Facial and object recognition

This project is an implementation of the novel yolov3 architecture in OpenCV.

Output exampe

You need to first download the yolov3 model weights.

Once installed, place them in the main folder (.../TM)

pip install -r requirements.txt

First launch the gui, then select an image/video.

This will create a processed copy of the original file in the same folder as the original was.

Ethz paper on non-maximum suppression
Link to the explanation of the darknet function
Link to a visual representation of haar cascades used in convectional image processing.