A Novel Human Interaction Game-like Application to Learn,

Perform and Evaluate Modern Contemporary Singing

“Human Beat Box”

S. K. Al Kork

2,3

, D. Uğurca

, C. Şahin

, P. Chawah

4,5

, L. Buchman

4,5

, M. Adda-Decker

4.5

, K. Xu

2,3

B. Denby

2,3

P. Roussel

2,3

B. Picart

, S. Dupont

, F. Tsalakanidou

, A. Kitsikidis

, F. M. Dagnino

M. Ott

F. Pozzi

, M. Stone

and E. Yilmaz

Argedor Information Technologies, Ankara, Turkey

Universite Pierre Maire Curie, Paris, France

Signal Processing and Machine Learning Lab ESPCI, Paris, France

Phonetics and Phonology Laboratory, LPP-CNRS, Paris, France

University Paris 3 Sorbonne Nouvelle, Paris, France

Unversity of Mons, Mons, Belgium

Information Technologies Institute, Centre for Research and Technology Hellas, Thessaloniki, Greece

Institute for Educational Technology, National Research Council (ITD-CNR), Palermo, Italy

Vocal Tract Visualization Lab, University of Maryland Dental School, Baltimore, U.S.A.

Keywords: Human Computer Interaction, Interactive Game, 3D Model, Ultrasound Sensor, Portable System, Human

Beat Box, Game based Learning.

Abstract: The paper presents an interactive game-like application to learn, perform and evaluate modern contemporary

singing. The Human Beat Box (HBB) is being used as a case study. The game consists of two main modules.

A sensor module that consists of a portable helmet based system containing an ultrasonic (US) transducer to

capture tongue movements, a video camera for the lips, Kinect camera for face gestures, and a microphone

for sound. A 3D environment game module is used to visualize a 3D recording studio as game world with all

of its unique elements like guitars, mixer, amplifier, speakers and a microphone in front of the 3D avatar to

simulate the recording ambience. The game also features a 2D virtual tutor to help the learner by giving oral

and written feedback during the game. He also gives feedbacks during the practice session to improve the

student’s performance. The game is still at its early stages of development and it is been tested using simple

HBB plosive sounds for percussion such as “PTK”.

1 INTRODUCTION

The adoption of Serious Games (SGs), namely games

expressly designed with explicit educational purposes

(Breuer & Bente, 2010), in the teaching practice is

a quite well consolidated trend in the Technology

Enhanced Learning field. At present digital games are

increasingly adopted to sustain learning and training

in a variety of educational fields: school education as

well as military and medical training, etc. (Bellotti, et

al., 2012) (Graafland, et al., 2012) (Chatam, 2009).

This is done for a wide range of target populations,

ranging from children to adults (Charlier, et al.,

2012).

The use of digital games also is not new in the

field of Cultural Heritage (Ott & Pozzi, 2008). As

could be expected, games are more widespread in the

Tangible Cultural Heritage (TCH) area, but a number

of games have also been developed in the field of

Intangible Cultural Heritage (ICH), where they are

also considered very promising (Mortara, et al.,

2014). The educational potential of games has been

widely explored and highlighted by researchers

within the wider research area of Game Based

Learning. Beside other important advantages, such as

promoting learners motivation (Garris, et al., 2002)

engagement (Susi, et al., 2007) and self-regulated

learning (Zap & Code, 2009), games are also widely

recognized as potentially adaptive to support the

learning of procedures and gestures (and also of

sequences of gestures, physical actions); in this flow,

they are widely adopted in professional training

640

Al Kork S., U

gurca D., ¸Sahin C., Chawah P., Buchman L., Adda-Decker M., Xu K., Denby B., Roussel P., Picart B., Dupont S., Tsalakanidou F., Kitsikidis

A., Dagnino F., Ott M., Pozzi F., Stone M. and Yilmaz E..

A Novel Human Interaction Game-like Application to Learn, Perform and Evaluate Modern Contemporary Singing - "Human Beat Box".

DOI: 10.5220/0005429506400650

In Proceedings of the 10th International Conference on Computer Vision Theory and Applications (MMS-ER3D-2015), pages 640-650

ISBN: 978-989-758-090-1

 2015 SCITEPRESS (Science and Technology Publications, Lda.)

(Martínez-Durá, et al., 2001). The i-Treasures project

(i-Treasures, 2013) aims to exploit this potential of

SGs in the ICH field: their capability of training

motor skills and then the possibility to support

sensorimotor learning for the teaching/learning of

intangible cultural expressions, which are based

mainly on procedural knowledge (namely, knowing

how to do something).

Performing arts such as traditional or modern

singing is usually learned under the supervision of a

master or an expert. So it can be said that such

features requires physical actions to survive as they

are transferred through a master apprentice approach.

A major objective of the i-Treasures project is to

provide students with innovative multi-media human

game-like application for sensorimotor learning

feedback to train specific articulatory strategies for

different type of rare singing, considered an

endangered Intangible Cultural Heritage (UNESCO,

2012).

The i-Treasures project also includes research

about a newly expanding contemporary singing style:

the “Human Beat Box”, where the vocalist imitates

percussive and drum instrument sounds. Beatboxing

is a form of vocal percussion primarily involving the

art of producing drum beats, rhythm, and musical

sounds using one's mouth, lips, tongue, and voice. It

may also involve singing, vocal imitation of

turntables, and the simulation of horns, strings, and

other musical instruments. Beatboxing today is

connected with hip-hop culture, being one of "the

elements", although it is not limited to hip-hop music.

The proposed game-like application is expected to

help those who want to learn the basic of singing or

even make a practice. The game will use 3D

visualization techniques and tools to support the users

in learning or mastering different type of singing

styles. The domain knowledge of real masters will be

transferred to the virtual tutor. Thus, the virtual tutor

will be able to evaluate the user as a real master does.

2 GAME ARCHITECTURE

The game architecture of the system is shown in

Figure 1.The system is composed of the following

modules: Sensor input capture module,

Communication module and 3D game environment

module.

Figure 1: Game System Architecture.

2.1 Data Acquisition Module

One of the challenging tasks in developing the

proposed game apprentice is to interface little known

sensors in the game industry such as ultrasound. To

meet the requirement, a portable helmet based system

has been used to capture motor behaviour during

singing and other oral-motor functions in a non-

laboratory experimental environment (Al Kork, et al.,

2014) (Chawah, et al., 2014) (A Jaumard-Hakoun, et

al., 2013). The system, based on vocal tract sensing

methods developed for speech production and

recognition, consists of a lightweight “hyper-helmet”

containing an ultrasonic (US) transducer to capture

tongue movement, a video camera for the lips, and a

microphone for audio capturing as seen in Figure 2.

Moreover, facial data are captured using a Kinect

v1 sensor. Synchronized colour and depth image

streams are used for facial feature tracking and

recognition of facial action units based on an

improved version of the approach presented in

(Tsalakanidou & Malassiotis, 2010).This data is used

to visualize the facial movements of the 3D avatars.

2.2 Architecture of the Communication

Module

The communication module is responsible for

providing bilateral data transfer between the sensors

and the game .The application communicates with

various external entities such as the hyper-helmet,

audio comparison and evaluation server. Figure 3

illustrates the communication architecture.

The module is capable of acquiring multiple video

streams in real-time from the sensors of the hyper-

helmet by using a third party sensor integration and

controlling software called as RTMaps

(S.A.-RTMaps, 2012). Thus the system actually

ANovelHumanInteractionGame-likeApplicationtoLearn,PerformandEvaluateModernContemporarySinging-

"HumanBeatBox"

641

Figure 2: Multi-sensor Hyper-Helmet.

communicates with RTMaps via Transmission

Control Protocol. Besides video, the module also

supports real-time transfer of the 3D Tongue Model’s

vertex positions. The module supports up to 60 fps

data transfer either for video or 3D animation. The

game also gets facial data from the facial expression

data capture and analysis module using TCP/IP

messaging. The communication module also lets the

player control the data capture by sending relevant

commands to the data capture system as well.

The sensor capture module also streams audio data

from leaners to an external server. This

communication is done by a web service. This server

provides the result of audio comparison to the game

via web service. 3DSML communicates with

LMS/CMS via web services. This communication

covers user authentication, expert data download and

learner performance upload. Currently only user

authentication is supported in this first version.

2.3 3D Game Environment Module

The game module will basically provide an interface

for students to learn, and practice HBB and it will also

evaluate the performance and progress of the student.

The performance of the student will be assessed based

on his/her tongue movements and the pitch of the

student’s sound as compared to the expert’s.

2.3.1 Learning Scenario

The Human Beatbox is an artistic form of human

sound production in which the vocal organs are used

Figure 3: Communication Architecture of Singing Game-like Application.

VISAPP2015-InternationalConferenceonComputerVisionTheoryandApplications

642

to imitate percussion instruments, but also wind and

string instruments. Therefore, facial and intra-oral

movements of the tongue, lips and jaw are involved

in sound production. The apprentice will be guided by

some description of the sound to be produced and

listening to the sounds to imitate them. These

descriptions can be related to natural sounds such as

imitating animal sounds or mouth noise.

The apprentice will also look at the mouth and

tongue of the expert to try to produce different

sounds. The rhythm is an important dimension in

HBB and hand movement or a metronome can be

added in the scenario. The HBB game will consist

initially of a single lesson, where each isolated sound

will be described and will be demonstrated by the

expert and then imitated by the learner. The

apprentice will go from one step to another until he

succeed and finishes all the stages of the lesson. The

Stages are described as follows:

Stage 1: Imitate the plosive sounds for percussion

“PTK”.

During this stage the student will repeat several

times the three consonants at different rhythm, slow

Pp; Tt; Kk…and fast… prolong each consonant, try

to make an “echo” (Ppppp, Tttttt…). In reverse order

KTP aiming in maintaining the rhythm and the speed

of the articulators’ movement and quality of sound

Play with loudness P(loud) tk (whisper) and alternate

pTk ; ptK The time it takes the student to perform this

task will also be taken into account.

Stage 2: Introduce vowels.

During this stage the student will add a vowel

(PTKKi; Peu; Ti…). The vowels will be produced in

loud and whispered mode.

Stage 3: Add sound effects and nasality.

During this stage the student will add other

consonants (PTK e PT one = Wan ; Wa a wan…)

make a fricative sound (Tsss ; Tsi ; Pfff ; Kiwi …)

Stage 4: Succession of sounds, rhythms and

loudness.

The learner should practice each stage a number

of times, and finally perform all the stages at once.

Once the evaluation score of the student increases, he

will be asked to make other sounds from other

instruments and try to combine then in a harmonious

way. The game will provide auditory and visual

feedback. Figure 4 shows the learning progress

approach of this game.

ivy

1Figure 4: Learning Progress of the HBB Game.

2.3.2 Game Play

In order to develop the human beat box game, the

following assets were prepared: a 2D virtual tutor, 3D

avatars for expert and learner, a 3D tongue model and

a 3D sound recording studio environment.

2.3.2.1 2D Virtual Tutor

The virtual tutor is a 2D illustrated NPC (Non Player

Character), who is designed as a young male beat box

artist. His role is to navigate and help the student by

giving oral and written feedback during the game. He

also gives feedback during the practice phase to

improve the student’s performance. Besides, he

explains the sensors used in this game such as the

components of the hyper-helmet during the getting

started phase. The Virtual Tutor has five different

states for different moods such as happy, explanatory,

unsatisfied, neutral and not bad as shown in Figure 5.

Figure 5: 2D Virtual Tutor Moods in HBB.

The mood posture is selected by the system based on

the student’s performance during the practice phase.

In addition, all the written feedback provided by the

virtual tutor can be configured via XML files, which

ensures which ensures easy manipulation and

multilingual capabilities. It is also possible to set the

visual style of the expert throughout the game.

2.3.2.2 3D Expert and Learner Avatars

The avatars of the HHB expert and learner are

prepared by using 31 bones for the face and a simple

skeleton for the body. It is possible to map Face

Action Units to these facial animation bones.

Currently, we have prepared 52 blend shapes for the

face. Blend Shapes, which is also known as morph

target animation, is a popular computer animation

technique mainly used for facial animation (Joshi,

2005). Even though this technique is very labour-

intensive, it is useful for generating good looking

animations. For the following versions of the game,

Activity1

Activity2

Activity3 Activity4

ANovelHumanInteractionGame-likeApplicationtoLearn,PerformandEvaluateModernContemporarySinging-

"HumanBeatBox"

643

we plan to mainly focus on lips, since the lip

movement is quite important in singing, and produce

more blend shapes for lips. In this way, it will be

possible to visualize the lips more realistically.

