Available via license: CC BY-NC 4.0
Content may be subject to copyright.
Virtual Art Exhibition System: An
Implementation Method for Creating an
Experiential Museum System in a Virtual
Space
Tsukasa FUKUDAa,1 and Naoki ISHIBASHI
a
a
Graduate school of Data Science, Musashino University
Abstract. In order to provide art exhibitions in a virtual space which integrates
various data of an art museum and gives an emotional experience, the Virtual Art
Exhibition System is proposed. This research uses a multi-database called Artizon
Cloud to display museum data, combinate with technologies called Data Sensorium
and Torus Treadmill to project images and enable visitors to walk around the virtual
museum. Moreover, the virtual museum will exploit the user’s intentions and be
personalized, automatically generating further art exhibitions.
Keywords. museum systems, virtual reality, virtual museum
1. Introduction
The social transformation with information technologies is rapidly widening in the world
even for public facilities. It is also the same for the museums. Museums in the world
have started importing the transformation, such as digitizing the art information archives
[1], and photographs of artworks beginning to be digitally archived [2,3,4]. As an
example, Google's "Google Art Project" is an archive of extremely high-resolution
digital images, with about 600 museums and galleries participating worldwide [5]. For
another example, archiving the museum layouts through the analysis of visitor behavior
[6], archiving the traditional culture and the staging of new experiences through digital
technology [7] are the examples of the current entries of digitization in museums. As
shown here, there are numerous things that digital implementation can do to museums.
The virtual (VR) museums are also getting lots of attention. Although there are
various definitions of the term "virtual museum" [9], in this paper, a virtual museum is
defined as making digitized objects available online or using high-tech equipment such
as VR glasses to make the visitors feel as if they are in a real-world museum [8].
With the increasing popularity and acceptance of virtual reality technology,
especially in the field of museums, virtual museums have once again become the focus
of attention [9]. Until today, there have been art museums reproduced on websites such
as the online museum services named HASARD [10], which is a service that displays
archives and details of artworks on the website and VIRTUAL MUSEUMS [11], which
1
Tsukasa Fukuda, Musashino University, 3-3-3, Ariake, Kotoku, Tokyo, 135-8181 Japan; E-mail:
g2150002@stu.musashino-u.ac.jp.
Information Modelling and Knowledge Bases XXXIV
M. Tropmann-Frick et al. (Eds.)
© 2023 The authors and IOS Press.
This article is published online with Open Access by IOS Press and distributed under the terms
of the Creative Commons Attribution Non-Commercial License 4.0 (CC BY-NC 4.0).
doi:10.3233/FAIA220491
38
is a service that museums around the world are displayed on the map, and the user can
watch tour videos and 360-degree images of each museum. In this way, the benefits of
digitization have taken us to a stage where visitors can enjoy artworks regardless of the
artwork’s location and the user’s location. Not just an advantage for users, while many
museums around the world have been forced to close temporarily due to the spread of
COVID19, a "virtual museum" that allows visitors to experience the artworks displayed
in the museum via internet can be said to be a lifeline for serving museum services under
the current pandemic circumstance. Although, while there are many virtual museums as
the examples above, there is still no museum that projects the artworks in a form of real
size. In general, real-world museums display multiple artworks in the room and provide
the space and the experiences to the visitors. The concept of a VR museum with the
advantage of being able to view artworks on the internet is very positive, yet we believe
that including more sense of reality to this concept would create an even better
experience for users. According to [12], the author advocates how important a sense of
reality is to the museum experience. The author claims: On a video screen, a painting has
no texture, and when three dimensions become two, something is always lost. Therefore,
realism could be a major keyword in the construction of a virtual museum. Thus, we
have chosen to create a VR museum that provides newly designed art experiences and to
also pursue a sense of reality.
In this paper, we will propose a virtual museum which will be called a Virtual Art
Exhibition System.
