3D space coding using virtual object surface

In order to describe and transmit the entire visual data of a 3-D space, the authors have proposed the concept of the “3-D Space Coding.” We applied a DCT-based method to the data space in which light ray data was recorded. Our method could compress light ray data more effectively than the video coding method like MPEG2. This paper discusses the “3-D Space Coding” using the structural properties of scene objects, and proposes “Virtual Object Surface” which make the compression rate better. Experimental results show the “Virtual Object Surface” can improve the compression rate.