Inspiration

Amidst our current pandemic, it is imperative that everyone wears their mask correctly. Doing so will cause a tremendous impact. Maskker allows people to check if they are wearing their masks correctly or not.

What it does

This program uses a deep learning model in order to check if the user is wearing their mask correctly or not.

How I built it

We built it by using Android Studio for the front end and the action events, and by using python and deep learning (using Kaggle's services) to detect if the person was wearing their mask correctly.

Challenges I ran into

Since the images were 1024x1024, we weren't able to use much data and as a result, the algorithm wasn't very accurate. Additionally, we had difficulties connecting the camera to the emulator, though we resolved the issue at the end.

Accomplishments that I'm proud of

We are really proud of how much we learned throughout this process, since none of us had used android studio before. Also, I'm really proud of how awesome the end result looked and how it worked.

What I learned

Through this experience, we learned a lot about Android app development and machine learning.

What's next for Maskker

We'd like to integrate more functionality to the maskker app for more safety features, such as an app that checks if somebody is within 6 feet of you. We would also like to train on more data and make the classifier more accurate.

Built With

Share this project:

Updates