Webcam-based eye tracking offers a practical approach for monitoring reading behavior in classroom environments. Nascimento et al. demonstrated the feasibility of using the Webcam-based eye tracking to observe comic reading patterns among children and showed that it effectively captures fixation time and reading dynamics. Despite the data quality challenges, Webcam-based eye tracking provided an accuracy comparable to that of infrared-based systems.
This study highlights the potential of the WBET technology as an accessible tool for classroom research. Despite its slightly lower precision compared to infrared-based systems, WBET proved effective in tracking gaze behavior during comic reading tasks, revealing meaningful attention patterns. Limitations, such as data loss, environmental distractions, and lower image quality, underscore the need for methodological improvements. However, the ecological validity of conducting research in a natural classroom setting demonstrated of WBET as a cost-effective alternative for educational studies. Future studies should refine its accuracy and explore broader applications in learning environments.
https://www.scielo.br/j/eins/a/Tc9PN5LdQdTPdW698jsYShp/
Eye-tracking equipment
Two eye tracking devices were used in this study. the The GazeRecorder software was used on two notebooks for data acquisition via the webcam, in which the resolution of the cameras was 640×480 at 30fps and 1280×720 at 30fps for the validation task and comic reading task, respectively.
The IR eye-tracking data were captured using Tobii Glasses Pro II, which is a portable eye-tracking device that uses corneal reflection and dark pupil methods. The system consists of several components, each serving a specific purpose. The head unit, which is a wearable eye tracker in the form of glasses, features two sensor cameras per eye that record the eye orientation, an IR illuminator that illuminates the eyes to support the sensor cameras, and a high-definition camera that records what is in front of the user. Data is captured at a sampling rate of 100Hz and a video resolution of 1920×1080 pixels at 25fps.
Webcam-based eye-tracking validation test
As a preliminary step, we conducted a validation test to assess the reliability of WBET used in this study. Three authors-two men and one woman-volunteered for this validation test, with ages ranging between 24-34 years. All three participants underwent the procedure without wearing glasses, including those who required vision correction lenses, to aid eye identification by the hardware. A screen with nine small black circles was shown to the participants, who received verbal instructions to look at the next reference point at intervals of approximately 3s (controlled with a stopwatch). The sequential order was followed from the top left corner to the bottom right, as that in the left to write (top to bottom) reading. Thus, the eye-tracker gaze estimate could be compared with the actual reference position. The validation task lasted approximately 30s. Each volunteer performed the task twice: once with the WBET GazeRecorder32 software and once with the Tobii Glasses Pro II33 device. Before the validation task, calibration was performed based on the eye tracker used. In the WBET, facial recognition is automatically performed by the software as the participant gazes at a red circle in the center of the screen. The calibration process involves nine red circles that appear and disappear individually at various locations on the screen. In contrast, in IR eye tracking, calibration was conducted while the participant focused on a circular figure placed at a distance of one outstretched arm.