Quantifying the Impact of Motion on 2D Gaze Estimation in Real-World Mobile Interactions

2025-02-14

AI Summary

Motion-induced degradation of 2D gaze estimation accuracy remains poorly quantified in real-world mobile scenarios. Method: We conducted two user studies, collecting protracted facial videos and high-precision ground-truth gaze labels under diverse motion states—including supine rest and maze navigation—and applied statistical regression to disentangle and quantify the independent contributions of three dynamic error sources: eye-to-camera distance, head pose, and device orientation. Contribution/Results: Motion increases gaze estimation error by up to 48.91%. We thus propose a novel, motion-robust evaluation paradigm for adaptive eye tracking, establishing an empirical benchmark for gaze model design, system calibration, and behavioral modeling in dynamic environments. This work provides actionable insights for improving gaze estimation reliability during natural user movement.

Technology Category

• Computer Vision: Motion & Tracking
• Intelligent Robots: State Estimation
• Machine Learning: Calibration & Uncertainty Quantification

Application Category

• User Modeling, Personalization and Recommendation: Metrics for user behavior and evaluating success
• Security and Privacy: Large-scale security measurements
• Graph Algorithms and Modeling for the Web: Efficient manipulation of static and dynamic Web-related graphs

Abstract

Mobile gaze tracking involves inferring a user's gaze point or direction on a mobile device's screen from facial images captured by the device's front camera. While this technology inspires an increasing number of gaze-interaction applications, achieving consistent accuracy remains challenging due to dynamic user-device spatial relationships and varied motion conditions inherent in mobile contexts. This paper provides empirical evidence on how user mobility and behaviour affect mobile gaze tracking accuracy. We conduct two user studies collecting behaviour and gaze data under various motion conditions - from lying to maze navigation - and during different interaction tasks. Quantitative analysis has revealed behavioural regularities among daily tasks and identified head distance, head pose, and device orientation as key factors affecting accuracy, with errors increasing by up to 48.91% in dynamic conditions compared to static ones. These findings highlight the need for more robust, adaptive eye-tracking systems that account for head movements and device deflection to maintain accuracy across diverse mobile contexts.

Problem

• Quantifies impact of motion on mobile gaze tracking accuracy.
• Identifies head distance, pose, device orientation as key factors.
• Highlights need for adaptive systems for varied mobile contexts.

Innovation

• Mobile gaze tracking
• User mobility impact
• Adaptive eye-tracking systems