Smooth motion parallax and high resolution display based on visually equivalent light field 3D

Abstract

The light field concept can correctly and completely describe the distribution of rays in 3D space within theory of geometrical optics. However, the data quantity is huge, and not easy to capture or process. Though light field 3D displays are almost ideal in principle, they are not really practical given the huge number of pixels required. To compress data quantity, we proposed the visually equivalent light field (VELF), which uses a characteristic of human vision. Though several cameras are needed, VELF can be captured by a camera array. Reconstructing the ray distribution involves linear blending, but this process is so simple that we can realize this calculation optically in the VELF3D display. It produces high image quality as its high pixel usage efficiency overcomes the tradeoff between resolution and directional density of rays. In this paper, we summarize the relationship between the characteristics of human vision and VELF. We introduce the VELF3D display that consists of a horizontal RGB stripe LCD panel and a parallax barrier, whose spacing width is almost the same as pixel pitch. Though it is similar to the conventional parallax barrier type autostereoscopic 3D display, it can reproduce correct rays for human vision. High feel of existence is induced by the display’s smooth and exact motion parallax; its resolution is high enough to display characters. Head tracking allows the viewing zone to be greatly expanded while maintaining smooth motion parallax. Since image capture and display are very simple, VELF is suitable for realtime live action applications with high image quality.