Eye movement velocity and gaze data generator for evaluation, robustness testing and assess of eye tracking software and visualization tools

TL;DR

This paper introduces a novel eye movement simulator that probabilistically models saccades and smooth pursuits, capable of generating realistic gaze data for testing eye tracking software and visualization tools across various conditions.

Contribution

It presents a new probabilistic eye movement simulation method that models saccades and pursuits, adaptable to different sampling rates and scene types, for evaluation purposes.

Findings

Simulator accurately reproduces real eye movement data

Capable of generating data for static and dynamic scenes

Supports various sampling rates and conditions

Abstract

Eye movements hold information about human perception, intention, and cognitive state. We propose a novel eye movement simulator that i) probabilistically simulates saccade movements as gamma distributions considering different peak velocities and ii) models smooth pursuit onsets with the sigmoid function. Additionally, it is capable of producing velocity and two-dimensional gaze sequences for static and dynamic scenes using saliency maps or real fixation targets. Our approach is also capable of simulating any sampling rate, even with uctuations. The simulation is evaluated against publicly available annotated data. The simulator can be used in EyeTrace or downloaded at http://ti.unituebingen. de/Projekte.1801.0.html.