SlideShare a Scribd company logo
1 of 44
Semantics for visual resources
Use Cases from E-Culture
Guus Schreiber
Free University Amsterdam
schreiber@cs.vu.nl
2
Purpose
 Analyze a number of use cases from e-culture
domain
– Multimedia plays key role
 Required technology
– Typically combination of technologies
 Relation to state of the art
Acknowledgements: This presentations contains
slides and images provided by Laura Hollink,
Giang Nguyen and Cees Snoek. Also thanks to
the MultimediaN E-Culture team
3
Use case: Asian chairs
User has found an image of an Asian chair
Annotation:
ex:image vra:stylePeriod aat:Guangxu .
How can we find images of Asian chairs from
the same historical period?
4
AAT info on Guangxu
5
Importance of time and space
information
 Many queries require time/space
knowledge, either absolute or abstracted
 For the chair image we can establish
– Country = China (link Chinese => China)
– Period = 1644-1911 (from Qing description)
 Technology requirements:
– Thesuari relating time/space concepts
– NLP for unstructured descriptions
– Time/space reasoning techniques
6
7
8
Sample place information in TGN
<tgn:AdministrativePlace rdf:about="&tgn;1000111"
tgn:standardLatitude="35"
tgn:standardLongitude="105“>
<vp:parentPreferred rdf:resource="&tgn;1000004"/>
……..
</tgn:AdministrativePlace>
9
Issues when searching for
“nearby” Asian chairs
 Close in space:
– Other country in (East) Asia
– Latitude/longitude
 Close in time:
– Links between style periods
– Match time periods (and
handle incomplete
information)
10
11
Use case: painting style
Find paintings of a similar style
MATISSE, Henri
Le bonheur de vivre (The Joy of Life)
1905-1906
Oil on canvas, 69 1/8 x 94 7/8 in. (175 x 241 cm)
Barnes Foundation, Merion, PA
12
How can we find this other Fauve
painting?
DERAIN, Andre
The Turning Road, L'Estaque, 1906
Oil on canvas, 51 x 76 3/4 in. (129.5 x 195 cm)
Museum of Fine Arts, Houston, Texas
13
Issues
 Parse annotation to find matches with thesauri
terms
– E.g. match artists to ULAN individuals
 Artists-style links
– AAT contains styles; ULAN contains artists, but there
is no link
• Learn link from corpora
• Derive it from other annotations
– Domain-specific rules/reasoning needed
• see example in SWRL doc
• Painters may have painted in multiple styles
14
15
16
Search: WordNet patterns that increase recall
without sacrificing precision (Hollink)
17
Issues w.r.t. thesauri
 Public availability!
 RDF/OWL representation
 Learning/specifying term/concept mapping
– owl:equivalentClass, owl:sameAs,
rdf:type, rdfs:subClassOf
– Domain-specific links
 Managing the evolution of the thesauri and
the mappings
18
Use case: find images with the
same subject
Find another painting which portrays dancing
19
Issues
 Same subjects can be visually very
different
 Subject is often missing from the
annotation
 Mismatch: users often search for subjects
of images
20
Conceptual subject descriptions
85% of the user queries:
General Descriptions of generally known items. Only general,
everyday knowledge is necessary. Descriptions are at the
level of the Natural categories of E. Rosch (1973), or more
general. E.g An ape eating a banana.
Specific Descriptions of objects or scenes that can be identified
and named. Specific domain knowledge is necessary to
recognize the objects or scenes. E.g. The old male gorilla
Kumba, born in Cameroon and now living in Artis, Amsterdam
Abstract Descriptions for which interpretative knowledge is
used. This category is subjective. E.g An animal threatened
with extinction.
21
Example concepts in image
 Specific
– Fall of the Berlin Wall
 General
– People walking at night
 Abstract
– Fall of the Iron Curtain
22
Use of conceptual categories by
people searching for images
Conceptual level: 83%
0%
20%
40%
60%
80%
100%
event time place relation scene object
Characteristics
Nuberofelementsin%of
conceptualelements
Abstract
Specific
General
23
Thesauri for scenes: Iconclass
24
25
26
Annotation of image content
 Template for subject description
Agent Action Object Recipient
 Guidelines for manual annotation
– Annotate as specific as possible
 Default reasoning
 CBIR support:
– Object identification
– Spatial relations
27
28
29
Some forms of image content are
well suited to image analysis
Collection of clothes
Abstract painting
30
The semantic gap
 The distance between Content-Based Image
Retrieval and semantics:
– Smeulders, Worring, Santini, Gupta, Jain. Content-
based image retrieval at the end of the early years.
IEEE Transactions on Pattern Analysis and Machine
Intelligence, 22(12), December 2000.
 Direct links between visual features and
semantic concepts become more difficult when
the domain is broader / more general
31
Example semantic bridge:
microscopic cell images
mpeg7 : StillRegion(region) ^
mpeg7x : Dense(region) ^
mpeg7 : DominantColor(region, col) ^
swrlb : lessThan(col, 100)
=> mpeg7 : Depicts(region, mesh : MatureGranule)
32
Segmentation often requires
user interaction
33
Automatic detection of concepts can be
difficult even in “easy” cases
What is the color
of this ape?
34
Image analysis useful for
collection navigation
35
Bridging the semantic gap:
CBIR and ontologies
Visual WordNet (GE paper)
– Adding knowledge about visual characteristics
to WordNet: mobility, color, …
– Build detectors for the visual features
– Use visual data to prune the tree of categories
when analyzing a visual object
36
Sample visual features and their
mapping to WordNet
37
Experiment: pruning the search
for “conveyance” concepts
6 concepts found
Including taxi cab
12 concepts found
Including passenger train
and commuter train
Three visual features: material, motion, environment
Assumption is that these work perfectly
38
Bridging the semantic gap:
concept detectors
 Snoek et al., TRECVID2004
– 185 hours of news video
 32 detectors for concepts in news video
– Through machine learning
 Similarity detectors based on keywords
and visual analysis
 Query interface in which these functions
can be combined
39
“Concepts” for which visual
detectors were built
40
LSCOM lexicon: 229 - Weather
 Context-specific (i.e.
news broadcast)
interpretation:
“Weather forecast”
41
LSCOM lexicon: 110 – Female Anchor
 Composite concept
 Alignment needed for
semantic search, e.g.
with WordNet
42
Natural-lang proc.
automatic annotation
text stings → concepts
Distributed
cultuurwijzer.nl collections
OAI-based access
Reasoning support
time/space reasoning
Web interface
support for web collections
Presentation facilities
semantic presentation
device-specific
Interoperability
XML/RDF/OWL
Scalability
> 10,000,000 triples
Ontologies
WordNet, AAT, TGN
ULAN, Dutch labels
Search strategies
sibling search
semantic distance
Dublin Core
specializations
dumb-down
semantic annotation
DIGITAL HERITAGE
COLLECTIONS
semantic search
BASELINEENHANCEDENHANCED
FEATURESFEATURES
NEWNEW
FEATURESFEATURES
43
44
Main observation
A combination of many different techniques
is needed to be able to cope with the
complexity of multimedia semantics
– NLP, segmentation, CBIR, visual feature
detectors, visual ontologies, publicly available
thesauri, thesauri mappings, dedicated
reasoning techniques (time, space, default),
personalization, presentation generation
 Key role for user studies

