SlideShare a Scribd company logo
Semantics Meets UX:
Mediating Intelligent Indexing of Consumers’
Multimedia Collections for Multifaceted Visualization
and Media Creation
Stacie Hibino, Alex Loui, Mark D. Wood, Sam Fryer, Cathleen Cerosaletti
Eastman Kodak Company
{hibino, alexander.loui, mdw, samuel.fryer, cathleen.cerosaletti}@kodak.com

ISM 2008 Conference Presentation
16 December 2008
Overview
  Problem Statement and Approach
  SSDF Architecture
  Semantic Indexing: Automatically Derived Metadata
  Inferencing Engine
  Workflow and User Experience: Browse and Search
  GUI Based on Semantic Indexing
  Media Creation
  Summary




©Eastman   Kodak Company, 2008   2
Problem Statement and Approach
  Problem
    • Increasing size of consumer photo and video clip collections
    • Desire by users to experience their collections, not organize them
  Approach
    • Semantic System Demonstration Framework (SSDF)
       – Flexible and extensible framework
       – Combines multiple semantic indexing algorithms for consumer
         photo and video clip collections into one integrated system
    • Koi, an SSDF desktop client application
       – Creates a user experience designed to mediate and leverage the
         intelligent indexing incorporated in SSDF
    • Together, Koi and SSDF empower users to experience their personal
      multimedia in novel and sophisticated ways
©Eastman   Kodak Company, 2008       3
SSDF Architecture
                                                     …            Client
                                      Client
                                                   Network


                                                Web Server
                                                                 …
                                 Asset       Query
                 Upload
                                 Servlet   Processor
                                              Operating System


                  System
                  Manager

                           Semantic
                          Indexer(s)
                                                                           Asset
                                                               Database
                                                      Triple
                                       Inference                           Store
                                                      Store
                                        Engine



©Eastman   Kodak Company, 2008                        4
Semantic Indexing Overview
  Goal
    • Create rich semantic descriptions of multimedia content


  Technology Needs
    • Detection and recognition of faces and people
    • Detection and recognition of typical consumer events
    • Detection of location of an event
    • Search and retrieve images and videos based on content
    • Sorting of images and videos based on specific image characteristics
    • Classification of scene for image retrieval



©Eastman   Kodak Company, 2008         5
People Recognition
                                 • Unconstrained imaging conditions and pose
                                 • Repeated examples of same small set under different conditions
                                 • Family members spanning all ages




                                                                                 Facial
            Detect Faces
                                                                               Features

     Feature Extraction


      Recognize Faces                                                      Classification
                                                                              Results
                   Face
                                         Madeline              Alicia
                 Database

©Eastman   Kodak Company, 2008                 6
Event Clustering and Auto-Labeling
Unorganized collection
                                             Use Date, Time, Metadata,
                                            and Image Content Analysis
                                         to Automatically Group and Label


                                                              Organized event grouping with label

                                                               Label    May 28, 2002
                                 Super event




                                                                                         Label      Memorial Day




©Eastman   Kodak Company, 2008                            7
CBIR Workflow
                                 Primitive Features: color histogram,
                                 Histogram of color composition, etc.



                                              ROI
                                            Primitive
                                            Features

                                                                        Sorted
                                                        Compare         Display
                                                                          List

                                   Feature Sets
                                   from global,
                      Image          tiles, and
                     Database       tile groups


©Eastman   Kodak Company, 2008          8
Image Value Index (IVI)
To automatically compute a set of probabilistic
measures of image content and semantic characteristics
comprising the following elements

                                 Measures based on technical quality (sharpness
 Technical
                                 and contrast)
                                 Measures based on aesthetic quality
 Aesthetic
                                 Measures based on social significance of people
 Social
                                 in image (via face recognition and social
                                 relationship)
                                 Measures based on occasion of image acquisition
 Event
                                 Measures based on user responses and image
 Usage
                                 use


©Eastman   Kodak Company, 2008                9
Inferencing Engine                                                        A1
                                           ContainsPerson

                                                               Soccer
  Higher order reasoning                                                       CapturedOn
  supported by modeling                                                                2008-12-15
                                                                  Likes
  people and asset metadata as                    P1
                                                                                   HasIVI
  a semantic network
                                                          HasGender
                                      ParentOf
    • Enables inference of                                                         4
                                                              Female
      relationships between people
                                          P2
    • Enables intelligent, semi-                                               HasEventType
                                                          ParentOf
      automated media creation
                                                                               Sports
  Semantic network
                                      SpouseOf
  represented using RDF data
                                                             P3
  model
                                                                          HasName
                                               ParentOf
  Inferencing based on Prolog-
                                                                                       Alice
  like query language                            P4
                                                                  HasGender

                                                                          Female



©Eastman   Kodak Company, 2008       10
Workflow and User Experience Overview
  The Koi client user experience was designed to
  mediate the results of automated indexing by
    • Presenting multiple views of the same data and corresponding
      interactions to leverage strengths of individual and combined
      algorithm results
    • Supporting multifaceted browsing
    • Enabling user correction in a way that is not disruptive to users’
      current activities




©Eastman   Kodak Company, 2008        11
Screen Layout
                          quick filters
   current view                                                   navigation bar
                                                  notifications
              current filter



                                                                            sort order




                                          main viewing area
      side menus
©Eastman   Kodak Company, 2008               12
Koi: Set Views
  Drifting View                               Calendar Views
                                                – All Years
    • Displays 12−20 media at a time,
      drifting horizontally across              – Single Year
      screen
                                                – Month over Multiple Years
    • Size, speed, and vertical
                                                – Single Month
      placement of media are
                                                – Day View
      randomly determined
                                              • Users can navigate between
    • Video clips auto play without
                                                calendar views by clicking a
      audio as they drift across the
                                                year, month, or day
      screen, adding surprise
                                              Grid View
  Slide Show View
                                              • Standard “light table” layout for
    • Presents a slide show of the
                                                efficiently viewing many media
      currently filtered set of assets


©Eastman   Kodak Company, 2008           13
Set View: Calendar View – All Years




©Eastman   Kodak Company, 2008   14
Set View: Calendar View – Month Over Yrs




©Eastman   Kodak Company, 2008   15
Koi: Pop-Up Views
  1. Single Picture View
      Expanded view of image or video clip

  2. Details View
      Extracted metadata with links for facet-based navigation

  3. Connection View
      Related media across four dimensions (People, Event,
      Image Similarity, Scene Type), visually depicting
                                                                 Single Picture View
      connections between media




Connection View
                                             Details View

©Eastman   Kodak Company, 2008                    16
Pop-Up View: Single Picture Details
  Extracted metadata with links for facet-based navigation




