The code for Bi-Mix: Bidirectional Mixing for Domain Adaptive Nighttime Semantic Segmentation arxiv

Framework: visualization results:

• scipy==1.2.2
• kornia
• scikit-image

Cityscapes: Please follow the instructions in Cityscape to download the training set.

Dark-Zurich: Please follow the instructions in Dark-Zurich to download the training/val/test set.

Nightdriving：Please follow the instructions in Nightdriving to download the training/val/test set.

If you want to train your own models, please follow these steps:

Step1: download the [pre-trained models](https://www.dropbox.com/s/3n1212kxuv82uua/pretrained_models.zip?dl=0) and put it in the root.
Step2: change the data and model paths in configs/train_config.py
Step3: run "python train.py"

To reproduce the reported results in our paper (on Dark-Zurich val or Nightdriving), please follow these steps:

Step1: change the data and model paths in configs/evaluate_config.py
Step2: run "python eva_ep.py"

To reproduce the reported results in our paper (on Dark-Zurich test), please follow these steps:

Step1: change the data and model paths in configs/test_config.py
Step2: run "python test.py"

To evaluate your methods on the test set, please visit this challenge for more details.

The code is based on DANNet and Zero-DCE.

• MGCDA
• GCMA
• DANNet
• AdaptSegNet

If you think this paper is useful for your research, please cite our paper:

@article{yang2021bi,
  title={Bi-Mix: Bidirectional Mixing for Domain Adaptive Nighttime Semantic Segmentation},
  author={Yang, Guanglei and Zhong, Zhun and Tang, Hao and Ding, Mingli and Sebe, Nicu and Ricci, Elisa},
  journal={arXiv preprint arXiv:2111.10339},
  year={2021}
}

• Guanglei Yang (yangguanglei@hit.edu.cn)