More Related Content

What's hot

NoTube: integrating TV and Web with the help of semantics
NoTube: integrating TV and Web with the help of semanticsNoTube: integrating TV and Web with the help of semantics
NoTube: integrating TV and Web with the help of semanticsGuus Schreiber
 
Fri schreiber key_knowledge engineering
Fri schreiber key_knowledge engineeringFri schreiber key_knowledge engineering
Fri schreiber key_knowledge engineeringeswcsummerschool
 
Mdst3703 2013-10-01-hypertext-and-history
Mdst3703 2013-10-01-hypertext-and-historyMdst3703 2013-10-01-hypertext-and-history
Mdst3703 2013-10-01-hypertext-and-historyRafael Alvarado
 
Mdst3703 2013-10-08-thematic-research-collections
Mdst3703 2013-10-08-thematic-research-collectionsMdst3703 2013-10-08-thematic-research-collections
Mdst3703 2013-10-08-thematic-research-collectionsRafael Alvarado
 
UVA MDST 3703 Thematic Research Collections 2012-09-18
UVA MDST 3703 Thematic Research Collections 2012-09-18UVA MDST 3703 Thematic Research Collections 2012-09-18
UVA MDST 3703 Thematic Research Collections 2012-09-18Rafael Alvarado
 
Agora User Committee Meeting 2013
Agora User Committee Meeting 2013Agora User Committee Meeting 2013
Agora User Committee Meeting 2013Lora Aroyo
 
Bloggen dhd (von Laurent Romary)
Bloggen dhd  (von Laurent Romary)Bloggen dhd  (von Laurent Romary)
Bloggen dhd (von Laurent Romary)MaxWeberStiftung
 
