Skip to content
/ CLAN Public

( TPAMI2022 / CVPR2019 Oral ) Taking A Closer Look at Domain Shift: Category-level Adversaries for Semantics Consistent Domain Adaptation

License

Notifications You must be signed in to change notification settings

RoyalVane/CLAN

Repository files navigation

Taking A Closer Look at Domain Shift: Category-level Adversaries for Semantics Consistent Domain Adaptation (CVPR2019)

This is a pytorch implementation of CLAN.

Oral Presentation Video

Watch the video

Prerequisites

  • Python 3.6
  • GPU Memory >= 11G
  • Pytorch 1.0.0

Getting started

The data folder is structured as follows:

├── data/
│   ├── Cityscapes/     
|   |   ├── gtFine/
|   |   ├── leftImg8bit/
│   ├── GTA5/
|   |   ├── images/
|   |   ├── labels/
│   ├── SYNTHIA/ 
|   |   ├── RAND_CITYSCAPES/
│   └── 			
└── model/
│   ├── DeepLab_resnet_pretrained.pth
...

Train

CUDA_VISIBLE_DEVICES=0 python CLAN_train.py --snapshot-dir ./snapshots/GTA2Cityscapes

Evaluate

CUDA_VISIBLE_DEVICES=0 python CLAN_evaluate.py --restore-from  ./snapshots/GTA2Cityscapes/GTA5_100000.pth --save ./result/GTA2Cityscapes_100000

Our pretrained model is available via Google Drive

Compute IoU

python CLAN_iou.py ./data/Cityscapes/gtFine/val result/GTA2Cityscapes_100000

Tip: The best-performance model might not be the final one in the last epoch. If you want to evaluate every saved models in bulk, please use CLAN_evaluate_bulk.py and CLAN_iou_bulk.py, the result will be saved in an Excel sheet.

CUDA_VISIBLE_DEVICES=0 python CLAN_evaluate_bulk.py
python CLAN_iou_bulk.py

Visualization Results

(a) (b)

(c) (d)

This code is heavily borrowed from the baseline AdaptSegNet

Citation

If you use this code in your research please consider citing

@article{luo2021category,
  title={Category-Level Adversarial Adaptation for Semantic Segmentation using Purified Features},
  author={Luo, Yawei and Liu, Ping and Zheng, Liang and Guan, Tao and Yu, Junqing and Yang, Yi},
  journal={IEEE Transactions on Pattern Analysis \& Machine Intelligence (TPAMI)},
  year={2021},
}

@inproceedings{luo2019Taking,
title={Taking A Closer Look at Domain Shift: Category-level Adversaries for Semantics Consistent Domain Adaptation},
author={Luo, Yawei and Zheng, Liang and Guan, Tao and Yu, Junqing and Yang, Yi},
booktitle={The IEEE Conference on Computer Vision and Pattern Recognition (CVPR)},
year={2019}
}

Related works

Seg-Uncertainty

About

( TPAMI2022 / CVPR2019 Oral ) Taking A Closer Look at Domain Shift: Category-level Adversaries for Semantics Consistent Domain Adaptation

Topics

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages