“…That is, given a face of the triangulation with vertices (x i , y i ), (x j , y j ), and (x k , y k ), and the corresponding sample values z i = f (x i , y i ), z j = f (x j , y j ), and z k = f (x k , y k ), we form the unique planar interpolant passing through the points (x i , y i , z i ), (x j , y j , z j ), and (x k , y k , z k ). In step 2, the Delaunay triangulation [30] is employed, due largely to its good properties for approximation. In many applications, it is desirable that the triangulation of S be uniquely determined by S alone, as this avoids the need for additional side information during the triangulation process.…”