CHIP Project: Personalized Museum Tour with Real-Time Adaptation on a Mobile ...
CHIP Project: Personalized Museum Tour with Real-Time Adaptation on a Mobile ...CHIP Project: Personalized Museum Tour with Real-Time Adaptation on a Mobile ...
CHIP Project: Personalized Museum Tour with Real-Time Adaptation on a Mobile ...Lora Aroyo
 
Digital Libraries, Digital Archives, Digital Humanities, Digital Scholarship:...
Digital Libraries, Digital Archives, Digital Humanities, Digital Scholarship:...Digital Libraries, Digital Archives, Digital Humanities, Digital Scholarship:...
Digital Libraries, Digital Archives, Digital Humanities, Digital Scholarship:...Jenn Riley
 
Digital Humanities: An Introduction
Digital Humanities: An IntroductionDigital Humanities: An Introduction
Digital Humanities: An IntroductionDilip Barad
 
Creating and Processing Digital Humanities Data
Creating and Processing Digital Humanities DataCreating and Processing Digital Humanities Data
Creating and Processing Digital Humanities DataAngela Zoss
 
MA in Digital Humanities
MA in Digital Humanities MA in Digital Humanities
MA in Digital Humanities Paul Spence
 
Digital Humanities and “Digital” Social Sciences
Digital Humanities and “Digital” Social SciencesDigital Humanities and “Digital” Social Sciences
Digital Humanities and “Digital” Social SciencesChantal van Son
 
Workset Creation for Scholarly Analysis Project presentation at CNI 2013
Workset Creation for Scholarly Analysis Project presentation at CNI 2013Workset Creation for Scholarly Analysis Project presentation at CNI 2013
Workset Creation for Scholarly Analysis Project presentation at CNI 2013Harriett Green
 
Balboa Park Commons: Collaborative Digitization for a Public Resource
Balboa Park Commons: Collaborative Digitization for a Public ResourceBalboa Park Commons: Collaborative Digitization for a Public Resource
Balboa Park Commons: Collaborative Digitization for a Public ResourceAnna Chiaretta Lavatelli
 
Building a Collaboration for Digital Publishing
Building a Collaboration for Digital PublishingBuilding a Collaboration for Digital Publishing
Building a Collaboration for Digital PublishingHarriett Green
 
Columbia.lippincott.2012
Columbia.lippincott.2012Columbia.lippincott.2012
Columbia.lippincott.2012JoanLippincott
 

What's hot (20)

NoTube: integrating TV and Web with the help of semantics
NoTube: integrating TV and Web with the help of semanticsNoTube: integrating TV and Web with the help of semantics
NoTube: integrating TV and Web with the help of semantics
 
Fri schreiber key_knowledge engineering
Fri schreiber key_knowledge engineeringFri schreiber key_knowledge engineering
Fri schreiber key_knowledge engineering
 
Mdst3703 2013-10-01-hypertext-and-history
Mdst3703 2013-10-01-hypertext-and-historyMdst3703 2013-10-01-hypertext-and-history
Mdst3703 2013-10-01-hypertext-and-history
 
Mdst3703 2013-10-08-thematic-research-collections
Mdst3703 2013-10-08-thematic-research-collectionsMdst3703 2013-10-08-thematic-research-collections
Mdst3703 2013-10-08-thematic-research-collections
 
UVA MDST 3703 Thematic Research Collections 2012-09-18
UVA MDST 3703 Thematic Research Collections 2012-09-18UVA MDST 3703 Thematic Research Collections 2012-09-18
UVA MDST 3703 Thematic Research Collections 2012-09-18
 
Agora User Committee Meeting 2013
Agora User Committee Meeting 2013Agora User Committee Meeting 2013
Agora User Committee Meeting 2013
 
20080606 VöGler GöTtingen E Humanities
20080606 VöGler GöTtingen E Humanities20080606 VöGler GöTtingen E Humanities
20080606 VöGler GöTtingen E Humanities
 
Bloggen dhd (von Laurent Romary)
Bloggen dhd  (von Laurent Romary)Bloggen dhd  (von Laurent Romary)
Bloggen dhd (von Laurent Romary)
 
CHIP Project: Personalized Museum Tour with Real-Time Adaptation on a Mobile ...
CHIP Project: Personalized Museum Tour with Real-Time Adaptation on a Mobile ...CHIP Project: Personalized Museum Tour with Real-Time Adaptation on a Mobile ...
CHIP Project: Personalized Museum Tour with Real-Time Adaptation on a Mobile ...
 
Digital Libraries, Digital Archives, Digital Humanities, Digital Scholarship:...
Digital Libraries, Digital Archives, Digital Humanities, Digital Scholarship:...Digital Libraries, Digital Archives, Digital Humanities, Digital Scholarship:...
Digital Libraries, Digital Archives, Digital Humanities, Digital Scholarship:...
 
