We're currently still in the midst of a global pandemic that has permanently changed the lives of people across the globe. We wanted to make something that would touch on this issue, which is why we developed Mask On!, a product that has applications in both the healthcare and business industries, in hopes of helping to combat COVID-19, even if it's just a little.

What it does

Mask On! is an image recognition web application that detects if someone is wearing a mask and following social distancing protocols. It takes an image that the user uploads, and classifies it accordingly. Mask On! has many potential applications, including screening customers/hospital patients without having them even setting a foot inside the premises.

How we built it

Mask On! utilizes the Google Cloud Vision api and a 19-layer deep neural network to detect masks on faces. The website utilizes a Django backend for image uploading and an intuitive JS/HTML/CSS frontend for the UI.

Challenges we ran into

As this was our first time ever working with ML/AI on a project of this scale, we naturally ran into several logistical and understanding problems, which we ultimately overcame.

Accomplishments that we're proud of

We're extremely proud of shipping a relatively polished product, as this was our first/second time ever participating in a hackathon. In addition, we were proud to work with ML/AI for the first time, as this was a space we've been interested in for a long time.

What we learned

We learned about basic ML/AI concepts, how to apply them in practice on a medium sized project, as well as some new things about full-stack development through our development of our website.

What's next for Mask On!

The technologies and concepts applied in Mask On! can be applied to a plethora of different global issues. We plan to continue to learn and grow, and create more products that can benefit the world.

Share this project: