Virtual-_Augmented_and_Mixed_Reality

104 C. Roca-Gonzalez et al. 3. Lohman, D.F.: Spatial abilities as traits, processes, and knowledge (1988) 4. McGee, M.G.: Human spatial abilities: Psychometric studies and environmental, genetic, hormonal, and neurological influences. Psychol. Bull. 86, 889–918 (1979) 5. Maier, P.H.: Spatial geometry and spatial ability–How to make solid geometry solid, pp. 63–75 (1996) 6. Mafalda, R.: Efeitos do uso de diferentes métodos de representaçao gráfica no desenvol- vimento da habilidade de visualizaçao espacial, Sao Paulo (2000) 7. Carroll, J.B.: Human cognitive abilities. Cambridge University Press, Cambridge (1993) 8. Hegarty, M., Waller, D.: Individual differences in spatial abilities. The Cambridge hand- book of visuospatial thinking, pp. 121–169 (2005) 9. Golledge, R.G.: Spatial Behavoir: A Geographical Perspective. Guilford Press (1997)10. Hegarty, M., Richardson, A.E., Montello, D.R., Lovelace, K., Subbiah, I.: Development of a self-report measure of environmental spatial ability. Intelligence 30, 425–447 (2002)11. Montello, D.R.: Scale and multiple psychologies of space. In: Campari, I., Frank, A.U. (eds.) COSIT 1993. LNCS, vol. 716, pp. 312–321. Springer, Heidelberg (1993)12. Allen, G.L.: Functional families of spatial abilities: Poor relations and rich prospects. In- ternational Journal of Testing 3, 251–262 (2003)13. Allen, G.L., Kirasic, K.C., Dobson, S.H., Long, R.G., Beck, S.: Predicting environmental learning from spatial abilities: An indirect route. Intelligence 22, 327–355 (1996)14. Montello, D.R., Lovelace, K.L., Golledge, R.G., Self, C.M.: Sex Related Differences and Similarities in Geographic and Environmental Spatial Abilities. Ann. Assoc. Am. Geogr. 89, 515–534 (1999)15. Kozhevnikov, M., Hegarty, M.: A dissociation between object manipulation spatial ability and spatial orientation ability. Mem. Cognit. 29, 745–756 (2001)16. Hegarty, M., Waller, D.: A dissociation between mental rotation and perspective-taking spatial abilities. Intelligence 32, 175–191 (2004)

Staging Choreographies for Team Training in Multiple Virtual Worlds Based on Ontologies and Alignments Emanuel Silva1,2, Nuno Silva2, and Leonel Morgado3 1 University of Trás-os-Montes e Alto Douro, Vila Real, Portugal 2 School of Engineering, Polytechnic of Porto, Porto, Portugal {ecs,nps}@isep.ipp.pt 3 INESC TEC (formerly INESC Porto) / Universidade Aberta, Lisbon, Portugal [email protected] Abstract. In this paper we present an approach that makes possible the staging of choreographies for education and training purposes in potentially any virtual world platform. A choreography is seen here as the description of a set of ac- tions that must or may be executed by a group of participants, including the goals to be achieved and any restrictions that may exist. We present a system- architecture and the formalization of a set of processes that are able to transform a choreography from a platform-independent representation into a specific vir- tual world platform’s representation. We adopt an ontology-based approach with distinct levels of abstraction for capturing and representing multi-actors and multi-domain choreographies to be staged in virtual world platforms with distinct characteristics. Ontologies are characterized according to two comple- mentary dimensions – choreography’s domain (independent and dependent) and virtual world platform (independent and dependent) – giving rise to four ontol- ogies. Ontology mappings between these ontologies enable the automatic gen- eration of a choreography for virtually any target virtual world platform, thus reducing the time and effort of the choreography development. Keywords: virtual worlds, training, choreography, multi-user, model-driven, ontology, mapping.1 IntroductionVirtual worlds have achieved significant levels of interest for supporting teaching andlearning activities [1], [2] since they provide the creation of immersive environmentswhere multiple elements of a team sharing a common virtual space can developcompetencies in a simulated context [3], [4]. Choreographies of virtual actors are aspecific type of content that represent the set of actions that can be performed simul-taneously by human-users and virtual computer-controlled actors thus enablinghuman trainees/students to play roles as part of teams or within a simulated socialcontext. In this sense, a choreography is the description of a set of actions that must ormay be executed by a group of participants, including the goals to be achieved andany restrictions that may exist.R. Shumaker and S. Lackey (Eds.): VAMR 2014, Part II, LNCS 8526, pp. 105–115, 2014.© Springer International Publishing Switzerland 2014

