The digital age offers a lot of challenges in the eld of visualization. Visual imagery has been effectively used to communicate messages through the ages, to express both abstract and concrete ideas. Today, visualization has ever-expanding applications in science, engineering, education, medicine, entertainment and many other areas. Different areas of research contribute to the innovation in the eld of interactive visualization, such as data science, visual technology, Internet of things and many more. Among them, two areas of renowned importance are Augmented Reality and Visual Analytics. This thesis presents my research in the fields of visualization and human-machine interaction. The purpose of the proposed work is to investigate existing solutions in the area of Augmented Reality (AR) for maintenance. A smaller section of this thesis presents a minor research project on an equally important theme, Visual Analytics. Overall, the main goal is to identify the most important existing problems and then design and develop innovative solutions to address them. The maintenance application domain has been chosen since it is historically one of the first fields of application for Augmented Reality and it offers all the most common and important challenges that AR can arise, as described in chapter 2. Since one of the main problem in AR application deployment is reconfigurability of the application, a framework has been designed and developed that allows the user to create, deploy and update in real-time AR applications. Furthermore, the research focused on the problems related to hand-free interaction, thus investigating the area of speech-recognition interfaces and designing innovative solutions to address the problems of intuitiveness and robustness of the interface. On the other hand, the area of Visual Analytics has been investigated: among the different areas of research, multidimensional data visualization, similarly to AR, poses specific problems related to the interaction between the user and the machine. An analysis of the existing solutions has been carried out in order to identify their limitations and to point out possible improvements. Since this analysis delineates the scatterplot as a renowned visualization tool worthy of further research, different techniques for adapting its usage to multidimensional data are analyzed. A multidimensional scatterplot has been designed and developed in order to perform a comparison with another multidimensional visualization tool, the ScatterDice. The first chapters of my thesis describe my investigations in the area of Augmented Reality for maintenance. Chapter 1 provides definitions for the most important terms and an introduction to AR. The second chapter focuses on maintenance, depicting the motivations that led to choose this application domain. Moreover, the analysis concerning open problems and related works is described along with the methodology adopted to design and develop the proposed solutions. The third chapter illustrates how the adopted methodology has been applied in order to assess the problems described in the previous one. Chapter 4 describes the methodology adopted to carry out the tests and outlines the experimental results, whereas the fifth chapter illustrates the conclusions and points out possible future developments. Chapter 6 describes the analysis and research work performed in the eld of Visual Analytics, more specifically on multidimensional data visualizations. Overall, this thesis illustrates how the proposed solutions address common problems of visualization and human-machine interaction, such as interface de- sign, robustness of the interface and acceptance of new technology, whereas other problems are related to the specific research domain, such as pose tracking and reconfigurability of the procedure for the AR domain.

Visualization and Human-Machine Interaction / Manuri, Federico. - (2017). [10.6092/polito/porto/2673784]

Visualization and Human-Machine Interaction

MANURI, FEDERICO
2017

Abstract

The digital age offers a lot of challenges in the eld of visualization. Visual imagery has been effectively used to communicate messages through the ages, to express both abstract and concrete ideas. Today, visualization has ever-expanding applications in science, engineering, education, medicine, entertainment and many other areas. Different areas of research contribute to the innovation in the eld of interactive visualization, such as data science, visual technology, Internet of things and many more. Among them, two areas of renowned importance are Augmented Reality and Visual Analytics. This thesis presents my research in the fields of visualization and human-machine interaction. The purpose of the proposed work is to investigate existing solutions in the area of Augmented Reality (AR) for maintenance. A smaller section of this thesis presents a minor research project on an equally important theme, Visual Analytics. Overall, the main goal is to identify the most important existing problems and then design and develop innovative solutions to address them. The maintenance application domain has been chosen since it is historically one of the first fields of application for Augmented Reality and it offers all the most common and important challenges that AR can arise, as described in chapter 2. Since one of the main problem in AR application deployment is reconfigurability of the application, a framework has been designed and developed that allows the user to create, deploy and update in real-time AR applications. Furthermore, the research focused on the problems related to hand-free interaction, thus investigating the area of speech-recognition interfaces and designing innovative solutions to address the problems of intuitiveness and robustness of the interface. On the other hand, the area of Visual Analytics has been investigated: among the different areas of research, multidimensional data visualization, similarly to AR, poses specific problems related to the interaction between the user and the machine. An analysis of the existing solutions has been carried out in order to identify their limitations and to point out possible improvements. Since this analysis delineates the scatterplot as a renowned visualization tool worthy of further research, different techniques for adapting its usage to multidimensional data are analyzed. A multidimensional scatterplot has been designed and developed in order to perform a comparison with another multidimensional visualization tool, the ScatterDice. The first chapters of my thesis describe my investigations in the area of Augmented Reality for maintenance. Chapter 1 provides definitions for the most important terms and an introduction to AR. The second chapter focuses on maintenance, depicting the motivations that led to choose this application domain. Moreover, the analysis concerning open problems and related works is described along with the methodology adopted to design and develop the proposed solutions. The third chapter illustrates how the adopted methodology has been applied in order to assess the problems described in the previous one. Chapter 4 describes the methodology adopted to carry out the tests and outlines the experimental results, whereas the fifth chapter illustrates the conclusions and points out possible future developments. Chapter 6 describes the analysis and research work performed in the eld of Visual Analytics, more specifically on multidimensional data visualizations. Overall, this thesis illustrates how the proposed solutions address common problems of visualization and human-machine interaction, such as interface de- sign, robustness of the interface and acceptance of new technology, whereas other problems are related to the specific research domain, such as pose tracking and reconfigurability of the procedure for the AR domain.
2017
File in questo prodotto:
File Dimensione Formato  
Thesis.pdf

accesso aperto

Descrizione: Doctoral Thesis
Tipologia: Tesi di dottorato
Licenza: PUBBLICO - Tutti i diritti riservati
Dimensione 60.28 MB
Formato Adobe PDF
60.28 MB Adobe PDF Visualizza/Apri
Pubblicazioni consigliate

I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.

Utilizza questo identificativo per citare o creare un link a questo documento: https://hdl.handle.net/11583/2673784
 Attenzione

Attenzione! I dati visualizzati non sono stati sottoposti a validazione da parte dell'ateneo