Learning Dynamic Models of Gaze Point Mapping from Eye Movement
Dawei Cheng, Mingyang An, Weiqian Yu, Liang Fang · 2014
The paper presents a high accuracy learning dynamic models system for gaze point mapping based on capturing eye movement from front camera of mobile devices. In the stage of detect pupil, we used a Haar feature-based cascade classifier for real-time eye detection, and connect region detect algorithm for pupil tracking. In the stage of gaze point mapping, we constructed a dynamic learning model by pupil coordinate data collected, denoising them by density-based cluster algorithm, and learning them with a mapping mechanism. Finally, the leaning models were evaluated by a set of experiments, and the results demonstrated the effectiveness of this approach.