106 E. Silva, N. Silva, and L. Morgado Because designing a choreography is a resource-intensive effort, it would be de-sirable for the result not to be hostage to a specific virtual world platform (VWP) butrather deployable in any VWP. However, as virtual platforms are very heterogeneousin terms of (e.g.) functionalities, data models, execution engines and program-ing/scripting languages or APIs, deploying a platform-based choreography intoanother VWP is difficult and time-consuming [5]–[8]. We believe that the approach presented in this paper provides a contribution thatfacilitates the development, sharing and adaptation of choreographies aimed to bestaged in different virtual platforms. For this, we suggest an approach where the con-ceptual representation model of the choreography is captured in the form of ontolo-gies, and its adaptation to a particular virtual world follows a set of models transfor-mation processes, similar to that suggested by the Model Driven Architecture (MDA)paradigm [9]. The proposed ontology-based definition of choreographies can capturenot only the physical aspects as objects but more complex content such as procedures,consisting of sets of actions and conditions in which the actors can perform them. Thus, this paper presents an approach that deals with the design and representationof platform-independent multi-user choreographies, and their deployment to differentVWPs with minimal effort and time using a set of transformation processes based onontologies and alignments. The rest of the paper comprehends four more sections. Insection 2 we present the proposed approach and the description of the system archi-tecture. Section 3 describes the developed experiments. Section 4 compares the re-lated work with the proposed ideas. Finally, Section 5 summarizes the proposal andsuggests future directions.2 Proposed ApproachTo deal with VWP with different characteristics, we argue that choreographies shouldbe clearly separated from the technical characteristics of the execution in the VWP. To this end, the core of the proposal is a “generic high-level ontology” that cap-tures the choreography in a conceptual and abstract fashion, so it is independent fromthe staging/deployment VWP. Thus, the data model of every virtual world must becaptured/represented by the so-called “platform-specific ontology”, and a mappingbetween the generic high-level ontology and the platform-specific ontology must bedefined. The mapping will provide the means to transform the platform-independentchoreography into a platform-dependent choreography. To address this the MDA software-development paradigm [9] is adopted andadapted. MDA specifies three default models of a system corresponding to differentlayers of abstraction: a Computation Independent Model (CIM) the most abstract,which represents the view of the system without any computational complexities; aPlatform Independent Model (PIM) that describes the behavior and structure of thesystem, but without technological details, and a Platform Specific Model (PSM) thatcombines the specifications in PIM with the specific details of a specific platform. Based on the concept of model independence and model transformation ofMDA, we adopt an approach based on two first-class citizen dimensions: the VWP

Staging Choreographies for Team Training in Multiple Virtual Worlds 107dimension and the choreography’s domain dimension. In fact, unlike in MDA, in ourapproach the model is not only independent from the VWP but also independent fromthe (choreography’s) domain. Fig. 1 depicts the MDA one-dimensional approach(Fig.1 a) in comparison with the two-dimensional envisaged approach (Fig. 1 b).The nomenclature refers to Ontology, Platform and Domain, with “x”assuming “d” and “i” values for “dependent” and “independent”, respectively. E.g.stands for “Ontology, Platform-independent, Domain-dependent”.Taking into account the characteristics of ontologies, and considering they are thebest way to represent conceptual information in order to bring the intelligent systems Fig. 1. Model-Driven representation according to: a) MDA and b) Our approachFig. 2. The system architecture with the processes of authoring, mappings, transformation andexecution

108 E. Silva, N. Silva, and L. Morgadocloser to the human conceptual level [10], we claim that ontologies are then the ade-quate knowledge representation model for bridging the gap between the human re-quirements and the computational requirements [10]–[13], thus able to play the role ofboth Computation-Independent Model (CIM) and Platform-Independent Model(PIM). Following MDA, the ontology-based choreography representations are succes-sively transformed through a set of processes until the final, platform-specific choreo-graphy ( ) that is executed in the specific VWP. The proposed architecture isdepicted in Fig. 2, and comprehends four ontologies and five processes.2.1 OntologiesThe following four ontologies are representational models of the different choreogra-phy abstractions:• (platform-independent and domain-independent) is the generic high-level ontology representing the core concepts of a choreography independent of any im- plementation environment, also designated as the foundational ontology. Fig. 3 presents its current status, whose motivations and design decisions have been de- scribed in a previous paper [14].• (platform-dependent and domain-independent) represents the core concepts of a choreography for a specific VWP. Despite this is a platform-dependent ontology, it remains independent from any application domain, and therefore is developed only once (eventually requiring adaptations in face of mutations in the VWP due to any evolution). Each virtual world can have their own interpretation of a choreography describing the concepts in a private way in order to best fit its characteristics, but this ontology must capture and represent every concept corresponding to those defined in the foundational ontology to capture the same semantic knowledge, so that is possi- ble to establish semantic relations between them. Thus, we consider that in order to apply this approach, it is necessary to develop, for each target VWP, an ontology to represent that virtual world platform’s particular interpretation of the fundamental ontology. Moreover, this ontology can incorporate additional concepts considering the specific characteristics of that virtual world and using its own terminology.Fig. 3. The Foundational Choreography Ontology ( ): representation of concepts, proper-ties and relations between concepts

