A Generative Approach to Virtual Museums

Daniel Sacher

, Daniel Biella

and Wolfram Luther

INKO, University of Duisburg-Essen, Lotharstr. 63, 47057 Duisburg, NRW, Germany

ZIM, University of Duisburg-Essen, Forsthausweg 2, 47048 Duisburg, NRW, Germany

Keywords:

ViMEDEAS, ViMCOX, Replicave, Metadata, Virtual Museum, 3D Framework, X3D, X3Dom, WebGL,

LIDO.

Abstract:

In this paper we introduce an automated approach for generating virtual museum exhibitions based on meta-

data. We will give an insight into the incorporation progress of X3Dom, as an additional export format sup-

ported by the 3D content generation framework, Replicave. The feasibility of these concepts will be tested by

providing new application examples within the scope of our ongoing cooperation with the Salomon Ludwig

Steinheim Institute.

1 INTRODUCTION

In three recent papers (Biella et al., 2012), (Biella

et al., 2010b) and (Biella et al., 2010a), we described

the Virtual Museum Exhibition Designer using an En-

hanced ARCO Standard (ViMEDEAS). ViMEDEAS

is a set of authoring tools and frameworks to cover

the entire design process of virtual museum planning,

creation, archiving, dissemination and presentation.

The metadata set ViMCOX (Virtual Museum and

Cultural Object Exchange Format) version 1.1 (Wolf

et al., 2012), included in ViMEDEAS, describes vir-

tual museums containing both classical and contem-

porary art, interactive exhibition content, assets, spa-

tial exhibition design and outdoor areas. ViMCOX

is based on international metadata standards and uses

LIDO version 1.0 as interchange and harvesting for-

mat for cultural objects (Coburn et al., 2010).

The Replicave framework is used to present exhi-

bitions locally or online (Biella et al., 2012). Repli-

cave is a Java based X3D toolkit for programmatic

modeling of 3D scenes. It generates 3D content at

runtime, based on exhibition templates or ViMCOX

metadata instances. Recent Web3D developments,

such as HTML5, WebGL and X3Dom, overcome the

disadvantages of plugin-based rendering and facili-

tate the use of widely adopted web techniques like

CSS, JQuery and Ajax as well as providing content

interactivity with existing web APIs and services. To

this end, we propose a new development track using

X3Dom as an additional visualization framework. It

is our goal to ease development of user interfaces us-

ing HTML5 and JavaScript to enable more suitable

illustrations of metadata or larger text-based content

embedded into HTML markup instead of displaying

text-based metadata within the 3D scene.

In addition we will introduce our generative ap-

proach for generating and publishing virtual muse-

ums. This addresses metadata-based modeling, pro-

cedural generation of content and 3D visualization us-

ing the rendering platforms X3D and X3Dom.

2 PORTING REPLICAVE (X3Dom)

The framework Replicave was developed to provide

a cost-efﬁcient way to create virtual museum exhibi-

tions by reusing 3D models and dynamically gener-

ated content based on room templates or ViMCOX

metadata. The framework uses Java and X3D to gen-

erate 3D scenes. Replicave is roughly divided into

three work packages: X3D markup generation, a tem-

plate engine and factory classes for room creation as

well as ViMCOX → X3D metadata mapping. X3D

is an ISO standard 3D format, which is not natively

supported by web browsers and requires plugin in-

stallation. X3Dom is a open source framework for

rendering X3D content in web browsers. X3Dom

uses WebGL and OpenGL ES as a rendering plat-

form, which requires a WebGL capable browser such

as Google Chrome, Safari or Firefox. The Adobe

Flash Player version 11 plugin is used as a fallback

renderer for non-WebGL browsers or if WebGL sup-

port is not available. X3Dom is developed and main-

tained by Fraunhofer IGD. X3Dom proposes a new

HTML proﬁle, which is an extension of X3D’s inter-

change proﬁle. The HTML proﬁle does not include

internal script nodes, pointing device sensor nodes or

274

Sacher D., Biella D. and Luther W..

A Generative Approach to Virtual Museums.

DOI: 10.5220/0004356102740279

In Proceedings of the 9th International Conference on Web Information Systems and Technologies (WEBIST-2013), pages 274-279

ISBN: 978-989-8565-54-9

 2013 SCITEPRESS (Science and Technology Publications, Lda.)

support for X3D ProtoTypes (Jung et al., 2012). The

embeddedX3D scene graph can be manipulated using

the X3Dom integration model (Behr et al., 2009) and

JavaScript to access the HTML5 DOM tree, rather

than using the Scene Access Interface (SAI) and X3D

plugin interfaces (Web3D-Consortium, 2004).

To accomplish the adaption of X3Dom we con-

sidered the use of an additional interposed XSLT

layer to convert the generated X3D markup on the ﬂy.

The X3D-Edit authoring tool already provides XSLT

stylesheets to transform X3D markup into X3Dom

web pages with xhtml markup (Savage-NPS, 2012).

Partially or unsupported node types are detected dur-

ing transformation, but are not omitted (cf. (X3D-

Graphics, 2012)). Furthermore, this approach is suit-

able for converting static scenes but not in terms of

dynamic scenes or interactive content generation and

to utilize the advantages and new features of X3Dom

as part of a SDK. Therefore, we decided to reuse and

extend the existing content generation classes as well

as the X3D writer package of the Replicave frame-

work rather than just convertingX3D content after ex-

port. Replicave’s X3Dom extension is an independent

content generator and SDK. Rapid porting and exten-

sion of source code was possible due to the markup

similarity and almost identical node speciﬁcation of

X3D and X3Dom. Extensions to load Shader, create

CubeMaps and to add JavaScript-EventHandler were

implemented and are accessible through the Repli-

cave API. X3Dom allows adding of JavaScript event

handlers to the following X3D nodes: Group, Trans-

form and Shape. It also supports these mouse events:

click, mousedown, mouseup, mousedown, mouseover,

mouseout and mousemove (Behr et al., 2011). Attach-

ing of inline events is supported via the Replicave

API by utilizing corresponding X3D node attributes

(DOM Level 0). In addition, events can be attached

or removed through the HTML markup (DOM Level

2).

Additional ports have been made to be conform

with X3Dom. This includes the conversion of 23

pre-assembled assets from X3D ProtoTypes to native

X3D Documents, because X3Dom currently doesn’t

provide support to load external or internal X3D Pro-

toTypes. Reordering of vertices for generating ge-

ometric window recesses of exhibition rooms, was

also required to be conform with the ear clipping

algorithm used to triangulate concave polygons in

X3Dom.

We dispensed with the port of the exist-

ing exhibition-templates, room-templates and corre-

sponding room connectors because a similar feature

set and automatized generation of exhibition space

can be adapted more ﬂexibly using the proposed

metadata-based modeling approach described in sec-

tion 3. The reuse of the software package to pro-

cess ViMCOX XML instances, required modiﬁcation

of the 3D metadata mapping procedure described in

(Biella et al., 2012). There are currently two differ-

ent approaches for specifying rooms using ViMCOX

metadata (Wolf et al., 2012):

1. To create rectangular room shapes, it is sufﬁcient

to specify two side lengths and the height of the

room. Each wall is aligned to a compass direction.

Referenced objects can be assigned to walls by

specifying compass directions.

2. To design custom room shapes, it requires at least

three wall deﬁnitions. Each wall needs to be

speciﬁed separately by providing absolute posi-

tion data of the bottom two corner points as well

as wall heights for construction. Objects are as-

signed to walls by specifying the corresponding

wall number. This allows arbitrary wall position-

ing to form the desired room shape.

The ﬁrst method described, maps the ViMCOX meta-

data to a rectangular room template and uses the pa-

rameters for length and height to construct the room,

while the second approach uses factory classes to gen-

erate walls and rooms based on the mapped meta-

data input (Sacher, 2011). It is sufﬁcient to focus

only on the second approach because of equal model-

ing capabilities and less effort is required to maintain

two implementations. Therefore, we implemented

an interface which calculates absolute wall positions

and transforms the compass directions of referenced

exhibits to wall numbers. This ensures that older

museum versions remain executable without post-

processing.

3 GENERATIVE EXHIBITIONS

On-the-ﬂy content creation based on exhibition room

templates (mediaroom, gallery, entrancehall, cloak-

room) was presented in (Biella et al., 2012) and

(Biella et al., 2010b). These templates can be con-

ﬁgured using XML conﬁguration ﬁles or program-

matic through the Replicave API. The previously pro-

posed template-based approach is rather static and

limited due to the fact that only three templates are

customizable in terms of individual positioning of ex-

hibition content. Moreover, room appearances can-

not be customized without manual programming, and

the ability to create individual sophisticated room

shapes is limited. Our metadata-based generative ap-

proach facilitates more ﬂexible generation of exhibi-

tion space and enables automatic distribution of con-

AGenerativeApproachtoVirtualMuseums

275

tents, metaphoric room connectors and camera view-

points marking points of interest.

The generative modeling paradigm, or procedu-

ral modeling, refers to parameters or lists of opera-

tions to generate 3D meshes and content using algo-

rithms rather than modeling interactively with CAD

software and exporting 3D models. Unlike the ap-

proach proposed in GML (Generative Modeling Lan-

guage) (TU Graz, 2010), our parameter design refers

not to low-level abstractions of components to de-

scribe 3D shapes such as points, lines or polygons,

but to an abstract representation of stylistic devices

such as (partition-)walls, windows, doorways or the

complete room shape. Parameters are speciﬁed in

the descriptive metadata sets of ViMCOX XML in-

stances, which are used by Replicave as conﬁguration

for 3D content generation. Architectural design can

be speciﬁed as 2D ﬂoorplan in the ViMCOX meta-

data set. The 2D data will be transformed to 3D when

the virtual museum is generated. Floorplan data can

be acquired using different techniques. Modeling ap-

proaches based on 2D ﬂoorplans–using shape gram-

mar extensions and ontologies with export for dif-

ferent renderer (Tomas Trescak, 2012), transforma-

tion of Industry Foundation Classes (IFC) of build-

ings into CityGML XML instances as explicit geom-

etry (J. Benner, A. Geiger, K. Leinemann, 2005) or

2D ﬂoorplanning using graph theory (Ancona et al.,

2007)–can be found in the literature. Our concept

involves two means of generating content—metadata

and 3D content generation. This facilitates the ex-

change of authoring tools, the use of different 3D

visualisation techniques based on client capabilities

and dynamic generation of exhibitions based on larger

collections. Our metadata processor, the MetaToolkit

(Biella et al., 2012), provides interfaces and model

classes for automatic distribution of exhibits, addi-

tion of predeﬁned viewpoints for each exhibit and

spatial modeling using algorithms. The MetaToolkit

uses JAXB (Java Architecture for XML Binding) to

accomplish a three tier schema mapping XML ↔

POJO → X3D. Schema-derived data classes gener-

ated by JAXB are transformed into corresponding

model classes to achieve convenient use in partial ap-

plications of ViMEDEAS and also to serve as input

for eased mapping of metadata to actual 3D content,

generated by Replicave (Biella et al., 2012).

