A Multimedia Service with MPEG-7 Metadata and Context Semantics

1,383 views
1,283 views

Published on

Yiwei Cao, Ralf Klamma, and Maziar Khodaei

The 9th Workshop on Multimedia Metadata (WMM‘09)

20.03.2009
Toulouse, France

Published in: Technology
0 Comments
2 Likes
Statistics
Notes
  • Be the first to comment

No Downloads
Views
Total views
1,383
On SlideShare
0
From Embeds
0
Number of Embeds
38
Actions
Shares
0
Downloads
36
Comments
0
Likes
2
Embeds 0
No embeds

No notes for slide
  • A Multimedia Service with MPEG-7 Metadata and Context Semantics

    1. 1. A Multimedia Service with MPEG-7 Metadata and Context Semantics Yiwei Cao, Ralf Klamma, and Maziar Khodaei Informatik 5 (Information Systems), RWTH Aachen University 20.03.2009 Toulouse, France The 9 th Workshop on Multimedia Metadata (WMM‘09)
    2. 2. Agenda <ul><li>Motivation and Scenario </li></ul><ul><li>System Design of CA3M Service (Context-aware Mobile Multimedia Service) </li></ul><ul><li>A MPEG-7ToRDF Converter </li></ul><ul><li>Implementation and Evaluation </li></ul><ul><li>Conclusions and Outlook </li></ul>CA3M Mobility Multimedia Uncertainty
    3. 3. Motivation <ul><li>Context [Dey 1999] </li></ul><ul><ul><li>Interaction between users und applications </li></ul></ul><ul><ul><li>Context awareness: location, time, device, community, activities </li></ul></ul><ul><li>Context-aware Mobile Multimedia Systems </li></ul><ul><ul><li>Presentation of relevant information: right information in the right format to the right community at right time and place </li></ul></ul><ul><ul><li>Multimedia adaptation of 3D models, videos, audios, images, texts </li></ul></ul><ul><li>Data Uncertainty in Location Based Services (LBS) [Henrickson 2004] </li></ul><ul><ul><li>Inconsistencies between models und real world </li></ul></ul><ul><ul><li>Inconsistencies between local environment models </li></ul></ul><ul><ul><li>Representation: unknown; ambiguous; imprecise; erroneous </li></ul></ul>
    4. 4. Car Race Scenario Context-Aware Multimedia Sharing <ul><li>Communities produce and consume media at different places, times, etc. </li></ul><ul><li>Semantic mapping of context data on semantic base types needed </li></ul><ul><ul><li>GPS data -> Tribunes, Curves, etc. (Places) </li></ul></ul><ul><ul><li>Time data -> Lap, Start, Finish, Box stop, etc. (Events) </li></ul></ul><ul><ul><li>Racing car data –> Ferrari, McLaren-Mercedes, BMW-Sauber Community, etc. (Agents) </li></ul></ul><ul><li>Multimedia adaption needed because of different mobile devices </li></ul>
    5. 5. Concepts of CA3M <ul><li>Community-Aware Static Multimedia Adaptation based on Clustering Algorithms [KSCa06] </li></ul><ul><li>Context-Aware Adaptation based on Semantic Einrichment of Multimedia Information [CKHJ08] </li></ul><ul><li>How to support MPEG-7/MPEG-21 Semantic Enrichment for further Semantic Processing? </li></ul><ul><li>How to systematically handle Uncertain Information in further Semantic Processing? </li></ul>Context-Aware Adaptation System (OWL) [CKHJ08, MDM‘08] Media Adaptation System (MPEG-7/21) [KSCa06, ECTEL‘06] MPEG-7ToRDF Context Reasoner [CKHJ08, MDM‘08] Uncertainty Management
    6. 6. Selection of MPEG7ToRDF Converter Technologies <ul><li>MPEG-7 Ontology [García et al., 2006] </li></ul><ul><ul><li>Developed by Roberto Garcia Gonzales within Rhizomik </li></ul></ul><ul><ul><li>Based on MPEG-7 as of 2001 </li></ul></ul><ul><ul><li>In OWL-Full-Format </li></ul></ul><ul><li>Alternatives </li></ul><ul><li>TBD: MPEG-21 Conversion </li></ul>Multimedia Ontologies Format Mapping MPEG-7 Upper DMS [Hunter, 2001] OWL-Full - COMM [Bloehdorn et al., 2005] OWL-DL - aceMedia [ace, 2005] RDFS-DS - MPEG-7 Ontologie (Rhizomik) [ García et al., 20 06] OWL-Full XSD2OWL XML2RDF
    7. 7. The MPEG7ToRDF Converter <ul><li>The MPEG-7 Ontology based on MPEG-7 Schema as of 2004 </li></ul><ul><li>Mapping MPEG-7 XML Schema to MPEG-7 Ontology in OWL </li></ul><ul><li>Mapping MPEG-7 XML Documents to RDF </li></ul>Uncertainty Management Context Reasoner [CKHJ08] <rdf:RDF xmlns:rdf=&quot;http://www.w3.org/1999/02/22-rdf-syntax-ns#&quot; xmlns:mpeg7=&quot;http://manet.informatik.rwth-aachen.de/ ~khodaei/mpeg7-v2.owl#&quot;> <mpeg7:SemanticPlaceType rdf:about=&quot;http://manet.informatik.rwth- aachen.de/~khodaei/place.xml#SemanticPlaceType_ admin20080227090806191&quot;> <mpeg7:Place rdf:parseType=&quot;Resource&quot;> <mpeg7:GeographicPosition rdf:parseType=&quot;Resource&quot;> <mpeg7:Point rdf:parseType=&quot;Resource&quot;> <mpeg7:longitude>7.438253814591</mpeg7:longitude> <mpeg7:latitude>43.774566243600006</mpeg7:latitude> <mpeg7:altitude>29</mpeg7:altitude> </mpeg7:Point> </mpeg7:GeographicPosition> </mpeg7:Place> <mpeg7:Label rdf:parseType=&quot;Resource&quot;> <mpeg7:Definition rdf:parseType=&quot;Resource&quot;>curve-1</mpeg7:Definition> <mpeg7:Name>43:46.47397N, 7:26.29523E, 29m</mpeg7:Name> </mpeg7:Label> </mpeg7:SemanticPlaceType> </rdf:RDF> MPEG-7ToRDF Media Adaptation System (MPEG-7/21) [KSCa06] Context-Aware Adaptation System (OWL) [CKHJ08]
