Home

Awesome

<div align="center"> <img width="100%" src="com.png"> </div> <div align="center">

简体中文 | English | Paper

TextDiff: Mask-Guided Residual Diffusion Models for Scene Text Image Super-Resolution

这里是论文TextDiff: Mask-Guided Residual Diffusion Models for Scene Text Image Super-Resolution的官方复现仓库。TextDiff是一个场景文字超分辨率优化模型(详见论文).

</div>

网络结构

<div align="center"> <img width="100%" src="model.png"> </div>

News

使用指南

环境配置

深度学习环境

数据集

相关权重文件

训练

  1. 安装
git clone https://github.com/Lenubolim/TextDiff.git
  1. 参数配置 <br> 见config.yaml文件 <br>

  2. 训练

python train.py

推理

python test.py

To-do lists

效果图

<div align="center"> <img width="400" src="sota.png"></div>

感谢

References

<div id="refer-anchor-1"></div>

:book: Citation

If you use (part of) my code or find my work helpful, please consider citing

@article{liu2023textdiff,
  title={TextDiff: Mask-Guided Residual Diffusion Models for Scene Text Image Super-Resolution},
  author={Liu, Baolin and Yang, Zongyuan and Wang, Pengfei and Zhou, Junjie and Liu, Ziqi and Song, Ziyi and Liu, Yan and Xiong, Yongping},
  journal={arXiv preprint arXiv:2308.06743},
  year={2023}
}

Acknowledgement

This code is developed relying on <a href="https://github.com/Royalvice/DocDiff" title="DocDiff">DocDiff</a> and <a href="https://github.com/mjq11302010044/TATT" title="TATT">TATT</a>. Thanks for these great projects. Among them, DocDiff is the main research content of my classmate, and I participated in part of the research.