LOESS: Multivariate smoothing by moving least squares. LOESS is a method for smoothing scattered data in one or more variables by local fitting of linear or quadratic models with a neighborhood weighting that is moved across the domain. A program is available to accelerate the basic method by building a k-d tree, fitting only at vertices, and extending by cubic Hermite blending.
Keywords for this software
References in zbMATH (referenced in 2 articles , 1 standard article )
Showing results 1 to 2 of 2.
- Sales-Mayor, F.; Wyatt, R.E.: A two-stage filter for smoothing multivariate noisy data on unstructured grids (2004)
- Grosse, Eric: LOESS: Multivariate smoothing by moving least squares (1989)