VIRTUAL EXHIBITION BUILDER

Jorge Carvalho Gomes, Maria Beatriz Carmo and Ana Paula Cláudio

Faculdade de Ciências da Universidade de Lisboa, Lisboa, Portugal

Keywords: Virtual Exhibition, Virtual Museum, X3D, Xj3D.

Abstract: Virtual visits on the Web are an important means to publicize a museum collection and attract visitors.

Software applications for building virtual exhibitions, in addition to being tools to create content for virtual

visits, may help the staff of a museum to conceive and mount exhibitions. This paper presents the Virtual

Exhibition Builder, an interactive software tool aimed at the creation of virtual exhibitions, given a 3D

model of the exhibition space in an X3D file and a set of information about the artworks that are to be

exposed. The development of this tool involves inputs proposed by museum experts.

1 INTRODUCTION

Museums located around the world are open to

remote, on-line visitors through the World Wide

Web, spreading their collections and attracting

visitors to the museum itself.

Offering 3D virtual exhibitions may provide a

sense of immersion inside the exhibition space and

allow free navigation. They require the creation of

the virtual model of the exhibition space and the

placement of the artworks inside it. The latter task

should be performed by museum experts, which

seldom include information technology specialists.

Thus an interactive tool for assisting the placement

of artworks in selected locations, besides being

useful to create virtual exhibitions, can also be quite

a valuable auxiliary tool to mount a new exhibition.

Some tools have been proposed to help a

museum curator to conceive and mount an

exhibition (Hrk, 2001; Wojciechowski, 2004;

Katalabs).

In previous work (Semião, 2008), an application

was conceived aiming to support the creation of

virtual exhibitions by the museum curator. It was

based on Web3D technology (Web3D), generating

3D environments in X3D. This prototype had some

limitations and to overcome them, a new version of

the tool with additional features was implemented

(Gomes, 2010). This tool was presented to museum

experts to get advice about the functionalities that

should be provided and, as a result, it was updated

taking into consideration their observations. This

paper presents this revised version, called Virtual

Exhibition Builder: in section 2 are referred the

opinions given by museum experts; section 3

describes the application; and in section 4 are drawn

conclusions and is indicated the future work.

2 INPUT FROM A MUSEUM

TEAM

The new application was presented to the director of

the City Museum of Lisbon Town Council and to

her team. Throughout the year, this museum presents

several temporary exhibitions, along with its

permanent collection.

The functionalities of the new application

comprised: the selection of the surfaces that could

exhibit artworks; the placement of 2D and 3D

artworks and removable structures, such as dividing

walls to dispose paintings or plinths to display 3D

objects; the refinement of previously built

exhibitions; the creation of an X3D file with the

resulting virtual exhibition.

All members of the museum team considered

that this kind of tool can help them to prepare an

exhibition. It can be used to try different approaches

to arrange the artworks within the exhibition space.

Afterwards, these preliminary proposals can be

discussed both among the elements of the team and

with the authors of the artworks.

It was concluded that photorealistic images are

not an important issue in this stage, since the main

concern is to decide about the harmonious

330

Gomes J., Carmo M. and Cláudio A..

VIRTUAL EXHIBITION BUILDER.

DOI: 10.5220/0003366503300333

In Proceedings of the International Conference on Computer Graphics Theory and Applications (GRAPP-2011), pages 330-333

ISBN: 978-989-8425-45-4

 2011 SCITEPRESS (Science and Technology Publications, Lda.)

arrangement of the artworks. Therefore, the ability

to dispose and simulate light sources is not required.

The tool preserves the relative dimensions of the

artworks and of the exhibition space which is a

significant feature according to the team.

The creation of virtual exhibitions to be available

on the Web was also considered an interesting

functionality.

They proposed some new features, namely, to

provide a new kind of removable structure that

simulates the limits of a glass case to cover 3D

artworks, giving a better perception of the available

space. They also suggested that simplified models of

3D artworks could be a reasonable approach to

speed up the preparation of temporary exhibitions.

Besides improving the application’s performance,

they avoid the time consuming task of creating

detailed 3D models.

It was noticed that navigation in 3D space can

raise some problems in achieving the correct

position of the user. This can be corrected by

defining viewpoints at key points allowing the user

to move among them.

3 VIRTUAL EXHIBITION

BUILDER

The development of this tool was initially guided by

the basic concern of producing a platform to assist

museum teams, most probably non-informatics, to

prepare an exhibition and create a virtual exhibition.

Moreover, it was intended to be used for different

physical spaces.