3D expert and learner avatars are shown in Figure

6. Figure 7 shows the blend shapes and various

interpolations used for the current version. The expert

avatar is designed as a professional musician while

the learner avatar is designed with casual clothes.

Figure 6: Expert Avatar (Left) and Learner Avatar (Right).

Figure 7: Blend Shapes Prepared for Face Modelling.

2.3.2.3 3d Studio Environment

A 3D recording studio is designed as the game’s

world with all of its unique elements like guitars,

mixer, amplifier, speakers and a microphone in front

of the 3D avatar to simulate the recording ambience

as shown in Figure 8.

Figure 8: 3D Environment of HBB.

2.3.2.4 User Interface

Distinctive parts of the beat box game are explained

in this section.

2.3.2.4.1 Main Screen

As explained in section 2.3.1, there are four Activities.

Each activity has two different options, which are

“observe” and “practice”, and also a “Getting

Started” activity, which serves as game tutorial, as

shown in Figure 9. The user can click the “Observe”

button to observe the expert performing or click the

“Practice” button to start practicing the beat box

himself. A short description of each activity is

provided on each activity panel, so that the learners

can track the content of the selected activity. Under

each activity description, there is a progress bar that

represents the success rate of the student. Activity 2,

3 and 4 are locked and this is represented with a key

lock on the related panels, because of the fact that the

progress of the student is not enough to unlock further

activities. Activity 1 is also locked at the beginning.

However, when the user completes the “Getting

Started” the first activity is unlocked without any

further process. All of the description texts, buttons,

labels etc. are configurable via the configuration

XML.

2.3.2.4.1.1 Getting Started

When the student clicks on the “Begin” button of the

“Getting Started” panel of the Main Menu, the tutorial

of the beat box game starts. The main purpose of the

“Getting Started” is to teach the basics of the game to

the student. To do so, the 2D virtual tutor explains

VISAPP2015-InternationalConferenceonComputerVisionTheoryandApplications

644

each user interface element and how to use them to

play the game. The 2D virtual tutor comes up from

bottom left of the screen to say what will happen next

as shown in Figure 10.

Afterwards, tutorial continues with the

explanation of the techniques, which are used for

giving feedbacks to the student in beat boxing,

respectively; (a) Tongue position, (b) Using

ultrasound for tongue movement, (c) Description of

the tongue-air interface, (d) Ultrasound image of

tongue, (e) The tongue model. This is shown in Figure

11.

The tutorial proceeds to observe screen

as illustrated. Virtual tutor continues with the

Figure 9: Game Main Menu Screen of HBB Game.

Figure 10: Tutorial beginning of HBB Game with 2D virtual avatar.

ANovelHumanInteractionGame-likeApplicationtoLearn,PerformandEvaluateModernContemporarySinging-

"HumanBeatBox"

645

Figure 11: Illustrations of HBB Game Components.

explanation of observe screen elements respectively;

(a) 3D avatar, (b) Animation player, (c) Expert video

clip, (d) Ultrasound image of the tongue, (e) 3D

tongue model. Then 2D virtual tutor navigates student

to practice tutorial. This tutorial works exactly like

observe tutorial. 2D virtual tutor explains each

element on the screen in plain text and verbally. In

practice tutorial, 2D virtual tutor describes each

element respectively: (a) Animation controller of

practice section, (b) All other views on the screen.

Thereafter the virtual tutor explains the usage and

functionality of the sensors and software used in this

game. It is planned to include video tutorial for the

sensors that are not well-known and not easy to use.

In this way it is believed that the user will have the

required knowledge. When the “Getting Started”

ends, it is expected from student to continue to play

the game-like application to learn and improve his

knowledge about human beat box.

2.3.2.4.1.2 Observe Screen

All of the four activities in this game have an observe

scene, which shows the performance of the expert

captured by various sensors. The aim is to let the user

observe the details of each performance before s/he

starts the practice. All of the sensor outputs and

recordings are synchronized. These outputs are all

controlled by the animation controller, which is

located at the bottom center of the screen as shown in

Figure 12. The Observe screen consists of different

visual and functional elements such as the animation

player, expert lip video (window on right top corner),

ultrasound image of the tongue (second window on

the right) and 3D tongue model view (window on

bottom right corner) as shown in Figure 12.There are

also several windows on the right part of the screen

that show the outputs of different sensors.

