<!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>Where Can I Add This Annotation?</article-title>
      </title-group>
      <contrib-group>
        <contrib contrib-type="author">
          <string-name>Author Keywords Authoring</string-name>
          <xref ref-type="aff" rid="aff0">0</xref>
          <xref ref-type="aff" rid="aff1">1</xref>
        </contrib>
        <contrib contrib-type="author">
          <string-name>Hypervideo</string-name>
          <xref ref-type="aff" rid="aff0">0</xref>
          <xref ref-type="aff" rid="aff1">1</xref>
        </contrib>
        <contrib contrib-type="author">
          <string-name>Annotation</string-name>
          <xref ref-type="aff" rid="aff0">0</xref>
          <xref ref-type="aff" rid="aff1">1</xref>
        </contrib>
        <contrib contrib-type="author">
          <string-name>Scene Graph</string-name>
          <xref ref-type="aff" rid="aff0">0</xref>
          <xref ref-type="aff" rid="aff1">1</xref>
        </contrib>
        <aff id="aff0">
          <label>0</label>
          <institution>Britta Meixner Chair of Distributed Information Systems University of Passau Passau</institution>
        </aff>
        <aff id="aff1">
          <label>1</label>
          <institution>Martin Kuchler University of Passau Passau</institution>
        </aff>
      </contrib-group>
      <pub-date>
        <year>2015</year>
      </pub-date>
      <abstract>
        <p>Annotated interactive non-linear videos normally demand a two-stage authoring process. The structure of the video is defined in the first step, where video scenes are arranged in a scene graph. Additional information (annotations) are added to the scenes in a second step. Therefore a timeline-based editor has to be opened which enables the author to add images, videos, audio files, or text to a selected scene. User tests showed that users do not find the entry point to the annotation editor currently placed as a button on a scene in the scenegraph. This work proposes solutions for a better and more intuitive workflow in the tool to add annotations to scenes.</p>
      </abstract>
    </article-meta>
  </front>
  <body>
    <sec id="sec-1">
      <title>-</title>
      <p>INTRODUCTION
Annotated interactive non-linear videos consist of video
scenes which are linked with each other in a scene graph.
Each scene or the whole video can be extended with
annotations (text, images, audio, or videos) which may add
additional information to the video contents. These videos are for
example very well suited for sports or physiotherapy training
videos. They can furthermore be used for e-learning or tours
through cities. The creation of this type of video comprises a
series of steps in the authoring process. First, scenes need to
be created and linked with each other in a scene graph. Then
the annotations have to be linked either to single scenes (local
annotation) or to the whole annotated interactive non-linear
video (global annotation). Global annotations are shown
during the whole playback while local annotations are faded in
and out during single scenes. This affords a two-stage
authoring process, whereby the navigational structure of the
annotated interactive non-linear video is created first. In the
second step, annotations are added.</p>
      <p>
        RELATED WORK
Annotated interactive non-linear videos can be seen as a
subset of hypervideos (for an example definition see Chambel et
al. [
        <xref ref-type="bibr" rid="ref2">2</xref>
        ]). These are often compared to interactive multimedia
presentations. We analyzed authoring tools from both areas.
They either use only videos which are linked with each other,
or they treat each media element (image, video, text, audio,
etc.) equally and do not have the two-stage authoring process
and the same structures that we have. A detailed overview of
related work can be found in [
        <xref ref-type="bibr" rid="ref5">5</xref>
        ].
      </p>
      <p>
        USER EVALUATION
Based on the knowledge that the button to add an
annotation to a scene in the scene graph is rarely found, we
conducted a user test combined with think-aloud protocols ([
        <xref ref-type="bibr" rid="ref3">3</xref>
        ])
and a screen capture software. Twenty test users (N = 20)
were split into two groups of ten users each (NA = NB =
10). The users all worked with a computer often to very often
and had average to very good knowledge about computers in
general. None of them worked with video annotations before
and had bad to very bad knowledge in this area accordingly.
