When we were curating and brainstorming ideas one thing that got our attention was to see that there weren't many free face mask detection apps and that actually inspired us to create one that was simple, fast, and reliable.
What it does
Ani-mask has a simple yet attractive UI and one of the most prominent features is the mask detection using the device camera.
How we built it
Ani-mask uses machine learning concepts integrated with TensorFlow and a KNN classifier. Also to improve user experience the UI has been built keeping in mind the element hierarchy and color theory.
Challenges we ran into
To start two of our team members dropped out. When we started coding the first challenge was integrating the tensorFlow addon. I was originally passing through too many images but it took a while to figure that out. Another problem was getting the webcam positioned properly. It doesn't sound like too hard of a problem but it just wasn't doing what we wanted so we scrapped it - then re built it and it worked fine.
Accomplishments that we're proud of
Getting the project working on time. We felt very accomplished seeing the working results. This is our first machine learning project.
What we learned
Since none of us had previously used Tensorflow and machine learning in our projects so analyzing and working on that was quite an experience
What's next for Ani-mask
We want Ani-mask as an open-source project where people can not only use it but also contribute to it secondly features as instant notifications will also be considered since we could not add those because of the limited time frame.