SlideShare a Scribd company logo
Metadata for video search
       Trouvaille


          Karel Braeckman
          6 maart 2009
Overview

    Challenges
    Example situation
    Search and annotation
        Search: Trouvaille
                – Sources
                – Demo
        Annotation
                – Problems
                – Demo




2    VRT-medialab: onderzoek en innovatie
Challenges

    Indexing and retrieving video material
        Growing database of video material
        But video is not self-descriptive → metadata
        Annotation: assist archivist with modern tools
                – Feature extraction
                – Timecoded attributes
        Search: User-friendly application to search through the archive




3    VRT-medialab: onderzoek en innovatie
Example: situation at VRT
                                           Search And Annotation




                                            Search results




4   VRT-medialab: onderzoek en innovatie
Example: situation at VRT

    Annotation and search with Basisplus
        Basisplus database contains 625 000 items
        65 fields (tfw, dienst, formaat, personen, …)
        Some fields should be chosen from controlled list:
         the VRT video thesaurus
                – Some hierarchy:
                     ‘lead’, ‘broad’ / ‘narrow’ and ‘use for’ terms
                – More than 300 000 lead terms




5    VRT-medialab: onderzoek en innovatie
Basisplus limitations

    Annotation by archivist
       - Knowledge of all fields and thesaurus
       - Familiar with annotation process




    But a program maker
       - Does not know in what field to type
       - Might use a different term, not contained in the thesaurus
       - Textual results → limited representation of video




6    VRT-medialab: onderzoek en innovatie
The ideal search application

    We want all relevant search results
    Without any false positives
      In practice a balance between the two
    User friendly
    Clear list of results
    Find desired result fast
    Jump to relevant part in video
        Time coded metadata




7    VRT-medialab: onderzoek en innovatie
Sources used for Trouvaille


       Current metadata archive
              (Basisplus)




                                                        Media Asset
                                                        Management
News feed
                                                          System
                                                         (Ardome)


                        News broadcasts
                           (Ardome)

                                                   + VRT video-Thesaurus
                                                   + List of Belgian cities
8           VRT-medialab: onderzoek en innovatie
Trouvaille Demo




9   VRT-medialab: onderzoek en innovatie
Annotation

     Manual annotation is a lot of work             assist archivist with
     modern tools
                 – Annotate efficiently
                 – As much automation as possible

     Need for time coded metadata
         Metadata describes the complete video item
         Time codes could mark the time range on which the metadata
          applies.

     A lot of metadata is lost during production process
         Teletext subtitles
         News rundown
         Scripts with shots / scenes and their contents

10    VRT-medialab: onderzoek en innovatie
annotation




11   VRT-medialab: onderzoek en innovatie
annotation

     Assist archivist by partial automation
         Split video in shots, audio segments, scenes
                 – Archivist can easily add time coded metadata to a scene
                 – Relevant thumbnail for every shot




         Detection, even recognition of faces




12    VRT-medialab: onderzoek en innovatie
annotation

     Assist archivist by partial automation
         Split video in shots, audio segments, scenes
         Detection, even recognition of faces
         Detect reuse and duplicates




13    VRT-medialab: onderzoek en innovatie
Conclusions

     New search interface Trouvaille
         User friendly
         Get desired result fast
         Range of views
     Result is a complete item in Trouvaille
         Annotation with time coded metadata (per scene)
         Search results should use this time codes




14    VRT-medialab: onderzoek en innovatie
Conclusions

     Annotation application
         Partial automation → efficient annotation
         Time coded metadata



     Too often reverse engineering of metadata that was
     readily available at some stage during production process




15    VRT-medialab: onderzoek en innovatie

More Related Content

Similar to Metadata for video search: Trouvaille

Separate Pasts, Common Futures: Digital film preservation in a broadcast en...
Separate Pasts,  Common Futures: Digital film preservation in a  broadcast en...Separate Pasts,  Common Futures: Digital film preservation in a  broadcast en...
Separate Pasts, Common Futures: Digital film preservation in a broadcast en...
Erwin Verbruggen
 
