EyeFocus

Framing the Problem

What is the problem you are trying to solve? Who does it affect?

Many viewers struggle to stay focused during long or educational YouTube videos due to distractions, multitasking, or fatigue. As attention drops, they miss key information and lose learning efficiency. This project addresses that challenge by transforming raw video input into an interactive, AI-enhanced experience that detects when focus is lost and generates concise recaps, helping users stay engaged and retain more knowledge.

Idea Explanation

What is your idea? How does it fix the problem?

Our idea, Eye-Focus, is a Chrome extension that helps users stay engaged while watching videos. It combines eye-tracking and AI summarization to automatically pause the video when the user looks away for a certain period of time and provide a short recap when they return about the iformation they may have missed. They can also ask questions to the extension to clarify ideas they may have missed. It also includes important topic alerts, notifying users right before key moments appear in the video. This ensures that users stay attentive and don’t miss crucial information.

Implementation

How do all the pieces fit together? Does your frontend make requests to your backend? Where does your database fit in?

The Chrome extension acts as the frontend, built using HTML, CSS, and JavaScript, and manages user settings like popup delay and alert timing. It interacts with the YouTube Player API to control playback and display recaps. A Python backend runs an eye-tracking model that detects when the user looks away and sends a signal to the extension. The extension connects with Groq AI, which generates quick summaries of the last 20 seconds and identifies key topics using video transcripts. There is no database required; all data is processed locally and temporarily cached for performance.

Challenges

What did you struggle with? How did you overcome it?

The biggest challenge was integrating the eye-tracking Python backend with the Chrome extension, since they operate in separate environments. The extension runs in the browser , while the eye tracker runs locally on the computer. We solved this by creating a local communication bridge using an HTTP server built with Python’s Flask library. The extension sends requests to the backend through a local endpoint that continuously updates the user’s gaze status. When the backend detects that the user looks away for more than 5 seconds, it triggers a signal to pause the YouTube video. This approach allowed real-time synchronization between the camera-based gaze detection and the browser interface, making the experience seamless and efficient.

Accomplishments

What did you learn? What did you accomplish?

• Developed a fully functional Chrome extension that integrates real-time eye-tracking with YouTube playback.

• Implemented Groq AI summarization for quick, context-aware video recaps.

• Added customizable user settings for popup timing, alerts, and AI preferences.

• Improved focus and productivity for users watching educational videos.

Next Steps

What are the next steps for your project? How can you improve it?

• Enhance gaze tracking precision using ML calibration and facial landmark detection.

• Add focus analytics to measure attention duration and engagement levels.

• Expand compatibility to Netflix, Coursera, and Udemy.

• Introduce voice-based interactions, allowing users to verbally request summaries or explanations.