Analytic evaluation of the FCC Voronoi-splines (Q6141541)

This interesting paper deals with the subject of analytic evaluation of splines, more specifically FCC Voronoi-Splines. Splines are widely used in approximation theory, computer graphics, vision, tomography, Magnetic Resonance Imaging (MRI), just to name a few areas. The paper under review centers on a specific class of splines called Voroni-splines (V-splines). For example, it is known that V-splines can be defined with arbitrary approximation order on any lattice with larger support and higher degree than comparable box-splines. The authors propose an analytic evaluation algorithm for splines and their derivatives (gradient and Hessian) of FCC volume datasets based on the V-splines of order two and three as well as various other interesting results. For example, they analyze the V-splines to show their wide improvement of performance on various GPUs. The paper is well written with a good set of references.