TOWARDS A TEMPLATE-BASED GENERATION OF VIRTUAL

3D MUSEUM ENVIRONMENTS

D. Biella and W. Luther

Institute of Informatics and Cognitive Sciences,University of Duisburg-Essen, Lotharstr. 65, Duisburg, Germany

Keywords: Metadata standards, Comparison of metadata standards, Virtual museum, Virtual 3D environments.

Abstract: This paper focuses on the question of how metadata and existing metadata standards can be used for the

administration, layout, storage, retrieval and visualization of Web-based virtual 3D museum environments.

We present enhanced metadata concepts that encompass the infrastructure of a virtual museum or laboratory

using stationary or mobile interfaces to communicate with information sources or interact with artifacts.

1 INTRODUCTION

In recent papers, we have presented Web-based

virtual 3D museum environments featuring photo-

realistic 3D models, nondeterministic simulations

and special user interaction (Biella, 2006; Biella &

Luther, 2007, 2008). We have also addressed the

question of whether 3D or 2D virtual environments

are preferable, depending on the type of exhibits

involved. How to create virtual 2D exhibitions is

discussed in (Nesland et al., 2005). Concepts and

design aspects of virtual exhibition systems using

XML-based conceptual hypermedia document mod-

els are addressed in (Foo, 2008).

The research presented in this paper focuses on

the question how metadata and existing metadata

standards can be used for the administration, layout,

storage, retrieval and visualization of such environ-

ments.

First, we provide a definition of the virtual mu-

seum and discuss the concept of a virtual museum as

an informal learning environment. Then, we high-

light the theoretical aspect of metadata and its poten-

tial functions in an application context. Next, various

existing metadata standards are introduced and

compared with regard their capabilities for 3D vir-

tual museum environments. Finally, we focus on the

ARCO metadata standard AMS, present a case study

in which it has been used to describe a virtual mu-

seum and highlight the strengths and weaknesses of

AMS when it is used to create virtual exhibitions in

specific environments providing meaningful user

interactions.

2 VIRTUAL MUSEUM

In this paper, we focus on virtual museums. We

follow the definition given by McKenzie, Sola and

Keene that defines a “virtual museum” as “an orga-

nized collection of electronic artifacts and informa-

tion resources—virtually anything which can be

digitized” (McKenzie, 1997), which “uses as the

means its collections, related information, knowled-

geable people, and the museum itself with its galle-

ries and displays of objects” (Keene, 1997).

According to Sola, the “traditional museum

piece, an item, a three-dimensional fact, is only a

datum among a complex of museum information, of

a message. We do not have museums because of the

objects they contain but because of the concepts that

these objects help to convey.“ (Sola, 1997)

In summary, we regard a virtual museum envi-

ronment as a combination of replicated or “born

digital” exhibits, ideas and concepts. Furthermore,

ideas and concepts can be conveyed through means

of interaction with objects. Although this approach

is commonly used in “hands-on” museums, there are

few virtual museums that follow this approach.

Despite the properties given above, a virtual mu-

seum is expected to support the following features:

 Modification of exhibits with regard to position,

form and content, even with the aim of creating

new, enhanced instances of a cultural object,

 Interaction with exhibits via adequate interfaces,

 Reversibility to the original state after a user’s

interaction, and

 Simulation of a kind defined by a discrete or

continuous process model.

399

Biella D. and Luther W.

TOWARDS A TEMPLATE-BASED GENERATION OF VIRTUAL 3D MUSEUM ENVIRONMENTS.

DOI: 10.5220/0001823103990402

In Proceedings of the Fifth International Conference on Web Information Systems and Technologies (WEBIST 2009), page

ISBN: 978-989-8111-81-4

We want to cite predominant forms of interaction

supported by various cultural object types (cf. Table

1).

Table 1: Object and interaction types.