2. An Experiential Museum System
2.1. Art Sensorium Project
To enhance the art experiences of art museums, the Art Sensorium Project was launched
at Musashino University in 2020(Fig.1). Current research goals of the project are as
follows: 1) system design and implementation of a multidatabase system that integrates
multiple digital archives of art collections, and 2) system design and implementation of
a virtual exhibition environment that provides more experiential and emotional impact
to an individual. In this paper, we propose an experimental system design and
implementation of a Data Sensorium application for Artizon Cloud [1] that 1)
automatically displays artworks, 2) extracts users' interests, and 3) automatically re-
displays artworks based on the extracted interests.
Figure 1. An overview of Art Sensorium Project
T. Fukuda and N. Ishibashi / Virtual Art Exhibition System 39
2.2. Virtual Art Exhibition System
The goal of the Virtual Art Exhibition System is to provide art experiences to each user
according to their interests and intentions. Toward the goal, we aim to pursue reality and
provide a dynamic virtual museum experience in cyberspace based on the idea
“Emotional Excitements” which we defined as something that visitors can truly enjoy
and be moved by.
By using the multidatabase system, where many artworks are converted into image
data, it is possible to display numerous artworks in a virtual space which would not be
possible in a physical museum. Furthermore, the Virtual Art Exhibition System will
exhibit dynamic and automatic curation according to each visitor’s intention and interests.
No matter the user’s and artwork location, this could provide an “Emotional Excitement”
that goes beyond the world. Below shows the features of the proposed system:
1) Personalized Exhibition generated based on the user's physical characteristics
Each user will be asked to input the height information before viewing. Then the system
will be displaying artworks adjusted to each user's height and the height of the eyes so
the users could easily view the artworks in the best position.
2) Personalized Exhibition generated from the user's intension and interests
The concept of personalization is extracted from the user's interests and intentions. The
definition of "interest" and "intention" is complex, and there are various ways to extract
user interest. Such as letting the user input, recording the user’s walking trajectory, the
time the user stopped in front of the artwork, user eye tracking, etc. In this system, we
defined "user interest" as the time the user spent in front of the artwork.
By combining these factors, visitors will be able to experience an even more
immersive experience in the virtual museum, and a unique “Emotional Experience” that
could only be achieved in virtual space.
2.3. Exploiting users’ intentions and generating art exhibitions
As a premise, we define that the more the user spends time in front of the artworks, the
more the user prefers the artwork. In other words, it can be said that the user’s behavior
shows the user's preferences. In this system, we succeeded creating 2) Personalized
Exhibition in which generating and exhibiting the artworks related to the user's intention
by calculating the user’s staying time in front of the artworks. Firstly, a set of artworks
will be randomly exhibited in the exhibition room. While the user is walking around, the
artwork that the user viewed the longest will be extracted. Secondly, the new set of
related artworks will be searched from the database, based on the chosen artwork (which
was most viewed from the previous set) (Figure 2). Related artworks are defined as:
artwork that is from the same period, author, and country to the chosen artwork. Finally,
the new set of the related artworks will be exhibited in the virtual museum. This curation
function is designed as a database user interface in the virtual space, so the intension is
converted to a query in SQL form. When a user spends more time in the virtual space for
viewing many sets of artworks, more precise interest for artworks could be extracted for
searching in detail. There are many possible approaches to re-design the curation
function such as semantic computation or spatio-temporal computation, although these
more advanced methods will be discussed in the future.
T. Fukuda and N. Ishibashi / Virtual Art Exhibition System40
Figure 2. An implementation of the art exhibition system.
3. Implementation Environment
To implement the actual size exhibition for the proposed system, the proposed system
uses a dynamic full-scale projection system called Data Sensorium which can provide
realistic virtual exhibitions of artworks. The next section will explain Data Sensorium,
and how it will be used in the Virtual Art Exhibition system.
3.1. Data Sensorium
Data Sensorium [13] is a Full-scaled and omnidirectional screened technology designed
by Iwata, H. of Tsukuba University in 2020 which can project 3D images on the wall
and floor using a total of 12 projectors. The proposed system also uses a VR locomotive
device called a Torus Treadmill [14] to let the visitors actually walk around the virtual
museum with the user's own body. By using these systems in which projecting the
museum space and allowing users to walk around using these technologies, it will be
possible to view artworks in their actual size surrounding the users and provide a
dynamical virtual art experience.