2007 EBU Training VRT Newsroom interoperability
2007 EBU Training VRT Newsroom interoperability2007 EBU Training VRT Newsroom interoperability
2007 EBU Training VRT Newsroom interoperability
European Broacasting Union
 
Rosinski ibm ai overview with several examples of projects in the media and l...
Rosinski ibm ai overview with several examples of projects in the media and l...Rosinski ibm ai overview with several examples of projects in the media and l...
Rosinski ibm ai overview with several examples of projects in the media and l...
FIAT/IFTA
 
Dennis Povshedny.Mediamosa intro.DrupalCamp Kyiv 2011
Dennis Povshedny.Mediamosa intro.DrupalCamp Kyiv 2011Dennis Povshedny.Mediamosa intro.DrupalCamp Kyiv 2011
Dennis Povshedny.Mediamosa intro.DrupalCamp Kyiv 2011
camp_drupal_ua
 
OOMEN MEZARIS ReTV
OOMEN MEZARIS ReTVOOMEN MEZARIS ReTV
OOMEN MEZARIS ReTV
FIAT/IFTA
 
Implementing artificial intelligence strategies for content annotation and pu...
Implementing artificial intelligence strategies for content annotation and pu...Implementing artificial intelligence strategies for content annotation and pu...
Implementing artificial intelligence strategies for content annotation and pu...
ReTV project
 
Implementing Artificial Intelligence Strategies for Content Annotation and Pu...
Implementing Artificial Intelligence Strategies for Content Annotation and Pu...Implementing Artificial Intelligence Strategies for Content Annotation and Pu...
Implementing Artificial Intelligence Strategies for Content Annotation and Pu...
ReTV project
 
Cebit-2008: Content Aggregation
Cebit-2008: Content AggregationCebit-2008: Content Aggregation
Cebit-2008: Content Aggregation
David Nuescheler
 
Of Communities and Practices: Digital Preservation Innovation & Research
Of Communities  and Practices: Digital Preservation Innovation & ResearchOf Communities  and Practices: Digital Preservation Innovation & Research
Of Communities and Practices: Digital Preservation Innovation & Research
Erwin Verbruggen
 
Digital Media Production
Digital Media ProductionDigital Media Production
Digital Media Production
Maarten Verwaest
 
RAI Archives: Looking to the future. Alberto Messina, Laurent Boch, RAI.
RAI Archives: Looking to the future. Alberto Messina, Laurent Boch, RAI.RAI Archives: Looking to the future. Alberto Messina, Laurent Boch, RAI.
RAI Archives: Looking to the future. Alberto Messina, Laurent Boch, RAI.
FIAT/IFTA
 
Fiat 20080921 results PISA
Fiat 20080921 results PISAFiat 20080921 results PISA
Fiat 20080921 results PISA
vrt-medialab
 
Video Indexing And Retrieval
Video Indexing And RetrievalVideo Indexing And Retrieval
Video Indexing And Retrieval
Yvonne M
 
Re-using Media on the Web tutorial: Media Fragment Creation and Annotation
Re-using Media on the Web tutorial: Media Fragment Creation and AnnotationRe-using Media on the Web tutorial: Media Fragment Creation and Annotation
Re-using Media on the Web tutorial: Media Fragment Creation and Annotation
MediaMixerCommunity
 
LORENZ Building an integrated digital media archive and legal deposit
LORENZ Building an integrated digital media archive and legal depositLORENZ Building an integrated digital media archive and legal deposit
LORENZ Building an integrated digital media archive and legal deposit
FIAT/IFTA
 
LYCKE Artificial intelligence, hype or hope?
LYCKE Artificial intelligence, hype or hope?LYCKE Artificial intelligence, hype or hope?
LYCKE Artificial intelligence, hype or hope?
FIAT/IFTA
 
Mpeg7
Mpeg7Mpeg7
YouTube and the National Library of Scotland
YouTube and the National Library of ScotlandYouTube and the National Library of Scotland
Digital Presentation Best Practices: Lessons Learned From Across the Pond
Digital Presentation Best Practices: Lessons Learned From Across the PondDigital Presentation Best Practices: Lessons Learned From Across the Pond
Digital Presentation Best Practices: Lessons Learned From Across the Pond
ULB - Bibliothèques
 