Staging Choreographies for Team Training in Multiple Virtual Worlds 109• (platform independent and domain dependent) is a choreography resulting from the authorship of . It captures the representation of a complete choreo- graphy for a specific application domain, without any concern about the technical specificities of any platform.• (platform dependent and domain dependent) is a choreography represented in/for a specific VWP. This is the choreography that is intended to serve as a refer- ence for the staging in the virtual world.2.2 ProcessesIn the proposed architecture we apply five processes to conduct a successive trans-formation of models representing the various abstractions of a choreography to adaptits specification to a particular virtual world. To illustrate the explanation, we will consider a generic Instant Messaging plat-form for which a choreography will be adapted. This is a very simple platform (de-veloped by the authors for testing purposes, cf. Fig. 7) where there is only the repre-sentation of an action called write.Authoring. Authoring is a user-based process in which the choreographer authors adomain dependent choreography in the form of an ontology. This process is typicallyperformed by an expert that manually builds the ontology. But end-user tools can alsobe developed to allow people without knowledge or training in ontologies to specifythe choreography through simple and intuitive interfaces. That is, end-user tools canin a transparent manner, build ontologies instead of human users directly, facilitatingthe authoring process.The foundation ontology ( ) is extended and refined semantically to describethe choreography entities specific to an application domain, giving rise to .Theauthoring process must ensure the set of changes applied does not change the seman-tics defined in . For that, the following assumptions must be guaranteed:• No axioms defined in can be removed;• No contradictions can be added, i.e., must be logically consistent;• No new root elements are allowed. I.e. new entities (classes and properties) aredefined as sub entities of those existing in and should not create new rootelements.Fig. 4 depicts an excerpt of the ontology resulting from authorship ( ), and moreformally its representation using Description Logics (DL) syntax. New concepts are added as well as restrictions that will define boundaries to theability of actions execution by the actors. The restrictions are based on the definitionof associations between roles (that actors can play) and actions. Commonly, the fol-lowing two types of restrictions are defined:1. To constrain the actors allowed to perform an action based on the user’s role, i.e. toperform an action, the actor must have a specific role previously defined. Therelation between the concepts referenced by the action and role is defined by theproperty ;

110 E. Silva, N. Silva, and L. Morgado2. To assign the role(s) of an actor based on the action(s) s/he performs. Thus, the roles are dynamically assigned during the choreography according to the actions performed by the actor. This association can be seen in a dual perspective. Using a small example: on the one hand, if an actor plays the role Role1 and performs ac- tions Action1 and Action2, s/he shall automatically plays the role Role2 thereafter. On the other hand, to play the role Role2 it is a necessary condition that the actor plays the role Role1 and performs the actions Action1 and Action2. Fig. 4. An excerpt of the ( ) Ontology resulting from the authoring process During the authoring process, the author can take advantage of all the semantic ex-pressiveness of the DL language to elaborate more complex choreographies. For ex-ample, if an actor during the choreography cannot play two roles, one can specifyusing DL that these two roles are disjoint.Mapping1. Mapping1 is the process that establishes correspondences between therepresentations of the two choreographies abstractions represented by and(Alignment1).An alignment is a set of correspondences (semantic bridges in this paper). A se-mantic bridge is a set of information elements describing what entities from bothsource and target ontology are semantically related, and what conditions must hold inorder to be considered [15]. Two types of semantic bridges are considered:1. Concept Bridge (CB), used to describe the semantic relations between (source and target) concepts;2. Property Bridge (PB), used to specify the semantic relations between (source and target) properties, either relations or attributes.In this process all the core concepts (concepts that are direct subclasses of the Thingconcept, depicted in Fig. 3) of the foundational ontology ( ) should be mapped toconcepts of the platform specific ontology ( ) to ensure that there is full corres-pondences between both ontologies. Correspondences between properties are definedbetween properties of two mapped concepts. In order to facilitate understanding,

Staging Choreographies for Team Training in Multiple Virtual Worlds 111a relation is established between the concept-concept correspondence and the proper-ty-property correspondence (Fig. 5).Mapping2. Mapping2 is the process that establishes correspondences between thedomain choreography ( ) and a VWP ontology ( ), i.e. Alignment2. Align-ment2 profits from (and extends) Alignment1, thus promoting reuse and reducingefforts (Fig. 5).Fig. 5. A partial view of the alignments resulting from Mapping1 and Mapping2Transformation. Transformation is the process that creates the VWP choreography( ) from and and according to the Alignment2.This is a fully automatic process that “copies” the classes and constraints(Fig. 4) to the (Fig. 6).Despite this being an automatic process, choreographers can intervene and edit theresulting ontology for additional adjustments. Thus, the ontology maybe further finely tuned to better fit the implementation platform. Fig. 6. An extract of ontology

112 E. Silva, N. Silva, and L. MorgadoExecution. Execution is the process that stages the choreography in a VWP through aPlayer (computer program) compatible with the VWP who has the ability to scheduleactions according to the choreography and control its execution by virtual characters.Further, it monitors the human-user performance by comparing the executed actionswith those described by the choreography, and reacts accordingly. This process uses areasoner mechanism to evaluate whether it is possible to perform the actions, verify-ing if all necessary conditions are met. When virtual-users are present, a planner isused to calculate a plan of actions for them.3 ExperimentsFor the evaluation of our approach we deployed several real-world choreogra-phies that were staged in two different multiuser platforms with very distinct cha-racteristics with human-users only. We used the VWP OpenSimulator1 (OpenSim)to create a realistic multiuser 3D environment; as a counterpart system, wedeveloped for testing purposes the aforementioned messaging platform. It is asimplified version of text-based virtual worlds of the Multi-User Dungeons era,following Morgado’s definition [16]. This messaging platform has very differentcharacteristics from the OpenSim, since it does not allow the representation ofscene objects, but enables the development of a team’s choreography nonetheless.Its interface provides the actions of the choreography in the form of buttons, theinteraction is done by pressing buttons, and when an action is performed success-fully by each team member, it is communicated to all other team members bymeans of a text log (Fig. 7). Authoring is obviously the most time-consuming and creative process, whilesemi-automatic Mapping1 and Mapping2 processes require reduced time and effort.Once these processes are done, the transformation and execution processes are fullyautomatic. Fig. 7. Staging the choreography in a) OpenSim and b) messaging platform1 opensimulator.org/

