Tracking the Progression of Reading Using Eye-Gaze Point Measurements and Hidden Markov Models

Document Type

Article

Publication Date

10-1-2020

Publication Title

IEEE Transactions on Instrumentation and Measurement

Volume

69

Issue

10

First Page

7857

Keywords

eye tracking, Eye-gaze measurements, hidden Markov models, information fusion, Kalman filter, machine learning

Last Page

7868

Abstract

In this article, we consider the problem of tracking the eye gaze of individuals while they engage in reading. In particular, we develop the ways to accurately track the line being read by an individual using commercially available eye-tracking devices. Such an approach will enable futuristic functionalities, such as comprehension evaluation, interest level detection, and user-assisting applications such as hand-free navigation and automatic scrolling. Furthermore, the proposed approach will pave the way to develop technology that may generate valuable feedback to content makers, such as web designers, authors, educators, and social media users. The existing commercial eye trackers provide an estimated location of the eye-gaze points every few milliseconds. However, these estimated gaze points are not sufficient to quantify reading progression - a specific eye-gaze activity. In this article, we propose algorithms to bridge the commercial gaze tracker outputs and informative eye-gaze patterns while reading. The proposed system consists of Kalman filters and hidden Markov models to parameterize these statistical models and to accurately detect the line being read. The proposed approach is shown to yield an improvement of 27.1% in line detection accuracy over line tracking using estimated eye-gaze points alone by the eye tracker.

DOI

10.1109/TIM.2020.2983525

ISSN

00189456

E-ISSN

15579662

Link to Full Text

Share

COinS