Digital Humanities: An Introduction
Digital Humanities: An IntroductionDigital Humanities: An Introduction
Digital Humanities: An Introduction
 
Creating and Processing Digital Humanities Data
Creating and Processing Digital Humanities DataCreating and Processing Digital Humanities Data
Creating and Processing Digital Humanities Data
 
MA in Digital Humanities
MA in Digital Humanities MA in Digital Humanities
MA in Digital Humanities
 
Digital Humanities and “Digital” Social Sciences
Digital Humanities and “Digital” Social SciencesDigital Humanities and “Digital” Social Sciences
Digital Humanities and “Digital” Social Sciences
 
Digital Humanities
Digital HumanitiesDigital Humanities
Digital Humanities
 
Granada0611 digital humanities
Granada0611 digital humanitiesGranada0611 digital humanities
Granada0611 digital humanities
 
Workset Creation for Scholarly Analysis Project presentation at CNI 2013
Workset Creation for Scholarly Analysis Project presentation at CNI 2013Workset Creation for Scholarly Analysis Project presentation at CNI 2013
Workset Creation for Scholarly Analysis Project presentation at CNI 2013
 
Balboa Park Commons: Collaborative Digitization for a Public Resource
Balboa Park Commons: Collaborative Digitization for a Public ResourceBalboa Park Commons: Collaborative Digitization for a Public Resource
Balboa Park Commons: Collaborative Digitization for a Public Resource
 
Building a Collaboration for Digital Publishing
Building a Collaboration for Digital PublishingBuilding a Collaboration for Digital Publishing
Building a Collaboration for Digital Publishing
 
Columbia.lippincott.2012
Columbia.lippincott.2012Columbia.lippincott.2012
Columbia.lippincott.2012
 

Similar to Semantics for visual resources: use cases from e-culture

E-Culture semantic search pilot
E-Culture semantic search pilotE-Culture semantic search pilot
E-Culture semantic search pilotGuus Schreiber
 
A socio-cultural ontology for urban development
A socio-cultural ontology for urban development A socio-cultural ontology for urban development
A socio-cultural ontology for urban development Stefan Trausan-Matu
 
Art and Architecture Thesaurus
Art and Architecture ThesaurusArt and Architecture Thesaurus
Art and Architecture Thesaurushuberannaj
 
Conceptual Organization And Retrieval Of Text By Historians
Conceptual Organization And Retrieval Of Text By HistoriansConceptual Organization And Retrieval Of Text By Historians
Conceptual Organization And Retrieval Of Text By Historiansjgerber
 
Improving Image Discovery for Art Scholars
Improving Image Discovery for Art ScholarsImproving Image Discovery for Art Scholars
Improving Image Discovery for Art ScholarsElizabeth Meyers
 
Between  information  retrieval  services  and bibliometrics  research. New  ...
Between  information  retrieval  services  and bibliometrics  research. New  ...Between  information  retrieval  services  and bibliometrics  research. New  ...
Between  information  retrieval  services  and bibliometrics  research. New  ...Andrea Scharnhorst
 
A Formal Modeling Proposal
A Formal Modeling ProposalA Formal Modeling Proposal
A Formal Modeling ProposalDarian Pruitt
 
Cross domain knowledge discovery, complex system theory and semantic web
Cross domain knowledge discovery, complex system theory and semantic webCross domain knowledge discovery, complex system theory and semantic web
Cross domain knowledge discovery, complex system theory and semantic webAndrea Scharnhorst
 
InfoVis 2010 Lecture 1
InfoVis 2010 Lecture 1InfoVis 2010 Lecture 1
InfoVis 2010 Lecture 1sankazim
 
How and why study big cultural data v2
How and why study big cultural data v2How and why study big cultural data v2
How and why study big cultural data v2Lev Manovich
 
Ontologies and the humanities: some issues affecting the design of digital in...
Ontologies and the humanities: some issues affecting the design of digital in...Ontologies and the humanities: some issues affecting the design of digital in...
Ontologies and the humanities: some issues affecting the design of digital in...Toby Burrows
 
Semantic Web and Linked Data for cultural heritage materials - Approaches in ...
Semantic Web and Linked Data for cultural heritage materials - Approaches in ...Semantic Web and Linked Data for cultural heritage materials - Approaches in ...
Semantic Web and Linked Data for cultural heritage materials - Approaches in ...Antoine Isaac
 