Staging Choreographies for Team Training in Multiple Virtual Worlds 1134 Related WorkThere is prior relevant related work addressing the description of plans to representtraining procedures to be staged by a single actor as well as by teams with severalelements and how actions are distributed among them. But, most approaches design achoreography aiming it to be staged on a particular VWP. This creates strong depen-dencies with this VWP, making it difficult or even impossible to apply to other virtualworlds. Thus, related work can be categorized according to three dimensions: model-ing independence, VWP independence and number and type of the actors. Some approaches use separate models to represent the specification of proceduresand scene [7], [17], [18]. They address team training scenarios but they are stronglydependent on the characteristics of the VWP. Some other approaches attempt tobridge the gap between the representation of procedures and its execution in distinctVWP. However, such approaches are only focused on a single user not allowing therepresentation of teamwork [19]–[21]. Instead, our approach is capable of representing teamwork choreographies involv-ing multi-users played either by human and virtual-characters. Also, the actions andscene are captured conceptually using a unique choreography model that is convertedto potentially any VWP.5 Conclusions and Future WorkIn this paper we propose an approach that allows the development of choreographiesand its adaptation and staging in potentially any VWP. For that, based on the conceptof MDA and the assumption that the use of ontologies is the best way to represent theconceptual information to approximate the intelligent systems to the human concep-tual level, we propose an ontology to capture the semantics of a generic choreographyindependent of any application domain and VWP. Further, for each VWP is adoptedan ontology representing its specific terminology and functionalities, and is mappedwith the generic one. Using a set of alignments between the ontologies we describe a complete sequenceof processes that allow adapting a choreography of a specific domain (but indepen-dent of any VWP) into a choreography suitable and capable of being staged into aspecific virtual world. We also describe the execution process that monitors, managesthe staging of the choreography, and uses reasoning engines to aid in the evaluationand validation of actions. Moreover, ontologies allow the integration in the same model all the modeling in-formation related to the choreography, i.e. the definition of procedures related toteamwork and the information about the scene. Using alignments between ontologies enables the automation of adaptation of thegeneric ontology to the specific target ontology, hence contributing to reduce devel-opment time and resources. In future work the Mapping1 and Mapping2 processes can be refined to incorpo-rate automatic matching mechanisms. So, it would be possible to increase the abili-ty to automate these processes while at the same time it reduces the need for userintervention.

114 E. Silva, N. Silva, and L. MorgadoAcknowledgments. This work is supported by FEDER Funds through the “ProgramaOperacional Factores de Competitividade - COMPETE” program and by NationalFunds through FCT “Fundação para a Ciência e Tecnologia” under the projectAAL4ALL (QREN13852).References 1. De Freitas, S.: Serious virtual worlds, Scoping Guide JISC E-Learn. Programme Jt. Inf. Syst. Comm. JISC UK (2008) 2. Morgado, L., Varajão, J., Coelho, D., Rodrigues, C., Sancin, C., Castello, V.: The attributes and advantages of virtual worlds for real world training. J. Virtual Worlds Educ. 1(1) (2010) 3. Kapahnke, P., Liedtke, P., Nesbigall, S., Warwas, S., Klusch, M.: ISReal: An Open Plat- form for Semantic-Based 3D Simulations in the 3D Internet. In: Patel-Schneider, P.F., Pan, Y., Hitzler, P., Mika, P., Zhang, L., Pan, J.Z., Horrocks, I., Glimm, B. (eds.) ISWC 2010, Part II. LNCS, vol. 6497, pp. 161–176. Springer, Heidelberg (2010) 4. Pinheiro, A., Fernandes, P., Maia, A., Cruz, G., Pedrosa, D., Fonseca, B., Paredes, H., Martins, P., Morgado, L., Rafael, J.: Development of a Mechanical Maintenance Training Simulator in OpenSimulator for F-16 Aircraft Engines. Procedia Comput. Sci. 15, 248–255 (2012) 5. Media Grid: Open File Formats Technology Working Group (OFF.TWG) Charter, http://mediagrid.org/groups/technology/OFF.TWG/ (accessed: October 14, 2013) 6. Mollet, N., Arnaldi, B.: Storytelling in Virtual Reality for Training. In: Pan, Z., Aylett, R., Diener, H., Jin, X., Göbel, S., Li, L. (eds.) Edutainment 2006. LNCS, vol. 3942, pp. 334–347. Springer, Heidelberg (2006) 7. Gerbaud, S., Mollet, N., Ganier, F., Arnaldi, B., Tisseau, J.: GVT: A platform to create vir- tual environments for procedural training. In: IEEE Virtual Reality Conference, VR 2008, pp. 225–232 (2008) 8. Vernieri, T.M.: A web services approach to generating and using plans in configurable ex- ecution environments (2006) 9. Alhir, S.: Methods & Tools - Understanding the Model Driven Architecture (MDA). Mar- tinig & Associates, fall (2003)10. Obrst, L., Liu, H., Wray, R.: Ontologies for Corporate Web Applications. AI. Mag. 24(3), 49 (2003)11. Gruber, T.R.: A translation approach to portable ontology specifications. Knowl. Ac- quis. 5(2), 199–220 (1993)12. Fensel, D.: Ontologies: A silver bullet for knowledge management and electronic com- merce. Springer, Heidelberg (2004)13. Gruninger, M., Lee, J.: Ontology Applications and Design-Introduction. Commun. ACM 45(2), 39–41 (2002)14. Silva, E., Silva, N., Paredes, H., Martins, P., Fonseca, B., Morgado, L.: Development of platform-independent multi-user choreographies for virtual worlds based on ontology combination and mapping. In: 2012 IEEE Symposium on Visual Languages and Human- Centric Computing (VL/HCC), pp. 149–152 (2012)15. Silva, N., Rocha, J.: MAFRA–an ontology MApping FRAmework for the semantic web. In: Proceedings of the 6th International Conference on Business information Systems (2003)