An abstract architectural overview is presented in

Figure 1. Multimedia content; exported cultural ob-

ject metadata (LIDO); 3D models like (interactive)

exhibits, 3D buildings or 3D interior rooms; 2D pic-

tures; or ﬂoorplans as 2D point set can be used in

ViMCOX. Algorithms can facilitate 2D ﬂoorplanning

and automatic determination of exhibit distribution or

room layout. Authoring tools can modify the content

base as well as ViMCOX metadata instances. The dis-

semination and visualization layers are middleware

for assembling and publishing virtual museum in-

stances, locally or on the web, as well as interpreting

and rendering ViMCOX metadata instances on dif-

ferent visualization platforms. ViMEDEAS currently

only support X3D and X3Dom. Our generative mod-

eling approach becomes more efﬁcient when larger

content bases need to be processed, which will be de-

scribed in the following application example as part of

our ongoing research (cf. section 3.1). The generated

show rooms will serve as an initial starting point for

further manual processing and visual design. Current

research addresses the development of API interfaces

to provide support for automatized generation of vir-

tual exhibition rooms based on simpliﬁed XML con-

ﬁguration ﬁles and virtual objects stored as ViMCOX

instances. The exploratory stage focuses on paintings

and basic room shapes—rectangular shapes (aligned

to compass-directions) or room shapes in form of reg-

ular polygons. Our algorithm requires a list of walls

for each room and an ordered list of exhibitsand doors

to be placed on the corresponding walls. Parameters

for deﬁning left and right spacing between objects are

optional and were initially set to 1.0m. The same

applies for heights of walls and vertical positioning

of the individual virtual objects. Initial textures for

walls, ceiling and ﬂoor are set to the Replicave pre-

set. The algorithm then calculates the individual wall

lengths and determines two side lengths (North-South

and West-East) for the construction of rectangular

rooms. For rooms in shape of regular polygons, the

largest wall will be used as the construction parame-

ter for calculating all wall coordinates. These calcula-

tions are done to preserve symmetric room shapes and

to determine additional spacing parameters to center

the objects as groups on walls, if desired. To calcu-

late wall lengths, the algorithm refers to the virtual

object dimension stored in the linked ViMCOX meta-

data instances. In the case of the ViMCOX object

type painting, the corresponding dimensions are the

image area, frame size and passe-partout size. Af-

terwards, the algorithm determines the positions of

each virtual painting or door and calculates a default

viewpoint for each object. The result is a single as-

sembled ViMCOX metadata instance, containing the

generated room instances with viewpoints, thresholds

(doors) and referenced objects as well as copies of the

initial virtual object instances. The generated virtual

rooms can then be visualized, locally or online, using

the Replicave framework (cf. section 3.2).

WEBIST2013-9thInternationalConferenceonWebInformationSystemsandTechnologies

276

3.1 Application Example

The virtual Fleischhacker museum hosts 196 pictorial

exhibits; 3D assets like plants, pillars, glass vitrines,

benches and information tableaus; as well as 29 re-

constructed tombstones and one greatly enlarged re-

construction of a signet, which serves as eye-catcher

in the entrance hall of the virtual museum. The exhi-

bition design, texts and room arrangementswere elab-

orated by Dr. Barbara Kaufhold and will be themati-

cally arranged throughout fourteen exhibition rooms.

All rooms are in rectangular shape except for the oc-

tagonal entrance hall, an oval hallway and room four-

teen, where we have design freedom to demonstrate

stylistic devices for vivid room design. Room four-

teen will host tombstones and provides access to out-

door areas.

Fleischhacker’s pictorial estate was scanned and

recorded as ViMCOX XML instances and has been,

where applicable, enrichedwith additional descriptive

metadata. For this purpose, we provided a simplistic

metadata editor to automatize data acquisition: Gen-

erating thumbnail and texture, assigning unique iden-

tiﬁers, determining virtual dimensions of each digital

copy and creating a suitable folder structure for fur-

ther automated-processing. The virtual dimensions

were predeﬁned to have an area of no more than one

square meter while preserving the original aspect ra-

tio of each image. The metadata editor was imple-

mented in C#. To map the ViMCOX schema deﬁni-

tion to runtime classes, we used the Microsoft XML

Schema Deﬁnition Tool included in the .Net frame-

work (MSDN, 2011). The EPIDAT database devel-

oped and hosted by the Steinheim Institute, provides

epigraphic information about Jewish tombstones, in-

cluding descriptiveinformation, inscriptions and tran-

scriptions (Steinheim Institute, 2011). These addi-

tional metadata will be stored with the virtual tomb-

stone objects to be visualized in the virtual world.

