our work is useful for your research, please consider citing:

@article{yue2024InvSR,
  title={Arbitrary-steps Image Super-resolution via Diffusion Inversion},
  author={Yue, Zongsheng and Kang, Liao and Loy, Chen Change},
  journal = {arXiv preprint arXiv:2412.09013},
  year={2024},
}

📋 License

This project is licensed under S-Lab License 1.0. Redistribution and use for non-commercial purposes should follow this license.

📧 Contact

If you have any questions, please feel free to contact me via zsyzam@gmail.com.