3D Eye Position Based Interaction within Hierarchically Represented Images

In this paper, we present a methodology on how users might interact with hierarchically represented collections of images using 3D eye position and eye blinks. The movement of a user´s eye which is detected by color and circular shape information can provide a convenient, natural, and high-bandwidth source of Human-Computer Interaction for navigating and traversing up and down a hierarchical structure of an image set. Our proposed hierarchical representation scheme of an image set is composed of multiple layers by calculating the 3D position and orientation with automatic feature detection, matching and summary through feature clustering. Applications of eye tracking and hierarchical image representation can be used for view navigation within the future 3DTV system and offers an additional HCI methodology.