©Eastman   Kodak Company, 2008   17
Pop-Up View: Connection View
  Displays related media by: People, Event, Image Similarity, Scene Type
  Visually depicts connections between media
    • Supports connection-based browsing (click outer media to place in center)




©Eastman   Kodak Company, 2008              18
Koi: Other Views
  People Views
    All People               Shows one face thumbnail per people cluster, each
                             cluster initially set by people-clustering algorithm
    Family                   Displays family relationships
    Edit Person Enables users to enter in profile and relationship
                information for individuals
  Event View
                             Displays assets by event, using event-clustering
                             algorithm
  Everyday View (for Groups )
                             Summarizes group-based media, activities (e.g., rating,
                             notes), and presence


©Eastman   Kodak Company, 2008                  19
Media Creation and Notification
  System-Generated Media Creation
    • System looks for opportunities to create composited media “stories”
    • Story generation triggered by date or recent uploads
       – Date example: Mother’s Day is in a week; system automatically
         produces an album featuring mother and kids
       – Event example: Pictures recently uploaded from a sporting event;
         system produces sport-themed multimedia creation
    • Triggers use inferencing engine


  Notification of New Stories
    • RSS-based notification mechanism
    • Notifier provides user with story name, click to preview

©Eastman   Kodak Company, 2008          20
Summary
   SSDF architecture and Koi client provide a new
   approach to a flexible, semantically aware client-server
   architecture
   System supports multifaceted search, browse, and
   creation based on automatically extracted and
   algorithmically derived metadata
   Koi employs a novel user interface for searching and
   browsing collections
   SSDF architecture provides an extensible platform for
   adding additional intelligence
   Potential future work includes evolving existing
   algorithms to increased performance levels as well as
   integrating new user-tested algorithms

©Eastman   Kodak Company, 2008   21
Backup Slides




©Eastman   Kodak Company, 2008   22
Set View: Grid View




©Eastman   Kodak Company, 2008   23
Search Menu




©Eastman   Kodak Company, 2008   24
2008_12 ISM2008 Semantics Meets UX

More Related Content

What's hot

Expendables E-AppStore
Expendables E-AppStoreExpendables E-AppStore
Expendables E-AppStore
lobalint
 
Overview of JPA (Java Persistence API) v2.0
Overview of JPA (Java Persistence API) v2.0Overview of JPA (Java Persistence API) v2.0
Overview of JPA (Java Persistence API) v2.0
Bryan Basham
 
Sanjay Mirchandani’s KeyNote – EMC Forum India – Mumbai November 17, 2011
Sanjay Mirchandani’s KeyNote – EMC Forum India – Mumbai November 17, 2011Sanjay Mirchandani’s KeyNote – EMC Forum India – Mumbai November 17, 2011
Sanjay Mirchandani’s KeyNote – EMC Forum India – Mumbai November 17, 2011
EMC Forum India
 
Taxonomy Assessments - Part Two
Taxonomy Assessments - Part TwoTaxonomy Assessments - Part Two
Taxonomy Assessments - Part Two
Access Innovations, Inc.
 
Web Performance Acceleration with Strangeloop AS1000
Web Performance Acceleration with Strangeloop AS1000Web Performance Acceleration with Strangeloop AS1000
Web Performance Acceleration with Strangeloop AS1000
Thomas Stensitzki
 
Coveo Search - Product Overview
Coveo Search - Product OverviewCoveo Search - Product Overview
Coveo Search - Product Overview
Amplexor
 

What's hot (6)

Expendables E-AppStore
Expendables E-AppStoreExpendables E-AppStore
Expendables E-AppStore
 
Overview of JPA (Java Persistence API) v2.0
Overview of JPA (Java Persistence API) v2.0Overview of JPA (Java Persistence API) v2.0
Overview of JPA (Java Persistence API) v2.0
 
Sanjay Mirchandani’s KeyNote – EMC Forum India – Mumbai November 17, 2011
Sanjay Mirchandani’s KeyNote – EMC Forum India – Mumbai November 17, 2011Sanjay Mirchandani’s KeyNote – EMC Forum India – Mumbai November 17, 2011
Sanjay Mirchandani’s KeyNote – EMC Forum India – Mumbai November 17, 2011
 
Taxonomy Assessments - Part Two
Taxonomy Assessments - Part TwoTaxonomy Assessments - Part Two
Taxonomy Assessments - Part Two
 
Web Performance Acceleration with Strangeloop AS1000
Web Performance Acceleration with Strangeloop AS1000Web Performance Acceleration with Strangeloop AS1000
Web Performance Acceleration with Strangeloop AS1000
 
Coveo Search - Product Overview
Coveo Search - Product OverviewCoveo Search - Product Overview
Coveo Search - Product Overview
 

Viewers also liked

2013_10 BayCHI Welcome Slides
2013_10 BayCHI Welcome Slides2013_10 BayCHI Welcome Slides
2013_10 BayCHI Welcome Slides
Stacie Hibino
 
2012_08 BayCHI Welcome Slides
2012_08 BayCHI Welcome Slides2012_08 BayCHI Welcome Slides
2012_08 BayCHI Welcome Slides
Stacie Hibino
 
2009 02 BayCHI Intro Slides
2009 02 BayCHI Intro Slides2009 02 BayCHI Intro Slides
2009 02 BayCHI Intro Slides
Stacie Hibino
 
2009_04_15 BayCHI Welcome Slides
2009_04_15 BayCHI Welcome Slides2009_04_15 BayCHI Welcome Slides
2009_04_15 BayCHI Welcome Slides
Stacie Hibino
 
2012_05 BayCHI Welcome Slides
2012_05 BayCHI Welcome Slides2012_05 BayCHI Welcome Slides
2012_05 BayCHI Welcome Slides
Stacie Hibino
 
Dec 2009 BayCHI Welcome Slides
Dec 2009 BayCHI Welcome SlidesDec 2009 BayCHI Welcome Slides
Dec 2009 BayCHI Welcome Slides
Stacie Hibino
 
2014_01 BayCHI Welcome Slides
2014_01 BayCHI Welcome Slides2014_01 BayCHI Welcome Slides
2014_01 BayCHI Welcome Slides
Stacie Hibino
 
2011_04 BayCHI Welcome Slides
2011_04 BayCHI Welcome Slides2011_04 BayCHI Welcome Slides
2011_04 BayCHI Welcome Slides
Stacie Hibino
 
2008_12 ISM2008 Reminiscing View presentation
2008_12 ISM2008 Reminiscing View presentation2008_12 ISM2008 Reminiscing View presentation
2008_12 ISM2008 Reminiscing View presentation
Stacie Hibino
 
2013_09_10 BayCHI Welcome Slides
2013_09_10 BayCHI Welcome Slides2013_09_10 BayCHI Welcome Slides
2013_09_10 BayCHI Welcome Slides
Stacie Hibino
 
