Uncertainty Boundaries for Complex Objects in Augmented Reality

摘要: Registration errors between the physical world and computer- generated objects are a central problem in Augmented Reality (AR) systems. Some existing AR systems have demonstrated how to dynamically estimate registration based on estimates of spatial system. Using these error estimates, also number ways ameliorating effects error. One part this previous work was creation use regions around objects; unfortunately, analytic methods used only created accurate for simple convex objects. In paper, we present stable algorithm generating uncertainty complex objects, including non-convex with interior holes. We demonstrate our approach can be create set more error-based highlights presence error, as general highlighting mechanism.