Information Technology Reference
In-Depth Information
Chapter 15
Three-Dimensional Video Contents Exploitation
in Depth Camera-Based Hybrid Camera System
Sung-Yeol Kim, Andreas Koschan, Mongi A. Abidi, and Yo-Sung Ho
*
Abstract.
Video-plus-depth
is an image sequence of synchronized color and depth
images. As importance of video-plus-depth increases as an essential part of the
next-generation multimedia applications, it is crucial to estimate accurate depth
information from a real scene and to find a practical framework to use the immer-
sive video in industry. In this chapter, we introduce a hybrid camera system
composed of a stereoscopic camera and a time-of-flight depth camera to generate
high-quality and high-resolution video-plus-depth. We also handle a hierarchical
decomposition method of depth images to render a dynamic 3D scene represented
by video-plus-depth rapidly. Finally, we present a method to generate streamable
3D video contents based on video-plus-depth and computer graphic models in the
MPEG-4 multimedia framework. The MPEG-4-based 3D video contents can sup-
port a variety of user-friendly interactions, such as free viewpoint changing and
free composition with computer graphic images.
1 Introduction
As immersive multimedia services are expected to be available in the near future
through a high-speed optical network, a three-dimensional (3D) video is recog-
nized as an essential part of the next-generation multimedia applications. As one
of 3D video representations, it is widely accepted that an image sequence of syn-
chronized color and depth images, which is often called as
video-plus-depth
[1],
provides the groundwork for the envisaging 3D applications. For a practical use of
the immersive video in future interactive 3D applications, such a 3D TV [2], it is
very important to estimate accurate depth information from a real scene.
In order to obtain reliable depth data, a variety of depth estimation algorithms
have been presented in the fields of computer vision and image processing [3].