Extending the knowledge level of cognitive architectures with Conceptual Spac...
Extending the knowledge level of cognitive architectures with Conceptual Spac...Extending the knowledge level of cognitive architectures with Conceptual Spac...
Extending the knowledge level of cognitive architectures with Conceptual Spac...Antonio Lieto
 
Multimodal Searching and Semantic Spaces: ...or how to find images of Dalmati...
Multimodal Searching and Semantic Spaces: ...or how to find images of Dalmati...Multimodal Searching and Semantic Spaces: ...or how to find images of Dalmati...
Multimodal Searching and Semantic Spaces: ...or how to find images of Dalmati...Jonathon Hare
 
chinchor_nvac_may06
chinchor_nvac_may06chinchor_nvac_may06
chinchor_nvac_may06webuploader
 
Teaching & Learning with Technology TLT 2016
Teaching & Learning with Technology TLT 2016Teaching & Learning with Technology TLT 2016
Teaching & Learning with Technology TLT 2016Roy Clariana
 
Martha Kellogg Smith
Martha Kellogg SmithMartha Kellogg Smith
Martha Kellogg Smithvonjobi
 
Spatial Groundings for Meaningful Symbols
Spatial Groundings for Meaningful SymbolsSpatial Groundings for Meaningful Symbols
Spatial Groundings for Meaningful SymbolsVlad Tanasescu
 

Similar to Semantics for visual resources: use cases from e-culture (20)

Esad 12may2010
Esad 12may2010Esad 12may2010
Esad 12may2010
 
E-Culture semantic search pilot
E-Culture semantic search pilotE-Culture semantic search pilot
E-Culture semantic search pilot
 
A socio-cultural ontology for urban development
A socio-cultural ontology for urban development A socio-cultural ontology for urban development
A socio-cultural ontology for urban development
 
Temporal
TemporalTemporal
Temporal
 
Art and Architecture Thesaurus
Art and Architecture ThesaurusArt and Architecture Thesaurus
Art and Architecture Thesaurus
 
Conceptual Organization And Retrieval Of Text By Historians
Conceptual Organization And Retrieval Of Text By HistoriansConceptual Organization And Retrieval Of Text By Historians
Conceptual Organization And Retrieval Of Text By Historians
 
Improving Image Discovery for Art Scholars
Improving Image Discovery for Art ScholarsImproving Image Discovery for Art Scholars
Improving Image Discovery for Art Scholars
 
Between  information  retrieval  services  and bibliometrics  research. New  ...
Between  information  retrieval  services  and bibliometrics  research. New  ...Between  information  retrieval  services  and bibliometrics  research. New  ...
Between  information  retrieval  services  and bibliometrics  research. New  ...
 
A Formal Modeling Proposal
A Formal Modeling ProposalA Formal Modeling Proposal
A Formal Modeling Proposal
 
Cross domain knowledge discovery, complex system theory and semantic web
Cross domain knowledge discovery, complex system theory and semantic webCross domain knowledge discovery, complex system theory and semantic web
Cross domain knowledge discovery, complex system theory and semantic web
 
InfoVis 2010 Lecture 1
InfoVis 2010 Lecture 1InfoVis 2010 Lecture 1
InfoVis 2010 Lecture 1
 
How and why study big cultural data v2
How and why study big cultural data v2How and why study big cultural data v2
How and why study big cultural data v2
 
Ontologies and the humanities: some issues affecting the design of digital in...
Ontologies and the humanities: some issues affecting the design of digital in...Ontologies and the humanities: some issues affecting the design of digital in...
Ontologies and the humanities: some issues affecting the design of digital in...
 
Semantic Web and Linked Data for cultural heritage materials - Approaches in ...
Semantic Web and Linked Data for cultural heritage materials - Approaches in ...Semantic Web and Linked Data for cultural heritage materials - Approaches in ...
Semantic Web and Linked Data for cultural heritage materials - Approaches in ...
 
Extending the knowledge level of cognitive architectures with Conceptual Spac...
Extending the knowledge level of cognitive architectures with Conceptual Spac...Extending the knowledge level of cognitive architectures with Conceptual Spac...
Extending the knowledge level of cognitive architectures with Conceptual Spac...
 
