Virtual reality is being considered as a new human-computer interface in various applications such as teleoperation, information visualization, and entertainment. One of the key challenges in virtual environment systems is to effectively reveal spatial relationships among objects within a three-dimensional (3D) space, particularly in the depth dimension, so that the user can perceive, locate and manipulate such objects with respect to each other effortlessly. In order to present 3D information to users effectively, selection and combination of proper depth cues are important.
In this study, a series of two experiments were conducted to investigate the effectiveness of stereopsis and partial occlusion. The first experiment was prepared for the second experiment, in which the display/control (D/C) gain of 3D controller was optimized. In the second experiment, human performance was evaluated by examining task completion time and coordination of a 3D pointing task in a virtual environment.
The results indicated that stereopsis is superior to partial occlusion in terms of two measures of performance (task completion time and coordination). In order to evaluate the effects of depth cues on human performance according to index of difficulty (ID) of the task, we classified tasks into two groups by their IDs. In the case of tasks that have lower IDs, stereopsis is more powerful than partial occlusion in terms of task completion time; however, in the case of tasks that have higher IDs, partial occlusion is nearly equal to stereopsis in terms of task completion time.
The results of this study can be applied to the design of 3D pointing task in a virtual environment. The effects of depth cues on human performance are dependent on the ID of the task. Therefore, it is necessary to consider the ID of the task in selecting and combining depth cues in virtual environment systems.
