This project allows you to stream a live video feed of objects being detected.
TensorFlow is used with the help of a pre-trained model to detect objects in a live video feed. The TensorFlow Object Detection API is used alongside the SSD Mobilenet v1 Coco model, this pretrained model is one of the fastest to detect objects (as of late 2017).
Install pipenv if not installed:
pip install -U pipenv
Set up the environment:
cd src/
pipenv install
pipenv shell
Then to run:
python app.py