README_hf.md

---
title: MAXIM Multi-Axis MLP for Image Processing
emoji: 🖼️
colorFrom: blue
colorTo: purple
sdk: gradio
sdk_version: 4.44.0
app_file: app.py
pinned: false
license: apache-2.0
---

# MAXIM: Multi-Axis MLP for Image Processing

This Hugging Face Space demonstrates the MAXIM model for various image processing tasks.

## About MAXIM

MAXIM is a Multi-Axis MLP architecture for image processing that achieves state-of-the-art results on multiple tasks:

- **Image Enhancement**: Photo retouching and low-light enhancement
- **Image Denoising**: Removing noise from images
- **Image Deblurring**: Removing motion blur and defocus blur
- **Image Deraining**: Removing rain streaks and raindrops
- **Image Dehazing**: Removing haze and fog

## Model Performance

MAXIM achieves excellent results across different benchmarks:

| Task | Dataset | PSNR | SSIM |
|:---:|:---:|:---:|:---:| 
| Denoising | SIDD | 39.96 | 0.960 |
| Deblurring | GoPro | 32.86 | 0.961 |
| Deraining | Rain13k | 33.24 | 0.933 |
| Dehazing | RESIDE-Indoor | 38.11 | 0.991 |
| Enhancement | LOL | 23.43 | 0.863 |
| Enhancement | FiveK | 26.15 | 0.945 |

## Usage

1. Upload an image using the interface
2. Select the desired image processing task
3. Click "Process Image" to see the results

The model will automatically download the required checkpoints for the selected task.

## Citation

If you use this model in your research, please cite:

```bibtex
@article{tu2022maxim,
  title={MAXIM: Multi-Axis MLP for Image Processing},
  author={Tu, Zhengzhong and Talebi, Hossein and Zhang, Han and Yang, Feng and Milanfar, Peyman and Bovik, Alan and Li, Yinxiao},
  journal={CVPR},
  year={2022},
}
```

## Links

- [Paper](https://arxiv.org/abs/2201.02973)
- [Original Repository](https://github.com/google-research/maxim)
- [Colab Demo](https://colab.research.google.com/github/google-research/maxim/blob/master/colab_inference_demo.ipynb)

## License

This project is licensed under the Apache License 2.0 - see the [LICENSE](LICENSE) file for details.