Multimodal Searching and Semantic Spaces: ...or how to find images of Dalmati...
Multimodal Searching and Semantic Spaces: ...or how to find images of Dalmati...Multimodal Searching and Semantic Spaces: ...or how to find images of Dalmati...
Multimodal Searching and Semantic Spaces: ...or how to find images of Dalmati...
 
chinchor_nvac_may06
chinchor_nvac_may06chinchor_nvac_may06
chinchor_nvac_may06
 
Teaching & Learning with Technology TLT 2016
Teaching & Learning with Technology TLT 2016Teaching & Learning with Technology TLT 2016
Teaching & Learning with Technology TLT 2016
 
Martha Kellogg Smith
Martha Kellogg SmithMartha Kellogg Smith
Martha Kellogg Smith
 
Spatial Groundings for Meaningful Symbols
Spatial Groundings for Meaningful SymbolsSpatial Groundings for Meaningful Symbols
Spatial Groundings for Meaningful Symbols
 

More from Guus Schreiber

Ontologies: vehicles for reuse
Ontologies: vehicles for reuseOntologies: vehicles for reuse
Ontologies: vehicles for reuseGuus Schreiber
 
Linking historical ship records to a newspaper archive
Linking historical ship records to a newspaper archiveLinking historical ship records to a newspaper archive
Linking historical ship records to a newspaper archiveGuus Schreiber
 
CommonKADS project management
CommonKADS project managementCommonKADS project management
CommonKADS project managementGuus Schreiber
 
UML notations used by CommonKADS
UML notations used by CommonKADSUML notations used by CommonKADS
UML notations used by CommonKADSGuus Schreiber
 
Advanced knowledge modelling
Advanced knowledge modellingAdvanced knowledge modelling
Advanced knowledge modellingGuus Schreiber
 
CommonKADS design and implementation
CommonKADS design and implementationCommonKADS design and implementation
CommonKADS design and implementationGuus Schreiber
 
CommonKADS communication model
CommonKADS communication modelCommonKADS communication model
CommonKADS communication modelGuus Schreiber
 
CommonKADS knowledge modelling process
CommonKADS knowledge modelling processCommonKADS knowledge modelling process
CommonKADS knowledge modelling processGuus Schreiber
 
CommonKADS knowledge model templates
CommonKADS knowledge model templatesCommonKADS knowledge model templates
CommonKADS knowledge model templatesGuus Schreiber
 
CommonKADS knowledge modelling basics
CommonKADS knowledge modelling basicsCommonKADS knowledge modelling basics
CommonKADS knowledge modelling basicsGuus Schreiber
 
CommonKADS knowledge management
CommonKADS knowledge managementCommonKADS knowledge management
CommonKADS knowledge managementGuus Schreiber
 
CommonKADS context models
CommonKADS context modelsCommonKADS context models
CommonKADS context modelsGuus Schreiber
 
Semantic Web: From Representations to Applications
Semantic Web: From Representations to ApplicationsSemantic Web: From Representations to Applications
Semantic Web: From Representations to ApplicationsGuus Schreiber
 
The Semantic Web: status and prospects
The Semantic Web: status and prospectsThe Semantic Web: status and prospects
The Semantic Web: status and prospectsGuus Schreiber
 
Ontology Engineering: Ontology Use
Ontology Engineering: Ontology UseOntology Engineering: Ontology Use
Ontology Engineering: Ontology UseGuus Schreiber
 
Ontology engineering: Ontology alignment
Ontology engineering: Ontology alignmentOntology engineering: Ontology alignment
Ontology engineering: Ontology alignmentGuus Schreiber
 
Ontology Engineering: Ontology evaluation
Ontology Engineering: Ontology evaluationOntology Engineering: Ontology evaluation
Ontology Engineering: Ontology evaluationGuus Schreiber
 
Ontology Engineering: ontology construction II
Ontology Engineering: ontology construction IIOntology Engineering: ontology construction II
Ontology Engineering: ontology construction IIGuus Schreiber
 

More from Guus Schreiber (20)

Ontologies: vehicles for reuse
Ontologies: vehicles for reuseOntologies: vehicles for reuse
Ontologies: vehicles for reuse
 
Linking historical ship records to a newspaper archive
Linking historical ship records to a newspaper archiveLinking historical ship records to a newspaper archive
Linking historical ship records to a newspaper archive
 
CommonKADS project management
CommonKADS project managementCommonKADS project management
CommonKADS project management
 
UML notations used by CommonKADS
UML notations used by CommonKADSUML notations used by CommonKADS
UML notations used by CommonKADS
 
Advanced knowledge modelling
Advanced knowledge modellingAdvanced knowledge modelling
Advanced knowledge modelling
 
CommonKADS design and implementation
CommonKADS design and implementationCommonKADS design and implementation
CommonKADS design and implementation
 
CommonKADS communication model
CommonKADS communication modelCommonKADS communication model
CommonKADS communication model
 
CommonKADS knowledge modelling process
CommonKADS knowledge modelling processCommonKADS knowledge modelling process
CommonKADS knowledge modelling process
 