2009_08_11 BayCHI Welcome slides
2009_08_11 BayCHI Welcome slides2009_08_11 BayCHI Welcome slides
2009_08_11 BayCHI Welcome slides
Stacie Hibino
 
2015_06_09 BayCHI Welcome Slides
2015_06_09 BayCHI Welcome Slides2015_06_09 BayCHI Welcome Slides
2015_06_09 BayCHI Welcome Slides
Stacie Hibino
 
2009_04_14 BayCHI Welcome Slides
2009_04_14 BayCHI Welcome Slides2009_04_14 BayCHI Welcome Slides
2009_04_14 BayCHI Welcome Slides
Stacie Hibino
 
2009_06_09 BayCHI Welcome Slides
2009_06_09 BayCHI Welcome Slides2009_06_09 BayCHI Welcome Slides
2009_06_09 BayCHI Welcome Slides
Stacie Hibino
 
January 2010 BayCHI Welcome Slides
January 2010 BayCHI Welcome SlidesJanuary 2010 BayCHI Welcome Slides
January 2010 BayCHI Welcome Slides
Stacie Hibino
 
2015_08 BayCHI Welcome Slides
2015_08 BayCHI Welcome Slides2015_08 BayCHI Welcome Slides
2015_08 BayCHI Welcome Slides
Stacie Hibino
 
2011_07_12 BayCHI Welcome Slides
2011_07_12 BayCHI Welcome Slides2011_07_12 BayCHI Welcome Slides
2011_07_12 BayCHI Welcome Slides
Stacie Hibino
 
2011_11_08 BayCHI Welcome Slides
2011_11_08 BayCHI Welcome Slides2011_11_08 BayCHI Welcome Slides
2011_11_08 BayCHI Welcome Slides
Stacie Hibino
 
04 Bay Chi Welcome 14 Apr2009
04 Bay Chi Welcome 14 Apr200904 Bay Chi Welcome 14 Apr2009
04 Bay Chi Welcome 14 Apr2009
Stacie Hibino
 

Viewers also liked (19)

2013_10 BayCHI Welcome Slides
2013_10 BayCHI Welcome Slides2013_10 BayCHI Welcome Slides
2013_10 BayCHI Welcome Slides
 
2012_08 BayCHI Welcome Slides
2012_08 BayCHI Welcome Slides2012_08 BayCHI Welcome Slides
2012_08 BayCHI Welcome Slides
 
2009 02 BayCHI Intro Slides
2009 02 BayCHI Intro Slides2009 02 BayCHI Intro Slides
2009 02 BayCHI Intro Slides
 
2009_04_15 BayCHI Welcome Slides
2009_04_15 BayCHI Welcome Slides2009_04_15 BayCHI Welcome Slides
2009_04_15 BayCHI Welcome Slides
 
2012_05 BayCHI Welcome Slides
2012_05 BayCHI Welcome Slides2012_05 BayCHI Welcome Slides
2012_05 BayCHI Welcome Slides
 
Dec 2009 BayCHI Welcome Slides
Dec 2009 BayCHI Welcome SlidesDec 2009 BayCHI Welcome Slides
Dec 2009 BayCHI Welcome Slides
 
2014_01 BayCHI Welcome Slides
2014_01 BayCHI Welcome Slides2014_01 BayCHI Welcome Slides
2014_01 BayCHI Welcome Slides
 
2011_04 BayCHI Welcome Slides
2011_04 BayCHI Welcome Slides2011_04 BayCHI Welcome Slides
2011_04 BayCHI Welcome Slides
 
2008_12 ISM2008 Reminiscing View presentation
2008_12 ISM2008 Reminiscing View presentation2008_12 ISM2008 Reminiscing View presentation
2008_12 ISM2008 Reminiscing View presentation
 
2013_09_10 BayCHI Welcome Slides
2013_09_10 BayCHI Welcome Slides2013_09_10 BayCHI Welcome Slides
2013_09_10 BayCHI Welcome Slides
 
2009_08_11 BayCHI Welcome slides
2009_08_11 BayCHI Welcome slides2009_08_11 BayCHI Welcome slides
2009_08_11 BayCHI Welcome slides
 
2015_06_09 BayCHI Welcome Slides
2015_06_09 BayCHI Welcome Slides2015_06_09 BayCHI Welcome Slides
2015_06_09 BayCHI Welcome Slides
 
2009_04_14 BayCHI Welcome Slides
2009_04_14 BayCHI Welcome Slides2009_04_14 BayCHI Welcome Slides
2009_04_14 BayCHI Welcome Slides
 
2009_06_09 BayCHI Welcome Slides
2009_06_09 BayCHI Welcome Slides2009_06_09 BayCHI Welcome Slides
2009_06_09 BayCHI Welcome Slides
 
January 2010 BayCHI Welcome Slides
January 2010 BayCHI Welcome SlidesJanuary 2010 BayCHI Welcome Slides
January 2010 BayCHI Welcome Slides
 
2015_08 BayCHI Welcome Slides
2015_08 BayCHI Welcome Slides2015_08 BayCHI Welcome Slides
2015_08 BayCHI Welcome Slides
 
2011_07_12 BayCHI Welcome Slides
2011_07_12 BayCHI Welcome Slides2011_07_12 BayCHI Welcome Slides
2011_07_12 BayCHI Welcome Slides
 
2011_11_08 BayCHI Welcome Slides
2011_11_08 BayCHI Welcome Slides2011_11_08 BayCHI Welcome Slides
2011_11_08 BayCHI Welcome Slides
 
04 Bay Chi Welcome 14 Apr2009
04 Bay Chi Welcome 14 Apr200904 Bay Chi Welcome 14 Apr2009
04 Bay Chi Welcome 14 Apr2009
 

Similar to 2008_12 ISM2008 Semantics Meets UX

Wpf Tech Overview2009
Wpf Tech Overview2009Wpf Tech Overview2009
Wpf Tech Overview2009
Our Community Exchange LLC
 
Bynet2.3 Adobe Flex builder 4
Bynet2.3 Adobe Flex builder 4Bynet2.3 Adobe Flex builder 4
Bynet2.3 Adobe Flex builder 4
Транслируем.бел
 
Process Project Mgt Seminar 8 Apr 2009(2)
Process Project Mgt Seminar 8 Apr 2009(2)Process Project Mgt Seminar 8 Apr 2009(2)
Process Project Mgt Seminar 8 Apr 2009(2)
avitale1998
 
Good Data: Collaborative Analytics On Demand
Good Data: Collaborative Analytics On DemandGood Data: Collaborative Analytics On Demand
Good Data: Collaborative Analytics On Demand
zsvoboda
 
Leadership Symposium on Digital Media in Healthcare
Leadership Symposium on Digital Media in HealthcareLeadership Symposium on Digital Media in Healthcare
Leadership Symposium on Digital Media in Healthcare
setstanford
 
