A comparison of tracking- and controller-based input for complex bimanual interaction in virtual environments

We describe a user study comparing a two-handed controller-based input device to a two-handed tracking solution, both offering the control space of six degrees of freedom to each hand. For benchmarking the different input modalities we implemented a set of evaluation tasks requiring viewpoint navigation, selection and object manipulation in a maze-like virtual environment. The results of the study reveal similar overall performance for both input modalities for compound tasks. However significant differences with respect to the involved subtasks were found. Furthermore we can show that the integral attributes of a subtask do not necessarily need to be manipulated by a single hand. Instead, the simultaneously required degrees of freedom for operating integrally perceived subtasks may also be distributed to both hands for better control.