From traditional to interactive playspaces: Automatic analysis of player behavior in the interactive tag playground

Alejandro Manuel Moreno Celleri

Research output: ThesisPhD Thesis - Research UT, graduation UTAcademic

114 Downloads (Pure)

Abstract

Play is an essential activity for the physical, cognitive and social development of children. Studies have shown that, through play, children can learn what their bodies are capable of, or develop positive social relationships with their peers. With the emergence of digital games, the way in which games are played has changed significantly. Many digital games promote sedentary gaming habits, or are played in such a way that meaningful social interactions cannot occur. On the other hand, digital games can be more fun than traditional games, capable of keeping players engaged for prolonged periods of time. Nowadays, new types of games are being developed that aim to promote the positive behavior associated with traditional play, as well as to retain the benefits of digital games. This is accomplished by employing sensors and actuators such as cameras, projection screens and accelerometers. Would it be possible to leverage these technological elements to design better games and provide enhanced game experiences? Could they be used to automate or improve the way in which we currently study how games are played? In this thesis, we answer these questions. We explore the use of technology to automatically and unobtrusively analyze player behavior in an interactive game installation. We analyzed recordings of children playing traditional tag games to identify ways to improve or automate the process by which the behavior of players is studied. The information derived from the analysis was used to design an interactive playground that enhances the tag game experience while supporting the physical and social aspects of play that are exhibited by players during traditional tag. This installation, the Interactive Tag Playground (ITP), uses cameras to track players in the playing field and projectors to display game elements on the floor. This allows players to move freely during the game. Results from a user study showed that our interactive version of tag was more enjoyable and immersive than the traditional game of tag, while still allowing players to exhibit physically active, social behavior. Besides being an entertainment installation, the ITP doubles as a game research platform. The ITP automatically logs the players' positions and roles. We used this information to automatically measure two aspects of play behavior that are important in the study of interactive playgrounds: physical activity and social interactions. We found that physical activity measured as speed from tracked players correlated well with exertion measurements from heart rate sensors. We also found that differences in social play behavior between children could be measured using social cues, such as the distance between players. Finally, we were interested in the automatic recognition of roles during tag games, as this could be used to find anomalous behavior such as cheating or bullying. Our results showed that automatically estimating players' roles during tag games is possible. In conclusion, the ability to automatically track players makes it possible to derive useful play behavior information. The work presented in this thesis showcases the potential benefits and applications of improving how play behavior is studied. Specifically, the use of automated, quantitative methods complements the qualitative methods currently used to study games. Furthermore, the automated analysis of player behavior can help the design of adaptive game installations and more engaging game experiences.
Original languageUndefined
Awarding Institution
  • University of Twente
Supervisors/Advisors
  • Heylen, Dirk K.J., Supervisor
  • Poppe, Ronald Walter, Advisor
Thesis sponsors
Award date21 Apr 2016
Place of PublicationEnschede, The Netherlands
Publisher
Print ISBNs978-90-365-4101-5
DOIs
Publication statusPublished - 21 Apr 2016

Keywords

  • Entertainment installations
  • Play
  • IR-100257
  • METIS-316455
  • Interactive Playgrounds
  • EWI-27429
  • Behavior understanding
  • Ambient Intelligence