Keynote at Depsa07 - architectural view of event processing
Keynote at Depsa07 - architectural view of event processingKeynote at Depsa07 - architectural view of event processing
Keynote at Depsa07 - architectural view of event processing
Opher Etzion
 
Aras PLM Roadmap
Aras PLM RoadmapAras PLM Roadmap
Aras PLM Roadmap
Aras
 
Practical Functional Javascript
Practical Functional JavascriptPractical Functional Javascript
Practical Functional Javascript
Oliver Steele
 
6.Live Framework 和Mesh Services
6.Live Framework 和Mesh Services6.Live Framework 和Mesh Services
6.Live Framework 和Mesh Services
GaryYoung
 
Unified big data architecture
Unified big data architectureUnified big data architecture
Unified big data architecture
DataWorks Summit
 
Session #1: Development Practices And The Microsoft Approach
Session #1: Development Practices And The Microsoft ApproachSession #1: Development Practices And The Microsoft Approach
Session #1: Development Practices And The Microsoft Approach
Steve Lange
 
Infomation models for agile bi
Infomation models for agile biInfomation models for agile bi
Infomation models for agile bi
Ehtisham Rao
 
PCI Geomatics Overview
PCI Geomatics OverviewPCI Geomatics Overview
PCI Geomatics Overview
Pci Geomatics
 
HCLT Brochure: E-Discovery and Document Review Solutions
HCLT Brochure: E-Discovery and Document Review SolutionsHCLT Brochure: E-Discovery and Document Review Solutions
HCLT Brochure: E-Discovery and Document Review Solutions
HCL Technologies
 
Identifying Auxiliary Web Images Using Combinations of Analyses
Identifying Auxiliary Web Images Using Combinations of AnalysesIdentifying Auxiliary Web Images Using Combinations of Analyses
Identifying Auxiliary Web Images Using Combinations of Analyses
Tewson Seeoun
 
Model Driven Architecture (MDA): Motivations, Status & Future
Model Driven Architecture (MDA): Motivations, Status & FutureModel Driven Architecture (MDA): Motivations, Status & Future
Model Driven Architecture (MDA): Motivations, Status & Future
elliando dias
 
JasperSoft and GlassFish
JasperSoft and GlassFishJasperSoft and GlassFish
JasperSoft and GlassFish
Eduardo Pelegri-Llopart
 
ICSM08a.ppt
ICSM08a.pptICSM08a.ppt
ICSM08a.ppt
Ptidej Team
 
Finding Performance Bottleneck Instantly.
Finding Performance Bottleneck Instantly.Finding Performance Bottleneck Instantly.
Finding Performance Bottleneck Instantly.
Kiran Badi
 
Modular applications with montage components
Modular applications with montage componentsModular applications with montage components
Modular applications with montage components
Benoit Marchant
 

Similar to 2008_12 ISM2008 Semantics Meets UX (20)

Wpf Tech Overview2009
Wpf Tech Overview2009Wpf Tech Overview2009
Wpf Tech Overview2009
 
Bynet2.3 Adobe Flex builder 4
Bynet2.3 Adobe Flex builder 4Bynet2.3 Adobe Flex builder 4
Bynet2.3 Adobe Flex builder 4
 
Process Project Mgt Seminar 8 Apr 2009(2)
Process Project Mgt Seminar 8 Apr 2009(2)Process Project Mgt Seminar 8 Apr 2009(2)
Process Project Mgt Seminar 8 Apr 2009(2)
 
Good Data: Collaborative Analytics On Demand
Good Data: Collaborative Analytics On DemandGood Data: Collaborative Analytics On Demand
Good Data: Collaborative Analytics On Demand
 
Leadership Symposium on Digital Media in Healthcare
Leadership Symposium on Digital Media in HealthcareLeadership Symposium on Digital Media in Healthcare
Leadership Symposium on Digital Media in Healthcare
 
Keynote at Depsa07 - architectural view of event processing
Keynote at Depsa07 - architectural view of event processingKeynote at Depsa07 - architectural view of event processing
Keynote at Depsa07 - architectural view of event processing
 
Aras PLM Roadmap
Aras PLM RoadmapAras PLM Roadmap
Aras PLM Roadmap
 
Practical Functional Javascript
Practical Functional JavascriptPractical Functional Javascript
Practical Functional Javascript
 
6.Live Framework 和Mesh Services
6.Live Framework 和Mesh Services6.Live Framework 和Mesh Services
6.Live Framework 和Mesh Services
 
Unified big data architecture
Unified big data architectureUnified big data architecture
Unified big data architecture
 
Session #1: Development Practices And The Microsoft Approach
Session #1: Development Practices And The Microsoft ApproachSession #1: Development Practices And The Microsoft Approach
Session #1: Development Practices And The Microsoft Approach
 
Infomation models for agile bi
Infomation models for agile biInfomation models for agile bi
Infomation models for agile bi
 
PCI Geomatics Overview
PCI Geomatics OverviewPCI Geomatics Overview
PCI Geomatics Overview
 
HCLT Brochure: E-Discovery and Document Review Solutions
HCLT Brochure: E-Discovery and Document Review SolutionsHCLT Brochure: E-Discovery and Document Review Solutions
HCLT Brochure: E-Discovery and Document Review Solutions
 
Identifying Auxiliary Web Images Using Combinations of Analyses
Identifying Auxiliary Web Images Using Combinations of AnalysesIdentifying Auxiliary Web Images Using Combinations of Analyses
Identifying Auxiliary Web Images Using Combinations of Analyses
 
Model Driven Architecture (MDA): Motivations, Status & Future
Model Driven Architecture (MDA): Motivations, Status & FutureModel Driven Architecture (MDA): Motivations, Status & Future
Model Driven Architecture (MDA): Motivations, Status & Future
 
JasperSoft and GlassFish
JasperSoft and GlassFishJasperSoft and GlassFish
JasperSoft and GlassFish
 
ICSM08a.ppt
ICSM08a.pptICSM08a.ppt
ICSM08a.ppt
 
Finding Performance Bottleneck Instantly.
Finding Performance Bottleneck Instantly.Finding Performance Bottleneck Instantly.
Finding Performance Bottleneck Instantly.
 
