Inspiration

COLLEGE STUDENT STRUGGLES - Imagine you have 5 assignments due at midnight and your mental health is at its all time low. You feel stressed and overwhelmed but don't want to go out to save time. You still don't know how to start being productive but still want to have a good study environment.

What it does

StudiStic provides study space with focus tools and widgets like an embedded Spotify, to-do list, notes, timer, monitors health levels and Gemini builds personalized study plan and recommendations based on that. All the widgets are built in one screen to help user avoid distractions.

rPPG (Remote Photoplethysmography)

rPPG is a scientific model used to analyze heart rate and perspiration based on the way light reflects to your skin, which detects rPPG signals. Face detection is done using OpenCV, then it selects specific areas to extract rPPG signals. It's a bit of a complicated process, and our steps are more simplified than the actual algorithm. Here's a link to read more on the topic: https://noldus.com/blog/what-is-rppg We used an already existing research project and clone their repo to use their algorithm for our project. Here's the link: https://github.com/shubham-rateria/rppg-js

How we built it

We used Next.js and React for the frontend, Firebase for authentication and real-time data, and Google Calendar API for effortless schedule syncing. The webcam data is processed entirely in-browser with TinyFaceDetector model that detects facial signals, and rPPG extraction that estimates heart rate, respiratory rate, and oxygen saturation without hardware. Finally, Gemini Flash analyzes your biometrics and tasks to generate real-time recommendations and personalized study plans.

Challenges we ran into

We faced inaccurate health monitoring data, issues of database integration and Presage implementation. We were especially stressed out about Presage, as there wasn't any API for web apps.

Accomplishments that we're proud of

We built an AI-based productivity tool that would help a lot of our peers. For most of us, this is our first hackathon, and we are proud to say that we made a tool that we would use willingly in our daily lives.

What we learned

How to utilize Gemini AI to analyze data sets, integrate our UI designs into our backend, and use rPPG and OpenCV to detect health issues.

What's next for Studistic

We would want to improve the sensitivity of our health detection as right now we have used open CV. We are planning to use presage by utilizing Fitbit such as Apple Watch, or any smart watch. Instead of just employing contactless wellness that monitor movement/emotion, we would want to add more accurate reading using user’s pulse. Other than, we want user to be able to choose their own video/picture as their study environment.

Built With

Share this project:

Updates