It is possible to determine which sensor output

windows will be displayed via configuration XML.

Another important aspect of the observe screen is the

way we can change the size of the windows. By

default, the 3D recording studio is displayed as the

big central view. However when the user double

clicks on any window, the views are changed and the

selected window is displayed at the big central

window as shown in Figure 13.

VISAPP2015-InternationalConferenceonComputerVisionTheoryandApplications

646

Figure 12: Observe Screen of HBB Game.

Throughout the observe phase, the 2D virtual tutor

comes up from the bottom left of the screen and

explains the HBB sounds to the student. The text of

the virtual tutor’s narration can be modified via the

XML file. The features of the observe screen are all

described in the following sections.

2.3.2.4.1.3 Animation Player

The student controls the animation of the expert

lip/face video, expert tongue ultrasound video, 3D

tongue model and 3D expert avatar through the

animation player. These are all synchronized. It

should also be noted that the audio is also

synchronized with these data. The user is free to

control both the 3D expert avatar and tongue model

by using a freely navigable camera. The animation

player has “Play”, “Pause” and “Stop” buttons like

most of the video players. The animation can be

rewind or forwarded. The speed of the animation can

be controlled as well. The total time of the animation

and video with the current time position on the

animation timeline is also provided as labels on the

animation player.

2.3.2.4.1.3.1 Expert Video Lips

The expert video is provided at the top right of the

observe screen. This video shows either the lips or

face of the expert. Currently, only the lips video is

provided. However, we can display any type of video

content. It is also possible to change the video content

in real-time if both the lip and face videos are

available.

Figure 13: Maximized window of 2D Tongue Ultrasound

mid-sagittal view.

2.3.2.4.1.3.2 Expert Video (Ultrasound)

The observe screen provides the ultrasound sensor

video output which is captured by the “hyper-helmet”

system. The user is expected to observe the upper

contour of the tongue on the mid-sagittal plane. The

video shows the movement of the tongue during

singing. In this window a face silhouette and

directional info are provided to help the user to easily

distinguish the tongue movements. These features are

drawn as an overlay.

2.3.2.4.1.3.3 3D Tongue Model Visualization

The observe screen provides the 3D interactive

ANovelHumanInteractionGame-likeApplicationtoLearn,PerformandEvaluateModernContemporarySinging-

"HumanBeatBox"

647

tongue model (Kele Xu, et al., 2014) which shows the

animation of the tongue of the expert Our simulations

use ArtiSynth (J. E. Lloyd, et al., 2012), a 3D finite

mesh model of the tongue consisting of 14,000 nodes.

In Figure 14, six nodes (yellow points) are chosen as

anchor nodes corresponding to the hyoid bone and the

short tendon as discerned in the ultrasound image.

These anchor nodes are held fixed during the

simulation. We must then set several constraint nodes

(green points) lying on the mid-sagittal contour in the

ultrasound images, in order to drive tongue motion.

The assumption is that the behaviour of the 3D model

will be approximately correct if the movement of the

driven 2D sagittal points is correct. Linear constraints

on 3D movements can be integrated into an Euler-

Lagrange equation using the Lagrange multiplier

method. In Figure 15, the smaller images next to the

main 3D display show two ultrasound frames, from

which displacements are calculated. These

displacements are then passed to the constraint nodes

on the 3D tongue model mid-sagittal surface in order

to drive the model.

Figure 14: An overview of interface for the proposed

system.

Figure 15: : Six Anchor nodes (yellow nodes) on the tongue

mesh at the rest position and Constraint Nodes (green

nodes) on the mid-sagittal contour at the rest position of

tongue model.

Currently, we visualized and presented the same

tongue model used in ArtiSynth in Unity 3D as an

interactive window. The learner can freely navigate

this tongue model. It is possible to change the view to

top, lateral etc, and to zoom in and out. The research

to transfer animation results to the game environment

is still under progress. However, the technical

implementation for tongue animation has been

completed; thus, in future versions the game will

display the full tongue animation The system has

been tested with simulation data and real time data

streaming. Figure 16 shows the highly detailed 3D

tongue model which is made of 14.000 vertices.