Staging Choreographies for Team Training in Multiple Virtual Worlds 11516. Morgado, L.: Technology Challenges of Virtual Worlds in Education and Training - Re- search Directions. In: 2013 5th International Conference on Games and Virtual Worlds for Serious Applications (VS-GAMES), pp. 1–5 (2013)17. Edward, L., Lourdeaux, D., Lenne, D., Barthes, J., Burkhardt, J.M.: Modelling autonom- ous virtual agent behaviours in a virtual environment for risk. IJVR Int. J. Virtual Real. 7(3), 13–22 (2008)18. Lopes, A., Pires, B., Cardoso, M., Santos, A., Peixinho, F., Sequeira, P., Morgado, L.: Sys- tem for Defining and Reproducing Handball Strategies in Second Life On-Demand for Handball Coaches’ Education19. Young, R.M., Riedl, M.O., Branly, M., Jhala, A., Martin, R.J., Saretto, C.J.: An architec- ture for integrating plan-based behavior generation with interactive game environments. J. Game Dev. 1(1), 1–29 (2004)20. Young, R.M., Thomas, J., Bevan, C., Cassell, B.A.: Zócalo: A Service-Oriented Architec- ture Facilitating Sharing of Computational Resources in Interactive Narrative Research (2011)21. Cash, S.P., Young, R.M.: Bowyer: A Planning Tool for Bridging the gap between Declara- tive and Procedural Domains. Artif. Intel., 14–19 (2009)

“Make Your Own Planet”: Workshop for Digital Expression and Physical Creation Hiroshi Suzuki, Hisashi Sato, and Haruo Hayami Kanagawa institute of technology Faculty information Technology 1030 shimoogino Atsugi Kanagawa, Japan [email protected] Abstract. We propose the “Make Your Own Planet” workshop, which com- bines handicraft and digital representation tools (3DCG effects). In this work- shop, a child uses a USB camera to select textures freely in the process of mak- ing an original 3DCG planet. All 3DCG planets are then placed in a simulated universe for public viewing. By watching this universe, viewers can appreciate the planet of each child. Further, the texture of each 3DCG planet is translated to a polyhedron template and printed out as a paper-craft template. In this process, children employ computers to transform their planets into physical ob- jects that they can bring home. We first describe the workshop concept and then the method by which it was implemented. Finally, we evaluate the workshop. Keywords: Digital workshop, 3DCG, Unity, I/O device.1 IntroductionWorkshops are currently viewed as opportunities for experimental learning. As such,various workshops are held every weekend at educational facilities, such as museumsand universities. In Japan, workshops have attracted attention as places of learning. CANVAS [1] is unique in that it promotes activities that link technology to the ex-pression of children. A non-profit organization (NPO) holds a “Workshop Collection”every March at Keio University’s Hiyoshi Campus. In Japan, CANVAS develops andhosts workshops for children at various educational facilities. This expo, now in itsninth year, has grown into a big event, attracting about 100,000 parents and childrenover two days. Not all the workshops in the Workshop Collection use digital technol-ogy, but the number of those that do is increasing. Most of the systems used in these workshops, require operations, such thoseprovided by keyboards and digital mice. For the reasons described above, older ele-mentary school children are targeted in these workshops.2 The Concept of Digital Workshop2.1 The Trend of the Digital WorkshopTypical examples of workshops that use technology are those for creating handmadecrafts through computer-aided activities. Many universities research and developR. Shumaker and S. Lackey (Eds.): VAMR 2014, Part II, LNCS 8526, pp. 116–123, 2014.© Springer International Publishing Switzerland 2014

