A wide-view VR system with multi-level description and rendering of 3D scenes
Yoshiaki Kuno, Tohru Yagi, Y. Uchikawa · 1996
Abetract- It ie known that a wide-view of an image increaeee a true eenee of reality in VR. The amount of the dieplaying data, however, become8 enormous, and dieplaying epeed will become much slower. In order to eolve theee problems, we have focueed on the characterietic of human vieion, of which the reeolution falle from the center to the periphery. If we emulate such a VR eyetem ueing non-uniform 3D ecene deecription and rendering, it will be necessary to detect eubject'e eye movemente in order to dieplay the high reeolution image always to hielher central viaion. In thia etudy, we propoee eye tracking eyetem ueing EOG (Electro-OculoGram) for that purpoee, and explain how it worke in an actuator control eyetem. Then, we ehow a rudimental result of EEG (Electro-EncephaloGram) during eye movements, which would be a ueeful tool to predict eye movemente. peripheral vision. If the high resolution image I. INTRODUCTION is displayed at the central area only, the data size will get smaller. Moreover, the displaying Some statistics show that our humans gain speed will be increased without decreasing a