Multiview Imaging and 3DTV Akira Kubota, Aljoscha Smolic, Marcus Magnor, Masayuki Tanimoto, Tsuhan Chen, Cha Zhang

Akira Kubota, Aljoscha Smolic, Marcus Magnor, Masayuki Tanimoto, Tsuhan Chen, and Cha Zhang: "Multiview Imaging and 3DTV", IEEE Signal Processing Magazine, vol. 24, no. 6, pp. 10–21, November 2007. [pdf] [bib]

Abstract

3DTV/FTV is a state-of-the-art imaging system. In the past, image systems such as photography, film,
and TV were distinct systems. At present, they are being digitized more and more, allowing them to be
handled on the same platform as pixel-based systems. These pixel-based systems are undergoing rapid development toward increasing the number of pixels. Although super high-definition TV has about 100 times the number of pixels of standard-definition TV, still only one view is used. In the future, the demand for more pixels will level off and will be replaced with a demand for more views (i.e., more light rays in 3D space). This is an evolution from pixel-based systems with a single image to ray-based systems with multi-view images. The technologies for light ray capturing and display are making rapid progress and have created a huge opportunity for 3DTV/FTV to enter the consumer mass market in the near future. MVI will open the way to ray-based image engineering that provides breakthrough technologies to treat rays one by one.
In the following sections, this article provides the reader with a brief introduction to the sequence of
fundamental processing steps required for 3DTV and FTV: 3D scene representation, multi-view image
capturing and rendering, coding, and displaying of a 3D scene. We hope that reading through these
sections will aid the reader in understanding the papers in this issue.

TU Braunschweig - Fakultät für Mathematik und Informatik - Computer Graphics - Publications - Multiview Imaging and 3DTV