Continuous metadata flows for distributed multimedia
Continuous metadata flows for distributed multimedia
The practical use of temporal multimedia has increased markedly in recent years as enabling technologies for the distribution and streaming of media have become available. As a part of this trend, hypermedia systems and models have adapted accordingly to incorporate such distributed multimedia for presentation. Structured interpretation of information has long been a fundamental feature of both open hypermedia systems and knowledge systems. Metadata, in its many forms, has become the cornerstone for providing this structured knowledge above and beyond basic data and information. This thesis presents the rationale and requirements for continuous metadata, which supports the metadata accompanying distributed multimedia throughout the lifecycle of streamed media, from generation, through distribution, to presentation. Throughout this process it is the temporal and continuous nature of the metadata which is paramount. A conceptual framework for continuous metadata is proposed to encapsulate these principles and ideas. Continuous metadata and the associated framework enable the development, in particular, of real-time, collaborative, semantically enriched distributed multimedia applications. Experience building one such system using continuous metadata is evaluated within the framework. An ontology is developed for the system to enable the collation, distribution, and presentation of structure aiding navigation of multimedia, and it is shown how continuous metadata utilising the ontology can be distributed using multicast
Page, Kevin R.
f9b006ae-e59e-4607-8279-487d80419f59
April 2011
Page, Kevin R.
f9b006ae-e59e-4607-8279-487d80419f59
De Roure, David
02879140-3508-4db9-a7f4-d114421375da
Page, Kevin R.
(2011)
Continuous metadata flows for distributed multimedia.
University of Southampton, School of Electronics and Computer Science, Doctoral Thesis, 164pp.
Record type:
Thesis
(Doctoral)
Abstract
The practical use of temporal multimedia has increased markedly in recent years as enabling technologies for the distribution and streaming of media have become available. As a part of this trend, hypermedia systems and models have adapted accordingly to incorporate such distributed multimedia for presentation. Structured interpretation of information has long been a fundamental feature of both open hypermedia systems and knowledge systems. Metadata, in its many forms, has become the cornerstone for providing this structured knowledge above and beyond basic data and information. This thesis presents the rationale and requirements for continuous metadata, which supports the metadata accompanying distributed multimedia throughout the lifecycle of streamed media, from generation, through distribution, to presentation. Throughout this process it is the temporal and continuous nature of the metadata which is paramount. A conceptual framework for continuous metadata is proposed to encapsulate these principles and ideas. Continuous metadata and the associated framework enable the development, in particular, of real-time, collaborative, semantically enriched distributed multimedia applications. Experience building one such system using continuous metadata is evaluated within the framework. An ontology is developed for the system to enable the collation, distribution, and presentation of structure aiding navigation of multimedia, and it is shown how continuous metadata utilising the ontology can be distributed using multicast
Text
thesis_krp-029-final_submission.pdf
- Other
More information
Published date: April 2011
Organisations:
University of Southampton
Identifiers
Local EPrints ID: 183241
URI: http://eprints.soton.ac.uk/id/eprint/183241
PURE UUID: 7c410847-3c6e-4d34-93cb-314bb55cc38b
Catalogue record
Date deposited: 23 May 2011 12:53
Last modified: 14 Mar 2024 03:02
Export record
Contributors
Author:
Kevin R. Page
Thesis advisor:
David De Roure
Download statistics
Downloads from ePrints over the past year. Other digital versions may also be available to download e.g. from the publisher's website.
View more statistics