Modular applications with montage components
Modular applications with montage componentsModular applications with montage components
Modular applications with montage components
 

More from Stacie Hibino

2017_02 BayCHI welcome slides
2017_02 BayCHI welcome slides2017_02 BayCHI welcome slides
2017_02 BayCHI welcome slides
Stacie Hibino
 
2016_10 BayCHI Welcome Slides
2016_10 BayCHI Welcome Slides2016_10 BayCHI Welcome Slides
2016_10 BayCHI Welcome Slides
Stacie Hibino
 
2016_09 BayCHI Welcome Slides
2016_09 BayCHI Welcome Slides2016_09 BayCHI Welcome Slides
2016_09 BayCHI Welcome Slides
Stacie Hibino
 
2016_08 BayCHI Welcome Slides
2016_08 BayCHI Welcome Slides2016_08 BayCHI Welcome Slides
2016_08 BayCHI Welcome Slides
Stacie Hibino
 
2016_02 BayCHI Welcome Slides
2016_02 BayCHI Welcome Slides2016_02 BayCHI Welcome Slides
2016_02 BayCHI Welcome Slides
Stacie Hibino
 
2016_01 BayCHI welcome slides
2016_01 BayCHI welcome slides2016_01 BayCHI welcome slides
2016_01 BayCHI welcome slides
Stacie Hibino
 
2015_12 BayCHI Welcome Slides
2015_12 BayCHI Welcome Slides2015_12 BayCHI Welcome Slides
2015_12 BayCHI Welcome Slides
Stacie Hibino
 
Come Hack With Us: A Hardware Hackathon at GHC
Come Hack With Us: A Hardware Hackathon at GHCCome Hack With Us: A Hardware Hackathon at GHC
Come Hack With Us: A Hardware Hackathon at GHC
Stacie Hibino
 
2015_09_08 BayCHI Welcome slides
2015_09_08 BayCHI Welcome slides2015_09_08 BayCHI Welcome slides
2015_09_08 BayCHI Welcome slides
Stacie Hibino
 
2015_07 BayCHI Welcome Slides
2015_07 BayCHI Welcome Slides2015_07 BayCHI Welcome Slides
2015_07 BayCHI Welcome Slides
Stacie Hibino
 
2015_05 BayCHI Welcome Slides
2015_05 BayCHI Welcome Slides2015_05 BayCHI Welcome Slides
2015_05 BayCHI Welcome Slides
Stacie Hibino
 
2015_03_10 BayCHI Welcome Slides
2015_03_10 BayCHI Welcome Slides2015_03_10 BayCHI Welcome Slides
2015_03_10 BayCHI Welcome Slides
Stacie Hibino
 
2015_02 BayCHI Welcome Slides
2015_02 BayCHI Welcome Slides2015_02 BayCHI Welcome Slides
2015_02 BayCHI Welcome Slides
Stacie Hibino
 
2015_01_13 BayCHI Welcome slides
2015_01_13 BayCHI Welcome slides2015_01_13 BayCHI Welcome slides
2015_01_13 BayCHI Welcome slides
Stacie Hibino
 
2013_12 BayCHI Welcome Slides
2013_12 BayCHI Welcome Slides2013_12 BayCHI Welcome Slides
2013_12 BayCHI Welcome Slides
Stacie Hibino
 
2013_11 Monthly BayCHI Meeting
2013_11 Monthly BayCHI Meeting2013_11 Monthly BayCHI Meeting
2013_11 Monthly BayCHI Meeting
Stacie Hibino
 
2013_08 BayCHI Welcome Slides
2013_08 BayCHI Welcome Slides2013_08 BayCHI Welcome Slides
2013_08 BayCHI Welcome Slides
Stacie Hibino
 
2013_07 BayCHI Welcome Slides
2013_07 BayCHI Welcome Slides2013_07 BayCHI Welcome Slides
2013_07 BayCHI Welcome Slides
Stacie Hibino
 
2013_06 BayCHI Welcome Slides
2013_06 BayCHI Welcome Slides2013_06 BayCHI Welcome Slides
2013_06 BayCHI Welcome Slides
Stacie Hibino
 
2013_05 BayCHI Welcome Slides
2013_05 BayCHI Welcome Slides2013_05 BayCHI Welcome Slides
2013_05 BayCHI Welcome Slides
Stacie Hibino
 

More from Stacie Hibino (20)

2017_02 BayCHI welcome slides
2017_02 BayCHI welcome slides2017_02 BayCHI welcome slides
2017_02 BayCHI welcome slides
 
2016_10 BayCHI Welcome Slides
2016_10 BayCHI Welcome Slides2016_10 BayCHI Welcome Slides
2016_10 BayCHI Welcome Slides
 
2016_09 BayCHI Welcome Slides
2016_09 BayCHI Welcome Slides2016_09 BayCHI Welcome Slides
2016_09 BayCHI Welcome Slides
 
2016_08 BayCHI Welcome Slides
2016_08 BayCHI Welcome Slides2016_08 BayCHI Welcome Slides
2016_08 BayCHI Welcome Slides
 
2016_02 BayCHI Welcome Slides
2016_02 BayCHI Welcome Slides2016_02 BayCHI Welcome Slides
2016_02 BayCHI Welcome Slides
 
2016_01 BayCHI welcome slides
2016_01 BayCHI welcome slides2016_01 BayCHI welcome slides
2016_01 BayCHI welcome slides
 
2015_12 BayCHI Welcome Slides
2015_12 BayCHI Welcome Slides2015_12 BayCHI Welcome Slides
2015_12 BayCHI Welcome Slides
 
Come Hack With Us: A Hardware Hackathon at GHC
Come Hack With Us: A Hardware Hackathon at GHCCome Hack With Us: A Hardware Hackathon at GHC
Come Hack With Us: A Hardware Hackathon at GHC
 
2015_09_08 BayCHI Welcome slides
2015_09_08 BayCHI Welcome slides2015_09_08 BayCHI Welcome slides
2015_09_08 BayCHI Welcome slides
 
2015_07 BayCHI Welcome Slides
2015_07 BayCHI Welcome Slides2015_07 BayCHI Welcome Slides
2015_07 BayCHI Welcome Slides
 
2015_05 BayCHI Welcome Slides
2015_05 BayCHI Welcome Slides2015_05 BayCHI Welcome Slides
2015_05 BayCHI Welcome Slides
 
2015_03_10 BayCHI Welcome Slides
2015_03_10 BayCHI Welcome Slides2015_03_10 BayCHI Welcome Slides
2015_03_10 BayCHI Welcome Slides
 
2015_02 BayCHI Welcome Slides
2015_02 BayCHI Welcome Slides2015_02 BayCHI Welcome Slides
2015_02 BayCHI Welcome Slides
 
2015_01_13 BayCHI Welcome slides
2015_01_13 BayCHI Welcome slides2015_01_13 BayCHI Welcome slides
2015_01_13 BayCHI Welcome slides
 
2013_12 BayCHI Welcome Slides
2013_12 BayCHI Welcome Slides2013_12 BayCHI Welcome Slides
2013_12 BayCHI Welcome Slides
 
2013_11 Monthly BayCHI Meeting
2013_11 Monthly BayCHI Meeting2013_11 Monthly BayCHI Meeting
2013_11 Monthly BayCHI Meeting
 