“Make Your Own Planet”: Workshop for Digital Expression and Physical Creation 117systems that support handmade work, such as paper crafts [2], stencil designs [3], andpop-up cards [4]. They then hold workshops to disseminate the results of their re-search in society. An important purpose of these workshops is to have participantsand children experience the “joy of creation” by making their own works. In these creative workshops, computers support creative activities by providingspecialized knowledge, augmenting skills, and reducing and simplifying tasks. Inother words, computers serve as specialists or professionals. Here, the relationshipbetween children and computers is vertically structured. However, we attempt to pro-vide structures and devices that enable the active involvement of children in creativeactivities by using computers; thus, they can experience the “joy of creation.” In this paper, we report the on “Make Your Own Planet” workshop, which com-bines handicraft and digital representation tools (3DCG effects). In this workshop, achild uses a USB camera to select textures freely in the process of making an original3DCG planet. All 3DCG planets are then placed in a simulated universe for publicviewing. By watching this universe, viewers can appreciate the planet of each child.Further, the texture of each 3DCG planet is translated to a polyhedron template andprinted out as a paper-craft template. In this process, children employ computers totransform their planets into physical objects that they can bring home.2.2 Rerated WorkWorkshops that use computers as tools for handmade activities are quite common.Broadly speaking they, can be divided into “programming learning systems,” “sup-port systems,” “expression tool systems.” The planet maker proposed in this paper isan expression tool system. A programing learning system is the most general example of a workshop that em-ploys computers [5]. Workshop programs that design robots and determine theirmovements are being implemented all over the world. The purpose of these work-shops is to understand the features of sensor devices and programming languages. Anunderstanding of algorithms and complex operations are necessary; thus, they are notappropriate for younger children. A computer that provides knowledge and offers support system can reduce the dif-ficulty of shaping activities. Therefore, it is possible to produce complex handwork,even with children and beginners. In recent years, support systems for paper crafts[6], pop-up-cards [7], have been developed. An expression tool system provides to user with expressive activities on a comput-er. These systems can be seen especially in media art. With “Minimal Drawing” [8],one can draw pictures on a simulated canvas, which is rotated. “Body paint” [9] al-lows users to draw on walls and experience their own bodies as brushes. “I /O Brush” [10] is a system relevant to our system. It presents a heightenedeffect to children by ink drawing that takes pictures as real world objects, thus en-couraging youthful creativity. The difference between it and our system is that thelatter permits children to create a piece of a three-dimensional entity. Children areable to watch the work of other children at the public viewing. In our system, we liken

118 H. Suzuki, H. Sato, and H. Hayamia three-dimensional planet to drawing. A child’s planet is on public view as a 3DCGanimation that simulates the universe. Moreover, the system can print on the spot.3 System DevelopmentThe planet maker consists of three modules: the “Paint Module,” the “Space DisplayModule” and the “Mapping module.” In this section, we describe each method todevelop a module and each module’s functions. Figure 1 shows an overview of thesystem. Fig. 1. Overview of the system3.1 Paint ModuleChildren can paint the own planets with the Paint Module. This module can paint a3DCG spherical object using an image captured with a USB camera as ink. Figure 2shows the principle drawing method.Spherical InterfaceWe developed an original tangible interface so that children could paint texture easilyon a sphere. This interface consisted of Arduino and Potentiometer. There are a num-ber of buttons on the sphere interface. One performs screen transitions, another is ashutter release button on the USB camera, and a third adjusts volume to alter texture.

“Make Your Own Planet”: Workshop for Digital Expression and Physical Creation 119These mechanisms are controlled by Arduino, an I / O device. Figure 2 shows thePaint Module GUI and the Spherical interface. At first, a user takes a picture of an object to use as ink. Next, it is painted with thetexture 3DCG display, by specifying the location of any of the spherical interfaces.Paint locations on the sphere are specified by touching the guide above along thelongitude. A sphere type interface allows rotation with central axis; thus the user candraw as with a brush. (a) Shooting of texture (b)Painting module Decide ButtonGuide Pen size Volume Shatter Button (c)Spherical Interface Coordinate of Coordinate of Display (x, y) Texture (u,v) Processing image for texture drawing Fig. 2. Paint Module and Spherical Type Interface3.2 Space Display ModuleSpace display module is a public viewing module that display all planets designed bythe children. The texture of the planet made by the paint module is registered to atexture database. The space display module displays each planet with the texture datanewly added to the database. By watching this space, children can appreciate the pla-net of each child.3.3 Mapping ModuleThe texture of each 3DCG planet is translated into a polyhedron template with a map-ping module and printed out as a paper-craft template. In this process, children are able

120 H. Suzuki, H. Sato, and H. Hayamito transform the planets that they made with computers into physical objects that theycan bring home. In other words, children can make digital as well as physical works.3.4 Planet SheetIn a workshop, the name of the producer is described on a sheet, and a portion of anamed planet is acquired as a picture with a USB camera. This picture appears as a label on the preview screen of the planet in a space dis-play module. Figure 3 shows the Paint Module GUI and Sphere Interface. Figure 3shows the flow of making paper craft template.(a) Planet sheet (d) Paper Craft Making (c) Template of Paper Craft (b) Texture of Planet Fig. 3. Flow of Making a Paper Craft Template4 Make Your Own Planet WorkshopIn order to evaluate our design, we conducted a workshop at the Workshop Correction9 in Hiyoshi Yokohama. The target age range was six years or older, and the timeallowed to experience the workshop was about 60 minutes for each participant. The workshop was conducted by preparing four client terminals. A total of five in-structors were assigned as facilitators to move the workshop forward. After the work-shop, a survey was carried out in order to obtain evaluations of the workshop.5 DiscussionWe conducted a survey with five kinds of questionnaires evaluation of the interface,degree of work sharing, satisfaction with one’s work, and motivation for future work.Another questionnaire contains questions about the attractive elements of “Make yourown Planet.”

