Work

Quantifying tracking quality during occlusion with an integrated gaze metric anchored to task performance

Year: 2025

Type: article

Abstract: Visual tracking of moving objects, even when they are temporarily hidden from view, is a fundamental skill for humans. This study examines gaze behavior in an object tracking task with intermittent occlusions, highlighting the role of task goals. Participants tracked an object moving along a linear path with random occlusions, performing a visual discrimination task upon the object's brief reappearance. We analyzed gaze behavior and developed a model to quantify the relationship between gaze dynamics and task performance. The model yields a performance metric that takes into account the full gaze signal, i.e. both smooth pursuit and saccades. We introduce a position- and velocity-integrated tracking quality metric based on these gaze dynamics, providing an objective continuous measure of tracking performance that works also throughout occlusions.

Source: Scientific Reports

DOI: 10.1038/s41598-025-17519-8

Authors: Tuisku Tammi, Jami Pekkanen, Benjamin Ultan Cowley, Otto Lappi

Institution: University of Helsinki

Language: English

Topic: Gaze Tracking and Assistive Technology

Subfield: Human-Computer Interaction

Field: Computer Science

Domain: Physical Sciences

Open Access status: gold

Estimated APC paid: $2,690

Funders: Helsingin Yliopisto, Academy of Finland