Awesome

NIR-assisted Low-light Video Enhancement Using Unpaired 24-hour Data

In this repository, we provide source codes and pretrained checkpoint for run the test demo.

FMSVD Dataset

You can download the FMSVD dataset here (split) or here (one zip file).

Third-Party Dataset and NIR Relighting Codes

You can dowload the third-party dataset and NIR relighting codes from this link.

Environment Setup

conda create -n nir python==3.8
conda activate nir
conda install pytorch==1.8.0 torchvision==0.9.0 torchaudio==0.8.0 cudatoolkit=11.1 -c pytorch -c conda-forge
python -m pip install detectron2 -f https://dl.fbaipublicfiles.com/detectron2/wheels/cu111/torch1.8/index.html
pip install opencv-python
cd package_core
python setup.py install
cd ..
pip install timm

Download pretrained checkpoint

Download checkpoints from here, and put it into ./.

Run Test Script

chmod 777 test.sh
./test.sh

The results will be saved to ./test_results.