Coordinate System Transformations for Imitation of Goal-Directed Trajectories in Virtual Humans

Abstract

The goal of this work is to record human hand trajectories during goal-oriented reaching behaviors in immersive VR for later imitation by a virtual human. It is desirable that the used trajectory representation generalizes over different positions and orientations of the target object. To this end, this work investigates trajectory representations in two neurophysiologically inspired coordinate systems. Results show that natural looking movement can be preserved as real motion capture data is used as a basis for these calculations.