Since mounting the exhibition involves selecting

artworks and associating each one with the surface

where it will be exposed, one of the problems arising

for the generic treatment of any exhibition space is

the need to automatically add to the description of

the exhibition space, modelled in X3D format, the

capacity to choose these surfaces. In the case of X3D

this process involves touch sensors in areas that can

become suitable for the placement of objects.

Furthermore, it is necessary to store all the

information about the structure of the virtual

exhibition model and all amendments in order to

support future alterations. To achieve this goal it was

conceived a conversion mechanism that transforms

any X3D model of an exhibition into a generic,

geometric description enabling a uniform treatment

of any scene.

The following subsections present the

architecture of the application, its basic features and

the user interface.

3.1 The Architecture of the Application

The architecture of the application is based on a

modular design in order to allow easy

implementation of new features.

The functionalities of the application are split

into multiple edit modes. An edit mode is an

independent unit, associated with a specific action

that can be undertaken, such as, adding a 2D

artwork, adding a 3D artwork, adding a removable

structure, creating a viewpoint. It has the capacity to

change the scene in some particular way and record

these changes on the X3D file that stores the final

virtual exhibition.

The articulation with each module is done

through well defined interfaces that contain the

necessary methods for the application’s workflow,

such as switching edit mode, saving the current

logical state or obtaining the graphical user interface

associated with the edit mode. Through these

interfaces, the application core can glue everything

together without relying on the implemented

modules.

There is an abstract implementation of a module

for the generic placement of objects, which includes

all the common features concerning object

placement. For each specific type of objects that can

be added to the scene, a derived module of this

abstract module is implemented, taking into account

the particular characteristics of the object.

Information about the artworks is stored in a

database integrated into the application, using

SQLite. Although it would be more interesting to

link the application to existing databases, this

solution would certainly raise compatibility issues

due to the lack of standardization in museums

databases. Solving these issues would require some

implementation specific knowledge, which is not

desired in this application. Moreover, as stated by

the experts of the museum, when dealing with the

conception of temporary exhibitions, it is a more

reasonable approach to use a database separate from

the one that contains the museum collection.

The elementary steps the user must take in order

to build a virtual exhibition are: load a X3D model

with the rooms of the exhibition space; select

surfaces adequate to display artworks; place and

adjust the artworks in those surfaces; optionally

define specific viewpoints to help navigation around

the exhibition; export the new X3D model. These

features will be detailed in the next subsection.

The application is implemented on Java SE 1.6

and uses Xj3D API 2.0M1, which builds and

handles the scene graph. To access and modify the

contents of this graph the methods of the SAI (Scene

VIRTUAL EXHIBITION BUILDER

331

Access Interface) Web3D standard are used. We

resorted to Java3D API 1.3 for some phases of the

geometric processing. The graphical user interface is

built on Java Swing.

3.2 Basic Features

Standardization of the Scene Geometry. Despite

being a well-defined format, X3D allows multiple

internal representations for the same visible result,

due to the wide variety of nodes that exist to

describe the geometry of an object. Thus, analyzing

an X3D scene when we do not know its creation

process poses major difficulties for interpreting its

geometry in an easy, complete and correct way. To

overcome this difficulty, we devised a process that

does not change the initial description of the scene

but adds a new definition of its geometry, which

contains the information necessary to add new

objects to the scene.

This abstraction layer of the initial scene

geometry is composed of a set of invisible surfaces

which are placed in the top of the original geometry.

Each surface is a flat area characterized by a single

normal vector and comprising an arbitrary number

of adjacent triangles. All the modifications done to

the scene, including the addition of touch sensors or

the highlight of surfaces, are done in this layer,

leaving the original scene completely untouched.

Selection of Surfaces. Before starting the placement

of art objects, the user must explicitly choose which

areas are suitable for placement. This selection can

be interactive, choosing suitable surfaces one by one

or using pre-defined filters. These filters examine

each surface and decide whether it should be

selected and can also be combined and used to

perform the inverse operation, that is, to reject

surfaces instead of selecting them.

At the current development stage, there are: an

area filter that selects all the surfaces that have a

certain minimum area; and a normal filter that

selects the surfaces whose normal satisfy the

conditions imposed by the user.

Placement of Objects. To allow extending the type

of objects that can be inserted into the scene, we

defined a set of basic parameters that characterize an

object and give support to the generic mechanism of

object placement: the bounding box; the base of

contact with the surface where it is placed; the

normal vector of this surface (corresponding to the

face or the visible volume of the object); and the

orientation of its upper edge.

After selecting the object to be inserted into the

scene, the previously chosen surfaces become

selectable and the user must pick one with the

mouse. The centre of the contact surface of the

object is placed on the specific surface point selected

with the cursor. Then the object can be rotated, its

position can be adjusted or can also be placed on

another surface.

