Virtual Environments (VE) are being widely used in various research fields for several decades such as 3D visualization, education, training and games. VEs have the potential to enhance the visualization and act as a general medium for human-computer interaction (HCI). However, limited research has evaluated virtual reality (VR) display technologies, monocular and binocular depth cues, for human depth perception of volumetric (non-polygonal) datasets. In addition, a lack of standardization of three-dimensional (3D) user interfaces (UI) makes it challenging to interact with many VE systems. To address these issues, this dissertation focuses on evaluation of effects of stereoscopic and head-coupled displays on depth judgment of volumetric dataset. It also focuses on evaluation of a two-handed view manipulation techniques which support simultaneous 7 degree-of-freedom (DOF) navigation (x,y,z + yaw,pitch,roll + scale) in a multi-scale virtual environment (MSVE). Furthermore, this dissertation evaluates auto-adjustment of stereo view parameters techniques for stereoscopic fusion problems in a MSVE. Next, this dissertation presents a bimanual, hybrid user interface which combines traditional tracking devices with computer-vision based "natural" 3D inputs for multi-dimensional visualization in a semi-immersive desktop VR system. In conclusion, this dissertation provides a guideline for research design for evaluating UI and interaction techniques.
