Object detection with imageAI

Based on imageAI

Import conda environment with

conda env create -f environment.yml

If you encounter any problems during install with wrapt try the following:

conda update --all
conda remove wrapt
pip install tensorflow

If you get an error during execution like

ImportError: numpy.core._multiarray_umath failed to import

try the following:

pip install -U numpy

Download YoloV3 Model

wget https://github.com/OlafenwaMoses/ImageAI/releases/download/1.0/yolo.h5

imageobjectdetection.py

Uses all *.jpg Files from the pictures directory and performes object detection on them. Saves the result in *_new.jpg in same folder

videoobjectdetection.py

Uses Webcam to capture frames an then performs object detection. Shows result in a window.