Evaluation of Environment-Independent Techniques for 3D Position Marking in Augmented Reality

Specifying 3D positions in the real world is an essential step to create augmented reality content. However, this task can be challenging when information about the depth or geometry of the target location is not available. In this paper, we evaluate alternative techniques for 3D pointing at a distance without knowledge about the environment. We present the results of two studies evaluating the accuracy of techniques based on geometry and human depth perception. We find that geometric methods provide higher accuracy but may suffer from low precision due to pointing errors. We propose a solution that increases precision by combining multiple samples to obtain a better estimate of the target position.