Eye-tracking as a tool for researching user behavior

Abstract

This article discusses eye-tracking technology, one of the critical technologies in analyzing user behavior when interacting with computer systems, and has a wide range of practical applications. The current capabilities and limitations of existing methods and equipment for eye tracking were reviewed and analyzed, and problems and promising areas for further research were identified. The central hardware systems were studied, such as desktop trackers, mobile glasses for eye-tracking, systems for immersive environments, and systems based on embedded cameras. This research also includes analyzing and classifying eye localization methods: eye shape-based, feature-based, and appearance-based. Also were investigated modern gaze-tracking methods, which are crucial to analyzing eye movement and determining the user's point of view. Feature-based methods provide accuracy but have limitations in strong lighting conditions. Model-based methods require high image resolution and calibration to measure gaze direction using facial characteristics. Cross-ratio-based methods do not require calibration but require additional lighting. Appearance-based methods are effective with low-quality images but have limitations concerning head position. The study indicates significant potential for improving gaze-tracking technologies to enhance user interfaces and increase accessibility. It emphasizes the need to develop adaptive methods that function in different environmental conditions and hardware limitations.