Object type Interaction type

Geometric object Moving, modifying the form, cloning

Visual object Watching from a different viewpoint,

modifying appearance

Dynamic object Launching the dynamic process via a

concept keyboard (Baloian et al.,

2007)

Room or lighting Being a part of the installation

Experiment Parameterizing and executing

Historical object Documenting the historical context,

creating extensions

It is noteworthy that reversibility is much easier to

accomplish in virtual simulations than in reality and

that it offers the opportunity to learn and understand

concepts by modifying objects regardless of their

physical accessibility or monetary or cultural value.

The concept of a museum as an informal learning

environment using metadata for the description of

room-based layout, dynamic exhibit models and

interaction design is shown in Figure 1.

Figure 1: Concept of a virtual museum.

3 ENLARGED METADATA

SCHEMA

A metadata standard is chosen or developed based

on a list of requirements or required functions. This

standard must be able to describe the smallest in-

formation units, which are usually objects or groups.

Gilliland-Swetland introduced the notions of intrin-

sic and extrinsic metadata, which relate, respective-

ly, to the content and context of the information

object. Intrinsic data are associated with the charac-

teristics of the object, and extrinsic data with contex-

tual parameters. The following metadata categories

are mentioned (Gilliland-Swetland, 2000): Adminis-

trative Metadata (AM), Descriptive Metadata (DM),

Preservation Metadata (PM), Technical Metadata

(TM), and Use Metadata (UM).

Although the ARCO data model (cp. Section 4)

provides a metadata standard that can be applied in

general museum contexts, the visualization tools

seem to be proprietary add-on applications that

present items in augmented reality environments.

We found only partial counterparts to parameterized

exhibition rooms, presentation forms, X-VRML

templates or interaction patterns in the AMS (Patel

et al., 2005) (Walczak et al., 2006).

Even if the ARCO standard used some of the

categories introduced above, it did not distinguish

between intrinsic and extrinsic data and follows the

three-tier classification CO-AO-RO. Neither propos-

al explicitly supports sophisticated virtual museum

generation, typical room arrangements or illumina-

tion concepts, dynamically changing objects or user-

object interaction following a certain action and

interaction logic. Powerful metadata schemes should

support graphical tools to generate and modify

architectural designs that define exhibition rooms

including the ambient infrastructure, like libraries,

video viewing or information desks. Using parame-

terized room templates and predefined presentation

and interaction styles, the exhibition design process

can be accelerated by using a standardized search

engine, a repository containing the cultural object

data based on a complete metadata schema. Parame-

ters and data could be introduced using script lan-

guages or a graphical user interface (GUI) with

enhanced object search, editing and preview func-

tionalities.

We would like to emphasize the necessity that a

valuable metadata concept should encompass the

infrastructure of a virtual museum or laboratory with

stationary or mobile interfaces to communicate with

the information sources or to interact with the arti-

facts. This can be done via information terminals

and sensitive touch screens or silent digital compa-

nions. The goal of our ongoing research is to deter-

mine to what extent the interaction logic can be

automatically generated via template-based tools

together with the virtual learning and experimenting

environment and the human-machine interfaces.

Whereas the interaction logic is characterized

within the use metadata and serves to launch opera-

tions that change the situation of a virtual object, the

technical metadata describe the underlying action

logic. This metadata scheme could also be inspired

by certain elements of the actual learning object

metadata (LOM) standard, such as interactivity type,

intended end user role and entries to measure occu-

pation time.

WEBIST 2009 - 5th International Conference on Web Information Systems and Technologies

400

In augmenting the proposed ARCO metadata stan-

dard, administrative and descriptive metadata

should encompass all the environment information

(e.g.: exhibition objects, room installation, geometry

and appearance attributes, illumination models).

Preservation metadata define constraints and consis-

tency checks to allow reversibility and preserve

authenticity and integrity when users are interacting

with virtual objects. Technical metadata should

contain the creation and dynamization of (virtual)

replicas and are related to the creation of user groups

and access rights. Use metadata circumscribe the

range of user activities, including virtual object

creation and manipulation and types of expressivity.

