<!DOCTYPE article PUBLIC "-//NLM//DTD JATS (Z39.96) Journal Archiving and Interchange DTD v1.0 20120330//EN" "JATS-archivearticle1.dtd">
<article xmlns:xlink="http://www.w3.org/1999/xlink">
  <front>
    <journal-meta />
    <article-meta>
      <title-group>
        <article-title>Enhancing the Presentation of Multimedia using Extracted Semantics</article-title>
      </title-group>
      <contrib-group>
        <contrib contrib-type="author">
          <string-name>Hyowon Lee Guest Speech at</string-name>
          <xref ref-type="aff" rid="aff0">0</xref>
        </contrib>
        <contrib contrib-type="author">
          <string-name>st SEMPS Workshop (</string-name>
          <xref ref-type="aff" rid="aff0">0</xref>
        </contrib>
        <aff id="aff0">
          <label>0</label>
          <institution>Centre for Digital Video Processing Dublin City University</institution>
        </aff>
      </contrib-group>
      <abstract>
        <p>• Centre &amp; my role • Selection of multimedia applications and their presentation design issues • Some observations - Different applications, different design decisions - Applying general design principles</p>
      </abstract>
    </article-meta>
  </front>
  <body>
    <sec id="sec-1">
      <title>-</title>
      <p>Overview
Centre for Digital Video Processing
at Dublin City University
• Developing automatic indexing/retrieval tools
for managing large amount of image/video
information
– Object/Face Detection &amp; Tracking in Video
– Audio &amp; Video Event Detection
– Video Delivery on Mobile Devices
– Large-scale Distributed Web Image Search
– Search Engine Design for Collaborative Video Retrieval
– Hardware Accelerator Design for MPEG-4 Mobile Platform
– Personalisation &amp; Recommendation for Video
– Synergy between automatic &amp; manual indexing
– Fusion of multi-modal query results</p>
      <p>My Role: Usability &amp; User Issues
• Understand the research &amp; development of
Image/Video indexing/retrieval tools within
the Centre
• Think how these could be exploited
– Envision the use: scenarios &amp; future system use
– Prototyping user-interfaces
– Deploy (if possible)
– User testing: monitor usage &amp; guide future
development
Mobile
Físchlár-News</p>
      <p>Movie Browser
CCTV Search
System
BBC Rushes
Search System</p>
      <p>shot boundary detection
Camera
shot</p>
      <p>Keyframe Extraction</p>
      <p>Físchlár-News Archive
• Online archive of daily RTE1 9pm TV news
• Automatic video indexing:</p>
      <p>News Story Segmentation, based on:
– Anchorperson detection (by shot clustering)
– Face detection
– Advertisement detection
– Shot length
– Activity
measure</p>
    </sec>
    <sec id="sec-2">
      <title>News story linkage analysis</title>
    </sec>
    <sec id="sec-3">
      <title>News story database</title>
      <p>Story Segmentation - SVM
(Support Vector Machine) with:
• Speech vs. music discrimination
• Anchorperson shot clustering
• Face detection
• Shot length cue
• Activity measure
Shot segmented, advert
detected program</p>
    </sec>
    <sec id="sec-4">
      <title>Story segmented program Broadcast TV news</title>
    </sec>
    <sec id="sec-5">
      <title>MPEG-1 encoding</title>
    </sec>
    <sec id="sec-6">
      <title>An MPEG-1 encoded daily 9 o’clock news program (30 min)</title>
    </sec>
    <sec id="sec-7">
      <title>Shot Boundary Detection</title>
    </sec>
    <sec id="sec-8">
      <title>Shot segmented program</title>
      <p>Advertisement
Detection</p>
    </sec>
    <sec id="sec-9">
      <title>Story-based news</title>
      <p>browsing, searching,
streamed playback and…
…recommendation</p>
    </sec>
    <sec id="sec-10">
      <title>Oracle</title>
      <p>Video
Server</p>
      <p>Web
application</p>
    </sec>
    <sec id="sec-11">
      <title>User</title>
      <p>profile
User Evaluation of Físchlár-News: An Automatic Broadcast News Delivery System. Lee H, Smeaton A.F, O'Connor N
and Smyth B. TOIS - ACM Transactions on Information Systems, 24(2), 2006.</p>
      <sec id="sec-11-1">
        <title>Automatic news story segmentation as main back-end =&gt; story-based browsing, searching, recommendation</title>
      </sec>
      <sec id="sec-11-2">
        <title>Deployment effort... User studies to refine the UI</title>
        <p>• Application specific...</p>
      </sec>
      <sec id="sec-11-3">
        <title>Daily update, up-to-dateness of news =&gt; Calendar</title>
      </sec>
      <sec id="sec-11-4">
        <title>Anchorperson’s 2-line summary statement as story summary text</title>
      </sec>
      <sec id="sec-11-5">
        <title>Average #stories per day (1020 only) =&gt; Linear list most effective (no drop-down box or pagination necessary)</title>
        <p>• General design principles, guidelines, graphic
design, web design, etc. – knowledge &amp;
experience I have in general
– E.g. Overview first, details on demand</p>
      </sec>
      <sec id="sec-11-6">
        <title>Day list of the months (calendar)</title>
      </sec>
      <sec id="sec-11-7">
        <title>Story list of the day</title>
      </sec>
      <sec id="sec-11-8">
        <title>Shot list of the story</title>
      </sec>
      <sec id="sec-11-9">
        <title>Playback (full detail)</title>
        <p>• General design principles, guidelines, graphic
design, web design, etc. – knowledge &amp;
experience I have in general
– E.g. Overview first, details on demand
– E.g. Visual consistency
Whenever list of stories
appears...
