A Survey of Event Detection Methods for Eye Movements Classification in Smooth-Pursuit-Based Interactive Applications

摘要

Gaze-based interaction is an important research area in the field of human-computer interaction. Gaze-based input gives benefit for motion-impaired people and provides faster and intuitive interaction with an application content. Developing gaze-based interactive applications has to deal with several challenges: data processing, eye movements classification, and action. Smooth pursuit eye movement has been used as one of primary input modalities for gaze-based interactive applications. In developing smooth-pursuit-based interactive applications, challenge in the event classification is important to be addressed. However, there is no survey paper that studies various event detection methods in smooth-pursuit-based interactive applications. To tackle this research gap, we present a brief literature review of some event detection methods that consider smooth pursuit eye movement. This paper also presents a preliminary comparison of event detection methods based on some properties that are needed in smooth-pursuit-based interactive applications. From the preliminary comparison, there are some considerations to be taken in to account in order to develop smooth-pursuit-based interactive applications.