Learning saliency-based visual attention: A review
Signal Processing
| Hi-index | 0.00 |
An understanding of gaze control requires knowledge of the basic properties of eye movements during scene viewing. Because most of what we know about eye movement behavior is based on the viewing of images on computer screens, it is important to determine whether viewing in this setting generalizes to the viewing of real-world environments. Our objectives were to characterize eye movement behavior in the real world using head-mounted eyetracking technology and to illustrate the need for and development of automated analytic methods. Eye movements were monitored while participants searched for and counted coffee cups positioned within a cluttered office scene. Saccades were longer than typically observed using static displays, but fixation durations appear to generalize across viewing situations. Participants also made longer saccades to cups when a pictorial example of the target was provided in advance, suggesting a modulation of the perceptual span in accordance with the amount of information provided.