Multi-Object-Tracking-in-python

This repository contains implementation of various multi-object trackers.

In addition, it includes is tutorial with goal to demonstrate principles of work this trackers in educational proposes. It is unfluenced by the Multiple Object Tracking course in Chalmers and Edx youtube 📺 and MATLAB open sourced implementations.

You can dive into in tutorial notebook (locally or in colab). Or explore code explicitly.

Using in virtualenv

Firstly, you need install Eigen3 in your system.

python3.10 -m venv .venv

source .venv/bin/activate

python -m pip install -e .

Development

As a dependencies package manager this project use PDM.

Install pre-commit

pre-commit install

Single Object tracking

pytest tests/SOT/

Tracking with fixed number of objects

pytest tests/MOT

Multi object tracking with Probability Hypothesis Density filter

pytest tests/PHD

Multi object tracking with Poisson Multi Bernoulli Mixture filter

pytest tests/PMBM