LighTDiff: Surgical Endoscopic Image Low-Light Enhancement with T-Diffusion

Tong Chen†, Qingcheng Lyu†, Long Bai†, Erjian Guo, Huxin Gao, Xiaoxiao Yang, Hongliang Ren, and Luping Zhou

Medical Image Computing and Computer Assisted Intervention (MICCAI) 2024

Best Paper Runner-Up Award

[arXiv]	[Paper]

Update

[10/24/2024] Our Dataset is now available in here.

[10/9/2024] We received MICCAI2024 Best Paper Runner-Up!

[9/3/2024] Our work received Oral Presentation on MICCAI2024!

[7/20/2024] Fixed some bugs.

[5/13/2024] Our work got early accepted by MICCAI2024!

[5/17/2024] Our code is now available!

Schematics

Results

Pre-installation

conda create -n lightdiff python=3.10
conda activate lightdiff
conda install pytorch==2.0.1 torchvision torchaudio cudatoolkit==12.1 -c pytorch
cd BasicSR-light
pip install -r requirements.txt
BASICSR_EXT=True sudo $(which python) setup.py develop
cd ../LighTDiff
pip install -r requirements.txt
BASICSR_EXT=True sudo $(which python) setup.py develop

Test

python lightdiff/train.py -opt configs/test.yaml

Train

python lightdiff/train.py -opt configs/train.yaml