2013_08 BayCHI Welcome Slides
2013_08 BayCHI Welcome Slides2013_08 BayCHI Welcome Slides
2013_08 BayCHI Welcome Slides
 
2013_07 BayCHI Welcome Slides
2013_07 BayCHI Welcome Slides2013_07 BayCHI Welcome Slides
2013_07 BayCHI Welcome Slides
 
2013_06 BayCHI Welcome Slides
2013_06 BayCHI Welcome Slides2013_06 BayCHI Welcome Slides
2013_06 BayCHI Welcome Slides
 
2013_05 BayCHI Welcome Slides
2013_05 BayCHI Welcome Slides2013_05 BayCHI Welcome Slides
2013_05 BayCHI Welcome Slides
 

Recently uploaded

Implementations of Fused Deposition Modeling in real world
Implementations of Fused Deposition Modeling  in real worldImplementations of Fused Deposition Modeling  in real world
Implementations of Fused Deposition Modeling in real world
Emerging Tech
 
CiscoIconsLibrary cours de réseau VLAN.ppt
CiscoIconsLibrary cours de réseau VLAN.pptCiscoIconsLibrary cours de réseau VLAN.ppt
CiscoIconsLibrary cours de réseau VLAN.ppt
moinahousna
 
Amul milk launches in US: Key details of its new products ...
Amul milk launches in US: Key details of its new products ...Amul milk launches in US: Key details of its new products ...
Amul milk launches in US: Key details of its new products ...
chetankumar9855
 
Using LLM Agents with Llama 3, LangGraph and Milvus
Using LLM Agents with Llama 3, LangGraph and MilvusUsing LLM Agents with Llama 3, LangGraph and Milvus
Using LLM Agents with Llama 3, LangGraph and Milvus
Zilliz
 
Best Practices for Effectively Running dbt in Airflow.pdf
Best Practices for Effectively Running dbt in Airflow.pdfBest Practices for Effectively Running dbt in Airflow.pdf
Best Practices for Effectively Running dbt in Airflow.pdf
Tatiana Al-Chueyr
 
Litestack talk at Brighton 2024 (Unleashing the power of SQLite for Ruby apps)
Litestack talk at Brighton 2024 (Unleashing the power of SQLite for Ruby apps)Litestack talk at Brighton 2024 (Unleashing the power of SQLite for Ruby apps)
Litestack talk at Brighton 2024 (Unleashing the power of SQLite for Ruby apps)
Muhammad Ali
 
CHAPTER-8 COMPONENTS OF COMPUTER SYSTEM CLASS 9 CBSE
CHAPTER-8 COMPONENTS OF COMPUTER SYSTEM CLASS 9 CBSECHAPTER-8 COMPONENTS OF COMPUTER SYSTEM CLASS 9 CBSE
CHAPTER-8 COMPONENTS OF COMPUTER SYSTEM CLASS 9 CBSE
kumarjarun2010
 
RPA In Healthcare Benefits, Use Case, Trend And Challenges 2024.pptx
RPA In Healthcare Benefits, Use Case, Trend And Challenges 2024.pptxRPA In Healthcare Benefits, Use Case, Trend And Challenges 2024.pptx
RPA In Healthcare Benefits, Use Case, Trend And Challenges 2024.pptx
SynapseIndia
 
Google I/O Extended Harare Merged Slides
Google I/O Extended Harare Merged SlidesGoogle I/O Extended Harare Merged Slides
Google I/O Extended Harare Merged Slides
Google Developer Group - Harare
 
July Patch Tuesday
July Patch TuesdayJuly Patch Tuesday
July Patch Tuesday
Ivanti
 
The Role of IoT in Australian Mobile App Development - PDF Guide
The Role of IoT in Australian Mobile App Development - PDF GuideThe Role of IoT in Australian Mobile App Development - PDF Guide
The Role of IoT in Australian Mobile App Development - PDF Guide
Shiv Technolabs
 
Girls Call Churchgate 9910780858 Provide Best And Top Girl Service And No1 in...
Girls Call Churchgate 9910780858 Provide Best And Top Girl Service And No1 in...Girls Call Churchgate 9910780858 Provide Best And Top Girl Service And No1 in...
Girls Call Churchgate 9910780858 Provide Best And Top Girl Service And No1 in...
maigasapphire
 
Calgary MuleSoft Meetup APM and IDP .pptx
Calgary MuleSoft Meetup APM and IDP .pptxCalgary MuleSoft Meetup APM and IDP .pptx
Calgary MuleSoft Meetup APM and IDP .pptx
ishalveerrandhawa1
 
“Deploying Large Language Models on a Raspberry Pi,” a Presentation from Usef...
“Deploying Large Language Models on a Raspberry Pi,” a Presentation from Usef...“Deploying Large Language Models on a Raspberry Pi,” a Presentation from Usef...
“Deploying Large Language Models on a Raspberry Pi,” a Presentation from Usef...
Edge AI and Vision Alliance
 
High Profile Girls Call ServiCe Hyderabad 0000000000 Tanisha Best High Class ...
High Profile Girls Call ServiCe Hyderabad 0000000000 Tanisha Best High Class ...High Profile Girls Call ServiCe Hyderabad 0000000000 Tanisha Best High Class ...
High Profile Girls Call ServiCe Hyderabad 0000000000 Tanisha Best High Class ...
aslasdfmkhan4750
 
Active Inference is a veryyyyyyyyyyyyyyyyyyyyyyyy
Active Inference is a veryyyyyyyyyyyyyyyyyyyyyyyyActive Inference is a veryyyyyyyyyyyyyyyyyyyyyyyy
Active Inference is a veryyyyyyyyyyyyyyyyyyyyyyyy
RaminGhanbari2
 
Acumatica vs. Sage Intacct vs. NetSuite _ NOW CFO.pdf
Acumatica vs. Sage Intacct vs. NetSuite _ NOW CFO.pdfAcumatica vs. Sage Intacct vs. NetSuite _ NOW CFO.pdf
Acumatica vs. Sage Intacct vs. NetSuite _ NOW CFO.pdf
BrainSell Technologies
 
Opencast Summit 2024 — Opencast @ University of Münster
Opencast Summit 2024 — Opencast @ University of MünsterOpencast Summit 2024 — Opencast @ University of Münster
Opencast Summit 2024 — Opencast @ University of Münster
Matthias Neugebauer
 
(CISOPlatform Summit & SACON 2024) Digital Personal Data Protection Act.pdf
(CISOPlatform Summit & SACON 2024) Digital Personal Data Protection Act.pdf(CISOPlatform Summit & SACON 2024) Digital Personal Data Protection Act.pdf
(CISOPlatform Summit & SACON 2024) Digital Personal Data Protection Act.pdf
Priyanka Aash
 
