High-Performance Spatial Hybrid User Interfaces

点击翻译按钮获取中文摘要

Interaction with three-dimensional (3D), spatial data is increasingly common. Spatial data are important for many sectors, including interior design, architecture, urban planning, industrial planning and design, simulation, training, animation, the game and entertainment industry, and the new “maker” movement. Yet, user interfaces for spatial data are complex, difficult to learn, or limited in functionality. My research program investigates better spatial user interfaces for 3D content. The new spatial hybrid user interfaces are a combination of user interfaces, where the user manipulates virtual objects quickly in the air, with the best of desktop user interfaces. Working towards the vision of spatial user interfaces that are efficient as 2D interfaces, the research program involves three main thrusts: characterization of basic operations, accurate interaction, and new technologies. The first thrust investigates how well current, basic spatial interaction techniques, such as 3D rotations, docking, and travel, perform in terms of speed and accuracy. Moreover, the research also investigates how fast transitions between in-air and desktop interfaces are and how reliable gestures are as a substitute for buttons and keys. While in-air manipulation seems attractive, the available experimental data points to a lack of precision. Thus, we investigate hybrid combinations of in-air and desktop interfaces in the second thrust, where the desktop interface is used for fine-tuning of the final result. Such combinations will yield hybrid spatial user interfaces for 3D content creation and editing that are both fast and precise. Moreover, we study new techniques for resolving depth ambiguities and switching between mono- and stereoscopic viewing. The final thrust of this research program investigates new tracking technologies for fingers and pens, with the aim to further increase tracking accuracy. The fundamental contributions of my research program will be accurate quantifications of the throughput of existing and new spatial interaction techniques, new methodologies for evaluation of interaction methods, novel design guidelines, as well as new technologies. Together, the work constitutes a major step forward towards high-performance spatial user interfaces that are still easy to learn. Better spatial user interfaces will increase the accessibility of 3D modeling for non-specialists, which in turn will accelerate the adoption of 3D content manipulation and creation by the general public, e.g., driven by the new maker/3D printing movement. Moreover, 3D content creation and gaming are core Canadian strengths. Canada will benefit from the research program presented here through many globally competitive HQP in user interface design and 3D graphics, as well as a potential startup. Also, the broad applicability of the results will increase interest in science and technical disciplines.