Focus Feen

Focus Flow

Inspiration

In today’s world, distractions from smartphones and social media are constant. But blaming these distractions is like blaming junk food for a bad diet. Instead of blaming external factors, we believe focus can be trained like a muscle at the gym. The idea for Focus Feen was born out of the desire to help people strengthen their focus skills, but in a way that’s measurable and fun. We wanted to gamify the process of tracking focus, helping users develop an unbreakable concentration flow over time.

What it does

Focus Feen uses a camera-based neural network model to track whether the user is focused or distracted by analyzing head movements. The app records focused time and provides real-time feedback when distractions break the user’s focus. By tracking these periods, users can gradually improve their focus and work towards uninterrupted concentration sessions.

How we built it

We built Focus Feen with two core components. First, we used Google's Teachable Machine to train a neural network model that uses head pose detection to predict focus. Second, we developed the user interface with React, using the model’s predictions to update a timer that tracks focused time. If the user is distracted for too long, the app ends the session. The frontend was styled with HTML and CSS to create an engaging and intuitive experience. We tested different platforms along the way, including Android and React Native, but ultimately settled on React for its flexibility.

Challenges we ran into

Embedding our neural network model into the user interface was our biggest challenge. The model was complex and came with limited documentation, forcing us to explore and understand the intricacies of integration. We tried multiple platforms, including Android and React Native, before finding success with React. Debugging, researching, and constantly pivoting became key parts of our process, and pair programming helped us stay focused and motivated through these hurdles.

Accomplishments that we're proud of

We’re proud of how our team adapted to challenges by embracing change and continuously learning throughout the project. Despite facing difficulties with embedding the model and platform limitations, we never gave up. Instead, we pivoted when needed and implemented pair programming to ensure every team member could contribute. We’re especially proud of the seamless integration of our neural network model into a user-friendly interface that makes focus tracking accessible and fun.

What we learned

Our biggest lesson was learning to embrace change and avoid falling into the sunk cost fallacy. When a platform or tool wasn’t working, we pivoted instead of stubbornly sticking to it. We also learned new technical skills, such as building web applications with Node.js and React, setting up Android projects, and handling version control—all while growing stronger as a team through collaboration.

What's next for Focus Feen

Looking forward, we plan to expand Focus Feen by refining our neural network model for even more accurate focus detection and incorporating new features, such as personalized insights and suggestions based on user behavior. We also hope to explore additional platforms and user experiences to make Focus Feen accessible to a wider audience, helping even more people unlock their focus potential.