HAZY RE-ID: AN INTERFERENCE SUPPRESSION MODEL FOR DOMAIN ADAPTATION PERSON RE-IDENTIFICATION UNDER INCLEMENT WEATHER CONDITION.

This package contains the source code which is associated with the following paper:

Jian Pang, Dacheng Zhang, Huafeng Li, Weifeng Liu, Zhengtao Yu, “HAZY RE-ID: AN INTERFERENCE SUPPRESSION MODEL FOR DOMAIN ADAPTATION PERSON RE-IDENTIFICATION UNDER INCLEMENT WEATHER CONDITION.” accepted as oral on ICME 2021.

Edited by Jian Pang

Usage of this code is free for research purposes only.

Thank you.

Requirements:

CUDA  10.2
Python  3.8
Pytorch  1.6.0
torchvision  0.2.2
numpy  1.19.0

Get Started

1.Install:

download the code
git clone https://github.com/PangJian123/ISM-ReID.git
cd ISM-ReID

2.Datasets and the pre-trained

Prepare datasets and the pre-trained model please refer to https://github.com/PangJian123/fast-reid
Download the synthetic hazy datasets through the links below: Hazy-DukeMTMC-reID:Baidu Pan（password：7ec8） Hazy-Market1501:Baidu Pan （password：xs52）

3.Run the training file:

    sh pre_train.sh (Supervised training on source domain)
    sh train.sh (Training ISM)

Hazy image generation:

The hazy image is obtained by combining clear image and the corresponding depth map through the method of hazyimage_generator.
The clear image comes from the original dataset(Market-1501 or DukeMTMC-ReID).
The depth map is generated by monodepth2. Note that, we add 'disp = torch.exp(-disp)' after 'disp = outputs[("disp, 0")]' in test_simple.py to obtain reversed depth map.

Contact:

Don't hesitate to contact me if you meet any problems when using this code.

Jian Pang
Faculty of Information Engineering and Automation
Kunming University of Science and Technology                                                           
Email: pangjian@stu.kust.edu.cn

Acknowledgements

Our code is based on https://github.com/JDAI-CV/fast-reid,https://github.com/nianticlabs/monodepth2 and https://github.com/phoenixtreesky7/hazyimage_generator.