Digital Preservation Best Practices: Lessons Learned From Across the Pond
Digital Preservation Best Practices: Lessons Learned From Across the PondDigital Preservation Best Practices: Lessons Learned From Across the Pond
Digital Preservation Best Practices: Lessons Learned From Across the Pond
Benoit Pauwels
 

Similar to Metadata for video search: Trouvaille (20)

Separate Pasts, Common Futures: Digital film preservation in a broadcast en...
Separate Pasts,  Common Futures: Digital film preservation in a  broadcast en...Separate Pasts,  Common Futures: Digital film preservation in a  broadcast en...
Separate Pasts, Common Futures: Digital film preservation in a broadcast en...
 
2007 EBU Training VRT Newsroom interoperability
2007 EBU Training VRT Newsroom interoperability2007 EBU Training VRT Newsroom interoperability
2007 EBU Training VRT Newsroom interoperability
 
Rosinski ibm ai overview with several examples of projects in the media and l...
Rosinski ibm ai overview with several examples of projects in the media and l...Rosinski ibm ai overview with several examples of projects in the media and l...
Rosinski ibm ai overview with several examples of projects in the media and l...
 
Dennis Povshedny.Mediamosa intro.DrupalCamp Kyiv 2011
Dennis Povshedny.Mediamosa intro.DrupalCamp Kyiv 2011Dennis Povshedny.Mediamosa intro.DrupalCamp Kyiv 2011
Dennis Povshedny.Mediamosa intro.DrupalCamp Kyiv 2011
 
OOMEN MEZARIS ReTV
OOMEN MEZARIS ReTVOOMEN MEZARIS ReTV
OOMEN MEZARIS ReTV
 
Implementing artificial intelligence strategies for content annotation and pu...
Implementing artificial intelligence strategies for content annotation and pu...Implementing artificial intelligence strategies for content annotation and pu...
Implementing artificial intelligence strategies for content annotation and pu...
 
Implementing Artificial Intelligence Strategies for Content Annotation and Pu...
Implementing Artificial Intelligence Strategies for Content Annotation and Pu...Implementing Artificial Intelligence Strategies for Content Annotation and Pu...
Implementing Artificial Intelligence Strategies for Content Annotation and Pu...
 
Cebit-2008: Content Aggregation
Cebit-2008: Content AggregationCebit-2008: Content Aggregation
Cebit-2008: Content Aggregation
 
Of Communities and Practices: Digital Preservation Innovation & Research
Of Communities  and Practices: Digital Preservation Innovation & ResearchOf Communities  and Practices: Digital Preservation Innovation & Research
Of Communities and Practices: Digital Preservation Innovation & Research
 
Digital Media Production
Digital Media ProductionDigital Media Production
Digital Media Production
 
RAI Archives: Looking to the future. Alberto Messina, Laurent Boch, RAI.
RAI Archives: Looking to the future. Alberto Messina, Laurent Boch, RAI.RAI Archives: Looking to the future. Alberto Messina, Laurent Boch, RAI.
RAI Archives: Looking to the future. Alberto Messina, Laurent Boch, RAI.
 
Fiat 20080921 results PISA
Fiat 20080921 results PISAFiat 20080921 results PISA
Fiat 20080921 results PISA
 
Video Indexing And Retrieval
Video Indexing And RetrievalVideo Indexing And Retrieval
Video Indexing And Retrieval
 
Re-using Media on the Web tutorial: Media Fragment Creation and Annotation
Re-using Media on the Web tutorial: Media Fragment Creation and AnnotationRe-using Media on the Web tutorial: Media Fragment Creation and Annotation
Re-using Media on the Web tutorial: Media Fragment Creation and Annotation
 
LORENZ Building an integrated digital media archive and legal deposit
LORENZ Building an integrated digital media archive and legal depositLORENZ Building an integrated digital media archive and legal deposit
LORENZ Building an integrated digital media archive and legal deposit
 
LYCKE Artificial intelligence, hype or hope?
LYCKE Artificial intelligence, hype or hope?LYCKE Artificial intelligence, hype or hope?
LYCKE Artificial intelligence, hype or hope?
 