Ground plot sketches to roughly indicate the spa-

tial design as well as positional information of the

virtual objects were made available to us. Other ini-

tial parameters for object arrangementand interior de-

sign were deﬁned as follows: The bottom edge of

each image will be aligned using an invisible auxil-

iary line at 1.1m; information tableaux will be posi-

tioned at ground level; spacing between digital ob-

jects (images, tableaux) on each wall was set to 1.0m,

while the spacing between doors and objects was set

to 0.5m; and the initial wall height amounts 2.5m.

3.2 Dissemination

Automatic dissemination and deployment of virtual

museums on the web is part of our ongoing research.

We introduced an automatized approach to assem-

ble and distribute Web Application Archives (WAR)

interfacing Apache ANT programmatically on ap-

plication level at runtime in (Biella et al., 2012)

and (Sacher, 2011). The generated WAR ﬁle lay-

out is segmented in four parts: The Java libraries

needed to do the schema mapping and to generate

3D content—the XML processor and the Replicave

framework libraries for generating X3D and X3Dom,

ViMCOX XML instances specifying virtual exhibi-

tions; JavaServer Pages (JSP) to interface the Java

APIs to visualize X3D and X3Dom prospectively. It

also has all required resources, such as X3D mod-

els (exhibits and assets), textures, multimedia ob-

jects and other user content. The intended dissem-

ination and presentation workﬂow can be described

as follows: Automatic or user-generated ViMCOX

XML instances are gathered from the content cre-

ator’s workspace to determine and assemble all re-

quired resources. This includes normalization of path

statements to match web-application conventions and

relative ﬁle path statements. The virtual museum

will be generated on server side at runtime. Meta-

data instances are processed and the virtual museum

will then be generated and presented in accordance

with the the selected visualization method (X3D or

X3Dom).

4 CONCLUSIONS

In general, a metadata-based modeling approach has

the ﬂexibility to exchange authoring software, meta-

data processors or visualization frameworks. The

APIs of the MetaToolkit and Replicave enable the

generation of exhibition space and facilitate automatic

distribution of exhibits. Automatized spatial design is

not limited to rectangular or regular n-sided polyan-

gular room shapes but can also utilize point lists rep-

resenting ground plots or the topographic maps of

rooms and buildings. Procedural modeling based

on native metadata still requires knowledge of soft-

ware engineering and it is mainly designed for expe-

rienced users and IT experts. We believe that exhi-

bition design is a iterative and creative process. Our

(semi-)automatic approach is suitable for larger con-

tent bases and when regular polygonal room shapes

are desired. The supported room shapes are appro-

priate for virtual museums without extravagant archi-

tectural design. In addition, we support the combina-

tion of manually created, auto-generated, prototyped

or dummy content, which can be replaced during the

design process. The generated exhibition rooms serve

AGenerativeApproachtoVirtualMuseums

277

as an initial starting point for further manual process-

ing and visual design. This includes the assignment

of suitable textures, color coordination, ﬁne-tuning of

object positions and scale to create more lifelike exhi-

bition rooms and to suit the curator’s preferences and

perceptions. We are targeting to assess our approach,

by comparing the users experienceperceivedin differ-

ent types of museum exhibitions comprising the same

content on display: real museum exhibitions, equiva-

lent 3D reconstructionsand (semi-)automaticallygen-

erated museums with and without post-processing.

In future prospects we want to combine recom-

mender systems–taking into account user proﬁles in-

cluding demographic data, user’s content and archi-

tectural preferences, their experience in 3D worlds

(addressing navigation aid or modes as well as avail-

able object interaction modes) or their ability to

learn–to generate different exhibitions, assemble dif-

ferent virtual museum tours as well as adding suitable

educational or supplicant material to support the vir-

tual museum visitor. A architecture for adaptive vir-

tual museums was proposed in (Lepouras and Vassi-

lakis, 2004). Another application scenario is the gen-

eration of ﬁctional exhibitions using content ﬁltering

based on metadata categories such as epochs, artistic

style, object material, artists, collections or multime-

dia content.