2D artworks are stored as images in the database,

along with additional information about the piece.

One of the characteristics stored in the database is

the physical dimensions of the artworks, so they are

displayed using the correct scale.

3D artworks are stored as X3D models. A

simplified representation can be built based on the

physical dimensions and a single image of the

artwork. In this case, a parallelepiped is placed in the

scene

and the image of the object is applied as

texture to all faces. In Fig. 1 there is an example of

two different representations of the same 3D

artwork.

Figure 1: Two alternative representations of the same

artwork are shown.

In addition to artworks, it is also possible to add

removable structures, like dividing walls or plinths.

For this purpose the user can place in the scene

parallelepiped objects, choosing their size and

colour. Once placed in the scene, the areas that

compose the surfaces are also considered suitable for

the placement of artworks and can be chosen in

surface selection edit mode.

Viewpoints. X3D allows the definition of pre-

defined positions for the user, called viewpoints,

which are associated with a direction of sight. To

support viewpoint definition interactively, the

application is capable of automatically obtaining the

position from where the user is looking at the scene

and allows the user to define that view as a

viewpoint.

X3D Exhibition File Generation. The X3D file

containing the description of the virtual exhibition is

produced through a process where each edit mode

GRAPP 2011 - International Conference on Computer Graphics Theory and Applications

332

independently writes its changes to the output file.

The file containing the original scene graph is

modified using the DOM (Document Object Model)

interface provided by Java API for XML processing.

Note that the changes made to the original file are

mostly addition of new nodes, not modification of

existing ones.

Exhibition Re-edition. To make adjustments to a

virtual exhibition previously built, the X3D file

resulting from its construction does not have all the

required information. Hence it is necessary to know

which objects were introduced in the initial scene.

To enable the re-editing of virtual exhibitions, a

mechanism was designed based on a set of auxiliary

data structures that store information about all

objects added to the scene. The contents of these

structures can be saved in a file - called the state file

- which reflects the current editing state of a scene.

3.3 User Interface

The user interface is composed of two sets of

elements: one is shared by all edit modes while the

other depends on the active edit mode.

The shared elements are split into three different

zones (Fig. 2): (1) the menu bar, with common

options like file save and file load; (2) the edit mode

switch buttons and (3) the 3D browser, including the

display area and the navigation bar.

The edit mode dependent options (4) are

displayed in two lateral panels, which can be

detached from the main window, as shown in Fig. 2,

or hidden in order to maximize the 3D display area.

Figure 2: User interface for 2D object placement.

The added objects in the 3D scene are

highlighted using coloured bounding boxes whose

visual appearance indicates if they belong to the

active edit mode, if they are selected or if the cursor

is over them, as shown in Fig. 1. The selectable

surfaces also display similar behaviour, but using

colour overlays to achieve the different appearances.

4 CONCLUSIONS AND FUTURE

WORK

We have presented an interactive tool - Virtual

Exhibition Builder - for building virtual exhibitions.

This tool aims to assist museum teams in the initial

stages of mounting an exhibition. In addition, the

resulting virtual exhibition can be shown through the

Web for diffusion purposes.

The development of this tool received input from

the experience gained in a previous work and was

designed in order to overcome its limitations. This

tool was presented to specialists from a museum and

several improvements were introduced afterwards.

Future developments include the creation of

printed layouts of the surfaces, with artworks

annotated with the distances between them and the

floor; the ability to change the colour of the

exhibition surfaces; the projection of videos on

surfaces; the reproduction of music depending on the

site that is being viewed at the time; the capability of

displaying in the scene details from the selected

artworks, such as title and author, in order to provide

more information to visitors at the virtual exhibition.

ACKNOWLEDGEMENTS

We would like to thank to Foundation Amadeu Dias

(University of Lisbon) and to Dr. Ana Cristina Leite

and her team.

REFERENCES

Gomes, J. C., Carmo, M. B., Cláudio, A. P. 2010.

Construção Interactiva de Exposições Virtuais.

INFORUM 2010, in Portuguese

Hrk, S. 2001. Virtual Art Gallery. CESCG 2001.

Katalabs, http://www.katalabs.com/blog/

Semião, P. M., Carmo, M. B. 2008. Virtual Art Gallery

Tool. GRAPP 2008, pp 471-476

Web 3D, http://www.web3d.org/

Wojciechowski, R., Walczack, K, White, M., Cellary, W.

2004. Building Virtual and Augmented Reality

Museum Exhibitions. 3D Web Technology 2004, pp

135-144

(4)Edit mode specific options

(1)Menu bar

(3)Display area

(3)Navigation bar

(2)Edit mode switch buttons

VIRTUAL EXHIBITION BUILDER

333