Mpeg7
Mpeg7Mpeg7
Mpeg7
 
YouTube and the National Library of Scotland
YouTube and the National Library of ScotlandYouTube and the National Library of Scotland
YouTube and the National Library of Scotland
 
Digital Presentation Best Practices: Lessons Learned From Across the Pond
Digital Presentation Best Practices: Lessons Learned From Across the PondDigital Presentation Best Practices: Lessons Learned From Across the Pond
Digital Presentation Best Practices: Lessons Learned From Across the Pond
 
Digital Preservation Best Practices: Lessons Learned From Across the Pond
Digital Preservation Best Practices: Lessons Learned From Across the PondDigital Preservation Best Practices: Lessons Learned From Across the Pond
Digital Preservation Best Practices: Lessons Learned From Across the Pond
 

More from vrt-medialab

Multischermenonderzoek
MultischermenonderzoekMultischermenonderzoek
Multischermenonderzoekvrt-medialab
 
Browser as a broadcast medium
Browser as a broadcast mediumBrowser as a broadcast medium
Browser as a broadcast medium
vrt-medialab
 
Champ iMinds
Champ iMindsChamp iMinds
Champ iMinds
vrt-medialab
 
Taming your media chaos
Taming your media chaosTaming your media chaos
Taming your media chaos
vrt-medialab
 
Presentatie iMinds MediaCRM
Presentatie iMinds MediaCRMPresentatie iMinds MediaCRM
Presentatie iMinds MediaCRM
vrt-medialab
 
Evaluatiestudie VillaSquare
 Evaluatiestudie VillaSquare Evaluatiestudie VillaSquare
Evaluatiestudie VillaSquarevrt-medialab
 
iMinds VillaSquare evaluation IBBT-SMIT
iMinds VillaSquare evaluation IBBT-SMITiMinds VillaSquare evaluation IBBT-SMIT
iMinds VillaSquare evaluation IBBT-SMIT
vrt-medialab
 
Building second screen TV apps
Building second screen TV appsBuilding second screen TV apps
Building second screen TV apps
vrt-medialab
 
Multischermenonderzoek
MultischermenonderzoekMultischermenonderzoek
Multischermenonderzoekvrt-medialab
 
Exploring your media with the Semantic Web
Exploring your media with the Semantic WebExploring your media with the Semantic Web
Exploring your media with the Semantic Web
vrt-medialab
 
BDMA workshop presentation - Using the Second Screen - MediaSquare - MediaCRM
BDMA workshop presentation - Using the Second Screen - MediaSquare - MediaCRMBDMA workshop presentation - Using the Second Screen - MediaSquare - MediaCRM
BDMA workshop presentation - Using the Second Screen - MediaSquare - MediaCRM
vrt-medialab
 
Champ belgian broadcast_days
Champ belgian broadcast_daysChamp belgian broadcast_days
Champ belgian broadcast_days
vrt-medialab
 
Champ Pitch Celtic-Plus Event 2011
Champ Pitch Celtic-Plus Event 2011Champ Pitch Celtic-Plus Event 2011
Champ Pitch Celtic-Plus Event 2011
vrt-medialab
 
medialoep
medialoepmedialoep
medialoep
vrt-medialab
 
video for html5
video for html5video for html5
video for html5
vrt-medialab
 
html5 an introduction
html5 an introductionhtml5 an introduction
html5 an introduction
vrt-medialab
 
Boost your search with semantic technology
Boost your search with semantic technologyBoost your search with semantic technology
Boost your search with semantic technology
vrt-medialab
 
Media Square : platform for second screen experiences
Media Square : platform for second screen experiencesMedia Square : platform for second screen experiences
Media Square : platform for second screen experiences
vrt-medialab
 
MediaSquare - Check into your favourite media
MediaSquare - Check into your favourite mediaMediaSquare - Check into your favourite media
MediaSquare - Check into your favourite media
vrt-medialab
 
Transmedia
TransmediaTransmedia
Transmedia
vrt-medialab
 

More from vrt-medialab (20)

Multischermenonderzoek
MultischermenonderzoekMultischermenonderzoek
Multischermenonderzoek
 
