Enfoque estructurado para el desarrollo de interfaces de usuario 3D (A Structured Approach to the Development of 3D User Interfaces)

Three-Dimensional User Interfaces, or 3DUIs, refer to those interfaces where interaction takes place in three-dimensional space. This area of Human- Computer Interaction is not a novel one, as these interfaces have been studied for several years now, though they have often been presented under other names, such as interactive 3D graphics, Virtual Reality or Augmented Reality. In fact, one of the icons of these interfaces, the head-mounted display, was invented short after another popular icon, the mouse, this being of desktop interfaces. However, the latter have now a state of maturity that three-dimensional interfaces have not reached yet. Desktop interfaces rely on a set of well-known interaction styles and widgets, and their development is based on well-funded methods. Quite the opposite, 3D user interfaces lack of a standard set of interaction techniques and controls, and their development relies on the experience and intuition of the developer.

Despite that fact, progress has been done in identifying the “universal” tasks that can be found in almost every 3D user interface, and together with it, many interaction techniques have been proposed to perform such tasks in different environments, some of them becoming quite popular such as ray casting or gogo. However, experience remains the key to successful development, and in this sense programmers apply their knowledge in software engineering practices, user interface designers rely on their skills in human-computer interaction, and artists use their know-how in content creation. The problem is that many proposed methods approach this development from the view of just one of these roles, then offering a partial solution to that development, which is not only about code, interaction or content but all these things together. Thus, it is necessary to combine both an engineering approach with a creative one.

That is one of the objectives addressed by the TRES-D framework, which is presented in this doctoral dissertation, offering a structured approach that is oriented to task and interaction as well as to objects and content. This way, the proposal promotes the participation of user interface designers and programmers as well as artists and digital content creators. In addition to them, it also involves the client, the domain expert and, of course, the user. The whole process is characterized by two main phases. The aim of the first one –the previous studyis to propose a solution to a given problem and, only if it is accepted, then the second phase follows –the detailed study- to complete the proposed design, and to implement and deploy it. This division in two phases reminds of former approaches to software development that were meant to reduce risks, and that is exactly its purpose here, to reduce the risks that the development of experimental interfaces involves.

All in all, TRES-D is not simply a process model. A new digital-virtual-real continuum is also presented, which is meant to describe the design space of three-dimensional interfaces. Three meta-models, one for objects, one for interaction and a last one for space, offer developers a common language to better understand these interfaces, bring their ideas to paper, and communicate them to other developers. In order to help the designer when selecting the right interaction technique, two decision trees are given as guidance tools, one for object selection/manipulation tasks and another one for text input, both for immersive environments. To help the programmer when translating existing interfaces to three-dimensional environments, a new library is introduced, named VUIToolkit, which transforms plain widgets into 3D ones by adding true depth to them. Finally, the TRES-D proposal is demonstrated with four case studies, each showing a different application of 3D user interfaces.
Ph.D. thesis
University of Castilla-La Mancha, Albacete, Spain, 29 February 2008