Towards Fixation Extraction in Corneal Imaging Based Eye Tracking Data

Christian Lander, Marco Speicher, Frederic Kerber, Antonio Krüger

In: Extended Abstracts of the 2018 CHI Conference on Human Factors in Computing Systems. ACM International Conference on Human Factors in Computing Systems (CHI-2018) April 21-26 Montreal Quebec Canada ACM 2018.


Humans sense most of the environment through their eyes. Hence, gaze is a powerful way to estimate visual attention. Head-mounted or mobile eye tracking is an established tool to analyze the visual behavior of people. Since these systems usually require some kind of calibration prior to usage, a new generation of mobile eye tracking devices based on corneal imaging has been investigated. However, little attention has been given on how to analyze corneal imaging specific eye tracking data. A classic approach in state-of-the-art systems is to extract different eye movements (e.g., fixations, saccades and pursuits movements). So far, there is no approach for applying these methods to corneal imaging data. We present a proof-of-concept method for fixation extraction and clustering of corneal imaging data. With this method we can compress the eye tracking data and make it ready for further analysis (e.g., attention measurement and object detection).

Deutsches Forschungszentrum für Künstliche Intelligenz
German Research Center for Artificial Intelligence