Browser as a broadcast medium
Browser as a broadcast mediumBrowser as a broadcast medium
Browser as a broadcast medium
 
Champ iMinds
Champ iMindsChamp iMinds
Champ iMinds
 
Taming your media chaos
Taming your media chaosTaming your media chaos
Taming your media chaos
 
Presentatie iMinds MediaCRM
Presentatie iMinds MediaCRMPresentatie iMinds MediaCRM
Presentatie iMinds MediaCRM
 
Evaluatiestudie VillaSquare
 Evaluatiestudie VillaSquare Evaluatiestudie VillaSquare
Evaluatiestudie VillaSquare
 
iMinds VillaSquare evaluation IBBT-SMIT
iMinds VillaSquare evaluation IBBT-SMITiMinds VillaSquare evaluation IBBT-SMIT
iMinds VillaSquare evaluation IBBT-SMIT
 
Building second screen TV apps
Building second screen TV appsBuilding second screen TV apps
Building second screen TV apps
 
Multischermenonderzoek
MultischermenonderzoekMultischermenonderzoek
Multischermenonderzoek
 
Exploring your media with the Semantic Web
Exploring your media with the Semantic WebExploring your media with the Semantic Web
Exploring your media with the Semantic Web
 
BDMA workshop presentation - Using the Second Screen - MediaSquare - MediaCRM
BDMA workshop presentation - Using the Second Screen - MediaSquare - MediaCRMBDMA workshop presentation - Using the Second Screen - MediaSquare - MediaCRM
BDMA workshop presentation - Using the Second Screen - MediaSquare - MediaCRM
 
Champ belgian broadcast_days
Champ belgian broadcast_daysChamp belgian broadcast_days
Champ belgian broadcast_days
 
Champ Pitch Celtic-Plus Event 2011
Champ Pitch Celtic-Plus Event 2011Champ Pitch Celtic-Plus Event 2011
Champ Pitch Celtic-Plus Event 2011
 
medialoep
medialoepmedialoep
medialoep
 
video for html5
video for html5video for html5
video for html5
 
html5 an introduction
html5 an introductionhtml5 an introduction
html5 an introduction
 
Boost your search with semantic technology
Boost your search with semantic technologyBoost your search with semantic technology
Boost your search with semantic technology
 
Media Square : platform for second screen experiences
Media Square : platform for second screen experiencesMedia Square : platform for second screen experiences
Media Square : platform for second screen experiences
 
MediaSquare - Check into your favourite media
MediaSquare - Check into your favourite mediaMediaSquare - Check into your favourite media
MediaSquare - Check into your favourite media
 
Transmedia
TransmediaTransmedia
Transmedia
 

Recently uploaded

OpenID AuthZEN Interop Read Out - Authorization
OpenID AuthZEN Interop Read Out - AuthorizationOpenID AuthZEN Interop Read Out - Authorization
OpenID AuthZEN Interop Read Out - Authorization
David Brossard
 
Artificial Intelligence for XMLDevelopment
Artificial Intelligence for XMLDevelopmentArtificial Intelligence for XMLDevelopment
Artificial Intelligence for XMLDevelopment
Octavian Nadolu
 
Cosa hanno in comune un mattoncino Lego e la backdoor XZ?
Cosa hanno in comune un mattoncino Lego e la backdoor XZ?Cosa hanno in comune un mattoncino Lego e la backdoor XZ?
Cosa hanno in comune un mattoncino Lego e la backdoor XZ?
Speck&Tech
 
How to use Firebase Data Connect For Flutter
How to use Firebase Data Connect For FlutterHow to use Firebase Data Connect For Flutter
How to use Firebase Data Connect For Flutter
Daiki Mogmet Ito
 
GraphRAG for Life Science to increase LLM accuracy
GraphRAG for Life Science to increase LLM accuracyGraphRAG for Life Science to increase LLM accuracy
GraphRAG for Life Science to increase LLM accuracy
Tomaz Bratanic
 
Serial Arm Control in Real Time Presentation
Serial Arm Control in Real Time PresentationSerial Arm Control in Real Time Presentation
Serial Arm Control in Real Time Presentation
tolgahangng
 
