Intract

Inspiration

Recently, we were conducting basics of coding session for the juniors of our university. During that, we were'nt able to gauge whether they were able to follow, whether they are clear on the concepts or atleast if they were present in the class. Teachers must face this on a daily basis. This is a small effort to make their lives easier for all that they do for us!

What it does

Our website is able to accurately detect the emotion of the students to gauge their level of involvement and interaction. This data is then compiled and is made visible to the teacher so that he can monitor the involvement of the class and ensure maximum retention by the students.

How we built it

We used reactjs to create the web app which runs on the nodejs engine on chrome for javascript and did the mood recognition on the user's system in the browser with the help of tfjs and face-api.js We used firebase in order to get real-time data and have an easily scalable product we have used firebase as a backend for storage and authentication

Challenges we ran into

Creating a real-time classroom sharing the data of each user. Getting the mood detection on the browser was quite difficult. Our first approach was to have the model running on a server but that caused broadband issues, we were able to solve it by running the ML model on each student's browser and just sending the result.

Accomplishments that we're proud of

We are proud that we were able to create and deploy the entire website during the duration of this hack. Moreover we are very proud of the architecture we came up with which causes the server cost to be a minimum. Our whole web app is created in Reactjs which helps us expand to native desktop apps and even mobile apps.