The users were chosen for that reason, because the SIVA
Producer should be easy to use, especially for users with limited
to no experience in video editing. The test users were aged
between 20 and 60 (median 25 years). The test was used
to find out where the problems of comprehension are in the
existing version of the SIVA Producer, therefore this small
number of test users is enough ([
        <xref ref-type="bibr" rid="ref6">6</xref>
        ]). Both groups got a short
explanation about the SIVA Producer. Then they opened a
predefined project with a scene graph. Images were already
loaded into the media repository but not added to the scenes.
Group A had the task to first add a global annotation. After
that, they were asked to add a local annotation to a specified
scene. Group B was asked to add an annotation to a scene
without adding a global annotation first. Both groups had to
find the button with the pen positioned on a scene (see Figure
1).
      </p>
      <p>Group A eventually found the entries to open the editor for
the global annotations in the menu or the toolbar and was
able to add the annotation to the whole video. Based on the
knowledge that annotations can be added this way, they tried
adding the local annotation the same way. Nine out of ten
users were not able to open the annotation editor for local
annotations, because they did not find the button on the scene
in the scene graph. Group B also started searching in the
menu or the toolbar even without the knowledge of where to
add a global annotation. In this group eight out of ten users
did not find the button.</p>
      <p>Taking a look at the search areas, two groups of users could
be recognized: those who did understand the connection
between scenes and local annotations, and those who did not.
Some users tried more than one way of opening the editor.
Thereby we counted each attempt. Nine users did not
realize the connection between scenes and local annotations and
wanted to open the editor via the menu. Six users searched
for a button in the toolbar. Seven users recognized that a local
annotation has to be added to a scene and tried a right-click on
the scene and started searching in the appearing menu. Five
users tried a double-click on the scene. Only three users were
able to open the editor without help.</p>
      <p>
        TOWARDS A SOLUTION
The user test showed us the areas where users are searching
for the button to add a local annotation to a scene. If the
editor should be startable from the menu, a second step where
the scene is selected is needed. Based on the results of the
user test described in the previous section, we created a
questionnaire [
        <xref ref-type="bibr" rid="ref4">4</xref>
        ] which was carried out by 76 persons. They were
aged between 16 and 65 and mainly between 20 and 35 years
old. The questionnaire was finished and not clicked randomly
by 58 persons (tested with statistical methods). The
questionnaire contained four parts: a short introduction, ways to select
a scene, possible button placements, and statistical data.
Offered ways to select the scene where the annotation should be
added were:
      </p>
    </sec>
    <sec id="sec-2">
      <title>Mark the scene with the mouse.</title>
      <p>Open a pop-up with a drop-down box to select a scene.</p>
    </sec>
    <sec id="sec-3">
      <title>Select scene in the annotation editor.</title>
      <p>Except selecting the scene in the annotation editor, each of the
ways to select the scene to be edited can be combined with
each of the ways to place the button to open the annotation
editor:</p>
    </sec>
    <sec id="sec-4">
      <title>Entry in the menu (see Figure 3).</title>
    </sec>
    <sec id="sec-5">
      <title>Button in the toolbar (see Figure 4).</title>
      <p>Entry in the right-click menu of a scene (see Figure 2).
Different questions including control questions were asked
for each of the six solutions. Furthermore, pairs out of the
three solutions for each problem were opposed and the
preference for one solution was asked for each pair. As a result
for the selection of the scene to which the annotation should
be added, the pop-up with the drop-down box was not
considered as useful. Marking a scene with the mouse or
selecting the scene after opening the annotation editor was regarded
useful. Regarding the placement of the button, the entry in the
menu was not regarded useful. In contrast, using the a
rightclick menu of the scene was considered useful compared to
each other placement of the button.</p>
      <p>CONCLUSION
Based on the two-stage authoring process of interactive
nonlinear videos, we faced the problem that authors were able to
create a scene graph in the SIVA Producer, but did not find
the entry point for opening the editor to add annotations to a
single scene. In this work, we have proposed three methods
of selecting a scene to which an annotation should be added
as well as three placements of the button which opens the
annotation editor. The most suitable solution according to
the conducted questionnaire was an entry in the right-click
menu of the scene to which the annotation should be added.
