Special Section Trend Report >> Good News

• By Will Craig
• 05/26/05

AN ENVISION STUDENT uses the Chinese
Garden application in the FLEX environment
and is transported to an Eastern China
garden in the year 1708.

Purdue University’s remarkable center breaks away from the classroom experience and brings true multi-sensory discovery and learning to students.

By Krishna P.C. Madhavan, Laura L. Arns, and Gary R. Bertoline

There is no doubt that the last decade has seen an explosive increase in the quantity of data available for discovery and learning purposes. And this is a good thing—humans do, after all, process their data through a variety of senses: auditory, oral, and tactile senses, plus the sense of sight. Yet, the highly textual nature of traditional datasets has represented a purely visual data form that caters only to a small percentage of the number of methods in which humans can understand data. In other words, most of these large volumes of data are considered to be “perceptually inferior”; says Grok-It Concepts (Grok-It Science; www.medibolt.com/conbottom.htm). This, in essence, is the problem that the field of data perceptualization attempts to address.

What is data perceptualization? Data perceptualization deals with various theoretical and practical techniques that can translate large volumes of data into useful information by allowing, according to Stuart K. Card, “…a richer use of many [human] senses, including sound and touch, to increase the rate at which people can assimilate and understand information” (from “Visualizing Retrieved Information: A Survey,” IEEE Computer Graphics and Applications, March 1996). And Grok-It Science states that while the fundamental principle underlying data perceptualization is intuitive and powerful, it is the process of translating data to overcome “perceptual inadequacy” itself that represents a new and emerging frontier in science and technology.

Bridging Discovery and Learning

That’s where the NSF-funded Envision Center for Data Perceptualization (www.envision.purdue.edu) at Purdue University (IN) comes in. The Center is organized under the Office of the Vice President for Information Technology, and is an interdisciplinary, high-performance, perceptualization showcase facility to support discovery and learning at Purdue. After all, it has long been understood that “undergraduates need to become an active part of the audience for research” (Reinventing Undergraduate Education: A Blueprint for America’s Research Universities. The Boyar Commission on Educating Undergraduates in the Research University, 1998; naples.cc.sunysb.edu/Pres/boyer.nsf), and that “the basic idea of learning as inquiry is the same as the idea of research; even though advanced research occurs at advanced levels, undergraduates beginning in the freshman year can learn through research.” But the Envision Center is not designed for students alone: While students (undergraduates and graduates) are exposed to the latest technology in the field of data perceptual-ization computing through the center’s instruction, Purdue faculty members use the facility for displaying and interacting with scientific data in innovative ways. This dual focus of the Envision Center acts as a strong bridge between discovery and learning, as it encourages active student participation in research projects.

Catalysis Project: Understanding at a Rapid Pace. One of those projects is the Catalysis Project, currently in progress at the center. Catalysis is a prime example of the link that the Envision Center is forging between discovery and learning. Importantly, it provides undergraduates and graduate students the opportunity to learn and explore, and at the same time, contribute to real-world research. Certainly, the development of new materials for new applications is one of the outstanding technical challenges in material science and chemical engineering. And in fact, recent advances in high throughput experimentation have accelerated the testing of new materials by an order of magnitude; the availability of powerful supercomputers, coupled with the evolution of the Teragrid, is enabling high throughput of a large number of molecular structures.

The key challenge addressed by the Catalysis project is to use new perceptualization tools to rapidly focus the vast amount of data that is generated, in a form that a materials expert can perceive, understand, and analyze at a fast pace. Funded by the Department of Energy and a 21st Century Grant from the state of Indiana, students from the Envision Center partner with students in the Chemical Engineering department to integrate their classroom projects into mainstream research projects. In pedagogical terms, by allowing students to actively engage with the concepts they are learning, the Catalysis project promotes situated learning. David Stein, in the online Situated Learning in Adult Education (ERIC Digest No. 195, 2000; http://www.cete.org/acve/docgen.asp?tbl=dige sts&ID=48), states that “in the situated learning approach, knowledge and skills are learned in the contexts that reflect how knowledge is obtained and applied in everyday situations…By embedding subject matter in the ongoing experience and by creating opportunities for learners to live subject matter in the context of real-world challenges, knowledge is acquired and learning transfers from the classroom to the realm of practice.”

The infrastructure available at the Envision Center plays a critical role in bridging that gap between discovery and learning. Some of the exciting technologies in use at the Center are:

VR Theater. Virtual Reality (VR) refers to “immersive, interactive, multisensory, viewer-centered, three-dimensional, computer-generated environments and the combination of technologies required to build these environments” (Carolina Cruz-Neira, Projection-Based Virtual Reality: The CAVE and Its Applications to Computational Science. Ph.D. Thesis, University of Illinois, 1995). VR theaters immerse users in the environment they are viewing. With such theaters, users aren’t just passive observers in the computer-generated world, but are interacting with the various components of the environment in real-time. Envision’s VR Theater is a Fakespace FLEX System (www.fakespace.com), featuring three 8 x 10 foot panels for rear projection of large-scale 3D images. These movable screens can be easily and rapidly rearranged to form a semi-enclosed “room” with three walls plus a fourth panel as the floor—an arrangement that creates a 3D immersive virtual environment. The VR Theater is also equipped with a state-of-the-art tracking system that allows corrective perspective rendering and direct interaction with the virtual environment. A fivechannel speaker system in the corners of the facility adds surround-sound cues to the virtual reality environment. Depending on the task at hand, the VR Theater is driven either using a 32-processor SGI Oynx2 computer, or an 8-processor SGI Oynx4. A high-end Windows and Linux PC cluster has been added to the already powerful VR Theater.

Motion Capture. In the virtual reality world, the process of recording a person’s movements or other live motion event, and converting those movements into a digital, 3D representation of the motion is called motion capture. The Envision Center houses a portable STT Motion Captor optical motion capture system composed of six infrared cameras on tripods, and as many as three linked computers. The system is operated in collaboration with the Department of Visual and Performing Arts, and is capable of capturing the movements of two people simultaneously. During the last academic year, for instance, graduate and undergraduate student groups from the Computer Graphics Technology program at Purdue used this system to produce an interactive dance performance in association with the Department of Visual and Performing Arts. A solo dancer wore the motion capture suit on stage, during the performance (see above). Data gathered from the suit was then used to create imagery which was projected onto two large screens on the stage.

Tiled Wall. The tiled display wall at the Envision Center is a 7 x 12 foot high-resolution display made up of a grid of 12 smaller projection displays controlled by many computers working together. The wall is capable of displaying 4,096 pixels horizontally and 2,304 pixels vertically, for a total of 9.4 million pixels—about five times the resolution found on a typical desktop workstation. The display system is extremely versatile, and can be used for2D, 3D, or stereo production of single, large-scale continuous images (see the image below). This capability is particularly useful when working with satellite imagery of geographical areas, and thus, not surprisingly, has immediate implications for homeland security and environmental protection initiatives. Yet, the tiled wall can also be used to display different graphics on each individual screen, or a combination of screens in the grid can be utilized for a variety of media displays. Image generation for the tiled wall is driven by a cluster of 12 high-performance PCs with high-end graphics cards.

Access Grid. The Envision Center places a premium on collaborating with researchers and student groups throughout the world, and the Access Grid (AG) at the center helps to make that possible. The Access Grid is an ensemble of resources that includes multimedia large-format displays, presentation and interactive environments, and interfaces to grid middleware and visualization environments. (For more on AGs, go to www.accessgrid.org). The AG allows participants from many geographically distant sites to communicate and collaborate with each other easily and inexpensively (see the visual, right). But unlike traditional videoconferencing, it runs over Internet2, using open source software developed by Argonne National Laboratory (www.anl.gov). The software is quite flexible, allowing each site to operate using a variety of hardware setups that range from individual laptop-based nodes to large classrooms with multiple projected displays. And by creating shared applications and tools, the AG also allows participants to share more than just video streams. In Spring 2004, the Envision Center developed an Introduction to Virtual Reality course jointly taught by faculty at Purdue and Indiana University, via the Access Grid. The class was also offered in Fall 2004 by Purdue and Iowa State University. Notably, a student in the spring class launched the development of AGJuggler, a toolkit that allows sharing of virtual environments and stereographics over the Access Grid.

Other capabilities. Over and above these components, the Envision Center also features a variety of elements such as high-resolution displays and portable stereo displays that can be used in classrooms to enhance instruction through the use of stereo graphics, sensing and haptic devices (a haptic device allows a user to interact with a computer by receiving tactile feedback), nano-manipulation haptic devices, and other high-performance computing resources. These elements allow researchers at the Envision Center to focus on pure research activities also. For example, at the recently concluded Measured Response Workshop organized by the Purdue Homeland Security Institute ( www.purdue.edu/DiscoveryPark/phsi), researchers at the Envision Center unveiled a new type of large, high-resolution display that promises a wide range of immediate applications.

Future Directions

How do students who have been a part of Envision react to the experience? One student participant pretty much summed it up: “Doing this has really taught me much more than anything I’ve learned in class.” And in fact, an important goal of the Envision Center is to provide such real-world research experience to students grounded in critical examination of various scientific concepts learned in the classroom. To this end, the Envision Center is continuing to explore the possibilities of linking classroom education with real-world research efforts. Students currently enrolled in various courses at Purdue, including courses such as Introduction to Augmented Reality, continue to team with researchers to produce classroom products that have direct and immediate use as mainstream data perceptualization tools. The center has also launched the new program, Envision Discoverers, wherein Purdue students can engage in informal learning about visualization tools and gain hands-on experience with the center’s scientific capabilities—all without the constraints (deadlines and grades) that are associated with forcredit classes. With its myriad facets, the Envision Center expects to make a significant contribution to bridging discovery and learning in the areas of science, technology, engineering, and mathematics.