Cite this

Moreno Celleri, Alejandro Manuel. / From traditional to interactive playspaces: Automatic analysis of player behavior in the interactive tag playground. Enschede, The Netherlands : University of Twente, 2016. 142 p.
@phdthesis{73da8dd0f35b4b9aacbef182defd0a71,
title = "From traditional to interactive playspaces: Automatic analysis of player behavior in the interactive tag playground",
abstract = "Play is an essential activity for the physical, cognitive and social development of children. Studies have shown that, through play, children can learn what their bodies are capable of, or develop positive social relationships with their peers. With the emergence of digital games, the way in which games are played has changed significantly. Many digital games promote sedentary gaming habits, or are played in such a way that meaningful social interactions cannot occur. On the other hand, digital games can be more fun than traditional games, capable of keeping players engaged for prolonged periods of time. Nowadays, new types of games are being developed that aim to promote the positive behavior associated with traditional play, as well as to retain the benefits of digital games. This is accomplished by employing sensors and actuators such as cameras, projection screens and accelerometers. Would it be possible to leverage these technological elements to design better games and provide enhanced game experiences? Could they be used to automate or improve the way in which we currently study how games are played? In this thesis, we answer these questions. We explore the use of technology to automatically and unobtrusively analyze player behavior in an interactive game installation. We analyzed recordings of children playing traditional tag games to identify ways to improve or automate the process by which the behavior of players is studied. The information derived from the analysis was used to design an interactive playground that enhances the tag game experience while supporting the physical and social aspects of play that are exhibited by players during traditional tag. This installation, the Interactive Tag Playground (ITP), uses cameras to track players in the playing field and projectors to display game elements on the floor. This allows players to move freely during the game. Results from a user study showed that our interactive version of tag was more enjoyable and immersive than the traditional game of tag, while still allowing players to exhibit physically active, social behavior. Besides being an entertainment installation, the ITP doubles as a game research platform. The ITP automatically logs the players' positions and roles. We used this information to automatically measure two aspects of play behavior that are important in the study of interactive playgrounds: physical activity and social interactions. We found that physical activity measured as speed from tracked players correlated well with exertion measurements from heart rate sensors. We also found that differences in social play behavior between children could be measured using social cues, such as the distance between players. Finally, we were interested in the automatic recognition of roles during tag games, as this could be used to find anomalous behavior such as cheating or bullying. Our results showed that automatically estimating players' roles during tag games is possible. In conclusion, the ability to automatically track players makes it possible to derive useful play behavior information. The work presented in this thesis showcases the potential benefits and applications of improving how play behavior is studied. Specifically, the use of automated, quantitative methods complements the qualitative methods currently used to study games. Furthermore, the automated analysis of player behavior can help the design of adaptive game installations and more engaging game experiences.",
keywords = "Entertainment installations, Play, IR-100257, METIS-316455, Interactive Playgrounds, EWI-27429, Behavior understanding, Ambient Intelligence",
author = "{Moreno Celleri}, {Alejandro Manuel}",
note = "SIKS dissertation series no. 2016-21",
year = "2016",
month = "4",
day = "21",
doi = "10.3990/1.9789036541015",
language = "Undefined",
isbn = "978-90-365-4101-5",
publisher = "University of Twente",
address = "Netherlands",
school = "University of Twente",

}

From traditional to interactive playspaces: Automatic analysis of player behavior in the interactive tag playground. / Moreno Celleri, Alejandro Manuel.

Enschede, The Netherlands : University of Twente, 2016. 142 p.

Research output: ThesisPhD Thesis - Research UT, graduation UTAcademic

TY - THES

T1 - From traditional to interactive playspaces: Automatic analysis of player behavior in the interactive tag playground

