This repo contains the source code for our work "Central Feature Learning for Unsupervised Person Re-identification"

Prerequisites

1. Pytorch 0.4.1
2. Python 3.6

Preparation

• Data preparation

mkdir data

ln -s [PATH TO MSMT17_V1] ./data/MSMT17_V1
ln -s [PATH TO DUKE] ./data/DukeMTMC-reID
ln -s [PATH TO Market] ./data/Market

• set the path of ImageNet pretrained models

ln -s [THE PATH OF IMAGENET PRE-TRAINED MODELS] imagenet_models

Run the code

• For pretraining the model

cd ./train
python supervised_train.py --gpu [CHOOSE WHICH GPU TO RUN] --exp-name [YOUR EXP NAME]

mkdir ./snapshot
mkdir ./snapshot/MSMT17_PRE
cp [PATH TO PRETRAINED MODEL] ./snapshot/MSMT17_PRE/
# it means the name of the experiment of pretraining is 'MSMT17_PRE'  

• For unsupervised training

cd ./unsupervised

# for market
python unsupervised_train.py --data MARKET --gpu [CHOOSE WHICH GPU TO RUN] \
--pre-name [THE EXP NAME OF PRE-TRIANED MODEL] --exp-name [YOUR EXP NAME] \
--batch-size 42 --scale 15 --lr 0.0001 

 # for duke
python unsupervised_train.py --data DUKE --gpu [CHOOSE WHICH GPU TO RUN] \
--pre-name [THE EXP NAME OF PRE-TRIANED MODEL] --exp-name [YOUR EXP NAME] \
--batch-size 40 --scale 5 --lr 0.0001 

Code link

the link to the specific code of each comparison method ECN
PAUL
HHL
MAR
BUC
PCB-PAST \

Reference

If you find our work helpful in your research, please kindly cite our paper:

Qize Yang, Hong-Xing Yu, Ancong Wu, Wei-Shi Zheng, "Patch-based discriminative feature learning for unsupervised person re-identification", In CVPR, 2019.

Zhun Zhong, Liang Zheng, Shaozi Li and Yi Yang, "Generalizing a person retrieval model hetero-and homogeneously", ECCV, 2018.

Zhun Zhong and Liang Zheng and Zhiming Luo and Shaozi Li1 and Yi Yang, "Invariance matters: exemplar memory for domain adaptive person re-identification", CVPR, 2019.