What question did this study set out to answer?

The aim is to improve the classification of eye movement events like fixations and smooth pursuits.

March 18, 2026Scientific DataOpen Access

Eye movement benchmark data for smooth-pursuit classification

View Full Paper

Ask AI

Bookmark

Authors

Analysis of eye movement events using a benchmark dataset enhances classification algorithms' accuracy.

The aim is to improve the classification of eye movement events like fixations and smooth pursuits.
Created a benchmark dataset without human annotation as the gold standard.
Collected nearly four hours of eye movement data from ten participants.
Stimuli were designed to prevent co-occurrence of fixations and smooth pursuits, distinguishing them from saccades based on velocity.
Provided a Python package for data preprocessing and labeling.
Established plausible benchmark labels for eye movement classification.
Provided raw data for researchers to develop and validate classification algorithms.
The dataset aids in enhancing feature engineering for eye tracking analysis.

Korthals et al. (2026) studied this question.

View Full Paper

Ask AI

Bookmark