AU - Moreno Celleri, Alejandro Manuel

N1 - SIKS dissertation series no. 2016-21

PY - 2016/4/21

Y1 - 2016/4/21

N2 - Play is an essential activity for the physical, cognitive and social development of children. Studies have shown that, through play, children can learn what their bodies are capable of, or develop positive social relationships with their peers. With the emergence of digital games, the way in which games are played has changed significantly. Many digital games promote sedentary gaming habits, or are played in such a way that meaningful social interactions cannot occur. On the other hand, digital games can be more fun than traditional games, capable of keeping players engaged for prolonged periods of time. Nowadays, new types of games are being developed that aim to promote the positive behavior associated with traditional play, as well as to retain the benefits of digital games. This is accomplished by employing sensors and actuators such as cameras, projection screens and accelerometers. Would it be possible to leverage these technological elements to design better games and provide enhanced game experiences? Could they be used to automate or improve the way in which we currently study how games are played? In this thesis, we answer these questions. We explore the use of technology to automatically and unobtrusively analyze player behavior in an interactive game installation. We analyzed recordings of children playing traditional tag games to identify ways to improve or automate the process by which the behavior of players is studied. The information derived from the analysis was used to design an interactive playground that enhances the tag game experience while supporting the physical and social aspects of play that are exhibited by players during traditional tag. This installation, the Interactive Tag Playground (ITP), uses cameras to track players in the playing field and projectors to display game elements on the floor. This allows players to move freely during the game. Results from a user study showed that our interactive version of tag was more enjoyable and immersive than the traditional game of tag, while still allowing players to exhibit physically active, social behavior. Besides being an entertainment installation, the ITP doubles as a game research platform. The ITP automatically logs the players' positions and roles. We used this information to automatically measure two aspects of play behavior that are important in the study of interactive playgrounds: physical activity and social interactions. We found that physical activity measured as speed from tracked players correlated well with exertion measurements from heart rate sensors. We also found that differences in social play behavior between children could be measured using social cues, such as the distance between players. Finally, we were interested in the automatic recognition of roles during tag games, as this could be used to find anomalous behavior such as cheating or bullying. Our results showed that automatically estimating players' roles during tag games is possible. In conclusion, the ability to automatically track players makes it possible to derive useful play behavior information. The work presented in this thesis showcases the potential benefits and applications of improving how play behavior is studied. Specifically, the use of automated, quantitative methods complements the qualitative methods currently used to study games. Furthermore, the automated analysis of player behavior can help the design of adaptive game installations and more engaging game experiences.

AB - Play is an essential activity for the physical, cognitive and social development of children. Studies have shown that, through play, children can learn what their bodies are capable of, or develop positive social relationships with their peers. With the emergence of digital games, the way in which games are played has changed significantly. Many digital games promote sedentary gaming habits, or are played in such a way that meaningful social interactions cannot occur. On the other hand, digital games can be more fun than traditional games, capable of keeping players engaged for prolonged periods of time. Nowadays, new types of games are being developed that aim to promote the positive behavior associated with traditional play, as well as to retain the benefits of digital games. This is accomplished by employing sensors and actuators such as cameras, projection screens and accelerometers. Would it be possible to leverage these technological elements to design better games and provide enhanced game experiences? Could they be used to automate or improve the way in which we currently study how games are played? In this thesis, we answer these questions. We explore the use of technology to automatically and unobtrusively analyze player behavior in an interactive game installation. We analyzed recordings of children playing traditional tag games to identify ways to improve or automate the process by which the behavior of players is studied. The information derived from the analysis was used to design an interactive playground that enhances the tag game experience while supporting the physical and social aspects of play that are exhibited by players during traditional tag. This installation, the Interactive Tag Playground (ITP), uses cameras to track players in the playing field and projectors to display game elements on the floor. This allows players to move freely during the game. Results from a user study showed that our interactive version of tag was more enjoyable and immersive than the traditional game of tag, while still allowing players to exhibit physically active, social behavior. Besides being an entertainment installation, the ITP doubles as a game research platform. The ITP automatically logs the players' positions and roles. We used this information to automatically measure two aspects of play behavior that are important in the study of interactive playgrounds: physical activity and social interactions. We found that physical activity measured as speed from tracked players correlated well with exertion measurements from heart rate sensors. We also found that differences in social play behavior between children could be measured using social cues, such as the distance between players. Finally, we were interested in the automatic recognition of roles during tag games, as this could be used to find anomalous behavior such as cheating or bullying. Our results showed that automatically estimating players' roles during tag games is possible. In conclusion, the ability to automatically track players makes it possible to derive useful play behavior information. The work presented in this thesis showcases the potential benefits and applications of improving how play behavior is studied. Specifically, the use of automated, quantitative methods complements the qualitative methods currently used to study games. Furthermore, the automated analysis of player behavior can help the design of adaptive game installations and more engaging game experiences.

KW - Entertainment installations

KW - Play

KW - IR-100257

KW - METIS-316455

KW - Interactive Playgrounds

KW - EWI-27429

KW - Behavior understanding

KW - Ambient Intelligence

U2 - 10.3990/1.9789036541015

DO - 10.3990/1.9789036541015

M3 - PhD Thesis - Research UT, graduation UT

SN - 978-90-365-4101-5

PB - University of Twente

CY - Enschede, The Netherlands

ER -