TinyLUT: Tiny Look-Up Table for Efficient Image Restoration at the Edge (NeurIPS 2024)

paper

Environment

CUDA 11.1
CUDNN 8.5.0
torch 1.10
torchvision 0.8.0
opencv-python

Dataset

Train Dataset

DVI2K

Test Dataset

Set5
Set14
Urban100
BSD100
Manga109

Usage

Transferring LUT network into LUTs

python ./Transfer_TinyLUT.py

Test LUTs

python ./test_TinyLUT.py

Acknowledgments

Thanks the ACs and the reviewers for their insightful comments, which are very helpful to improve our paper!
This work is based on the following works, thank the authors a lot.
SRLUT
MULUT

Citation

@inproceedings{
      li2024tinylut,
      title={Tiny{LUT}: Tiny Look-Up Table for Efficient Image Restoration at the Edge},
      author={Huanan LI and Juntao Guan and Lai Rui and Sijun Ma and Lin Gu and Zhangming Zhu},
      booktitle={The Thirty-eighth Annual Conference on Neural Information Processing Systems},
      year={2024}
}