DANAE Project
DANAE = Dynamic
and distributed Adaptation of scalable multimedia coNtent in
a context Aware Environment)
IST-FP6-1-507113 - STREP Project
Research Overview:
DANAE proposes to address the dynamic and
distributed adaptation of scalable multimedia content in
a context-aware environment. Its objectives are to specify,
develop, integrate and validate in a testbed a complete
framework able to provide end-to-end quality of (multimedia)
service at a minimal cost to the end-user. An application
will be specifically developed and implemented on a demonstrator,
to illustrate the new service concepts pioneered by the
Project.
Our contributions are:
- Definition of means for semantic adaptation, i.e., how should meaningful
user context, like user interest in certain thematics, age, mood contribute
to video, audio and image adaptation
- Realization of an adaptation engine to the semantic context for making
a full featured distributed multimedia adaptation, integration
in the overall end-to-end system
The research project started in 01/2004. First major results are expected
for end of 2004 (see Abstract below). Running time: July 2006.
Team Member: Michael Zufferey (PhD-student).
Official DANAE Homepage
Abstract:
The increasing diversity of devices and heterogeneity of networks pose nowadays
a challenge in the delivery and consumption of multimedia content. In this
context, the Part 7 of
the MPEG-21 standard formally named Digital Item Adaptation (DIA) targets
the adaptation of multimedia content based on usage environment, such as network
characteristics, terminal capabilities and user characteristics. But, MPEG-21
DIA does not take into account MPEG-7 semantic description tools, which provide
means for a conceptual (semantic) description that is close to the human
understanding of multimedia content. Therefore, to fill this gap, we propose
an interactive and user-centric framework called Semantic Adaptation Framework
(SAF). The SAF provides facilities for the generation of all the required
semantic metadata and enables an MPEG-21 adaptation engine to semantically
adapt the multimedia content in order to provide the user with the best possible
experience.
The SAF specifies the following semantic metadata:
• MPEG-7 compliant ontologies for content annotation and
context representation.
• MPEG-7 semantic descriptions of the resource.
• Semantic generic bitstream syntax descriptions (semantic
gBSD) of the resource.
• Steering descriptions, which define all the possible
semantic adaptation for the resource.
• Semantic user preferences.
• Content Digital Items, which provide a flexible link
between the semantic gBSD and the steering description.
The modular architecture of the SAF (see below) allows an easy extensibility
and scalability of both software modules and semantic metadata.
The SAF consists of the following modules:
• SemanticGenerator: Provides means for the semantic annotation
of the resource and generation of the MPEG-7 semantic description. It generates
the semantic gBSD and the steering description.
• Content Digital Item Tool (CDITool): Builds a content
Digital Item (contentDI) by aggregating the semantic indexed gBSD, the steering
description and other content retrieved metadata.
• Semantic User Preferences Tool: Allows a user to select
her/his preferred topics, content structure and semantic entities with their
relations from the available ontologies.
Figure : Architecture of the SAF.
Maintained by : harald.kosch@itec.uni-klu.ac.at
Last updated 06/12/2004.