The main drawback of the current development

track is that X3Dom does not support all the X3D sen-

sor nodes that were available in native X3D, such as

the pointing device sensors SphereSensor, Cylinder-

Sensor and PlaneSensor. These sensor nodes were

used to implement user-to-object interaction, includ-

ing rotating, scaling and translating exhibits. Another

shortcoming in comparison with X3D plugins is the

switching and selecting of viewpoints or even the an-

imated transition between multiple viewpoints of a

scene (viewpoint tour/camera path animation). This

feature set was formerly accessible using the built-in

interfaces of X3D browser plugins. These features are

not part of the X3Dom speciﬁcation and are there-

fore designated to be implemented by web develop-

ers. Thus, we need to re-implement equivalent behav-

ior using JavaScript.

The proposed automatization process for publish-

ing exhibitions on the web requires manual selection

of the desired visualization method and format. A

more convenient way would be to detect client ca-

pabilities using JavaScript, i.e., to detect the WebGL

rendering context or X3D browser plugins to provide

the appropriate visualization method based on client

capabilities.

Moreover, Replicave generates 3D content server-

sided and requires a servlet environment to deliver

web-based exhibitions. We are not currently consid-

ering a JavaScript-based port of Replicave due to the

broad feature set of ViMCOX and the great effort con-

ceivably required to implement complete metadata

mapping using JavaScript and JSON.

When generating virtual museums, Replicave

serves the same generated content and provides

almost the same viewing experiences using X3D

or X3Dom. Nevertheless, X3D-capable browser

plugins, like BSContact (Bitmanagement Software

GmbH, 2012), support per-vertex lighting; thus, the

shading quality of lit objects will increase depending

on the object’s polygon count. On the other hand,

X3Dom supports per-pixel lighting and therefore pro-

vides different viewing experience when light sources

are added to the scene.

The new development track using X3Dom as a

second visualization method is a further step to-

wards multiplatform—desktop, web and mobile—

virtual museums. Initial tests to load generated virtual

museums on mobile platforms succeeded. For testing

purpose we used a Google Nexus 7 tablet running An-

droid 4.1.2 and Firefox 16.0.2. We observed that the

X3D navigation mode EXAMINE works well using

touch input. However, the required X3D navigation

mode WALK—to simulate walkthroughs—currently

does not match the user experience provided by X3D

plugins.

ACKNOWLEDGEMENTS

Many thanks to our project partners from the Stein-

heim Institute and our students for valuable sum-

maries of their diploma theses and for carefully im-

plementing the software, building virtual museums as

well as testing many models and examples.

REFERENCES

Ancona, M., Drago, S., Quercini, G., and Bogdanovych,

A. (2007). Rectangular dualization of bicon-

nected planar graphs in linear time and related

applications. http://citeseerx.ist.psu.edu/viewdoc/

summary?doi=10.1.1.67.3445. [Online; accessed

31.01.2013].

Behr, J., Eschler, P., Jung, Y., and Z¨ollner, M. (2009).

X3DOM: A DOM-based HTML5/X3D integration

model. In Proceedings of the 14th International Con-

ference on 3D Web Technology, Web3D ’09, pages

127–135, New York, NY, USA. ACM.

Behr, J., Jung, Y., Drevensek, T., and Aderhold, A. (2011).

Dynamic and interactive aspects of X3DOM. In

Proceedings of the 16th International Conference on

WEBIST2013-9thInternationalConferenceonWebInformationSystemsandTechnologies

278

3D Web Technology, Web3D ’11, pages 81–87, New

York, NY, USA. ACM.

Biella, D., Luther, W., and Baloian, N. (2010a). Beyond the

ARCO standard. In 16th International Conference on

Virtual Systems and Multimedia (VSMM), pages 184–

191.