Figure 16: 3D Interactive Tongue Model in Game

Environment.

2.3.2.4.1.4 Main Window-3D Expert Avatar

This 3D virtual environment represents a recording

studio. In this environment, the user is free to navigate

by using the virtual camera. The face of the avatar is

animated by using the data derived from the Kinect

sensor through the facial analysis software. This

animation is also synchronized with the rest of the

other sensor outputs.

2.3.2.4.1.4.1.1 Sound Progress Window

The sound that the expert imitates is shown at the

center top as shown in figure 12. Here the plosive

sounds P/T/K are visualized. It should be noted that

this text is also synchronized with the other outputs.

This text and the length are also defined in the

configuration XML. The played part is highlighted

with green to show the current location of the

performed sound.

2.3.2.4.1.5 Practice Screen

All of the four activities in this game have a practice

scene enabling the learner to practice the selected

activity. The practice scene shows the 3D avatar at the

center screen, the sensor data of the expert on the right

panel and the sensor data of the player on the left as

shown in Figure 17.

VISAPP2015-InternationalConferenceonComputerVisionTheoryandApplications

648

Similar to the observe scene, the sensor data

windows are all configurable. Thus, whenever it is

required to simplify the scene, some windows may

not be displayed. This can be easily set via the

configuration XML. The practice starts with the

introduction of the virtual tutor. After providing some

basic instructions, the virtual tutor asks the user to get

ready and counts down. The learner imitates the

requested sound.

Figure 17: Practice Screen of HBB Game.

After the performance is completed, the recorded

sound is sent to the remote server that is responsible

of audio analysis and comparison. This server

receives the audio file including the sound produced

by the learner and compares it with the corresponding

expert sound (which is part of the Human Beat Box

experts’ database). Following the completion of

comparison, the software generates a detailed text file

that contains the similarity ratio, sub-parts and

corresponding similarities etc. This text file is send

back to the game via web-service. This output is used

to evaluate the performance of the learner. It should

be noted that this remote evaluation takes very short

time and it does not introduce any lag or latency. This

evaluation approach was designed for the first version

of the game However, in future versions there will be

improvements regarding this approach and more

information will be taken into account by the

evaluation algorithm such as the position of the

tongue, the opening of the mouth, etc.

After getting the evaluation result, the virtual tutor

provides feedback about the performance of the

learner by using the corresponding grading text, e.g.

“You need to improve! Let’s do it again!”, “Good

performance! Just needs a little bit more!”,

“Outstanding performance! You are ready for the

next activity” etc. The animation controller in the

practice screen is used to record the learner’s

performance and repeat it whenever needed.

3 CONCLUSIONS

A first version of an interactive game-like application

to learn, perform and evaluate modern contemporary

singing focusing on the Human Beat Box case study

was developed in the context of the i-Treasures

project. The game integrates several state-of-the-art

sensors such as Kinect for facial data capture and a

prototype multi-sensor hyper-helmet for vocal tract

capture. It provides a user-friendly interface including

an engaging 3D recording studio environment and

singer avatars and visualizes a variety of sensor

outputs on user demand. A virtual tutor guides the

user through the game and provides real-time

feedback for the evaluation of the user‘s performance

in different singing activities.

The developed game framework is capable of

providing bilateral communication with external

sensors and third party sensor control software such

as RTMaps. The framework supports UDP and

TCP/IP based network communication as well as web

services. More sensors can be included in the system

without additional effort. Besides communication

utilities, the framework enables game designers and

domain experts to configure the application easily by

setting the fields of an XML file.

4 FUTURE WORK

An updated version of the HBB game is currently

under development. We plan to update the game

design, improve the game play and integrate learning

analytics so as to increase user engagement and

improve the game’s educational effectiveness. We

also plan to focus more on educational aspects, since

without meeting specific educational targets, the final

output will be more a technology demonstration

rather than an educational tool.

From a technical perspective, we plan to integrate

Kinect v2 in future versions. This sensor will help us

to better visualize and animate the lips and the face.

We also plan to develop a mobile version of this

game, mainly intended for observation and learning

rather than practicing HBB.

