Code for Paper "Learning Diverse Features with Part-Level Resolution for Person Re-Identification"

This code is developed based on pytorch framework.

Prerequisites

• Pytorch(1.1.0)

• yacs (0.1.6)

• torchvision(0.3.0)

• tb-nightly(2.0.0)

• Cython(0.29.12)

• pytorch-ignite(0.1.2)

SetUP

git clone https://github.com/AI-NERC-NUPT/PLR-OSNet.git

Dataset folder

• dukemtmc-reid

  • DukeMTMC-reID

    • bounding_box_train

    • bounding_box_test

    • query

• market1501

  • Market-1501-v15.09.15

    • bounding_box_train

    • bounding_box_test

    • query

• cuhk03

  • CUHK03_labeled
    • bounding_box_train
    • bounding_box_test
    • query
  • CUHK03_detected
    • bounding_box_train
    • bounding_box_test
    • query

Train

cd PLR-OSNet/scripts
vim train.sh
bash train.sh

Results(without re-rank)

	rank-1	mAP
Market1501	95.6	88.9
DukeMTMC	91.6	81.2
CUHK03-Labeled	84.6	80.5
CUHK03-Detected	80.4	77.2

Datasets

• Market1501 https://www.cvfoundation.org/openaccess/content_iccv_2015/papers/Zheng_Scalable_Person_Re-Identification_ICCV_2015_paper.pdf

• DukeMTMC-reID

  https://arxiv.org/abs/1701.07717

• CUHK03

  https://www.cvfoundation.org/openaccess/content_cvpr_2014/papers/Li_DeepReID_Deep_Filter_2014_CVPR_paper.pdf