How to Get CNIC Information System with Paksim Ga.pptx
How to Get CNIC Information System with Paksim Ga.pptxHow to Get CNIC Information System with Paksim Ga.pptx
How to Get CNIC Information System with Paksim Ga.pptx
danishmna97
 
June Patch Tuesday
June Patch TuesdayJune Patch Tuesday
June Patch Tuesday
Ivanti
 
Presentation of the OECD Artificial Intelligence Review of Germany
Presentation of the OECD Artificial Intelligence Review of GermanyPresentation of the OECD Artificial Intelligence Review of Germany
Presentation of the OECD Artificial Intelligence Review of Germany
innovationoecd
 
How to Interpret Trends in the Kalyan Rajdhani Mix Chart.pdf
How to Interpret Trends in the Kalyan Rajdhani Mix Chart.pdfHow to Interpret Trends in the Kalyan Rajdhani Mix Chart.pdf
How to Interpret Trends in the Kalyan Rajdhani Mix Chart.pdf
Chart Kalyan
 
AI 101: An Introduction to the Basics and Impact of Artificial Intelligence
AI 101: An Introduction to the Basics and Impact of Artificial IntelligenceAI 101: An Introduction to the Basics and Impact of Artificial Intelligence
AI 101: An Introduction to the Basics and Impact of Artificial Intelligence
IndexBug
 
Let's Integrate MuleSoft RPA, COMPOSER, APM with AWS IDP along with Slack
Let's Integrate MuleSoft RPA, COMPOSER, APM with AWS IDP along with SlackLet's Integrate MuleSoft RPA, COMPOSER, APM with AWS IDP along with Slack
Let's Integrate MuleSoft RPA, COMPOSER, APM with AWS IDP along with Slack
shyamraj55
 
UiPath Test Automation using UiPath Test Suite series, part 6
UiPath Test Automation using UiPath Test Suite series, part 6UiPath Test Automation using UiPath Test Suite series, part 6
UiPath Test Automation using UiPath Test Suite series, part 6
DianaGray10
 
National Security Agency - NSA mobile device best practices
National Security Agency - NSA mobile device best practicesNational Security Agency - NSA mobile device best practices
National Security Agency - NSA mobile device best practices
Quotidiano Piemontese
 
Columbus Data & Analytics Wednesdays - June 2024
Columbus Data & Analytics Wednesdays - June 2024Columbus Data & Analytics Wednesdays - June 2024
Columbus Data & Analytics Wednesdays - June 2024
Jason Packer
 
Energy Efficient Video Encoding for Cloud and Edge Computing Instances
Energy Efficient Video Encoding for Cloud and Edge Computing InstancesEnergy Efficient Video Encoding for Cloud and Edge Computing Instances
Energy Efficient Video Encoding for Cloud and Edge Computing Instances
Alpen-Adria-Universität
 
Generating privacy-protected synthetic data using Secludy and Milvus
Generating privacy-protected synthetic data using Secludy and MilvusGenerating privacy-protected synthetic data using Secludy and Milvus
Generating privacy-protected synthetic data using Secludy and Milvus
Zilliz
 
5th LF Energy Power Grid Model Meet-up Slides
5th LF Energy Power Grid Model Meet-up Slides5th LF Energy Power Grid Model Meet-up Slides
5th LF Energy Power Grid Model Meet-up Slides
DanBrown980551
 
Salesforce Integration for Bonterra Impact Management (fka Social Solutions A...
Salesforce Integration for Bonterra Impact Management (fka Social Solutions A...Salesforce Integration for Bonterra Impact Management (fka Social Solutions A...
Salesforce Integration for Bonterra Impact Management (fka Social Solutions A...
Jeffrey Haguewood
 
UI5 Controls simplified - UI5con2024 presentation
UI5 Controls simplified - UI5con2024 presentationUI5 Controls simplified - UI5con2024 presentation
UI5 Controls simplified - UI5con2024 presentation
Wouter Lemaire
 

Recently uploaded (20)

