#########################################################
This Repo. is used for our ACM MM2021 paper:
HAT: Hierarchical Aggregation Transformers for Person Re-identification
(https://arxiv.org/pdf/2107.05946.pdf, https://dl.acm.org/doi/abs/10.1145/3474085.3475202)
#########################################################
For using the codes, please download the public Market-1501 and Duke datasets. Besides, please install the Pytorch as the Official Suggestions.
Then, run the train.py and test.py with your GPUs. You can get the same results in the above paper.
The codes can be download at
Link：https://pan.baidu.com/s/10wt1ai0-zza7RohkIs60IQ
Extracting code：4ox2
More details can be found at https://github.com/gwenzhang/HAT
########################################################
If you have any problems. Please concat the first author:
########################################################
If you find this code help you, please cite our paper
@inproceedings{zhang2021hat,
title={Hat: Hierarchical aggregation transformers for person re-identification},
author={Zhang, Guowen and Zhang, Pingping and Qi, Jinqing and Lu, Huchuan},
booktitle={Proceedings of the 29th ACM International Conference on Multimedia},
pages={516--525},
year={2021}
}