The university classroom of the future may bear little resemblance to the instructor-centered environments common today. Distance education has been part of university curricula in the form of correspondence and telecourses in the past, but the maturity and affordability of such technologies as interactive compressed video and electronic mail offer the potential for highly interactive, student- centered learning environments which are as satisfying and engaging as "being there." This environment is established, in the view of one professor, through successful collaboration among students who are actively engaged in the construction of "something meaningful and substantial."* This paper is an evaluation of computer science course, "Virtual Reality, Telepresence and Beyond," offered to graduate students, both on- and off-site at the University of Maryland in the fall of 1993. The evaluation addresses the question of what makes a collaboration successful, assembling data from student surveys, class observations, a student journal, and electronic mail to test hypotheses that students who are more actively involved in collaborating on class projects will report a more satisfying experience than those who were not, and suggests there will be no significant difference in satisfaction based upon whether the student is enrolled locally or remotely. However, it is suggested that the use of interactive, compressed video will enhance the sense of presence of remote students over more traditional forms of distance learning.
*Ben Shneiderman, Professor, Department of Computer Science, University of Maryland
------------------------------------------------------------------------------------------------
In the Fall term of 1993, Ben Shneiderman of the University of Maryland presented a course titled "Virtual Reality, Telepresence, and Beyond." The course was offered to students at the University of Maryland and to remote students in 12 states across the United States. The remote students viewed the class through live video feeds and via satellite with a 2-5 day tape delay through National Technological University. The first project for the course was called EVE which stands for Encyclopedia of Virtual Environments.
EVE is a series of short articles authored by members of the class. Originally, EVE articles were envisioned to be written by pairs of students taking the class from different locations. Real world difficulties such as students dropping the course and getting pairs of authors at different locations to sign up for each EVE topic forced the articles to be written by both single authors and by co-located pairs of authors as well as pairs of authors at diffenent locations.
The topic of this paper is to evaluate the EVE class project from two perspectives. The first is in the evaluation of the production of EVE articles. This will be examined as a case study in Computer Supported Collaborative Work (CSCW). (This term also appears as Computer Supported Cooperative Work.)
The second is the evaluation of EVE itself as an encyclopedia for users to extract information in a quick and usable way. A variety of users completed a survey to evaluate the aspects of EVE as a tool. Recommendations for improvements to the EVE class project are then given based on the findings.
Both members of this project are remote students living in Seattle, Washington and Colorado Springs, Colorado.
------------------------------------------------------------------------------------------------
With the advent of affordable helmet-mounted display systems, many system designers are challenged with understanding the implications of employing such a device in their design. This type of implementation, to remain cost effective, is likely not to be generating a high-resolution image, nor is it likely to have a tracking capability. Typical environments where this system would be employed are teleremote operations of robotic assemblers, high-risk or contamination-prone work environments, or remote operations in spaces that will not permit the user to be manipulating the actuator assembly of a machine, such as underwater vehicle manipulators.
While a large amount of literature exists regarding the physiological issues of a helmet-mounted display systems, little empirical work has been published regarding the field-of-view issues. This is attributed to the technology traditionally found in these display systems, generally having such as high cost factor that tracking was likely to be a small additional cost, and the understanding that helmet-mounted displays introduced psychological and perceptual phenomena that required tracking to prevent reduced performance. Regardless of the past, the use of helmet-mounted displays can still have advantages over panel-mounted displays, if carefully implemented by the system designers.
This report examines the evidence of tracking a two-dimensional path using a telerobotic vision system guided by a remote joystick using either panel-mounted visual displays or helmet-mounted visual displays.
------------------------------------------------------------------------------------------------
The purpose of this experiment is to investigate performance on a perspective tactical display as compared with a plan-view display. The perspective display represents information about altitude and attitude directly, whereas the plan-view display necessitates the use of an auxiliary information panel to show this information. Participants performed tasks requiring them to identify all descending aircraft displayed, with two different numbers of descending aircraft. Time to completion and partial results were collected automatically for each participant in a 2x2 within-subjects design. The experiment revealed strengths and weaknesses of both types of displays: plan-view was more effective for finding every descending aircraft, while perspective view was better for quickly identifying most of the descending aircraft.
------------------------------------------------------------------------------------------------
The application of vertex-based representations for storing images and its potential as a compression technique is discussed. Vertex representations are data structures that have been successfully used in the domain of VLSI circuit design by virtue of being very compact. A review of vertex representations and of standard image compression schemes is presented. An algorithm for transforming pixel rasters into vertex rasters is described. This algorithm was applied to a series of test images and the resulting vertex structures were further compressed using two popular data compression techniques: Huffman encoding and the Lempel-Ziv-Welsh algorithm. Strengths and weaknesses of vertex rasters are inferred from these experiments and directions for further research are suggested.
------------------------------------------------------------------------------------------------
This project aims at using novel data access methods and efficient screen-drawing algorithms to facilitate rapid and flicker-free zooming of a 2-dimensional starfield display. A starfield display uses dots to represent database items; the location and color of each dot being a function of the attributes of the item.
The project uses an existing FilmFinder application that embodies these concepts. The X-axis shows the year of the film (ranges from 1920 to 1995) and the Y-axis has the popularity (0-9). Each axis has a double-box slider that changes the scale of the axis to perform a zoom - for example, using the x-axis slider, you can zoom in to view films made between 1950 and 1995. Zooming in causes the films' display rectangles to grow in size and also move to different parts of the display.
The third dimension is the length of each film, which is varied using a range slider. The display itself is still 2-dimensional, but the same principles that are used for the x and y ranges, apply.
For the zoom operation to appear continuous and smooth, three requirements have to be met.
We have achieved smooth zooming by changing the data structures and data access methods (#1) as well as by increasing the update rates (#2).
------------------------------------------------------------------------------------------------
This paper describes the development of algorithms for the recognition of (Army) Dismounted Infantry signal gestures. The set consists of seven static poses (two arms), and ten dynamic (one arm and two arm) coordinated movements. The tracker data was generated by 6-DOF tracking devices attached to the wrists and back of the subject. The approach for static getures is based on template matching. For the dynamic gestures, two approaches were used. One based on tracing 3d points over time, using a state machine like algorithm and the other using analysis of the trajectories of the sensors. The development of the recognition system is not yet complete, but partial results are reported.
A set of .tif files are on the ftp server, under pub/vrtp/Jove/Gestures which show the specific gesture set.
------------------------------------------------------------------------------------------------
With the arrival of the information age, we are provided with tools that allow us to share information easily along the information highway. What better way to make use of this technology than to provide a computer science departmental server that provides general information of our department and an easy access to the work we have done.
In this report, I shall describe the process it took for the design and implementation of the departmental server, and the kind of considerations that we had to go through.
------------------------------------------------------------------------------------------------
This paper proposes a three dimensional universe of tree-mapped data that uses unique navigation techniques and animation to provide a density increase in visualization. Some of the limitations of tree-maps and cone trees are addressed while preserving their benefits. In addition, "universal" navigation is presented in a way that promotes context identification rather than disorientation. The paper concludes with possible future extensions.