Tracking Vehicles Over Time

This project is aimed to track vehicles over time by fusing measurements from LiDAR and a camera.

Watch the Demo · Report Bug · Request Feature

Table of Contents

1. About The Project
   • Built With
2. Getting Started
   • Prerequisites
   • Installation
3. Usage
4. License
5. Contact
6. Acknowledgements

About The Project

In this project, I employed techniques to detect objects in 3D point clouds and then used an Extended Kalman Filter for sensor fusion and tracking. The extended Kalman filter (EKF) is an algorithm that allows us to estimate the state of a system and track it over time using noisy measurements. By combining the strengths of both LiDAR and camera sensors, we were able to improve the overall accuracy and robustness of the tracking system.

(back to top)

Built With

• 
• 
• 
• 

(back to top)

Getting Started

Clone the repository on your local machine and run the loop_over_dataset.py. Additional configurations are provided in the loop_over_dataset.py.

Prerequisites

I have provided the requirements.txt. I highly recommend you to create a virtual environment before installing the prerequsite libraries.

Installation

• create a virtual environment

  py -m venv env

• installing all requirements at the same time

  py -m pip install -r requirements.txt

(back to top)

Usage

You can try different deep learning models to evaluate the accuracy of the tracking system.(The model needs to be trained on open-source waymo dataset that can be found here) And most importantly you can use it to show the effects of different filters with adjusted hyper-parameters.

(back to top)

License

Distributed under the MIT License. See LICENSE.txt for more information.

(back to top)

Contact

• Deniz Temur - [email protected]
• 

(back to top)

Acknowledgements

• (C) 2020, Dr. Antje Muntzinger / Dr. Andreas Haja

(back to top)