UHDPromer [IJCV-26]

This is the official PyTorch codes for the paper
Neural Discrimination-Prompted Transformers for Efficient UHD Image Restoration and Enhancement
Cong Wang*, Jinshan Pan, Liyan Wang, Wei Wang*, and Yang Yang

---

Dependencies and Installation

• Ubuntu >= 18.04
• CUDA >= 11.0
• Other required packages in requirements.txt

# git clone this repository
git clone https://github.com/supersupercong/UHDPromer.git
cd UHDPromer 

# create new anaconda env
conda create -n uhdpromer python=3.8
source activate uhdpromer 

# install python dependencies
pip3 install -r requirements.txt
python setup.py develop

Train the model

Dataset

UHD-LL, UHD-Haze, UHD-Blur

Train model

bash train.sh

Test model

bash test.sh

Visual results

UHD-LL&UHD-Haze&UHD-Blur&LOL-GoPro-SOTS

Pretrained models

UHD-LL&UHD-Haze&UHD-Blur

Citation

@article{2026Neural,
  title={Neural Discrimination-Prompted Transformers for Efficient UHD Image Restoration and Enhancement},
  author={ Wang, Cong  and  Pan, Jinshan  and  Wang, Liyan  and  Wang, Wei  and  Yang, Yang },
  journal={International Journal of Computer Vision},
  volume={134},
  number={3},
  year={2026},
}

License

This work is licensed under a Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International License.

Acknowledgement

This project is based on BasicSR.