ACKNOWLEDGEMENTS

This work is funded by the European Commission via

the i-Treasures project (Intangible Treasures -

Capturing the Intangible Cultural Heritage and

Learning the Rare Know-How of Living Human

ANovelHumanInteractionGame-likeApplicationtoLearn,PerformandEvaluateModernContemporarySinging-

"HumanBeatBox"

649

Treasures FP7-ICT-2011-9-600676-i-Treasures). It is

an Integrated Project (IP) of the European Union's 7th

Framework Programme 'ICT for Access to Cultural

Resources.

REFERENCES

A Jaumard-Hakoun, et al., 2013. Capturing, Analyzing, and

Transmitting Intangible Cultural Heritage with the i-

Treasures Project. Edinburgh, UK, s.n.

Al Kork, S. K. et al., 2014. A Multi-Sensor Helmet to

Capture Rare Singing, An Intangible Cultural Heritage

Study. Cologne, Germany, s.n.

Bellotti, F. et al., 2012. Designing a course for stimulating

Entrepreneurship in higher education through Serious

Games. Procedia Computer Science, Volume 15, pp.

174-186.

Breuer, J. & Bente, G., 2010. Why so serious? On the

Relation of Serious Games and Learning - Eludamos.

Journal for Computer Game Culture,, 4(1), pp. 7-24.

Charlier, N., Ott, M., Remmele, B. & Whitton, N., 2012.

Not Just for Children: Game-Based Learning for Older

Adults. Cork,Ireland, In 6th European Conference on

Games Based Learning,, pp. 102-108.

Chatam, R., 2009. The 20th-century revolution in military

training. In: E. K, ed. Development of professional

expertise. Toward measurement of expert performance

and design of optimal learning environements. New

York: Cambridge University Press, pp. 27-60.

Chawah, P. et al., 2014. An educational platform to capture,

visualize and analyze rare singing. Singapore, s.n.

Garris, R., Ahlers, R. & Driskell, J. E., 2002. Games,

motivation, and learning: A research and practice

model. Simulation & Gaming, 33(4), pp. 441-467.

Graafland, M., Schraagen, J. M. & Schijven, M. P., 2012.

Systematic review of serious games for medical

education and surgical skills training.. British Journal

of Surgery, Volume 99, p. 1322–1330.

i-Treasures, 2013. The i-Treasures project. (Online)

Available at: www.i-treasures.eu/

J. E. Lloyd, Stavness, I. & S. Fels, 2012. “ArtiSynth: A fast

interactive biomechanical modeling toolkit combining

multibody and finite element simulation. Springer,, pp.

344-394.

Joshi, P., 2005. Learning controls for blend shape based

realistic facial animation. s.l., ACM SIGGRAPH 2005

Courses.

Kele Xu, et al., 2014. 3D tongue motion visualization based

on ultrasound image sequences. Singapore, s.n.

Martínez-Durá, M. et al., 2001. Serious Games for Health

and Safety Training. In: M. Prensky, ed. Digital game-

based learning. New York: Mc Graw-Hill.

Mortara, M. et al., 2014. Learning Cultural Heritage by

serious games. Journal of Cultural Heritage, 15(3), pp.

318-325.

Ott, M. & Pozzi, F., 2008. ICT and Cultural Heritage

Education: Which Added Value?. In: M. e. a. Lytras,

ed. Emerging Technologies and Information Systems

for the Knowledge Society - Lecture Notes in Computer

Science. Berlin: Springer , pp. 131-138.

S.A.-RTMaps, I., 2012. Intempora S.A.-RTMaps. (Online)

Available at: http://www.intempora.com/

Susi, T., Johanesson, M. & Backlund, P., 2007. Serious

Games - An Overview, Skövde, Sweden: University of

Skövde.

Tsalakanidou, F. & Malassiotis, S., 2010. Real-time 2D+3D

Facial Action and Expression Recognition. Pattern

Recognition, 43(5), pp. 1763-1775.

UNESCO, 2012. UNESCO. (Online)

Available at: http://www.unesco.org/culture/ich/en/

convention.

Zap, N. & Code, J., 2009. Self-Regulated Learning in Video

Game Environments. In: R. Ferdig, ed. Handbook of

Research on Effective Electronic Gaming in Education.

s.l.:Hershey, pp. 738-756.

VISAPP2015-InternationalConferenceonComputerVisionTheoryandApplications

650