Many applications, from planetary sciences to medicine, require practitioners to correlate and understand complex datasets through a range of dimensions, scales, resolutions, and temporal dynamics. Such complex nature is referred to as data heterogeneity. Clearly, how we organize, visualize, and correlate these heterogeneous datasets will influence the effectiveness of information interpretation and decision-making, and ultimately reflect on the success of a task. Many technical challenges persist in effectively dealing with complex visualization tasks involving large-scale heterogeneous data. To address some of the challenges, the PI aims to develop a heterogeneous display environment integrated with versatile data visualization and user interaction capabilities. The design of the visualization system is guided by a theory where the various aspects of complex controls on 3D visualization and user interaction are viewed as a multidimensional continuum space. The proposed system can potentially facilitate the process of visualizing and correlating large-scale heterogeneous data. Particularly, the PI will evaluate the effectiveness of the proposed technology using scenarios directly related to scientific activities underway by researchers in NASA's Mars Exploration Program. The system will also offer a platform to refine the continuum theory and investigate methods that could improve the effectiveness of complex data visualization and user interaction.
The PI further outlines a comprehensive plan for developing interdisciplinary education opportunities tightly integrated with the proposed research agenda for a wide range of learner groups. Examples include training opportunities embedded in research activities for diversified learner groups, curriculum development to address the educational needs for students in interdisciplinary areas of research; and the development of short courses and hands-on projects for summer schools and community outreach events. The PI will also explore the value of 3D visualization techniques in science and engineering education.
The success of these plans will set up a solid foundation for the PI to achieve her life-long goals of establishing first-class interdisciplinary research and educational programs.
Among the many technical challenges persisting in 3D visualization systems, we consider the following aspects, display heterogeneity, adaptation of data complexity, rich interaction techniques, and collaborative support, are particularly lacking when dealing with complex visualization tasks involving large-scale heterogeneous data. Through the support of this NSF CAREER award, the PI and her team started off by designing and carrying out a series of carefully-designed user-centered experiments that evaluated the capabilities and limitations of existing visualization and user interface techniques for supporting complex single-user or multi-user visualization tasks. These user-centered experiments led to the development of guidelines for developing a visualization platform more suitable for the needs of complex visualization tasks. In parallel, the team observed the daily operation of a range of application domains, from planetary science, GIS visualization, mining systems, to medical applications, analyzed the task characteristics of these different application domains, and established an array of application requirements. Driven by the guidelines developed through usability studies and domain knowledge through field studies, the team developed a heterogeneous display environment integrated with a rich range of data visualization and user interaction capabilities. The system integrates 2D and 3D display components across a range of dimensions and resolutions and simultaneously creates a variety of virtual or augmented views of multiple heterogeneous datasets. The system offered a platform for investigating theory and methods that could help to tackle the challenges of creating effective data visualization and user interaction techniques. Through the course of the project, the team utilized the developed infrastructure to study and develop a range of 2D and 3D interaction techniques that facilitate a small group of users effectively exploring, interacting with, and collaborating on multiple heterogeneous datasets in coherent and intuitive manners. One significant outcome of this grant was its impact on a new application domain, which is the Mining Safety Training Program developed by graduate student Leonard Brown collaboratively with the Department of Mining and Geological Engineering. Overall, the sponsored research activities output nearly 20 publications, a work in-progress PhD dissertation, and 4 patent applications. Besides the research outcomes, a wide range of educational activities were carried out, including the creation of interdisciplinary educational opportunities embedded in research activities for a wide range of learner groups, curriculum development to address the educational needs for students in interdisciplinary areas of research; and the development of short courses and hands-on projects for summer schools and community outreach events. During the entire project period (2007-2013), one Ph.D. student at the Department of Computer Sciences has been fully worked on this project for his dissertation. The student is expecting to graduate in Fall 2013. Besides, two other Ph.D. graduate students at the College of Optical Sciences) and two post-docs had worked to address specific problems related to this project. The two Optics students graduated with their Ph.D. degree in 2010. Additionally, two undergraduate students (Optical Science) and one undergraduate senior from a minority school participated in the projects as undergraduate research assistant or summer intern for research experience training. On the aspect of curriculum development, the PI developed a new graduate/undergraduate course on the topic of "Introduction to Display Science and Technology." The course was offered every year to both local graduate/undergraduate students as well as to distance learning students. Based on the material developed and taught for the course, the PI is writing a tutorial textbox. The PI also participated a range of outreach activities, from offering lectures at the Arizona Youth University, to host lab visits and demos to undergraduate high-school students, to host high school teachers for Research Experience.
