|
As digital broadcasting technologies have been rapidly progressed, users' expectations for realistic and interactive
broadcasting services also have been increased. As one of such services, 3D multi-view broadcasting has received much
attention recently. In general, all the view sequences acquired at the server are transmitted to the client. Then, the user
can select a part of views or all the views according to display capabilities. However, this kind of system requires high
processing power of the server as well as the client, thus posing a difficulty in practical applications. To overcome this
problem, a relatively simple method is to transmit only two view-sequences requested by the client in order to deliver a
stereoscopic video. In this system, effective communication between the server and the client is one of important
aspects.
In this paper, we propose an efficient multi-view system that transmits two view-sequences and their depth maps
according to user's request. The view selection process is integrated into MPEG-21 DIA (Digital Item Adaptation) so
that our system is compatible to MPEG-21 multimedia framework. DIA is generally composed of resource adaptation
and descriptor adaptation. It is one of merits that SVA (stereoscopic video adaptation) descriptors defined in DIA
standard are used to deliver users' preferences and device capabilities. Furthermore, multi-view descriptions related to
multi-view camera and system are newly introduced. The syntax of the descriptions and their elements is represented in
XML (eXtensible Markup Language) schema. If the client requests an adapted descriptor (e.g., view numbers) to the
server, then the server sends its associated view sequences. Finally, we present a method which can reduce user's visual
discomfort that might occur while viewing stereoscopic video. This phenomenon happens when view changes as well as
when a stereoscopic image produces excessive disparity caused by a large baseline between two cameras. To solve for
the former, IVR (intermediate view reconstruction) is employed for smooth transition between two stereoscopic view
sequences. As well, a disparity adjustment scheme is used for the latter.
Finally, from the implementation of testbed and the experiments, we can show the valuables and possibilities of our
system.
|
