The goal of scattered data interpolation is to construct a (typically smooth) function from a set of unorganized samples. These techniques have a wide range of applications in computer graphics. For instance they can be used to model a surface from a set of sparse surface samples, to reconstruct a BRDF from a set of measurements, to interpolate between keyframes, or to compute the physical properties of a fluid. The course will offer a best practice guide to the field. We will review the diverse techniques, illustrate them with examples drawn from the computer graphics literature, and contrast them. We will examine stability and computational properties with an eye on real-time applications. Although the course provides mostly an engineering perspective, to provide a deeper understanding of the techniques we will cover some of the underlying theory.