Eye movements are an integral and critical part of our normal visual behavior, and evaluating these may offer valuable insights for clinical interventions, diagnostics, and understanding visual perception. With raising interest in studying gaze behavior in freely moving participants and in natural conditions, new methods are required to automatically classify events in data collected under unrestricted conditions, for example using a mobile eye tracker. In a recent paper in the journal “Behavior Research Methods”, Ashkan Nejad and colleagues show how their artificial intelligence-based method “ACE-DNV” can help to determine the type of eye movements made at any one time during such tasks. The method uses video-analysis of the scene footage made by the mobile eyetracker, and works as well as using additional information from depth cameras or inertial measurement units, which are not typically integrated into mobile eye trackers. Want to know more? Find the article here.
Illustration of the of gaze events in natural viewing that the ACE-DNV method can distinguish between.