The enlarged metadata standard makes a clear

distinction between intrinsic and extrinsic metadata

and completes the topics artifact selection, digital

acquisition, storage/collection management, model

refinement and exhibition building and can be im-

plemented via an extension of a VRML model.

4 COMPARISON OF EXISTING

METADATA STANDARDS

The metadata standards Dublin Core, CIDOC-CRM,

VRA Core and ARCO have been compared based

on their 3D data capability.

The Dublin Core Metadata Initiative (DCMI) is

concerned with the development of metadata stan-

dards for the description of resources with a focus on

interoperability between heterogeneous document

management systems. Default DC metadata ele-

ments can be used in the description of artifacts and

resources in museums. However, interaction or

simulation models and extrinsic hierarchies cannot

be described.

The Comité international pour la documentation

Conceptual Reference Model (CIDOC CRM) is an

ISO standard for the formal semantic description of

cultural heritage information. In addition to DC, it

can describe the temporal and spatial properties of

an object-related event. CRM is limited in terms of

multimedia content description but is well suited for

physical objects, rather than for virtual objects.

Visual Resources Association (VRA) focuses on

the metadata classification for both physical and

virtual visual information objects. VRA Core’s level

of detail in object descriptions is similar to that of

CIDOC CRM and offers many relation types for

modeling relations between exhibits and between a

museum and its rooms. It has limited capacities for

“born-digital” 3D objects.

The ARCO project sought to develop a single

standard that was also suitable for virtual 3D objects

and their workflow (modeling, modification and

visualization). It concretizes the notion of an infor-

mation object by defining the abstract class cultural

object (CO), the physical artifact, and by deriving

two nonabstract instances, the acquired object (AO)

and the refined object (RO). The digital representa-

tion of the CO (as AO or RO) is the Media Object

(MO). Examples of MOs include 3D models and

images of various MIME types.

The ARCO metadata element set (AMS) is an

extension (partly based on DC, CIDOC-CRM) that

defines six metadata types for ARCO objects: Re-

source discovery metadata, Presentation Metadata,

Curatorial and Descriptive Metadata, Technical

Metadata, Themed Metadata and Administrative

Metadata.

The major advantages of the ARCO standard are

the ability to manage native 3D and to store chrono-

logical information. The results of the analysis of

ARCO’s features with regard to certain metadata

types are provided in Table 2.

Table 2: Support of certain metadata types in the ARCO

metadata standard.

Type Existing or partly existing feature Missing feature

AM Acquisition, rights, location,

digitization, metadata creation

DM Creation (production date or period,

location, creator, contribution),

source, type (material), geometry

and appearance, components,

accession, actual location, field

collection

Modification by

users, No

environment

description

PM Consistency

TM Object metadata concerning virtual/

digital manifestation, MIME-type,

media object instance in database,

data type and format extent, person

effort to produce, rights, skill level

UM Type-specific metadata, size,

resolution panorama, compression,

color depth, dimension, textures,

modeler software, language,

animation, algorithm, manipulation

Only 3ds Max,

VRML & Dy-

namic modeling,

No user impact,

no interaction

model, algorithm

only for rescaling

The ARCO metadata standard—despite some

minor exceptions—meets all the demands made by

Gilliland-Swetland. Nonetheless, some functions are

mapped to different metadata types (e.g.: rights). For

this reason, we have chosen this standard to design

3D virtual museums and laboratories based on room

templates, exhibit-specific navigation and interaction

techniques.

TOWARDS A TEMPLATE-BASED GENERATION OF VIRTUAL 3D MUSEUM ENVIRONMENTS

401

5 FEASIBILITY STUDY

The feasibility of our concept was demonstrated by

using an existing implementation of an African

Grassland museum (Mafo, 2007) that includes an

entrance hall and a separate hall for exhibits on the

topic of daily life (cf. Figures 2 and 3). The system

architecture consists of a PHP-capable Web server, a

MySQL database server and Altova XML Spy and a

Web browser as the client-sided interfaces. The

object descriptions are stored in the database. Data is

read and written through an ODBC connection. The

Web-based metadata and object visualization front-

end lists several objects in a selection list. When an

object is clicked, it is visualized, and several meta-

data can be displayed (Figure 4).

