Inspiration
We wanted to know how the emotions shown on one's face are related to one's mental wellness
What it does
Tracks emotions of users for a given time and outputs a score representing their mental wellness
How we built it
For the emotion detection, we used a model made by someone else link: https://github.com/Kakarot-2000/Project-Illumina For the score, we felt that emotions and mental wellness could be linearly related and hence used a linear regression model to get the score
Challenges we ran into
We could not display the score on the Flask web app and could only display it in the terminal. We could not use the start and stop buttons to run it and the code runs by a given time that starts when you visit the website. We tried using Replit initially but there wasn't enough storage there. Hence we had to do it locally using VS code. We could not find data set for this online and could only come up with it ourselves. Hence the output score may not be that accurate.
Accomplishments that we're proud of
We were able to make use of other people's model for our project even though we did not fully understand their code.
What we learned
Flask will not work on google collab and sometimes, the issue may be with the platform we are using.
What's next for EmoHealth
We will try to fix the issues if possible
Log in or sign up for Devpost to join the conversation.