“Make Your Own Planet”: Workshop for Digital Expression and Physical Creation 121 We obtained the results of the survey of 261 children at the workshop. Table 1shows the questionnaire and items. Figure 4 shows the result of the questionnaires 1-4. Table 1. Qestionnaire and Items Fig. 4. Result of Q1-Q4 questionnaires Concerning the usability of the system, 44.4% stated that it was “very easy” or“easy” to use. If “becoming easy with use” is included, the positive opinion was88.2%. For “work share,” “most children stated that they did not “refer at all” to thecreations of others. Nearly 70% of the children created an original planet, withoutreferencing those of other participants. An overwhelming 93.2% answered that they“very satisfied” and “satisfied” with their planets. As far as motivation for futurework, 75.9% of the children wished to repeat the experience. Since there were nonegative opinions, it is clear that the satisfaction with the workshop and the motiva-tion were high. This workshop thus appealed to the children.

122 H. Suzuki, H. Sato, and H. Hayami Fig. 5. Answers of Q5 questionnaire Figure 5 shows the results of the other questionnaire, in which the childrenranked activities. The found the ability to choose a pattern freely the most interest-ing aspect. They had a great interest in employing the planet’s spherical controller.Creating a paper planet received third place. From these results, it is clear that mak-ing physical objects as paper crafts increased the children’s motivation for creativeactivity. The proportion of children who created by referring to the works of others wasabout 30%. The question on the appreciation of the work of a friend had a score lowerthan those of other items. The children had, however, a positive feeling that theirworks were shown in a public place. The space module was received positively, but it did not efficiently function as atool to stimulate the ideas of children when comparing their works. However, al-though children liked their own creations, they also expressed a strong desire to viewthe planets of their peers. The operation of the spherical interface required some prac-tice, but we succeeded in providing an environment in which children employed com-puters. Thus, we designed and put into operation a fully functioning digital workshopthat offered a special creative activity.6 Future workFor the future, the following two points should be kept in mind. The first is the neces-sity of improving the spherical interface. The children found it hard to paint part ofthe pole area of the sphere. The pole area is narrower, since a mounting surface jointis part of the interface base and the sphere. This feature made it difficult to draw.Therefore, we will improve spherical interface, as shown in Figure 6. The secondpoint is the need to improve the space display module used in public viewing. Wefound that children found it difficult the view the works. To correct this fault, we arecurrently developing a system that allows a planet to be viewed at the WEB.

“Make Your Own Planet”: Workshop for Digital Expression and Physical Creation 123 Fig. 6. Image of Improved Spherical InterfaceReferences 1. CANVAS: Workshop Correction, http://www.wsc.or.jp 2. Mitani, J., Suzuki, H.: Making Papercraft Toys from Meshes using Strip-based Approx- imate Unfolding. ACM Transactions on Graphics (Proceeding of SIGGRAPH 2004) 23(3), 259–263 (2004) 3. Igarashi, Y., Igarashi, T.: Holly: A Drawing Editor for Designing Stencils. IEEE Computer Graphics and Applications 30(4), 8–14 (2010) 4. Mori, Y., Igarashi, T.: An Interactive Design System for Plush Toys. ACM Transactions on Graphics and Applications 30(4), 8–14 (2010) 5. Seymour, A.: Papert: Mindstorms: Children, Computers, and Powerful Ideas. Basic Books (1994) 6. Tama Software: Pepacra Desginer, http://www.tamasoft.co.jp/pepakura/ 7. Li, X.-Y., Shen, C.-H., Huang, S.-S., Ju, T., Hu, S.-M.: Popup: Automatic paper architec- tures from 3D models. In: SIGGRAPH 10 ACM SIGGRAPH, papers: Article No. 111 (2010) 8. Kusachi, E., Junji, W.: Minimal Drawing: Drawing Experience with both User’s Intension and Accidental Strokes. Journal of the Virtual Reality Society of Japan 12(3), 389–392 (2007) 9. Mehment Kten.: Body Paint, http://www.memo.tv/bodypaint/10. Ryokai, K., Marti, S., Ishii, H.: Designing the World as Your Palette. In: Proceedings of Conference on Human Factors in Computing Systems (2005)

Usability Evaluation of Virtual Museums’ Interfaces Visualization Technologies Stella Sylaiou, Vassilis Killintzis, Ioannis Paliokas, Katerina Mania, and Petros Patias School of Social Sciences, Hellenic Open University Lab or Medical Informatics, Aristotle University of Thessaloniki, Greece Democritus University of Thrace, Greece Laboratory of Photogrammetry and Remote Sensing, Aristotle University of Thessaloniki, Greece [email protected], [email protected], [email protected], [email protected] Abstract. This paper reports on a user-centered formative usability evaluation of diverse visualization technologies used in Virtual Museums. It initially presents the selection criteria and the five museum websites involved in the analysis. Then, it describes the evaluation process, in which a group of subjects explored the museums’ on-line resources and answered in two usability ques- tions concerning overall reaction to the website and the subjective satisfaction of the users. After user testing, quantitative and qualitative data have been col- lected and statistically analysed. However, much research remains to be done on future research in terms of larger sample, different methodologies and varied contexts. Keywords: history and culture, digital humanitis, cultural informatics.1 IntroductionThe London Charter encourages virtual museums to promote rigorous design of digi-tal heritage visualization however; it suggests that virtual museums should ensure thatembedded visualization paradigms follow a human-centric design so that they pro-mote the study and interpretation of cultural heritage assets. The Principle 2 of theLondon Charter states that computer-based visual media should be employed whenthey provide added value for the study of cultural assets compared to other methods.It stresses that in order to determine the suitability of each technologically-drivenvisualization method, a systematic evaluation of such methods should be carried outbased on specific evaluation criteria. Relevant research sources utilized should beidentified and evaluated in a structured and documented way taking into accountbest practice within communities of practice. The London Charter’s main goal isto encourage dissemination of computer-based visualization so that significant rela-tionships between cultural elements can be determined by visitors. Such dissemina-tion should target to strengthen the study, interpretation and preservation of culturalheritage.R. Shumaker and S. Lackey (Eds.): VAMR 2014, Part II, LNCS 8526, pp. 124–133, 2014.© Springer International Publishing Switzerland 2014

