This repo contains PyTorch code for ICCV19' paper: Deep Meta Metric Learning, including person re-identification experiments on Market-1501 and DukeMTMC-reID datasets.
- Python 3.6+
- PyTorch 0.4
- tensorboardX 1.6
To install all python packages, please run the following command:
pip install -r requirements.txt
After downloading the datasets above, move them to the datasets/
folder in the project root directory, and rename dataset folders to 'market1501' and 'duke' respectively. I.e., the datasets/
folder should be organized as:
|-- market1501
|-- bounding_box_train
|-- bounding_box_test
|-- ...
|-- duke
|-- bounding_box_train
|-- bounding_box_test
|-- ...
After adding dataset directory in demo.sh
, simply run the following command to train DMML on Market-1501:
bash demo.sh
Usage instructions of all training parameters can be found in config.py
.
To evaluate the performance of a trained model, run
python eval.py
which will output Rank-1, Rank-5, Rank-10 and mAP scores.
Please use the citation provided below if it is useful to your research:
Guangyi Chen, Tianren Zhang, Jiwen Lu, and Jie Zhou, Deep Meta Metric Learning, ICCV, 2019.
@inproceedings{chen2019deep,
title={Deep Meta Metric Learning},
author={Chen, Guangyi and Zhang, Tianren and Lu, Jiwen and Zhou, Jie},
booktitle={ICCV},
year={2019}
}