Gaze tracking: A sclera recognition approach.

摘要

Gaze tracking is the complex process of detecting where a person is looking. The main objective of this thesis is to provide an alternate approach that improves upon current techniques used for gaze determination. With the intention of unaided feature extraction, only a basic video camera is used to extract pertinent facial features. These features are found using a new technique that was developed in order to search for the sclera regions based on their characteristic color. The sclera, or the "white" of the eye, was found to be consistent across people, and was chosen because it can be used to locate the eye region within the face as well as to define a new feature that is used for gaze calculation. By using the sclera, some of the problems facing current gaze tracking techniques have been overcome.