Joint Generative and Contrastive Learning for Unsupervised Person Re-identification

Implement of paper:Joint Generative and Contrastive Learning for Unsupervised Person Re-identification.

Qualitative results

Installation

Install GCL

Requirements

• Python 3.6
• Pytorch 1.2.0

git clone https://github.com/chenhao2345/GCL
cd GCL
python setup.py develop

Prepare Datasets

cd examples && mkdir data

Download the raw datasets DukeMTMC-reID, Market-1501, MSMT17, and then unzip them under the directory like

ABMT/examples/data
├── dukemtmc-reid
│   └── DukeMTMC-reID
├── market1501
└── msmt17
    └── MSMT17_V1(or MSMT17_V2)

Install HMR for Mesh Estimation

Download our extracted meshes from Google Drive.

Or refer to HMR ro get meshes for ReID datasets.

Train GCL

Only support 1 GPU training for the moment.

Stage 1: Warm up identity encoder

Train a ResNet50 with an unsupervised method, for example, JVTC(or download our trained models from Google Drive) and MLC.

Stage 2: Warm up structure encoder and discriminator

Adjust path for dataset, mesh, pre-trained identity encoder.

sh train_stage2_market.sh

Stage 3: Joint training

sh train_stage3_market.sh

TensorBoard Visualization

Stage 2:

For example,

tensorboard --logdir logs/market_init_JVTC_unsupervised/

Stage 3:

For example,

tensorboard --logdir logs/market_init_JVTC_unsupervised/stage3/

Citation

@article{chen2020joint,
  title={Joint Generative and Contrastive Learning for Unsupervised Person Re-identification},
  author={Chen, Hao and Wang, Yaohui and Lagadec, Benoit and Dantcheva, Antitza and Bremond, Francois},
  journal={arXiv preprint arXiv:2012.09071},
  year={2020}
}