OpenID AuthZEN Interop Read Out - Authorization
OpenID AuthZEN Interop Read Out - AuthorizationOpenID AuthZEN Interop Read Out - Authorization
OpenID AuthZEN Interop Read Out - Authorization
 
Artificial Intelligence for XMLDevelopment
Artificial Intelligence for XMLDevelopmentArtificial Intelligence for XMLDevelopment
Artificial Intelligence for XMLDevelopment
 
Cosa hanno in comune un mattoncino Lego e la backdoor XZ?
Cosa hanno in comune un mattoncino Lego e la backdoor XZ?Cosa hanno in comune un mattoncino Lego e la backdoor XZ?
Cosa hanno in comune un mattoncino Lego e la backdoor XZ?
 
How to use Firebase Data Connect For Flutter
How to use Firebase Data Connect For FlutterHow to use Firebase Data Connect For Flutter
How to use Firebase Data Connect For Flutter
 
GraphRAG for Life Science to increase LLM accuracy
GraphRAG for Life Science to increase LLM accuracyGraphRAG for Life Science to increase LLM accuracy
GraphRAG for Life Science to increase LLM accuracy
 
Serial Arm Control in Real Time Presentation
Serial Arm Control in Real Time PresentationSerial Arm Control in Real Time Presentation
Serial Arm Control in Real Time Presentation
 
How to Get CNIC Information System with Paksim Ga.pptx
How to Get CNIC Information System with Paksim Ga.pptxHow to Get CNIC Information System with Paksim Ga.pptx
How to Get CNIC Information System with Paksim Ga.pptx
 
June Patch Tuesday
June Patch TuesdayJune Patch Tuesday
June Patch Tuesday
 
Presentation of the OECD Artificial Intelligence Review of Germany
Presentation of the OECD Artificial Intelligence Review of GermanyPresentation of the OECD Artificial Intelligence Review of Germany
Presentation of the OECD Artificial Intelligence Review of Germany
 
How to Interpret Trends in the Kalyan Rajdhani Mix Chart.pdf
How to Interpret Trends in the Kalyan Rajdhani Mix Chart.pdfHow to Interpret Trends in the Kalyan Rajdhani Mix Chart.pdf
How to Interpret Trends in the Kalyan Rajdhani Mix Chart.pdf
 
AI 101: An Introduction to the Basics and Impact of Artificial Intelligence
AI 101: An Introduction to the Basics and Impact of Artificial IntelligenceAI 101: An Introduction to the Basics and Impact of Artificial Intelligence
AI 101: An Introduction to the Basics and Impact of Artificial Intelligence
 
Let's Integrate MuleSoft RPA, COMPOSER, APM with AWS IDP along with Slack
Let's Integrate MuleSoft RPA, COMPOSER, APM with AWS IDP along with SlackLet's Integrate MuleSoft RPA, COMPOSER, APM with AWS IDP along with Slack
Let's Integrate MuleSoft RPA, COMPOSER, APM with AWS IDP along with Slack
 
UiPath Test Automation using UiPath Test Suite series, part 6
UiPath Test Automation using UiPath Test Suite series, part 6UiPath Test Automation using UiPath Test Suite series, part 6
UiPath Test Automation using UiPath Test Suite series, part 6
 
National Security Agency - NSA mobile device best practices
National Security Agency - NSA mobile device best practicesNational Security Agency - NSA mobile device best practices
National Security Agency - NSA mobile device best practices
 
Columbus Data & Analytics Wednesdays - June 2024
Columbus Data & Analytics Wednesdays - June 2024Columbus Data & Analytics Wednesdays - June 2024
Columbus Data & Analytics Wednesdays - June 2024
 
Energy Efficient Video Encoding for Cloud and Edge Computing Instances
Energy Efficient Video Encoding for Cloud and Edge Computing InstancesEnergy Efficient Video Encoding for Cloud and Edge Computing Instances
Energy Efficient Video Encoding for Cloud and Edge Computing Instances
 
Generating privacy-protected synthetic data using Secludy and Milvus
Generating privacy-protected synthetic data using Secludy and MilvusGenerating privacy-protected synthetic data using Secludy and Milvus
Generating privacy-protected synthetic data using Secludy and Milvus
 
5th LF Energy Power Grid Model Meet-up Slides
5th LF Energy Power Grid Model Meet-up Slides5th LF Energy Power Grid Model Meet-up Slides
5th LF Energy Power Grid Model Meet-up Slides
 
Salesforce Integration for Bonterra Impact Management (fka Social Solutions A...
Salesforce Integration for Bonterra Impact Management (fka Social Solutions A...Salesforce Integration for Bonterra Impact Management (fka Social Solutions A...
Salesforce Integration for Bonterra Impact Management (fka Social Solutions A...
 
