Idea
A browser-based eye-tracking platform delivering real-time, personalized gaze estimation for privacy-conscious web applications and researchers
Research Paper
Core Innovation
This paper presents WebEyeTrack, which uniquely combines lightweight gaze estimation models with on-device few-shot learning and head pose estimation. Unlike prior work, it enables accurate, real-time eye tracking directly in browsers with minimal calibration and enhanced privacy by avoiding cloud processing. This approach scales easily across devices without specialized hardware.
Market Size (TAM)
$2–10B TAM, $1–2B SAM; assumption: growing demand for eye-tracking in web UX, accessibility, and AR/VR interfaces.
Potential Customers & Pain Points
- Web Developers Needing Integrated Eye-Tracking
- UX Researchers Requiring Scalable Gaze Data
- Accessibility Tool Makers Seeking Low-Calibration Solutions
- Privacy-Focused Enterprises Avoiding Cloud Data Transfer
Business Model
SaaS platform with tiered API access for developers and enterprise licensing for privacy-sensitive clients
Competitive Landscape
- Tobii
- EyeTech Digital Systems
- Pupil Labs
Implementation Challenges
- Browser performance limitations
- User adoption of calibration steps
- Competition from specialized hardware vendors
Validation Strategy
- Develop browser plugin prototype and test accuracy
- Pilot with UX researchers for real-world feedback
- Partner with accessibility tool developers for integration
Research Paper Overview
WEBEYETRACK: Scalable Eye-Tracking for the Browser via On-Device Few-Shot Personalization
Summary
WebEyeTrack is a browser-based eye-tracking framework that integrates lightweight state-of-the-art gaze estimation models with on-device few-shot learning and head pose estimation, enabling accurate, real-time gaze tracking with minimal calibration and strong privacy.