Development and Evaluation of an Input Method Using Virtual Hand Models for Pointing to Spatial Objects in a Stereoscopic Desktop Environment with a Fitts’ Pointing Task

Abstract

Virtual Reality for desktop or table mounted display environments can be a strong tool in supporting users in sighting spatial information and give a better understanding in their spatial relationships. Visualization of spatial data incorporates spatial interaction and manipulation with six degrees of freedom for data processing which is part of current research objectives. Modern tracking devices for desktop applications allow capturing a model of the human hand with three translational and three rotatory degrees of freedom which can be used as an input system allowing the control over a virtual hand model in a stereoscopic desktop system. Interaction with complex data in rich environments using virtual hand models, such as three-dimensional geographic information systems with integrated spatial data, requires visualization with a low degree of occluding virtual entities in the virtual scene but a high degree of recognition. An experimental task using a Fitts’ pointing task is conducted to investigate different minimalistic modeling approaches of virtual hand models using the Leap Motion Controller in a stereoscopic desktop environment. Movement times are analyzed for pointing movements in three different depth layers. Results indicate that a kinematic visualization of a virtual hand model produces significantly better movement times in pointing over a minimalistic point cloud model.