3.2. Artizon Cloud
For the database, we use an art multidatabase system called Artizon cloud as target art
data. The system combines multiple databases and provides information about an art
museum and artworks which is collected from various archives of artworks and past
exhibitions. Currently, the Artizon cloud only handles data of Artizon Museum [15]. By
using this multidatabase system, we will be able to display heterogeneous content at the
Virtual Art Exhibition system.
T. Fukuda and N. Ishibashi / Virtual Art Exhibition System 41
Figure 3. An implementation of Artizon cloud
Figure 4. System Structure of the Virtual Exhibition System
T. Fukuda and N. Ishibashi / Virtual Art Exhibition System42
Figure 5. A Screenshot of an exhibition in Virtual Art Exhibition Renderer
4. System structure & Implementation Method
4.1. System Structure
Virtual Art Exhibition System extracts art data from the Art Collection Database, accepts
user interaction data obtained from the Data Sensorium, and sends graphics to Data
Sensorium (Fig.4). To achieve this, the system consists of the 4 following elements.
4.2. Implementation Method
a) Virtual Art Exhibition Renderer
The first step to set up a virtual museum is to create a room in virtual space (Figure 5).
After creating the room, two pictures are placed on each side, in a total of eight artworks
in the room (Figure 7). Outside the room (Figure 6), there are two bars which the user
can go through. The right one will be re-searching the related artwork based on the most
viewed artwork by the user. This means, if the user enters the right bar, the user will be
shown a new set of artworks related to the user's intention. The left bar will reset the
user’s viewed data. The exhibiting artwork’s size will be based on the actual size of each
artwork when loaded. When the user starts the program, each user is assigned an ID.
Based on the ID, the program will calculate and display which picture the user most liked
(which most stayed longer). Figure.8 shows a sphere embedded underneath each artwork
as a component to record the user’s viewing time, and the sphere is set to 2m radius and
invisible by default so that it will not disturb the exhibition. If the user gets close enough
to the artwork, the sphere will count as “IN”, and when the user gets away, it will count
as “OUT”. Timestamps of IN/OUT as well as an artwork identifier are sent to the logging
function. The virtual art exhibition renderer is implemented by using a cross-platform
game engine, unity [16]. There are many possible approaches to extract users’ intention
such as monitoring gaze, EEG detection and so on, however we decided to apply the
simple metric using sphere to study feasibility of the Virtual Art Exhibition system. More
intelligent metrics should be discussed in the future.
T. Fukuda and N. Ishibashi / Virtual Art Exhibition System 43
Figure.6 Outside Appearance of Virtual Exhibition Hall
Figure 7. The implem entation of personalized
exhibition (1)
Figure 8. The implementation of personalized
exhibition (2)
b) Virtual Art Exhibition Generator
Virtual Art Exhibition Generator receives data of the user behavior with the Logging
Function, then generates a set of artworks according to users’ interest from the Curation
Function. The collection data of the Artizon Museum are used as the target data. Since
the museum collection is managed by using the Artizon Cloud, a proprietary multi-
database system, this system uses the Artizon Cloud API to use the collection data.
c) Logging Function
Logging Function receives IN/OUT timestamps as well as the artwork identifiers and
registers these data in the database which is in the Virtual Art Exhibition Generator.
d) Curation Function
The method to calculate the total viewing time per each artwork will be calculated the
time the user entered the sphere (OUT) -(minus) the time the user left the sphere (IN).
The result time will be defined as the user’s staying time in front of the artwork, which
means the user has the most “intention” towards. After getting the staying time result,
the system will list it in order of the most-stayed and most-less-stayed. After the most-
stayed artwork gets extracted, the system will search the next set of related artworks and
send the set of the art to the Virtual Art Exhibition Renderer. Once again, related artworks
are defined as: artwork that is from the same period, created by the same author and same
country to the chosen artwork.
