An HMM-based framework for real-time EOG-based eye movement labelling
Year: 2026
Type: article
Abstract: Electrooculography (EOG) is a technique for tracking and recording eye movements based on the standing potential that exists across the human eye. Compared to traditional video-based eye gaze tracking techniques, the EOG signals recorded across the human eye contain rich information about the ocular movements humans perform. The process of identifying these different eye movements is a fundamental part of eye-movement data analysis and is essential for operating several human–computer interfaces (HCIs). This work presents the development of an eye movement detection technique which automatically detects and labels fixations, saccades and blinks in real-time providing temporal information on each event. The proposed technique yielded an average F-score exceeding 90% when applied to long segments of EOG data, which is well above the performance obtained using two state-of-the-art techniques.
Institution: University of Malta
Language: English
Subfield: Human-Computer Interaction
Field: Computer Science
Domain: Physical Sciences
Open Access status: hybrid
Estimated APC paid: $2,590
OpenAlex ID: W7155002981