UI5 Controls simplified - UI5con2024 presentation
UI5 Controls simplified - UI5con2024 presentationUI5 Controls simplified - UI5con2024 presentation
UI5 Controls simplified - UI5con2024 presentation
 

Metadata for video search: Trouvaille

  • 1. Metadata for video search Trouvaille Karel Braeckman 6 maart 2009
  • 2. Overview Challenges Example situation Search and annotation  Search: Trouvaille – Sources – Demo  Annotation – Problems – Demo 2 VRT-medialab: onderzoek en innovatie
  • 3. Challenges Indexing and retrieving video material  Growing database of video material  But video is not self-descriptive → metadata  Annotation: assist archivist with modern tools – Feature extraction – Timecoded attributes  Search: User-friendly application to search through the archive 3 VRT-medialab: onderzoek en innovatie
  • 4. Example: situation at VRT Search And Annotation Search results 4 VRT-medialab: onderzoek en innovatie
  • 5. Example: situation at VRT Annotation and search with Basisplus  Basisplus database contains 625 000 items  65 fields (tfw, dienst, formaat, personen, …)  Some fields should be chosen from controlled list: the VRT video thesaurus – Some hierarchy: ‘lead’, ‘broad’ / ‘narrow’ and ‘use for’ terms – More than 300 000 lead terms 5 VRT-medialab: onderzoek en innovatie
  • 6. Basisplus limitations Annotation by archivist - Knowledge of all fields and thesaurus - Familiar with annotation process But a program maker - Does not know in what field to type - Might use a different term, not contained in the thesaurus - Textual results → limited representation of video 6 VRT-medialab: onderzoek en innovatie
  • 7. The ideal search application We want all relevant search results Without any false positives In practice a balance between the two User friendly Clear list of results Find desired result fast Jump to relevant part in video  Time coded metadata 7 VRT-medialab: onderzoek en innovatie
  • 8. Sources used for Trouvaille Current metadata archive (Basisplus) Media Asset Management News feed System (Ardome) News broadcasts (Ardome) + VRT video-Thesaurus + List of Belgian cities 8 VRT-medialab: onderzoek en innovatie
  • 9. Trouvaille Demo 9 VRT-medialab: onderzoek en innovatie
  • 10. Annotation Manual annotation is a lot of work assist archivist with modern tools – Annotate efficiently – As much automation as possible Need for time coded metadata  Metadata describes the complete video item  Time codes could mark the time range on which the metadata applies. A lot of metadata is lost during production process  Teletext subtitles  News rundown  Scripts with shots / scenes and their contents 10 VRT-medialab: onderzoek en innovatie
  • 11. annotation 11 VRT-medialab: onderzoek en innovatie
  • 12. annotation Assist archivist by partial automation  Split video in shots, audio segments, scenes – Archivist can easily add time coded metadata to a scene – Relevant thumbnail for every shot  Detection, even recognition of faces 12 VRT-medialab: onderzoek en innovatie
  • 13. annotation Assist archivist by partial automation  Split video in shots, audio segments, scenes  Detection, even recognition of faces  Detect reuse and duplicates 13 VRT-medialab: onderzoek en innovatie
  • 14. Conclusions New search interface Trouvaille  User friendly  Get desired result fast  Range of views Result is a complete item in Trouvaille  Annotation with time coded metadata (per scene)  Search results should use this time codes 14 VRT-medialab: onderzoek en innovatie
  • 15. Conclusions Annotation application  Partial automation → efficient annotation  Time coded metadata Too often reverse engineering of metadata that was readily available at some stage during production process 15 VRT-medialab: onderzoek en innovatie