... to make obvious what a
piece of presentation on the
screen represents and doesn’t
require interpretation effort
Applications
Development
(interaction design +
software engineering)</p>
        <p>Físchlár-Nursing</p>
        <p>Físchlár-News
CCTV Search
System
BBC Rushes
Search System
Físchlár-TRECVid2004: Combined Text- and Image-Based Searching of Video Archives. O'Connor N, Lee H, Smeaton
A.F, Jones G, Cooke E, Le Borgne H and Gurrin C. ISCAS 2006 - IEEE International Symposium on Circuits and
Systems, Kos, Greece, 21-24 May 2006.
Applications
Development
(interaction design +
software engineering)</p>
        <p>Físchlár-Nursing</p>
        <p>Físchlár-News
Technology
Development for
automatic
extraction of
syntactic &amp;
semantic features
in image/video</p>
        <p>Físchlár-TV
Shot
Boundary
Detection
Keyframe
Extraction</p>
        <p>Mobile
Físchlár-News</p>
        <p>Movie Browser
TRECVid02
Interactive Search
System</p>
        <p>TRECVid03
Interactive Search
System</p>
        <p>TRECVid04
Interactive Search</p>
        <p>System
Face Detection
Building Detection
Indoor/Outdoor
Cityscape/Landscape
Advert Detection</p>
        <p>News Story</p>
        <p>Segmentation
Video
Recommendation</p>
        <p>Sports Summarisation</p>
        <p>Image-Image
Similarity RF
Potential screen complexity – use of main plain
vs. background plain, round edges, and
corresponding buttons
MediAssist (Personal</p>
        <p>Photo Manager)
CCTV Search
System
BBC Rushes
Search System
- all the detected Objects &amp; Events and link possibility are indicated
[unit]
Link to…
… and here is the ASR or Closed
Caption text that</p>
        <p>Link to…</p>
        <p>Link to…
[unit]s with linked</p>
        <p>OBJECT 3
… and here is the ASR or Closed
Caption text that
… and here is the ASR or Closed
Caption text that
… and here is the ASR or Closed
Caption text that</p>
        <p>Link to…</p>
        <p>[unit]s with similar text/image content
Selecting</p>
        <p>Search result
+</p>
        <p>Search result</p>
      </sec>
      <sec id="sec-11-10">
        <title>Start with a micro-interaction scheme (using</title>
        <p>buttons that represent objects), then develop it
further in an integrated interface
Detected objects have no labelling (meaning) but
only as spatial region or blob on the keyframe</p>
      </sec>
      <sec id="sec-11-11">
        <title>Applying this interaction scheme to objectbased Relevance Feedback applications...</title>
        <p>User-Interface to a CCTV Video Search System. Lee H, Smeaton A.F, O'Connor N and Murphy
N. ICDP 2005 - IEE International Symposium on Imaging for Crime Detection and Prevention,
London, U.K., 7-8 June 2005.</p>
      </sec>
      <sec id="sec-11-12">
        <title>Using the Unit Representation (enabling object-based interaction) at micro-level interaction (Conventional) Relevance Feedback idea</title>
      </sec>
      <sec id="sec-11-13">
        <title>Search result showing not only matched</title>
        <p>list of objects but also geographical map
summarising and highlighting the matched
object’s route (note the application’s main
purpose: chase a suspect)</p>
        <p>Observations
• Considerable amount of time &amp; effort in
manually hard-designing UIs
– For example...</p>
      </sec>
      <sec id="sec-11-14">
        <title>Design taking 4 months, 4 iterative</title>
        <p>refinements: starting with pen-and-paper
sketches, then Photoshop sketches,
discussing with technical team then me
resketching accordingly</p>
        <p>Observations
• Huge amount of time &amp; effort in manually
hard-designing
– For example...
• Different situations require different specific
design decisions
– For example, presentation of keyframes</p>
      </sec>
      <sec id="sec-11-15">
        <title>Linear list images and text</title>
      </sec>
      <sec id="sec-11-16">
        <title>Images overlayed with text</title>
      </sec>
      <sec id="sec-11-17">
        <title>Image in different sizes</title>
      </sec>
      <sec id="sec-11-18">
        <title>Which layout on which situation?</title>
        <p>– For example...
• Different situations require different specific design
decisions
– For example, presentation of keyframes</p>
      </sec>
    </sec>
  </body>
  <back>
    <ref-list />
  </back>
</article>