    8. 8. Context Reasoner <ul><li>Reasoning on context information </li></ul><ul><li>SPARQL </li></ul><ul><ul><li>RDF Query Language </li></ul></ul><ul><ul><li>Ontology Reasoning Language </li></ul></ul><ul><ul><li>Implementation using the Jena Framework </li></ul></ul>Uncertainty Management MPEG-7ToRDF Media Adaptation System (MPEG-7/21) [KSCa06] Context-Aware Adaptation System (OWL) [CKHJ08] PREFIX mpeg7: < http://manet.informatik.rwth-aachen.de/~khodaei/mpeg7-v2.owl#> PREFIX xsd: <http://www.w3.org/2001/XMLSchema#> SELECT ?agent ?place WHERE { ?agent mpeg7:name “FirstName LastName&quot; ?agent mpeg7:hasGeographicPosition ?loc1 ?loc1 mpeg7:longitude ?lon1 ?loc1 mpeg7:latitude ?lat1 ?place mpeg7:hasGeographicPosition ?loc2 ?place mpeg7:definition “curve-n” ?loc2 mpeg7:longitude ?lon2 ?loc2 mpeg7:latitude ?lat2 FILTER Distance(?loc1, ?loc2) <= distance } Context Reasoner [CKHJ08]
    9. 9. Uncertainty Management <ul><li>Clustering Algorithms based on User Preferences </li></ul><ul><ul><li>Calculation Euclidean Distance of Communities </li></ul></ul><ul><ul><li>Clusters are recalculated when new users come </li></ul></ul><ul><li>Rating/Ranking as User Query Feedback </li></ul>Context Reasoner [CKHJ08] MPEG-7ToRDF Media Adaptation System (MPEG-7/21) [KSCa06] Context-Aware Adaptation System (OWL) [CKHJ08] Uncertainty Management
    10. 10. Evaluation <ul><ul><li>Evaluating within the DFG UMIC Excellence Cluster 2006-2011 </li></ul></ul><ul><ul><ul><li>Concepts and demonstrators for smart, mobile, broadband, low-cost systems will be developed which support the demanding applications of the next-decade mobile Internet </li></ul></ul></ul><ul><ul><ul><li>Mobile Multimedia Information for </li></ul></ul></ul><ul><ul><ul><ul><li>Cultural Heritage Management (Virtual Campfire) </li></ul></ul></ul></ul><ul><ul><ul><ul><li>Virtual City Guides </li></ul></ul></ul></ul><ul><ul><ul><ul><li>Immersive Games </li></ul></ul></ul></ul><ul><ul><ul><li>Semantic Enrichment of Metadata (MPEG-7, MPEG-21, …) for Location-Based Services </li></ul></ul></ul><ul><ul><ul><li>Systematic Evaluation of Services with MoBSOS [RKSp08] </li></ul></ul></ul><ul><li>Process </li></ul><ul><ul><ul><li>Converting MPEG-7 / MPEG-21 Files into OWL/RDF </li></ul></ul></ul><ul><ul><ul><li>SPARQL Queries: Context Search on Semantic Base Types </li></ul></ul></ul>
    11. 11. Conclusions <ul><li>CA3M - A mobile context-aware multimedia service </li></ul><ul><ul><li>Integration into existing LAS architecture as a web service </li></ul></ul><ul><ul><li>Enhancement of information interoperability </li></ul></ul><ul><ul><li>Evaluation within UMIC Cluster of Excellence </li></ul></ul><ul><li>Technical Improvements </li></ul><ul><ul><li>Generating MPEG-7 ontology based on the 2004 MPEG-7 Schema </li></ul></ul><ul><ul><li>Realizing a converter mapping XML document to OWL/RDF </li></ul></ul><ul><ul><li>Integrated reasoning on multimedia and context information </li></ul></ul><ul><ul><li>Semantic query processing for multimedia information </li></ul></ul><ul><ul><li>Systematic handling of uncertainty for location based services (rating/ranking, clustering) </li></ul></ul>
    12. 12. Outlook <ul><li>Data Management </li></ul><ul><ul><li>Improvement of SPARQL queries for large amount of data </li></ul></ul><ul><ul><li>Performing SPARQL queries on relational databases </li></ul></ul><ul><ul><li>Field tests with mobile communities </li></ul></ul><ul><li>Uncertainty Management </li></ul><ul><ul><li>Improvement of context reasoning </li></ul></ul><ul><ul><li>Handling uncertainty with probabilistic methods </li></ul></ul><ul><ul><ul><li>Fuzzy logic </li></ul></ul></ul><ul><ul><ul><li>Bayesian network </li></ul></ul></ul><ul><li>Applications & User-Generated Content </li></ul><ul><ul><li>Mobile Storytelling </li></ul></ul><ul><ul><li>Cultural Heritage Management </li></ul></ul><ul><ul><li>Services on various mobile devices e.g. iPhone, Google G1 </li></ul></ul>

    ×