Skip to content

← ArchivePaper1986

Free-Form Deformation of Solid Geometric Models

Thomas W. Sederberg, Scott R. Parry

SIGGRAPHAcademic3352 cites52 descendantsSkinning

Introduces FFD lattices: deforming models by warping the space around them, a foundation for countless rig deformers.

Abstract

Presents free-form deformation (FFD), a technique for deforming solid geometric models using trivariate tensor product Bernstein polynomials. The method enables intuitive sculpting-like manipulation of objects embedded in a flexible lattice of control points, supporting local or global deformations while maintaining derivative continuity and optionally preserving volume.

How to read this

Category
Method: a space-warping deformation technique (free-form deformation)
Contributions
  • Free-form deformation (FFD) that deforms solid models by warping the space around them via a lattice of control points
  • Use of trivariate tensor-product Bernstein polynomials for intuitive, sculpting-like manipulation
  • Support for local or global deformations with derivative continuity and optional volume preservation
Context
Builds on the object-deformation operators of Barr and the lattice-embedding idea of Bezier, generalizing both into arbitrary free-form warps of the embedding space; it in turn underpins countless later rig deformers and free-form modeling tools.Builds on: Global and Local Deformations of Solid Primitives · General Distortion of an Ensemble of Biparametric Surfaces
Correctness
The method is general and mathematically grounded in Bernstein polynomial lattices; a reader should note the deformation quality depends on lattice resolution and placement, and that it warps embedding space rather than respecting surface or anatomical structure.
Clarity
Accessible at the concept level but math-dense in the formulation; a first pass conveys the lattice idea, a second pass is needed for the polynomial machinery.
How to read it
Read for the core idea that you deform the space, not the mesh: focus on the lattice and control-point setup; a second pass on the Bernstein-polynomial trivariate formulation pays off if you will implement it.

Builds on

Built upon by

Related work

Keywords

This page summarises the entry and links to its original source. The archive never hosts or redistributes the publication itself.Show it in the full archive list →