CommonKADS knowledge model templates
CommonKADS knowledge model templatesCommonKADS knowledge model templates
CommonKADS knowledge model templates
 
CommonKADS knowledge modelling basics
CommonKADS knowledge modelling basicsCommonKADS knowledge modelling basics
CommonKADS knowledge modelling basics
 
CommonKADS knowledge management
CommonKADS knowledge managementCommonKADS knowledge management
CommonKADS knowledge management
 
CommonKADS context models
CommonKADS context modelsCommonKADS context models
CommonKADS context models
 
Introduction
IntroductionIntroduction
Introduction
 
Semantic Web: From Representations to Applications
Semantic Web: From Representations to ApplicationsSemantic Web: From Representations to Applications
Semantic Web: From Representations to Applications
 
The Semantic Web: status and prospects
The Semantic Web: status and prospectsThe Semantic Web: status and prospects
The Semantic Web: status and prospects
 
Vista-TV overview
Vista-TV overviewVista-TV overview
Vista-TV overview
 
Ontology Engineering: Ontology Use
Ontology Engineering: Ontology UseOntology Engineering: Ontology Use
Ontology Engineering: Ontology Use
 
Ontology engineering: Ontology alignment
Ontology engineering: Ontology alignmentOntology engineering: Ontology alignment
Ontology engineering: Ontology alignment
 
Ontology Engineering: Ontology evaluation
Ontology Engineering: Ontology evaluationOntology Engineering: Ontology evaluation
Ontology Engineering: Ontology evaluation
 
Ontology Engineering: ontology construction II
Ontology Engineering: ontology construction IIOntology Engineering: ontology construction II
Ontology Engineering: ontology construction II
 

