Centrum voor Wiskunde en Informatica
INformation Systems
SampLe: Towards a Framework for System-supported
Multimedia Authoring
Kateryna Falkovych, Frank Nack,
Jacco van Ossenbruggen, Lloyd Rutledge
REPORT INS-E0302 AUGUST 31, 2003
INS
Information Systems
CWI is the National Research Institute for Mathematics and Computer Science. It is sponsored by the
Netherlands Organization for Scientific Research (NWO).
CWI is a founding member of ERCIM, the European Research Consortium for Informatics and Mathematics.
CWI's research has a theme-oriented structure and is grouped into four clusters. Listed below are the names
of the clusters and in parentheses their acronyms.
Probability, Networks and Algorithms (PNA)
Software Engineering (SEN)
Modelling, Analysis and Simulation (MAS)
Information Systems (INS)
Copyright © 2003, Stichting Centrum voor Wiskunde en Informatica
P.O. Box 94079, 1090 GB Amsterdam (NL)
Kruislaan 413, 1098 SJ Amsterdam (NL)
Telephone +31 20 592 9333
Telefax +31 20 592 4199
ISSN 1386-3681
SampLe: Towards a Framework for System-
supported Multimedia Authoring
ABSTRACT
Much current research on hypermedia generation accepts user input only at the start of an
otherwise fully-automated process. However, since multimedia presentation creation is often a
complex and creative process, it has multiple phases which would each benefit from human
intervention. This paper presents a hypermedia generation model that lets the user influence all
phases of this computer-assisted human-guided process. The main focus is on providing extra
support for helping the user find relevant media items and combine them meaningfully into a
rich and coherent multimedia presentation. Like fully-automated systems, our approach uses
explicit knowledge about the presentation's topic domain, narrative structures, hypermedia
presentation and distinctions between media modalities. This paper presents a motivating
scenario that is used to derive a number of system requirements and to discuss the pros and
cons of the presented approach.
Keywords and Phrases: Multimedia authoring, semantics, narrative structure, intelligent support
Samp
L
e: Towards a Framework for System-supported Multimedia Authoring
Kateryna Falkovych, Frank Nack, Jacco van Ossenbruggen, Lloyd Rutledge
CWI, Amsterdam
Firstname.Lastname@cwi.nl
Abstract
Much current research on hypermedia generation
accepts user input only at the start of an otherwise fully-
automated process. However, since multimedia
presentation creation is often a complex and creative
process, it has multiple phases which would each benefit
from human intervention. This paper presents a
hypermedia generation model that lets the user influence
all phases of this computer-assisted human-guided
process. The main focus is on providing extra support for
helping the user find relevant media items and combine
them meaningfully into a rich and coherent multimedia
presentation. Like fully-automated systems, our approach
uses explicit knowledge about the presentation's topic
domain, narrative structures, hypermedia presentation
and distinctions between media modalities. This paper
presents a motivating scenario that is used to derive a
number of system requirements and to discusses the pros
and cons of the presented approach.
1. Introduction
The production of multimedia presentations is a
complex, resource demanding, distributed, and creative
process. The aim of the process is to provide engaging
and relevant information by composing a multi-
dimensional network of relationships between different
kinds of audio-visual information units.
We divided this process into five stages: theme
identification, specification of presentation structure,
collection of material, arrangement of material, and
presentation creation. At the first stage, a topic of the
presentation is chosen within a domain of interest. The
scope of the presentation is outlined at this stage by
defining the genre of the presentation as well as main and
secondary characters. This information allows
specification of the logical structure of the presentation
at the second stage. At the third stage appropriate
material is selected and placed within each part of the
logical structure. The particular ordering of the selected
material is done at the fourth stage. At the fifth stage the
final presentation is created by determining spatial-
temporal relationships between selected items and
defining stylistic aspects.
Even though we presented the stages above in a
sequential manner, we find that the various stages are
typically iterative and mutually interrelating. While this
increases the complexity of the process it also leads to
semantically rich and engaging presentations.
Many multimedia presentations, such as those made
by pupils, students, researchers or managers, are
manually crafted presentations. The support in manual
production comes mainly in form of production
environments such as Director, Premiere, Photoshop,
Flash, FrontPage, PowerPoint and others. These tools
are, however, not equipped to support the complex
processes of content and design development, as they
assume that the user has a sufficient level of expertise [4].
In this paper, we present a framework that supports
the five steps of the presentation generation process. Our
approach combines the creative strength of humans with
the analytical and procedural power of machines. It
allows a way of presentation generation in which the user
has full control over the presentation creation process,
but at the same time is facilitated with ontology-based
and context-oriented information at those stages where
she lacks knowledge or skills.
Our system, called SampLe (Semi-Automatic
Multimedia Presentation generation Environment), is
connected to a large multimedia database. In order to
create a presentation, the material from the database is
used.
Since different users have different levels of expertise
in the domain as well as various experiences with
presentation building process, the system should be able
to support any type of a user by providing support at
each stage of presentation creation process. Depending
on the user's level of expertise, created presentations can
be included into the system repository, enriching it with
new points of view on the material and new interrelation
structures between media items. In such a way, stored
presentations might help novice users (such as students
having an assignment to build a presentation) get an
insight into the domain and get an idea about possible
topics and designs for their own presentations.
Additionally, the system is also in the position to offer
assistance on the presentational level by providing
various ways of structuring a presentation.
The paper is structured as follows. Section 2
discusses related work in which we show connections
between our approach and existing presentation
generation techniques. Further discussion follows the
five stage process described in the Introduction. Section
3 presents the theme identification stage. Section 4
discusses presentation structure specification. Section 5
describes the elaboration on the created structure by
facilitating content selection. Arrangement of selected
media items is presented in Section 6. Section 7 discusses
the last stage of the process where the presentation is
created according to the proper style and duration.
Finally Section 8 outlines conclusions and future work.
2. Related work
The development of the SampLe framework took
account of existing work in two related research domains,
namely authoring environments for the fully manual
generation of presentations and techniques from the field
of automatic presentation generation.
The usual computer support for manually crafted
presentations is manifested in form of production
environments, such as image editing tools (Photoshop,
Illustrator, GIMP, or Maya), new media authoring tools,
such as Director/Shockwave, Flash, Dreamweaver,
Frontpage, PowerPoint, GRiNS, Authorware, HyperCard,
or WWW presentation technology, such as HTML and
SMIL. Although these tools provide the user with much
freedom during the development process, they are not
equipped to support the complex process of content
development, because they assume that the user has a
sufficient level of expertise.
Systems that facilitate automated presentation
generation, on the other hand, are typically applied for
dynamic, interactive environments that do not allow the
intervention of a human during presentation generation at
any time after the initial request, such as web sites of
museums [8], real time instruction generation [1],
discourse driven hypermedia presentations [9] or user
tailored biographies of artists [6]. These systems provide
ontology-based description layers for content and
presentation aspects that permit reliable control
mechanisms to establish presentations that are flexible
enough to respond to changing individual user needs and
user groups. The complexity of logical structure
developments and the complex choice of stylistic aspects
regarding conveying certain information typically result,
however, in quite simple structures and styles and hence
unexciting presentations.
The blend of these two methodologies for multimedia
presentation generation can lead to an improvement of
mainly human generated presentations.
First attempts to make use of description structures and
mechanisms taken from automatic processes as the basis
for support in manual authoring environments are
described on the content level, such as for the authoring
of motion pictures [2], or on a task-oriented level, such as
in supporting the early exploration of design ideas [3]. The
advantage of this system-guided approach is that it
facilitates the creation of attractive presentations by
