![[Table of Contents]](/icons/button-contents.gif){kind=icon}
An Exploration of Techniques to Improve Relative Distance
Judgments within an Exocentric Display
by Gina Crvarich
An Exploration of Techniques to Improve Relative Distance
Judgments
within an Exocentric Display
by
Gina B. Crvarich
A thesis submitted in partial fulfillment
of the requirements for the
degree of
Masters of Science in Engineering
University of Washington
1995
Approved by _____________________________________________
Chairperson of Supervisory Committee
Program Authorized
to Offer Degree __________________________________________
Date _____________________________________________
Master's Thesis
In presenting this thesis in partial fulfillment of the requirements for a Master's degree at the University of Washington, I agree that the Library shall make its copies freely available for inspection. I further agree that extensive copying of this thesis is allowable only for scholarly purposes, consistent with "fair use" as prescribed in the U.S. Copyright Law. Any other reproduction for any purpose or by any means shall not be allowed without my written permission.
Signature ______________________________
Date ________________________________
University of Washington
An Exploration of Techniques to Improve Relative Distance
Judgments within
an Exocentric Display
by Gina B. Crvarich
Chairperson of the Supervisory Committee: Professor Thomas Furness III
Department of Industrial Engineering
The focus of this research was to determine how a variety of different computer graphics techniques could contribute to the accurate and rapid perception of relative distance within a three-dimensional (3D) computer generated environment. The specific 3D environment of interest was an exocentric, or god's-eye-view display of a set of target objects hovering over a patch of terrain. In addition to examining the speed and accuracy with which subjects made relative distance judgments, subjective evaluations of judgment confidence and strategy were also taken into consideration when evaluating the benefit of different techniques. The goal of this research was to provide interface design recommendations for those creating exocentric views within 3D spatial displays for a range of applications.
Three experiments were performed. The first experiment used a repeated measure design to evaluate the benefits of three display techniques on a specific relative distance judgment task. The three techniques were: stereoscopic viewing, control over image rotation and tracking an observer's head-motions in order to update the image viewpoint. Subjects made 18 relative distance judgments for each of the eight conditions being tested. Following each condition, subjects gave a confidence rating of their performance. The results indicated that only image rotation had a statistically significant, positive effect on task performance as well as on subjects' self-reported confidence in their performance. Subject-control over image rotation also increased the length of time subjects took to make a judgment.
Since image rotation increased both task accuracy and time per trial, the second experiment evaluated various rotation techniques. Based on subject comments in first experiment, the goal of the second experiment was to determine if subject-controlled rotation, as opposed to computer-controlled rotation, was needed to maintain high task accuracy. In addition, this experiment sought to decrease the length of time taken to make an accurate judgment for the same relative distance task as in the first experiment. Results from this experiment suggest that there is a performance advantage to providing subjects control over image rotation. However, subjects were able to significantly reduce the length of time to make a relatively accurate judgment using only one of the computer-controlled techniques. In addition, subjects stated a strong preference for a rotation technique which combined both subject and computer control capabilities.
Given that subjects performed well and stated a strong preference for the rotation technique which combined both subject and computer control, the third experiment, run as an exploratory study, sought to determine the length of judgment time sufficient for this technique to be effective for the same relative distance task. Results from this pilot study indicated that there is wide subject variance in the ability to make rapid relative distance judgments using this particular rotation technique. However, for decisions which can be made in more than 24 seconds, the particular rotation technique tested was effective.
The results show that in order to accurately perform a relative distance judgment task within an exocentric spatial display, a user should be given the ability to significantly change his or her perspective view of the 3D scene. Such capability also seems to increase a user's confidence in their distance judgments. In an effort to determine an effective method for providing the different perspective views, it was observed both quantitatively and qualitatively that users should be given the ability to control the changes in the world view. A particular rotation technique was preferred by the subjects and known within the experiment as "Discrete Views + Manual Rotation". This technique seemed to allow users the ability to employ a wide range of decision strategies, as well as make accurate relative distance judgments within the spatial display provided.
The research described in this thesis was supported by a contract from the Air Force Office of Scientific Research to Thomas A. Furness III (Contract # 92-NL-225 and INST PROP NO: 78216).
I would first like to thank Suzanne Weghorst for her guidance with this thesis. I truly valued her advice on and interest in this project; her contributions have helped me be very proud of this work. I would also like to thank the other members of my committee, Tom Furness, Woody Barfield and Mark Ganter. I am grateful for their input and have sincerely enjoyed working with each of them over the course of the last two years.
I would also like to thank the following people for each of their roles in this project. To Paul Schwartz and Paul Danset - these two came to the rescue more than once in order to enable the software and the hardware to cooperate with my work. I would like to thank Conrad Kraft - both for his advice during this project and for his illuminating questions during my thesis defense. Also, many thanks to Joey King - for his enthusiasm, interest and excellent advice during the course of my work at the HITLab. It was truly a pleasure to share with him the progression of this work.
I would like to acknowledge and thank Glenna Satalich (the other "g") for being such a true friend - she, most of all, will make my memories of graduate school a wonderful thing to remember. And finally, I would like to thank Bruce Anderson - for his encouragement, love and incredible insight into this amazing process. He truly made me believe I could accomplish this goal - I look so forward to spending the rest of our lives together.
I would like to dedicate this thesis to my parents, Gene and Concetta Crvarich. They have taught me so many wonderful things and have done it all by their loving examples.