Face Mask Detection and People Counting

Introduction:

Covid-19 pandemic has affected people all around the world. There is a need for technological solutions like this project that can help us curb the growth of this pandemic and help save lives.
Checking if people are wearing masks in closed areas and keeping a count of the number of people is extremely important to keep Covid-19 in check. That is why we chose to do this project and help the world in a small way.

Steps:

1. Convert a video stream or a video file to get individual frames (images) from it.
2. Use the PreTrained face detection model built by OpenVino to detect faces
3. Use the bounding boxes obtained from this model to run a binary classifier that classifies if someone is wearing a mask or no.
4. Draw boxes around the faces and label them (mask/no-mask)
5. Join all the images to create a video stream again

Flowchart:

Output:

Check Out the Project here:

1. Google Colab
2. GitHub

References:

1. OpenVino Face Detection Model

Contributors:

Abilash Nair & Rohit Rokde