Our future work includes the implementation of the proposed
solution as well as further user tests to confirm the results of
the questionnaire.</p>
    </sec>
  </body>
  <back>
    <ref-list>
      <ref id="ref1">
        <mixed-citation>
          1.
          <string-name>
            <surname>Bulterman</surname>
            ,
            <given-names>D. C. A.</given-names>
          </string-name>
          , and
          <string-name>
            <surname>Hardman</surname>
            ,
            <given-names>L.</given-names>
          </string-name>
          <article-title>Structured multimedia authoring</article-title>
          .
          <source>ACM Trans. Multimedia Comput. Commun. Appl</source>
          .
          <volume>1</volume>
          ,
          <issue>1</issue>
          (Feb.
          <year>2005</year>
          ),
          <fpage>89</fpage>
          -
          <lpage>109</lpage>
          .
        </mixed-citation>
      </ref>
      <ref id="ref2">
        <mixed-citation>
          2.
          <string-name>
            <surname>Chambel</surname>
            ,
            <given-names>T.</given-names>
          </string-name>
          ,
          <string-name>
            <surname>Zahn</surname>
            ,
            <given-names>C.</given-names>
          </string-name>
          , and
          <string-name>
            <surname>Finke</surname>
            ,
            <given-names>M.</given-names>
          </string-name>
          <article-title>Hypervideo design and support for contextualized learning</article-title>
          .
          <source>In Advanced Learning Technologies</source>
          ,
          <year>2004</year>
          . Proceedings. IEEE International Conference on (
          <year>Aug 2004</year>
          ),
          <fpage>345</fpage>
          -
          <lpage>349</lpage>
          .
        </mixed-citation>
      </ref>
      <ref id="ref3">
        <mixed-citation>
          3.
          <string-name>
            <surname>Ericsson</surname>
            ,
            <given-names>K. A.</given-names>
          </string-name>
          ,
          <article-title>and</article-title>
          <string-name>
            <surname>A</surname>
          </string-name>
          .,
          <string-name>
            <surname>S. H.</surname>
          </string-name>
          <article-title>How to study thinking in everyday life: Contrasting think-aloud protocols with descriptions and explanations of thinking</article-title>
          .
          <source>Mind, Culture, and Activity 5</source>
          ,
          <issue>3</issue>
          (
          <year>1998</year>
          ),
          <fpage>178186</fpage>
          .
        </mixed-citation>
      </ref>
      <ref id="ref4">
        <mixed-citation>
          4.
          <string-name>
            <surname>Gillham</surname>
            ,
            <given-names>B.</given-names>
          </string-name>
          <string-name>
            <surname>Developing</surname>
          </string-name>
          a Questionnaire.
          <source>Real world research. Bloomsbury Academic</source>
          ,
          <year>2000</year>
          .
        </mixed-citation>
      </ref>
      <ref id="ref5">
        <mixed-citation>
          5.
          <string-name>
            <surname>Meixner</surname>
            ,
            <given-names>B.</given-names>
          </string-name>
          <string-name>
            <surname>Annotated Interactive</surname>
          </string-name>
          Non-linear Video - Software
          <string-name>
            <surname>Suite</surname>
            , Download and
            <given-names>Cache</given-names>
          </string-name>
          <string-name>
            <surname>Management</surname>
          </string-name>
          .
          <source>PhD thesis</source>
          , Universita¨t Passau,
          <year>2014</year>
          .
        </mixed-citation>
      </ref>
      <ref id="ref6">
        <mixed-citation>
          6.
          <string-name>
            <surname>Nielsen</surname>
            ,
            <given-names>J.</given-names>
          </string-name>
          <article-title>How many test users in a usability study? Website</article-title>
          ,
          <year>June 2012</year>
          . (Nielsen Norman Group).
          <source>Retrieved May 15</source>
          ,
          <year>2014</year>
          , from http://www.nngroup.com/articles/how-many
          <string-name>
            <surname>-</surname>
          </string-name>
          test-users/.
        </mixed-citation>
      </ref>
    </ref-list>
  </back>
</article>