Model-free Marking
Specifying points in 3D space is fundamental operation for many AR tasks. If a model of the environment is known, a simple raycast can be used to determine the position over surfaces. Without a model of the environment, two rays can be used to triangulate a target. Naive triangulation techniques suffer from low precision due to user input, or hardware limitations. In this research we designed, evaluated, and tested two enhanced geometric techniques for 3D point marking. VectorCloud uses multiple rays to reduce pointing jitter, and Image Refinement allows users to refine 3D directions of the rays on a static image of the target area. Both techniques improve the precision of 3D point marking, and ImageRefinement is superior to VectorCloud overall. These techniques are relevant in the design of mobile AR systems for large outdoor areas.
|
Lages, W. S., Li, Y., Lisle, L., Lu, F., Höllerer, T., & Bowman, D. A. (2019, March). Enhanced Geometric Techniques for Point Marking in Model-Free Augmented Reality. In 2019 IEEE Conference on Virtual Reality and 3D User Interfaces (VR) (pp. 1036-1037). IEEE. Chicago. https://doi.org/10.1109/VR.2019.8798087
|
Team: Wallace Lages, Yuan Li, Lee Lisle, and Doug Bowman
Sponsored by: Office of Naval Research
Sponsored by: Office of Naval Research