Abstract

Head mounted cameras are well known and widely used in the fields of documentary movies and augmented reality. When combined with an eye tracker, these cameras can emphasize the point of gaze within a picture of a whole scene. A gaze-driven, head-fixed camera was described, with witch it is possible to capture pictures viewed by the eyes. The goal of this paper is to combine a wide-angle picture made by a head-mounted scene camera with a zoomed picture made by a gaze driven camera. It is not necessary to insert the gaze picture into a "clean" mono perspective. Hybrid perspective motivates to overlay these images while preserving their inherent perspective and to add several contrasts like size, resolution, color and brightness. Bi-foveated movies have been made with a wearable computer and presented to ten subjects observed by an eye tracking system. Throughout the whole movie 53% the viewer's gazes pointed directly at the gaze spot within the scene, 22% were in the spot vicinity, and 25% ignored the spot. These numbers varied for four selected scenes: They ranged from only 12% ignoring the spot in a low velocity, less complex scene up to 63% in a high speed, highly complex scene. The wearable system made it possible to shoot movies without external help and a guideline could be developed as of what kind of scenes are adequate for a hybrid compositing.

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call