Usability Evaluation of Virtual Museums’ Interfaces Visualization Technologies 125 Various researches have evaluated museum websites using design patterns [1],usability of virtual museum websites [2,3], utilizing both empirical and expert-basedmethods combining quantitative and qualitative research methods [4, 5], explored therelationship between the sense of Presence, previous user experience and enjoyment[6], the effect of various visualisation technologies to the sense of Presence [7], havedeveloped guidelines concerning issues ranging from design considerations to projectphilosophies [8], exploring requirements for online art exhibitions [9]. The main goalof this paper is to explore the usability parameters that can be used as reference forevaluating virtual museums, which often incorporate varied technological elements.After a short introduction to virtual museums and the selected cases for the purposesof research and the usability evaluation, the participants, the experimental procedureand the methods used for the statistical analysis are presented. In the last section ofthe paper, the research results are analysed and discussed.2 Virtual MuseumsA virtual museum [10] is a complex environment that according to the choices of thedesign team, determines the visitors' final experience and subsequent attitudes to-wards the use of digital media in museums. In order to cluster the wide range of exist-ing museum websites into specific representative categories, a team of four scientistsexperienced in interactive design and the use of Information and CommunicationTechnologies in culture and education, was assembled. Museum online resourceswere divided according to the presentation method employed for their visualizationand grouped/ classified according to that in five technologically-oriented categories ofmuseum sites mainly including: Panoramic images (QTVR), (2) Scalable images withtext, (3) Searchable databases, (4) 3D environments, (5) Videos. The experts shared a preselected pool of museum websites and worked indepen-dently to extract within these categories the factors that may influence the user's expe-rience according to their personal understanding and recent research literature onevaluation strategies for virtual museums. Subsequently, the factors were merged intoa set of five qualities or capacities: imageability, interactivity, navigability, virtualspatiality and narration as explained in Table 1. Of the five representative cases ofvirtual museums as presented below, four serve as extensions to existing physicalmuseums, while one is totally imaginary.Imageability: Panoramic Images. Imageability is defined as the “quality in a physicalobject that gives it a high probability of evoking a strong image in any given observer.It is shape, colour, or arrangement, which facilitate making of vividly identified, po-werfully structured, highly useful mental images of the environment” [11, p. 9]. In VEsof high imageability, users can experience the real museum space through panoramicimages that can be manipulated thanks to a set of interactive tools, such as rotate andpan, zoom in and out, and even navigate. The case selected for this study, labeled asM1, is the \"Virtual Exhibition Tours\" (http://www.nga.gov/onlinetours/index.shtm) ofthe National Gallery of Art in Washington. In this online environment, visitors can

126 S. Sylaiou et al.select specific works of art for larger image views, close-up details, streaming audiocommentary, and information about the object (Figure 1). Quality Table 1. Qualities of museum online resources.1. Imageability Definition Perceptual quality of a VE that makes it memorable2. Interactivity The HCI functionality that makes a VE able to communicate3. Navigability with its visitors4. Virtual Spatiality5. Narration The degree to which navigation capabilities are perceived from structural elements of the VE The extension of physical museum space and the metaphors of architecture to virtual space Narration via a collection of videos that engages the virtual visitors providing them the opportunity to investigate a theme in a variety of ways and construct their own meaning Fig. 1. Screenshot of the Van Gogh Virtual Exhibition Tour at the National Gallery of ArtInteractivity with Scalable Images and Texts. Image scalability provides the oppor-tunity to examine museum artifacts or parts of them in detail by applying zoom toolsover high resolution images. These zoom-on-demand features allow viewing aspectsof photos that are not visible to the naked eye because of their small size or because ofthe museums' spatial proximity restrictions. Image exploration tools make VEs highlyinteractive and enhance museum experience [6]. The selected case for this study,labeled as M2, is the Metropolitan Museum of New York (http://www.metmuseum.org/) (Figure 2).

Usability Evaluation of Virtual Museums’ Interfaces Visualization Technologies 127Fig. 2. Snapshot of an object contained in the section of Greek and Roman Art of the Metropol-itan Museum of ArtNavigability: Searching Utility for Images and Texts. This type of online museumenvironments offers multiple search options and enhanced image manipulation. Sear-chable databases typically contain 2D representations in the form of photos and flatscans of objects along with their corresponding metadata, which are uploaded to themuseum’s online database. The hallmark of these sites is a search engine, which al-lows searching by content, concept, or metadata, thanks to an entry point usually con-sisting of a text area in which visitors enter search criteria based on keywords. Thecase selected for this study, labeled as M3, is the Museum of Modern Art(http://www.moma.org/explore/collection/index). Through its database, visitors cannavigate the various thematic areas of the museum, and search its collections by artist,work or keyword. It also has an advanced search that allows adding refinement crite-ria such as period or object management status (Figure 3). Fig. 3. The advanced search engine of the Museum of Modern Art online database