This project was designed to act as a "traffic camera" for tracking vehicles and pedestrians to generate metrics based on the data. The idea is to be able to use a small hardware setup to run the models while having the data accessible from a website in the cloud. The hardware being used here is the Nvidia Jetson Nano. One of the main goals of this project is to provide an ethical way to count vehicle and pedestrian traffic. Many of the systems that exist do not put an emphasis on privacy and that is why it is one of our main goals.
Coming soon...
Python 3.8.x or newer (64-bit installation required)
pip install -r requirements.txt
Ensure that you are using the pip associated with Python 3.8.x
Running the tracker:
python TrafficTracker.py --video_path path/to/file.mp4
Running the Flask App (Heatmap Visualization):
cd app/
python flask_heatmap.py
Go to your browser and enter localhost:5000/heatmap
Running the container on Jetson Nano:
sudo docker run -it --rm --runtime nvidia --network host traffictracker:latest
Running development container:
sudo docker run -it --rm --runtime nvidia -v /path/to/TrafficTracker:/TrafficTracker traffictracker:dev
Distributed under the MIT License. See LICENSE for more information.