DMRVisNet

PyTorch implementation and pretrained model for DMRVisNet. For details, see the paper DMRVisNet: Deep Multihead Regression Network for Pixel-Wise Visibility Estimation Under Foggy Weather.

Preparation

Environment

Please make sure that you have installed torch, einops, tqdm, Pillow, opencv-python, progress and matplotlib.

Download Dataset

Please refer to FACI dataset.

Download pre-trained model weight

Please download from Google Drive.

File Structure

dmrvisnet
├── checkpoints
│   └── final
├── data_pkl
│   ├── 000_000.pkl
│   ├── 000_001.pkl
│   ├── ...
│   ├── 099_028.pkl
│   └── 099_029.pkl
├── src
├── train.sh
└── test.sh

Train

./train.sh

Test

./test.sh

Quick start

Please refer to src/quickstart.ipynb

Citation

Please cite this as:

@article{you2022dmrvisnet,
  title={Dmrvisnet: Deep multihead regression network for pixel-wise visibility estimation under foggy weather},
  author={You, Jing and Jia, Shaocheng and Pei, Xin and Yao, Danya},
  journal={IEEE Transactions on Intelligent Transportation Systems},
  volume={23},
  number={11},
  pages={22354--22366},
  year={2022},
  publisher={IEEE}
}