In the past 15 years scattered interpolation has been applied to nearly every area in computer graphics. These areas include morphing, skinning, facial animation, meshing, terrain generation, fluids, computational photography, rendering, and more. Despite this broad usage the relevant techniques are not generally covered in textbooks. This course fills the gap. It describes all of the major algorithms, several necessary background concepts, and (where possible) the relationships between approaches. Each technique also includes examples and references to how it has been applied in graphics. The techniques covered include kernel regression, Shepard’s interpolation and moving least squares, Gaussian processes, harmonic splines, and radial basis function (RBF) interpolation. The course presentation assumes knowledge of linear algebra, as will be familiar to computer graphics programmers and technical directors. We have developed novel explanations of concepts such as Greens functions using only linear algebra concepts. One ap- pendix introduces a functional analysis view of interpolation and requires calculus.
Scattered Data Interpolation for Computer Graphics
Proceeding SIGGRAPH '14 ACM SIGGRAPH 2014 Courses Article No. 27