High Profile Girls call Service Pune 000XX00000 Provide Best And Top Girl Ser...
High Profile Girls call Service Pune 000XX00000 Provide Best And Top Girl Ser...High Profile Girls call Service Pune 000XX00000 Provide Best And Top Girl Ser...
High Profile Girls call Service Pune 000XX00000 Provide Best And Top Girl Ser...
bhumivarma35300
 

Recently uploaded (20)

Implementations of Fused Deposition Modeling in real world
Implementations of Fused Deposition Modeling  in real worldImplementations of Fused Deposition Modeling  in real world
Implementations of Fused Deposition Modeling in real world
 
CiscoIconsLibrary cours de réseau VLAN.ppt
CiscoIconsLibrary cours de réseau VLAN.pptCiscoIconsLibrary cours de réseau VLAN.ppt
CiscoIconsLibrary cours de réseau VLAN.ppt
 
Amul milk launches in US: Key details of its new products ...
Amul milk launches in US: Key details of its new products ...Amul milk launches in US: Key details of its new products ...
Amul milk launches in US: Key details of its new products ...
 
Using LLM Agents with Llama 3, LangGraph and Milvus
Using LLM Agents with Llama 3, LangGraph and MilvusUsing LLM Agents with Llama 3, LangGraph and Milvus
Using LLM Agents with Llama 3, LangGraph and Milvus
 
Best Practices for Effectively Running dbt in Airflow.pdf
Best Practices for Effectively Running dbt in Airflow.pdfBest Practices for Effectively Running dbt in Airflow.pdf
Best Practices for Effectively Running dbt in Airflow.pdf
 
Litestack talk at Brighton 2024 (Unleashing the power of SQLite for Ruby apps)
Litestack talk at Brighton 2024 (Unleashing the power of SQLite for Ruby apps)Litestack talk at Brighton 2024 (Unleashing the power of SQLite for Ruby apps)
Litestack talk at Brighton 2024 (Unleashing the power of SQLite for Ruby apps)
 
CHAPTER-8 COMPONENTS OF COMPUTER SYSTEM CLASS 9 CBSE
CHAPTER-8 COMPONENTS OF COMPUTER SYSTEM CLASS 9 CBSECHAPTER-8 COMPONENTS OF COMPUTER SYSTEM CLASS 9 CBSE
CHAPTER-8 COMPONENTS OF COMPUTER SYSTEM CLASS 9 CBSE
 
RPA In Healthcare Benefits, Use Case, Trend And Challenges 2024.pptx
RPA In Healthcare Benefits, Use Case, Trend And Challenges 2024.pptxRPA In Healthcare Benefits, Use Case, Trend And Challenges 2024.pptx
RPA In Healthcare Benefits, Use Case, Trend And Challenges 2024.pptx
 
Google I/O Extended Harare Merged Slides
Google I/O Extended Harare Merged SlidesGoogle I/O Extended Harare Merged Slides
Google I/O Extended Harare Merged Slides
 
July Patch Tuesday
July Patch TuesdayJuly Patch Tuesday
July Patch Tuesday
 
The Role of IoT in Australian Mobile App Development - PDF Guide
The Role of IoT in Australian Mobile App Development - PDF GuideThe Role of IoT in Australian Mobile App Development - PDF Guide
The Role of IoT in Australian Mobile App Development - PDF Guide
 
Girls Call Churchgate 9910780858 Provide Best And Top Girl Service And No1 in...
Girls Call Churchgate 9910780858 Provide Best And Top Girl Service And No1 in...Girls Call Churchgate 9910780858 Provide Best And Top Girl Service And No1 in...
Girls Call Churchgate 9910780858 Provide Best And Top Girl Service And No1 in...
 
Calgary MuleSoft Meetup APM and IDP .pptx
Calgary MuleSoft Meetup APM and IDP .pptxCalgary MuleSoft Meetup APM and IDP .pptx
Calgary MuleSoft Meetup APM and IDP .pptx
 
“Deploying Large Language Models on a Raspberry Pi,” a Presentation from Usef...
“Deploying Large Language Models on a Raspberry Pi,” a Presentation from Usef...“Deploying Large Language Models on a Raspberry Pi,” a Presentation from Usef...
“Deploying Large Language Models on a Raspberry Pi,” a Presentation from Usef...
 
High Profile Girls Call ServiCe Hyderabad 0000000000 Tanisha Best High Class ...
High Profile Girls Call ServiCe Hyderabad 0000000000 Tanisha Best High Class ...High Profile Girls Call ServiCe Hyderabad 0000000000 Tanisha Best High Class ...
High Profile Girls Call ServiCe Hyderabad 0000000000 Tanisha Best High Class ...
 
Active Inference is a veryyyyyyyyyyyyyyyyyyyyyyyy
Active Inference is a veryyyyyyyyyyyyyyyyyyyyyyyyActive Inference is a veryyyyyyyyyyyyyyyyyyyyyyyy
Active Inference is a veryyyyyyyyyyyyyyyyyyyyyyyy
 
Acumatica vs. Sage Intacct vs. NetSuite _ NOW CFO.pdf
Acumatica vs. Sage Intacct vs. NetSuite _ NOW CFO.pdfAcumatica vs. Sage Intacct vs. NetSuite _ NOW CFO.pdf
Acumatica vs. Sage Intacct vs. NetSuite _ NOW CFO.pdf
 
Opencast Summit 2024 — Opencast @ University of Münster
Opencast Summit 2024 — Opencast @ University of MünsterOpencast Summit 2024 — Opencast @ University of Münster
Opencast Summit 2024 — Opencast @ University of Münster
 
(CISOPlatform Summit & SACON 2024) Digital Personal Data Protection Act.pdf
(CISOPlatform Summit & SACON 2024) Digital Personal Data Protection Act.pdf(CISOPlatform Summit & SACON 2024) Digital Personal Data Protection Act.pdf
(CISOPlatform Summit & SACON 2024) Digital Personal Data Protection Act.pdf
 
High Profile Girls call Service Pune 000XX00000 Provide Best And Top Girl Ser...
High Profile Girls call Service Pune 000XX00000 Provide Best And Top Girl Ser...High Profile Girls call Service Pune 000XX00000 Provide Best And Top Girl Ser...
High Profile Girls call Service Pune 000XX00000 Provide Best And Top Girl Ser...
 