Semantics for visual resources: use cases from e-culture

  • 1. Semantics for visual resources Use Cases from E-Culture Guus Schreiber Free University Amsterdam schreiber@cs.vu.nl
  • 2. 2 Purpose  Analyze a number of use cases from e-culture domain – Multimedia plays key role  Required technology – Typically combination of technologies  Relation to state of the art Acknowledgements: This presentations contains slides and images provided by Laura Hollink, Giang Nguyen and Cees Snoek. Also thanks to the MultimediaN E-Culture team
  • 3. 3 Use case: Asian chairs User has found an image of an Asian chair Annotation: ex:image vra:stylePeriod aat:Guangxu . How can we find images of Asian chairs from the same historical period?
  • 4. 4 AAT info on Guangxu
  • 5. 5 Importance of time and space information  Many queries require time/space knowledge, either absolute or abstracted  For the chair image we can establish – Country = China (link Chinese => China) – Period = 1644-1911 (from Qing description)  Technology requirements: – Thesuari relating time/space concepts – NLP for unstructured descriptions – Time/space reasoning techniques
  • 6. 6
  • 7. 7
  • 8. 8 Sample place information in TGN <tgn:AdministrativePlace rdf:about="&tgn;1000111" tgn:standardLatitude="35" tgn:standardLongitude="105“> <vp:parentPreferred rdf:resource="&tgn;1000004"/> …….. </tgn:AdministrativePlace>
  • 9. 9 Issues when searching for “nearby” Asian chairs  Close in space: – Other country in (East) Asia – Latitude/longitude  Close in time: – Links between style periods – Match time periods (and handle incomplete information)
  • 10. 10
  • 11. 11 Use case: painting style Find paintings of a similar style MATISSE, Henri Le bonheur de vivre (The Joy of Life) 1905-1906 Oil on canvas, 69 1/8 x 94 7/8 in. (175 x 241 cm) Barnes Foundation, Merion, PA
  • 12. 12 How can we find this other Fauve painting? DERAIN, Andre The Turning Road, L'Estaque, 1906 Oil on canvas, 51 x 76 3/4 in. (129.5 x 195 cm) Museum of Fine Arts, Houston, Texas
  • 13. 13 Issues  Parse annotation to find matches with thesauri terms – E.g. match artists to ULAN individuals  Artists-style links – AAT contains styles; ULAN contains artists, but there is no link • Learn link from corpora • Derive it from other annotations – Domain-specific rules/reasoning needed • see example in SWRL doc • Painters may have painted in multiple styles
  • 14. 14
  • 15. 15
  • 16. 16 Search: WordNet patterns that increase recall without sacrificing precision (Hollink)
  • 17. 17 Issues w.r.t. thesauri  Public availability!  RDF/OWL representation  Learning/specifying term/concept mapping – owl:equivalentClass, owl:sameAs, rdf:type, rdfs:subClassOf – Domain-specific links  Managing the evolution of the thesauri and the mappings
  • 18. 18 Use case: find images with the same subject Find another painting which portrays dancing
  • 19. 19 Issues  Same subjects can be visually very different  Subject is often missing from the annotation  Mismatch: users often search for subjects of images
  • 20. 20 Conceptual subject descriptions 85% of the user queries: General Descriptions of generally known items. Only general, everyday knowledge is necessary. Descriptions are at the level of the Natural categories of E. Rosch (1973), or more general. E.g An ape eating a banana. Specific Descriptions of objects or scenes that can be identified and named. Specific domain knowledge is necessary to recognize the objects or scenes. E.g. The old male gorilla Kumba, born in Cameroon and now living in Artis, Amsterdam Abstract Descriptions for which interpretative knowledge is used. This category is subjective. E.g An animal threatened with extinction.
  • 21. 21 Example concepts in image  Specific – Fall of the Berlin Wall  General – People walking at night  Abstract – Fall of the Iron Curtain
  • 22. 22 Use of conceptual categories by people searching for images Conceptual level: 83% 0% 20% 40% 60% 80% 100% event time place relation scene object Characteristics Nuberofelementsin%of conceptualelements Abstract Specific General
  • 24. 24
  • 25. 25
  • 26. 26 Annotation of image content  Template for subject description Agent Action Object Recipient  Guidelines for manual annotation – Annotate as specific as possible  Default reasoning  CBIR support: – Object identification – Spatial relations
  • 27. 27
  • 28. 28
  • 29. 29 Some forms of image content are well suited to image analysis Collection of clothes Abstract painting
  • 30. 30 The semantic gap  The distance between Content-Based Image Retrieval and semantics: – Smeulders, Worring, Santini, Gupta, Jain. Content- based image retrieval at the end of the early years. IEEE Transactions on Pattern Analysis and Machine Intelligence, 22(12), December 2000.  Direct links between visual features and semantic concepts become more difficult when the domain is broader / more general
  • 31. 31 Example semantic bridge: microscopic cell images mpeg7 : StillRegion(region) ^ mpeg7x : Dense(region) ^ mpeg7 : DominantColor(region, col) ^ swrlb : lessThan(col, 100) => mpeg7 : Depicts(region, mesh : MatureGranule)
  • 33. 33 Automatic detection of concepts can be difficult even in “easy” cases What is the color of this ape?
  • 34. 34 Image analysis useful for collection navigation
  • 35. 35 Bridging the semantic gap: CBIR and ontologies Visual WordNet (GE paper) – Adding knowledge about visual characteristics to WordNet: mobility, color, … – Build detectors for the visual features – Use visual data to prune the tree of categories when analyzing a visual object
  • 36. 36 Sample visual features and their mapping to WordNet
  • 37. 37 Experiment: pruning the search for “conveyance” concepts 6 concepts found Including taxi cab 12 concepts found Including passenger train and commuter train Three visual features: material, motion, environment Assumption is that these work perfectly
  • 38. 38 Bridging the semantic gap: concept detectors  Snoek et al., TRECVID2004 – 185 hours of news video  32 detectors for concepts in news video – Through machine learning  Similarity detectors based on keywords and visual analysis  Query interface in which these functions can be combined
  • 39. 39 “Concepts” for which visual detectors were built
  • 40. 40 LSCOM lexicon: 229 - Weather  Context-specific (i.e. news broadcast) interpretation: “Weather forecast”
  • 41. 41 LSCOM lexicon: 110 – Female Anchor  Composite concept  Alignment needed for semantic search, e.g. with WordNet
  • 42. 42 Natural-lang proc. automatic annotation text stings → concepts Distributed cultuurwijzer.nl collections OAI-based access Reasoning support time/space reasoning Web interface support for web collections Presentation facilities semantic presentation device-specific Interoperability XML/RDF/OWL Scalability > 10,000,000 triples Ontologies WordNet, AAT, TGN ULAN, Dutch labels Search strategies sibling search semantic distance Dublin Core specializations dumb-down semantic annotation DIGITAL HERITAGE COLLECTIONS semantic search BASELINEENHANCEDENHANCED FEATURESFEATURES NEWNEW FEATURESFEATURES
  • 43. 43
  • 44. 44 Main observation A combination of many different techniques is needed to be able to cope with the complexity of multimedia semantics – NLP, segmentation, CBIR, visual feature detectors, visual ontologies, publicly available thesauri, thesauri mappings, dedicated reasoning techniques (time, space, default), personalization, presentation generation  Key role for user studies

Editor's Notes

  1. &amp;lt;number&amp;gt; Criteria: Indirectly derived from the image Interpretation and domain knowledge are required
  2. &amp;lt;number&amp;gt;