Related Experiment Video
Updated: Jul 17, 2026
07:09
Gaze in Action: Head-mounted Eye Tracking of Children's Dynamic Visual Attention During Naturalistic Behavior
Published on: November 14, 2018
10.6K
A review of machine learning in scanpath analysis for passive gaze-based interaction
Abdulrahman Mohamed Selim1, Michael Barz1,2, Omair Shahzad Bhatti1
1German Research Center for Artificial Intelligence (DFKI), Interactive Machine Learning Department, Saarbrücken, Germany.
Frontiers in Artificial Intelligence
|June 21, 2024
Summary
This review explores machine learning for scanpath analysis in passive gaze-based interaction. It highlights common practices, research trends, and identifies future challenges in interpreting eye movements for improved human-computer interaction.
Area of Science:
- Human-Computer Interaction
- Machine Learning
- Eye Tracking
Background:
- Scanpaths, representing eye movements (fixations and saccades), are crucial in understanding user behavior.
- Passive gaze-based interaction utilizes implicit interpretation of eye movements to enhance user experience.
- Machine learning (ML) offers advanced capabilities for automatic scanpath analysis.
Purpose of the Study:
- To review ML applications in scanpath analysis for passive gaze-based interaction from 2012-2022.
- To identify research domains, learning tasks, and common ML practices.
- To highlight research gaps and challenges for future investigation.
Main Methods:
- Systematic literature review of 2,425 publications, focusing on 77 relevant studies.
- Analysis of ML techniques applied to scanpath data.
- Examination of data collection, preparation, model selection, and evaluation methodologies.
Main Results:
- Identified key research domains and common ML tasks within passive gaze-based interaction.
- Detailed common practices in ML pipeline for scanpath analysis.
- Highlighted areas for improvement and emerging ML trends.
Conclusions:
- ML is increasingly vital for interpreting scanpaths in passive gaze-based interaction.
- Standardized practices and addressing emerging ML challenges are needed.
- This review provides a roadmap for future research in scanpath analysis and gaze-based systems.
