What's NEXT? An interactive next best view approach

摘要

Shape-from-silhouettes algorithms use either a surface or a volumetric approach. A problem with the volumetric approach is that in general we know neither the accuracy of the reconstruction, nor where to locate new viewpoints for improving the accuracy. In this paper we present a general approach to interactive, object-specific volumetric algorithms, based on a necessary condition for the best possible reconstruction to have been performed. The outlined approach can be applied to any class of objects. As an example of this approach, an interactive algorithm has been implemented for convex polyhedra.