Figure 2: Grassland museum (3D entrance hall).

Figure 3: Grassland museum (“Daily Life” hall).

Figure 4: Metadata visualization (Entrance hall).

6 CONCLUSIONS

We have presented workflow-based system architec-

ture for the administration and visualization of me-

tadata for 3D virtual museums. Several metadata

standards have been evaluated. ARCO is the pre-

ferred standard due to its high degree of complete-

ness with regard to different metadata types and its

ability to handle 3D content. A case study has prov-

en the feasibility of this concept, an empirical vali-

dation of the enhanced metadata standard is planned.

Another focus of our future work concerns the

inclusion of conversational agents in analogy to

existing solutions for real museums (Kopp et al.,

2005) and agent descriptions in metadata standards.

REFERENCES

ARCO Data Model: http://www.arco-web.org/ams/

Baloian, N., Breuer, H., Luther, W. 2007. Concept key-

boards in the animation of standard algorithms, Jour-

nal of Visual Language and Computing, 2007

Biella, D., 2006. Replication of Classical Psychological

Experiments in Virtual Environments, Logos, Berlin.

Biella, D., Luther, W., 2007. Reusability of 2D and 3D

Course Content for Replicated Experiments in Virtual

3D Environments. In Proceedings of the 6

European

Conference on e-Learning, 2007. ACL Press.

Biella, D., Luther, W. 2008. A Parameterizable Frame-

work for Replicated Experiments in Virtual 3d Envi-

ronments. Proceedings of the 4th International Confe-

rence on Web Information Systems and Technologies

(WEBIST2008), Vol. 1, pp. 316-323

Foo, Schubert, 2008. Online Virtual Exhibitions: Concepts

and Design Considerations. DESIDOC Journal of Li-

brary and Information Technology, Vol. 28 (4), July

2008, pp. 22-34

Gilliland-Swetland, A., 2000. Introduction to metadata:

Setting the stage, online,

http://www.slis.kent.edu/~mzeng/metadata/Gilland.pdf

Keene, S., 1997. Becoming Digital. In: Museum Manage-

ment and Curatorship, Vol. 15 (3), pp. 299-313

Kopp, S., Gesellensetter, L., Krämer, N., Wachsmuth, I.,

2005. A conversational agent as museum guide—

design and evaluation of a real-world application. In

Panayiotopoulos et al. (eds.): Intelligent Virtual

Agents, LNAI 3661, Berlin, Springer, pp. 329-343.

LOM standard, http://ltsc.ieee.org/wg12/files/LOM_1484_

12_1_v1_Final_Draft.pdf

McKenzie, J., 1997. Building a Virtual Museum Commu-

nity, In Proceedings Museum & The Web Conference,

Los Angeles (USA)

Mafo, L., 2007. Interaktion mit Exponaten in virtuellen

Museen anhand von ausgewählten Beispielen. Thesis

(MA), University of Duisburg-Essen.

Nesland, G., Hove L.-J., & Nordbotten, J. C., 2005. The

Virtual Exhibit Creator—A presentation generator for

education use. Proc. of the ITI 2nd Intl. Digital Media

Conference in Egypt Cairo, Egypt.

Patel,

M., White, M., Mourkoussis, N., Walczak, K.,

Wojciechowski

R., Chmielewski, J., 2005. Metadata

Requirements for Digital Museum Environments. Int.

Journal of digital libraries. Vol. 5 (3), pp. 179-192.

Sola, T., 1997. Essays on Museums and Their Theory:

Towards the Cybernetic Museum. Helsinki: Museums

Association, Finland, p. 14

Walczak, K., Cellary, W., White, W., 2006. Virtual Mu-

seum Exhibitions. Computer, Vol. 39 (3), pp. 93-95.

WEBIST 2009 - 5th International Conference on Web Information Systems and Technologies

402