Efficient, Intuitive User Interfaces for Classroom-Based Immersive Virtual

Environments

Doug A. Bowman, Matthew Gracey & John Lucas

Virginia Tech, Dept. of Computer Science, Blacksburg, Virginia, USA

{bowman | mgracey | jolucas}@vt.edu

Abstract

The educational benefits of immersive virtualenvironments (VEs) have long been touted, but very fewimmersive VEs have been used in a classroom setting. Wehave developed three educational VEs and deployed themin university courses. A key element in the success ofthese applications is a simple but powerful user interface(UI) that requires no training, yet allows students tointeract with the virtual world in meaningful ways. Wediscuss the design of this UI and the results of anevaluation of its usability in university classrooms.

1. Introduction

For many years, researchers in the area of immersivevirtual environments (VEs) have pointed to education as akey application area [2, 3]. However, the use ofimmersive VEs in educational settings has been limited.Poor system usability and a lack of low-cost immersiveVE systems are two major obstacles.

In our research, we have been developing educationalapplications of VEs with a focus on usability and cost.We believe that educational VEs need to be highlyinteractive in order to succeed. If highly interactive VEsare to be used in the classroom, however, the UI must betransparent – instantly understandable and extremelyefficient. We describe a generic UI metaphor that hasproven to be highly usable in real classroom use withthree different educational VE applications.

2. Educational VE applications

We have developed three VE tools for classroom use.Virtual-SAP [1] is a VE tool for visualizing the

effects of earthquakes on building structures. Users canbuild 3D structures and can specify the size and materialproperties of each element. The response of the structureto an earthquake is then simulated, and the results areshown as a 3D animation within the VE. Students canuse Virtual-SAP by watching as pre-defined structures aresubjected to earthquakes, or by testing various “what-if”scenarios. We have used it in both undergraduate- andgraduate-level Building Structures classes.

The Vir tua l Env ironment NormalizingTransformation System (VENTS) helps computergraphics students learn the normalizing transformation for

3D perspective views. VENTS allows students tovisualize a 3D world coordinate system, a view frustumalong with related clipping planes, points, and vectors,and an object to be rendered (figure 1). Students can thenselect the steps of the transformation and see the objectsmove to show the effect of each step. Audio descriptionsof each step are available, and users can interactively setthe initial conditions for the transformation. VENTS hasbeen used in senior-level computer graphics classes.

Figure 1. VENTS (left) and NetViz (right)

The third educational VE, NetViz, is a visualization oflarge-scale network traffic data. NetViz (figure 3) uses a“city” metaphor to represent the data, with buildingsrepresenting a connection or set of connections. The cityis divided into “districts” representing different protocols.Within each district, the building’s location represents theaverage bandwidth and average packet size of theconnection, while its height represents the lifetime of theconnection. Students can watch the city change as timepasses, navigate to any point, and analyze patterns in thedata. NetViz has been used in both undergraduate andgraduate classes in computer networking.

3. User interface design

We have designed a standard UI for these threeapplications focused on the special usability requirementsof classroom-based VEs. Our goals were learnability,efficiency, multiple student use, whole class benefits, andease of use for instructors.

The hardware we chose for our system needed to beinexpensive and portable. We assembled a system usingonly commercial, off-the-shelf components, consisting of:

• Dual-boot (Windows 2000/Linux) PC• Daeyang i-visor head-mounted display• Intersense Intertrax2 3DOF tracker• Handykey Twiddler2 12-button chord keyboard• InFocus LCD projector

IEEE Virtual Reality 2004 March 27-31, Chicago, IL USA 0-7803-8415-6/04/$20.00©2004 IEEE.

219Proceedings of the 2004 Virtual Reality (VR’04) 1087-8270/04 $ 20.00 IEEE

• VGA video splitter• Rolling A/V cart

In this system, one immersed student wears the HMDand head tracker, and uses the chord keyboard for input.The rest of the class sees the immersed student’s viewprojected on a large screen. This complete immersive VEsystem cost less than $7000, orders of magnitude lessthan typical research lab systems.

a. b.