2008_12 ISM2008 Semantics Meets UX

  • 1. Semantics Meets UX: Mediating Intelligent Indexing of Consumers’ Multimedia Collections for Multifaceted Visualization and Media Creation Stacie Hibino, Alex Loui, Mark D. Wood, Sam Fryer, Cathleen Cerosaletti Eastman Kodak Company {hibino, alexander.loui, mdw, samuel.fryer, cathleen.cerosaletti}@kodak.com ISM 2008 Conference Presentation 16 December 2008
  • 2. Overview Problem Statement and Approach SSDF Architecture Semantic Indexing: Automatically Derived Metadata Inferencing Engine Workflow and User Experience: Browse and Search GUI Based on Semantic Indexing Media Creation Summary ©Eastman Kodak Company, 2008 2
  • 3. Problem Statement and Approach Problem • Increasing size of consumer photo and video clip collections • Desire by users to experience their collections, not organize them Approach • Semantic System Demonstration Framework (SSDF) – Flexible and extensible framework – Combines multiple semantic indexing algorithms for consumer photo and video clip collections into one integrated system • Koi, an SSDF desktop client application – Creates a user experience designed to mediate and leverage the intelligent indexing incorporated in SSDF • Together, Koi and SSDF empower users to experience their personal multimedia in novel and sophisticated ways ©Eastman Kodak Company, 2008 3
  • 4. SSDF Architecture … Client Client Network Web Server … Asset Query Upload Servlet Processor Operating System System Manager Semantic Indexer(s) Asset Database Triple Inference Store Store Engine ©Eastman Kodak Company, 2008 4
  • 5. Semantic Indexing Overview Goal • Create rich semantic descriptions of multimedia content Technology Needs • Detection and recognition of faces and people • Detection and recognition of typical consumer events • Detection of location of an event • Search and retrieve images and videos based on content • Sorting of images and videos based on specific image characteristics • Classification of scene for image retrieval ©Eastman Kodak Company, 2008 5
  • 6. People Recognition • Unconstrained imaging conditions and pose • Repeated examples of same small set under different conditions • Family members spanning all ages Facial Detect Faces Features Feature Extraction Recognize Faces Classification Results Face Madeline Alicia Database ©Eastman Kodak Company, 2008 6
  • 7. Event Clustering and Auto-Labeling Unorganized collection Use Date, Time, Metadata, and Image Content Analysis to Automatically Group and Label Organized event grouping with label Label May 28, 2002 Super event Label Memorial Day ©Eastman Kodak Company, 2008 7
  • 8. CBIR Workflow Primitive Features: color histogram, Histogram of color composition, etc. ROI Primitive Features Sorted Compare Display List Feature Sets from global, Image tiles, and Database tile groups ©Eastman Kodak Company, 2008 8
  • 9. Image Value Index (IVI) To automatically compute a set of probabilistic measures of image content and semantic characteristics comprising the following elements Measures based on technical quality (sharpness Technical and contrast) Measures based on aesthetic quality Aesthetic Measures based on social significance of people Social in image (via face recognition and social relationship) Measures based on occasion of image acquisition Event Measures based on user responses and image Usage use ©Eastman Kodak Company, 2008 9
  • 10. Inferencing Engine A1 ContainsPerson Soccer Higher order reasoning CapturedOn supported by modeling 2008-12-15 Likes people and asset metadata as P1 HasIVI a semantic network HasGender ParentOf • Enables inference of 4 Female relationships between people P2 • Enables intelligent, semi- HasEventType ParentOf automated media creation Sports Semantic network SpouseOf represented using RDF data P3 model HasName ParentOf Inferencing based on Prolog- Alice like query language P4 HasGender Female ©Eastman Kodak Company, 2008 10
  • 11. Workflow and User Experience Overview The Koi client user experience was designed to mediate the results of automated indexing by • Presenting multiple views of the same data and corresponding interactions to leverage strengths of individual and combined algorithm results • Supporting multifaceted browsing • Enabling user correction in a way that is not disruptive to users’ current activities ©Eastman Kodak Company, 2008 11
  • 12. Screen Layout quick filters current view navigation bar notifications current filter sort order main viewing area side menus ©Eastman Kodak Company, 2008 12
  • 13. Koi: Set Views Drifting View Calendar Views – All Years • Displays 12−20 media at a time, drifting horizontally across – Single Year screen – Month over Multiple Years • Size, speed, and vertical – Single Month placement of media are – Day View randomly determined • Users can navigate between • Video clips auto play without calendar views by clicking a audio as they drift across the year, month, or day screen, adding surprise Grid View Slide Show View • Standard “light table” layout for • Presents a slide show of the efficiently viewing many media currently filtered set of assets ©Eastman Kodak Company, 2008 13
  • 14. Set View: Calendar View – All Years ©Eastman Kodak Company, 2008 14
  • 15. Set View: Calendar View – Month Over Yrs ©Eastman Kodak Company, 2008 15
  • 16. Koi: Pop-Up Views 1. Single Picture View Expanded view of image or video clip 2. Details View Extracted metadata with links for facet-based navigation 3. Connection View Related media across four dimensions (People, Event, Image Similarity, Scene Type), visually depicting Single Picture View connections between media Connection View Details View ©Eastman Kodak Company, 2008 16
  • 17. Pop-Up View: Single Picture Details Extracted metadata with links for facet-based navigation ©Eastman Kodak Company, 2008 17
  • 18. Pop-Up View: Connection View Displays related media by: People, Event, Image Similarity, Scene Type Visually depicts connections between media • Supports connection-based browsing (click outer media to place in center) ©Eastman Kodak Company, 2008 18
  • 19. Koi: Other Views People Views All People Shows one face thumbnail per people cluster, each cluster initially set by people-clustering algorithm Family Displays family relationships Edit Person Enables users to enter in profile and relationship information for individuals Event View Displays assets by event, using event-clustering algorithm Everyday View (for Groups ) Summarizes group-based media, activities (e.g., rating, notes), and presence ©Eastman Kodak Company, 2008 19
  • 20. Media Creation and Notification System-Generated Media Creation • System looks for opportunities to create composited media “stories” • Story generation triggered by date or recent uploads – Date example: Mother’s Day is in a week; system automatically produces an album featuring mother and kids – Event example: Pictures recently uploaded from a sporting event; system produces sport-themed multimedia creation • Triggers use inferencing engine Notification of New Stories • RSS-based notification mechanism • Notifier provides user with story name, click to preview ©Eastman Kodak Company, 2008 20
  • 21. Summary SSDF architecture and Koi client provide a new approach to a flexible, semantically aware client-server architecture System supports multifaceted search, browse, and creation based on automatically extracted and algorithmically derived metadata Koi employs a novel user interface for searching and browsing collections SSDF architecture provides an extensible platform for adding additional intelligence Potential future work includes evolving existing algorithms to increased performance levels as well as integrating new user-tested algorithms ©Eastman Kodak Company, 2008 21
  • 22. Backup Slides ©Eastman Kodak Company, 2008 22
  • 23. Set View: Grid View ©Eastman Kodak Company, 2008 23
  • 24. Search Menu ©Eastman Kodak Company, 2008 24