T. Fukuda and N. Ishibashi / Virtual Art Exhibition System44
5. Feasibility Study
As a feasible study, we will explain in detail how the changes of the exhibits in the
exhibition room experienced by a user using the actual system are presented. Table 1
shows an initial set of artworks that is randomly exhibited. In this example, the user
stayed in front of Paul Klee's “garden in POT” for the longest time among other artworks.
After the user finishes viewing the exhibition and goes out of the room and re-search the
related art works, the next set of artworks is exhibited (Table 2). In this case, it shows
that the set of related artworks had been extracted by the same period and same author
to the chosen artwork in the first set. For the second exhibition, the user viewed the
artwork named “Sunrise” as the longest. Based on this result, the system exhibited the
third set of artworks based on the same author and the same country to the chosen artwork.
Overall, through this experiment, it was proven that the Virtual Art Exhibition
System had worked successfully and could automatically generate exhibits based on the
user's interests and intentions.
Table 1. The searched result of database (1)
Table 2. The searched result of database (2)
Henri MATISSE Odalisque 1926 Le Cateau-Cambresis
{now Le Cateau} 00:00:04.319401
KOIDE Narashige
N
ude 1926 Osaka 00:00:02.767603
Georges ROUAULT VIII. Satin III (from
Flowers of Evil) 1926 Paris 00:00:02.566769
Georges ROUAULT
XIII. "Debauchery
and death..."(from
Flowers of Evil)
1926 Paris 00:00:02.552296
YOSHIDA Hiroshi Sunrise (from Views
of Mt. Fuji) 1926 Kurume, Fukuoka 00:00:19.112251
Paul KLEE House in the Water 1930 Munchenbuchsee,
nr Berne 00:00:03.921128
Paul KLEE Garden of the Level-
Crossing Attendant 1934 Munchenbuchsee,
nr Berne 00:00:03.653234
Paul KLEE Jointed Doll Walking 1937 Munchenbuchsee,
nr Berne 00:00:06.153174
Marino MARINI Rider on an Ochre
Ground 1957 Pistoia 00:00:04.054004
Marino MARINI Waiting 1965 Pistoia 00:00:03.500778
KAWABATA Minoru Untitled 1993 Tokyo 00:00:08.23219
MASANOBU Masatoshi Work 1967 Susaki, Kochi 00:00:03.451733
FUKUSHIMA Hideko Displacement
Diagram 1974 Tokyo 00:00:04.386278
TAKIGUCHI Shuzo Toyama
Henri MATISSE Woman with
Checked Collar 1937 Le Cateau-Cambresis
{now Le Cateau} 00:00:07.014538
Paul KLEE Garden in POT 1926 Munchenbuchsee, nr
Berne 00:00:13.242213
Lee KRASNER Moontide 1961 New York 00:00:02.550423
T. Fukuda and N. Ishibashi / Virtual Art Exhibition System 45
AOKI Shigeru Bathers 1904 Kurume, Fukuoka 00:00:03.785742
AOKI Shigeru Landscape 1910 Kurume, Fukuoka 00:00:02.802216
TAKATA Rikizo Fountain at the Villa
Medici 1972 Kurume, Fukuoka 00:00:15.76204
KOGA Harue
Copy from Hans
Prinzhorn's Artistry of
the Mentally Ill
c. 1930 Kurume, Fukuoka 00:00:03.76265
TAKATA Rikizo Erechtheum in
Athens 1938 Kurume, Fukuoka 00:00:01.786504
YOSHIDA Hiroshi Matterhorn, Evening 1925 Kurume, Fukuoka 00:00:02.053703
SAKAMOTO Hanjiro
Enokidera Shrine
(from Five Views of
Tsukushi)
1918 Kurume, Fukuoka 00:00:04.18662
SAKAMOTO Hanjiro
Hatsuse Namiko
playing Akiyama
Sizuko (from Sketch
of Players on Stage)
1971 Kurume, Fukuoka 00:00:09.803794
6. Conclusion
In this paper, a virtual art museum system is proposed. The virtualization of museums
has a lot of advantages, such as creating a chance for users to encounter new, unknown,
spectacular artworks and art experience. This benefit is not just for the virtual museum,
but also for physical museums as well, which means physical and virtual museums can
ultimately reinforce each other. In addition, digitization of artwork has the advantage that
artworks can be permanently preserved and passed on to the next generation. As
mentioned in the beginning, The goal of the system is to create a realistic, full-scaled
virtual museum. Therefore, the proposed virtual museum used the multidatabase system
called art collection database (Artizon Cloud) [1] and a walking device Torus Treadmill
and Data Sensorium. In this case, we succeeded in generating a museum based on the
user's interests and intentions. Through the feasibility study, based on the key word of
“Emotional Excitements”, the virtual art museum system provided users an art
experience that can only be achieved through virtual technology such as personalized
exhibitions. As for the future works, more intelligent and dynamic sensoring metrics to
extract precise user’s intention, semantic computation functions for the dynamic curation,
storing and sharing functions of user experiences for art communication could be
challenged.
