Abstract

This paper describes the steps involved in generating a surface which interpolates a given set of input points. The approach presented departs from those reported in literature in two significant ways. The parametric values are assigned independently and not averaged over the row/column. This helps to localize changes during error optimization. The measure used for the interpolation error is the shortest distance between the input points and the surface and not the distance between the input point and its corresponding point as used in literature. This it is believed, is more intuitive physically and has been shown to result in a better interpolation.

This content is only available via PDF.
You do not currently have access to this content.