Yunpeng Zhai, Qixiang Ye, Shijian Lu, Mengxi Jia, Rongrong Ji, Yonghong Tian. "Multiple Expert Brainstorming for Domain Adaptive Person Re-identification", ECCV 2020
Paper PDF
Datasets (Market-1501 and DukeMTMC-reID).
- PyTorch 1.3.1
bash pretrain.sh <source dataset> <target dataset> <architecture>
For example, (duke->market):
bash pretrain.sh dukemtmc market1501 densenet
bash pretrain.sh dukemtmc market1501 resnet50
bash pretrain.sh dukemtmc market1501 inceptionv3
bash train.sh <source dataset> <target dataset> <architecture-1> <architecture-2> <architecture-3>
For example, (duke->market)
bash train.sh dukemtmc market1501 densenet resnet50 inceptionv3
bash test.sh <target dataset> <architecture> <checkpoint_path>
For example, (market1501, densenet)
bash test.sh market1501 densenet logs/xxxx/xxxx-MEB-Net/checkpoint.pt.pth
Our code is based on open-reid and MMT.
If you use this method or this code in your research, please cite as:
@article{zhai2020multiple,
title={Multiple Expert Brainstorming for Domain Adaptive Person Re-identification},
author={Zhai, Yunpeng and Ye, Qixiang and Lu, Shijian and Jia, Mengxi and Ji, Rongrong and Tian, Yonghong},
journal={arXiv preprint arXiv:2007.01546},
year={2020}
}
React
Typescript
Django
Laravel
Facebook
Microsoft
Google
Alibaba
D3
Tencent