Acknowledgement
This research is strongly supported by Asia AI Institute of Musashino University,
especially Prof. Hiroo Iwata for introducing Data Sensorium and Prof. Yasushi Kiyoki
for coordinating the Art Sensorium Project. The art data are provided by Artizon
Museum and Ishibashi Foundation, and a collaborative research relationship with the
museum have strengthened this research. We would like to show gratitude to everyone
in these organizations.
T. Fukuda and N. Ishibashi / Virtual Art Exhibition System46
References
[1]
Ishibashi, N.: “Artizon Cloud: A Multidatabase System Architecture for an Art
Museum,” The 31ST INTERNATIONAL CONFERENCE ON
INFORMATION MODELLING AND KNOWLEDGE BASES(EJC2021).
(2021)
[2]
The Metropolitan Museum of Art: The Met Collection, available via WWW,
https://www. metmuseum.org/art/collection (2021)
[3]
Muse e du Louvre: Atlas database of exhibits, available via WWW,
http://cartelen.louvre.fr/ (2021).
[4]
Paris Muse es: Les collections en ligne des muse es de la Ville de Paris, available
via WWW, https: //www.parismuseescollections.paris.fr/ (2021).
[5]
Muroya, T.: “A Digital Image Archive in a Museum” (In Japanese) available via
https://www.jstage.jst.go.jp/article/itej/70/1/70_160/_pdf/-char/ja (2016)
[6]
Kovavisaruch, L., Sanpechuda, T., Chinda, K., Kamolvej, P. and
Sornlertlamvanich, V.: “Museum Lay- out Evaluation based on Visitor
Statistical History,” Asian Journal of Applied Sciences, Vol.5, No.3, pp.615-
622 (2017)
[7]
Digitized Thailand: White Paper, NECTEC, available via WWW,
https://www.facebook.com/ digitizedthailand (2009).
[8]
Latham, K.F., J.E. Simmons.: “Foundations of Museum Studies. Evolving
Systems of Knowledge” Libraries Unlimited, Santa Barbara, CA. (2014)
[9]
SCHWEIBENZ, W.: “The virtual museum: an overview of its origins, concepts,
and terminology” (2019).
[10]
HASARD: https://wam-hasard.com/
[11]
VIRTUAL MUSEUMS: https://virtualmuseums.io/
[12]
Mintz, A.: “Media and Museums: A Museum Perspective” (1998)
[13]
IWATA, H., SASAKI, S., ISHIBASHI, N., Sornlertlamvanich, V., ENZAKI,
Y., KIYOKI, Y.: “Data Sensorium: Spatial Immersive Displays for
Atmospheric Sense of Place,” The 32ND INTERNATIONAL CONFERENCE
ON INFORMATION MODELLING AND KNOWLEDGE BASES(EJC2022).
(2022, submitted)
[14]
Iwata, H.: The Torus Treadmill: Realizing Locomotion in VEs, IEEE Computer
Graphics and Applications; Vol.19 No.6; 1999 pp.30-35
[15]
Artizon Museum: https://www.artizon.museum/en/
[16]
Unity: https://unity.com/
T. Fukuda and N. Ishibashi / Virtual Art Exhibition System 47