Figure 2. The “remote control” metaphor

We use a remote control metaphor, where eachapplication function is mapped to a specific button on theinput device. An on-screen interface representationreminds the user of this mapping (figure 2). When thereare more than 12 functions, we use a “tabbed” interface,where 3 buttons select a mode and the other 9 are used forfunctions in that mode. For 3D navigation we use gaze-directed steering and allow both forward and backwardmotion. Two buttons on the input device are permanentlymapped to navigation (figure 2b). 3D selection andmanipulation are performed via a ray-casting metaphorbased on the user’s gaze direction, and users can moveobjects toward or away from them along this ray. Finally,we provide a view-locking feature to allow instructors tofreeze the image momentarily, and pre-def inedviewpoints to which the user can be taken automatically.

4. Usability results

We have used our educational VE applications in fivedifferent undergraduate and graduate classes. The classesranged in size from 7 to 45 students. We collected variousmeasures of usability based on observations, surveys, andinterviews. Overall, the usability of the applications washigh. All students who used the applications directly wereable to navigate the 3D environment and perform taskswithin it. In our most recent classroom visit, studentswho used Virtual-SAP rated it at 6.67 on a 7-point scalefor ease of use.

The remote control metaphor was also successful. Allstudents understood the mapping between therepresentation and the buttons on the device, and evenwhen buttons changed based on the current mode,students had no difficulty adapting.

Gaze-directed flying was intuitive for most students.Several students familiar with video games suggested thatthe addition of a “strafing” feature, in which the user canalso move to the left/right, would be useful.

Students found gaze-based selection and manipulationto be easy and effective. It is less fatiguing than image-plane selection, and can be quite efficient.

The combination of one immersed student and an“audience” of observing students was a potential problemarea, but we found that the audience understood what theysaw on the screen quite well. They rated theirunderstanding of what they saw on the screen at averagesof 5.75 (Virtual-SAP) and 5.42 (NetViz).

The ability to lock the user’s view and to take the userto pre-defined viewpoints allowed instructors morecontrol over what the class saw. In Virtual-SAP, forexample, the instructor wanted students to view structuresfrom above, so view locking allowed users to relax whileseeing a bird’s-eye view.

The only major negative usability results relate to theVE hardware we used. The HMD was difficult to keep inplace and uncomfortable to wear. The narrow FOV madeit easy for users to become disoriented. The head trackerproduced unpredictable data when the head orientationwas out of its range. It appears that although one can puttogether an immersive VE system for under $10,000,usability may suffer as a result.

5. Conclusions

We have presented a usable and low-cost UI forimmersive VE applications used in a classroom. This UImeets our stated goals: it is learnable and efficient; itallows multiple students to experience the VE; it allowsthe entire class to benefit; and it is generally easy forinstructors to use as a teaching tool.

6. Acknowledgements

This work was partially supported by National ScienceFoundation Grant DUE-0127326. We would like to thankMehdi Setareh, Alex Kalita, Srinidhi Varadarajan, MarkHenry, Prasuna Chennupati, and Kristin Wheeler.

7. References

[1] Bowman, D., M. Setareh, M. Pinho, N. Ali, A.Kalita, Y. Lee, J. Lucas, M. Gracey, M. Kothapalli,Q. Zhu, A. Datey, and P. Tumati, “Virtual-SAP: AnImmersive Tool for Visualizing the Response ofBuilding Structures to Environmental Conditions,”Proceedings of IEEE Virtual Reality, 2003, 243-250.

[2] Durlach, N. and A. Mavor, Virtual Reality: Scientificand Technological Challenges, National AcademyPress, 1995.

[3] Sutherland, I. “The Ultimate Display,” presented atthe IFIP Congress, 1965.

220Proceedings of the 2004 Virtual Reality (VR’04) 1087-8270/04 $ 20.00 IEEE