Biella, D., Luther, W., and Baloian, N. (2010b). Vir-

tual Museum Exhibition Designer Using Enhanced

ARCO Standard. In XXIX International Conference of

the Chilean Computer Science Society (SCCC), pages

226–235.

Biella, D., Luther, W., and Sacher, D. (2012). Schema

migration into a web-based framework for generat-

ing virtual museums and laboratories. In Proceedings

of the International Conference on Virtual Systems

and Multimedia (VSMM 2012), Milan, Italy, IEEE CS

Press, pages 307–314.

Bitmanagement Software GmbH (2012). BS Con-

tact. http://www.bitmanagement.com/de/products/

interactive-3d-clients/bs-contact. [Online; accessed

05.11.2012].

Coburn, E., Light, R., Mckenna, G., Stein, R., and

Vitzthum, A. (2010). LIDO: Lightweight Information

Describing Objects, Speciﬁcation Version 1.0.

J. Benner, A. Geiger, K. Leinemann (2005). Flexible gen-

eration of semantic 3d building models. Technical

Services Quarterly, pages 17–22. [Online; accessed

05.11.2012].

Jung, Y., Keil, J., and Behr, J. (2012). Demo: X3DOM

Declarative (X)3D in HTML5. http://www2012.

wwwconference.org/proceedings/nocompanion/Dec3

D2012

01.pdf. [Online; accessed 05.11.2012].

Lepouras, G. and Vassilakis, C. (2004). Adaptive virtual

reality museums on the web. http:// www.uoc.edu/

humﬁl/ articles/ eng/ pujol0304/ pujol0304.pdf. [On-

line; accessed 31.01.2013].

MSDN (2011). XML Schema Deﬁnition Tool

(Xsd.exe). http://msdn.microsoft.com/en-us/library/

x6c1kb0s%28v=vs.100%29.aspx. [Online; accessed

05.11.2012].

Sacher, D. (2011). Ein webbasiertes Rahmenwerk f¨ur

virtuelle Laboratorien und Museen: Schemaintegra-

tion und Implementierung. Diploma thesis, University

of Duisburg-Essen.

Savage-NPS (2012). X3D to X3Dom Stylesheet. http://

www.web3d.org/x3d/stylesheets/X3dToX3dom.xslt.

[Online; accessed 05.11.2012].

Steinheim Institute (2011). Epidat - Epigraphical database.

http://www.steinheim-institut.de/cgi-bin/epidat. [On-

line; accessed 05.11.2012].

Tomas Trescak (2012). Intelligent Generation and Control

of Interactive Virtual Worlds. PhD thesis, Universitat

Autonoma de Barcelona.

TU Graz (2010). Generative Modeling Language (GML).

http://www.generative-modeling.org/. [Online; ac-

cessed 05.11.2012].

Web3D-Consortium (2004). Information technology: Com-

puter graphics and image processing – Extensible 3D

(X3D), Part 2: Scene access interface (SAI). http://

www.web3d.org/ﬁles/speciﬁcations/19775-1/V3.0/

index.html. [Online; accessed 05.11.2012].

Wolf, V., Song, Y., Sacher, D., Luther, W., and Biella, D.

(2012). ViMCOX: Virtual Museum and Cultural

Object Exchange Format, Speciﬁcation Version 1.1.

http://www.vimedeas.com/wordpress/?page

id=40.

[Online; accessed 05.11.2012].

X3D-Graphics (2012). http://x3dgraphics.com/examples/

X3dForWebAuthors/Chapter08-UserInteractivity/

Doors.xhtml. [Online; accessed 05.11.2012].

APPENDIX

Replicave

X3D

X3Dom

Dissemination

Visualization

Authoring

Tools

Renderer n

ViMCOX

Content

Algorithms

LIDO

2D / 3D

Multimedia

Figure 1: Abstract architectural overview.

AGenerativeApproachtoVirtualMuseums

279