YDDDDG / 3D2Unet

3D2Unet: 3D Deformable Unet for Low-Light Video Enhancement (PRCV2021)

3DDUNET

This is the code for 3D2Unet: 3D Deformable Unet for Low-Light Video Enhancement (PRCV2021) Conference Paper Link

Dataset

We use SMOID dataset from SMOID

Code

Prerequisites

• Python 3.6
• PyTorch 1.7 with GPU
• opencv-python
• scikit-image
• tensorboard
• For 3D deformable compile process please refer D3Dnet

Train and Test

Please run main.py to train and test the model

Citing

If you use any part of our research, please consider citing:

@inproceedings{zeng2021mathrm,
  title={3D2Unet:3D Deformable Unet for Low-Light Video Enhancement},
  author={Zeng, Yuhang and Zou, Yunhao and Fu, Ying},
  booktitle={Chinese Conference on Pattern Recognition and Computer Vision (PRCV)},
  pages={66--77},
  year={2021},
  organization={Springer}
}

Acknowledgement

Our work and implementations are inspired by following projects: ESTRNN SMOID D3Dnet