HRnet-reID

Strong And Friendly

A re-ID baseline constructed by High-Resolution Network. The rank-1 of Market-1501 is improved to 94.7%, which is much higher than the ResNet based method in the same dataset.

Installation

• Install Pytorch from http://pytorch.org/
• Install Torchvision from the source

git clone https://github.com/pytorch/vision
cd vision
python setup.py install

• [Optinal] You may skip it. Install apex from the source

git clone https://github.com/NVIDIA/apex.git
cd apex
python setup.py install --cuda_ext --cpp_ext

Because pytorch and torchvision are ongoing projects.

Here we noted that our code is tested based on Pytorch 1.10.0 and CUDA 11.3 on Windows 11.

Dataset & Preparation

Download Market1501 Dataset [Google] [Baidu]

Preparation: Put the images with the same id in one folder. You may use

python prepare.py

Remember to change the dataset path to your own path.

Train & Test

Just following HRNet-reID.ipynb~