Highly realistic video communication, which provide feel of existence of remote users and objects, is expected to be a next generation communication using broadband connection over the internet. To achieve such system, it is important to reproduce not only high quality image but also sense of directivity depending on user's position. However, viewpoint image synthesis is usually not easy for live action. In this paper, an approach for viewpoint image synthesis using multi-camera images and approximate depth information will be demonstrated. It can provide smooth motion parallax with high speed processing and high image quality.
We are conducting research on super multi-view 3D video systems for highly realistic communications. We developed a glasses-free auto-multiscopic 3D display, and a 3D capturing system to capture live real moving objects. We have started several field experiments in a public space with this system. In this study, we report on the basic analysis of the capability of an auto-multiscopic display as a digital museum media, in the public space.
The viewing zone of the conventional 3D display that uses a lenticular lens determined by the specifications of the lens, and it was difficult to expand because the lens with wider viewing zone has larger aberration. In this study, a method of enlarging the viewing zone by placing the frame on the real image region and narrowing the display width was tried. According to the configuration placing the mirrors on the sides, the viewing zone can be expanded to about b times by making the display width to 1/b. Where b is an integer of 2 or more, we experimented for the case of b=2, 3.
In the previous paper, it has been proposed an optical system to display the 3D animation consisting of a large number of frames by the printed image, but the apparatus was required a certain amount of depth for the backlight. In this study, a method to achieve similar function without backlight by using a concave/convex lens and a lenticular lens was tried. According to this method, significantly thinner apparatus can be realized.