Skip to content

Latest commit

 

History

History
47 lines (37 loc) · 1.35 KB

INSTALL.md

File metadata and controls

47 lines (37 loc) · 1.35 KB

Installation

PKF is built upon codebase of YOLOX and OC-SORT. I tested the code with Python 3.8.

1. Installing on the host machine

Step1. Install dependencies

git clone /~https://github.com/hwcao17/pkf.git
cd pkf
pip install -r requirements.txt
python setup.py develop

Install Eigen3, you may use

sudo apt install libeigen3-dev

Install pybind11, you may use

pip install pybind11

Step2. Install the probabilistic data association

cd trackers/pkf_tracker/data_association
python setup.py install

If you have problem with pybind11 at this step, you may consider installing pybind11 using

pip install "pybind11[global]"

Step3. Install pycocotools.

pip install cython; pip install 'git+/~https://github.com/cocodataset/cocoapi.git#subdirectory=PythonAPI'

Step4. Others

pip install cython_bbox pandas xmltodict

2. Download the pretrained models

Download the pretrained models from Google Drive and put them under <pkf root>/pretrained