Adaptive surface extraction from anisotropic volumetric data: contouring on generalized octrees

In this article, we present an algorithm to extract adaptive surfaces from anisotropic volumetric data. For example, this kind of data can be obtained from a set of segmented images from the sampling of an implicit function, or it can be built by using depth images produced by time-of-flight cameras. However, for many applications as geometry modeling, rendering, or finite elements, it is better to use an explicit surface representation. This surface must fit to the geometrical and topological features of the object in order to obtain a good approximation and to avoid topological artifacts. Our algorithm is able to extract adaptive surfaces that accurately approximate the geometry of the original object while minimizing aliasing effects. In addition, our solution is suitable to handle the anisotropy of volumetric representations. In comparison with relevant methods in the state of the art, ours offers a good compromise between mesh quality and precision in the geometrical approximation.