Principal Warps: Thin-Plate Splines and the Decomposition of Deformations

@article{Bookstein1989PrincipalWT,
  title={Principal Warps: Thin-Plate Splines and the Decomposition of Deformations},
  author={Fred L. Bookstein},
  journal={IEEE Trans. Pattern Anal. Mach. Intell.},
  year={1989},
  volume={11},
  pages={567-585}
}
  • F. Bookstein
  • Published 1 June 1989
  • Engineering
  • IEEE Trans. Pattern Anal. Mach. Intell.
The decomposition of deformations by principal warps is demonstrated. The method is extended to deal with curving edges between landmarks. This formulation is related to other applications of splines current in computer vision. How they might aid in the extraction of features for analysis, comparison, and diagnosis of biological and medical images in indicated. > 

Figures from this paper

Edgels and Tangent Planes in Image Warping

TLDR
An extension to the edgel method of Bookstein and Green (1993a,b) is outlined for use in three dimensions and an example of the results of using this formulation for three-dimensional magnetic resonance images of the human head is given.

Image Warping Using Few Anchor Points and Radial Functions

TLDR
This work decomposes transformations based on radial basis junctions into linear and radial terms, and shows examples where such a decomposition is advantageous.

Image warping using radial basis functions

TLDR
This paper will analyze interpolation by TPS, experiment with other radial basis functions, and suggest two alternative functions that provide better results.

Multiple Landmark Warping Using Thin-plate Splines

TLDR
An image warping algorithm using thin-plate splines that allows multilandmarks being specified and warped simultaneously simultaneously and QR decomposition techniques are used to stabilize the solution.

Groupwise Non-rigid Registration Using Polyharmonic Clamped-Plate Splines

TLDR
A novel groupwise data-driven algorithm for non-rigid registration that automatically constructs a low-dimensional, common representation of the warp fields to enable the analysis of groups of registered images.

Tracking Points on Deformable Objects Using Curvature Information

TLDR
A significant improvement is presented to the approach of Duncan et al. to analyze the deformations of curves in sequences of 2D images based on the paradigm that high curvature points usually possess an anatomical meaning, and are therefore good landmarks to guide the matching process.

Generalized Thin-Plate SplineWarps

TLDR
Three types of warps based on the thin-plate spline are proposed, including a flexible rigid warp that describes the optic flow field induced by a smooth rigid surface, and satisfies the affine epipolar geometry constraint.

Generalized Thin-Plate Spline Warps

TLDR
Three types of warps based on the Thin-Plate Spline are proposed, including a rigid flexible warp that describes the optic flow field induced by a smooth and rigid surface, and satisfies the affine epipolar geometry constraint.

Landmark methods for forms without landmarks: localizing group differences in outline shape

  • F. Bookstein
  • Psychology
    Proceedings of the Workshop on Mathematical Methods in Biomedical Image Analysis
  • 1996
Procrustes superposition and the thin-plate spline each principally developed within the context of discrete landmark data, can be combined in a novel adaptive filter for detecting localized group
...

References

SHOWING 1-10 OF 25 REFERENCES

Surface Spline Interpolation: Basic Theory and Computational Aspects

One of the most efficient methods to date for global interpolation of scattered data has come to be called “surface spline interpolation”. It turns out that the underlying mathematical theory has for

E. Morphometrics for Functional Imaging Studies

TLDR
Morphometries studies the empirical geometry of homology-variation in the relative locations of sets of homologous points over a sample of forms by extensive algebraic manipulation of the landmark coordinates to remove all traces of the Cartesian system in which the forms were digested.

Toward a Notion of Feature Extraction for Plane Mappings

My concern is the delineation of features for plane mappings as they express the relation of two biological images. Perhaps one picture represents a statistical norm for an organ, the other the form

Optimal registration of deformed images

Motivated by the need to locate and identify objects in three dimensional CT images, an optimal registration method for matching two and three dimensional deformed images has been developed. This

Multilevel computational processes for visual surface reconstruction

Regularization of Inverse Visual Problems Involving Discontinuities

  • Demetri Terzopoulos
  • Mathematics
    IEEE Transactions on Pattern Analysis and Machine Intelligence
  • 1986
TLDR
This paper proposes a general class of controlled-continuity stabilizers which provide the necessary control over smoothness in visual reconstruction problems that involve both continuous regions and discontinuities, for which global smoothness constraints fail.

Scattered data interpolation: tests of some methods

Absract. This paper is concerned with the evaluation of methods for scattered data interpolation and some of the results of the tests when applied to a number of methods. The process involves

A computerized system for the elastic matching of deformed radiographic images to idealized atlas images.

A system of computer programs is described which, for the first time, is able to use computerized tomographic data to automatically locate, measure, and describe anatomical structures of interest

The Measurement of Biological Shape and Shape Change.

TLDR
This chapter discusses the study of Shape Transformation after D'Arcy Thompson, which involves two types of transformations: functional craniology and craniometrics and multivariate morphometrics.

Multivariate interpolation at arbitrary points made simple

The concrete method of ‘surface spline interpolation’ is closely connected with the classical problem of minimizing a Sobolev seminorm under interpolatory constraints; the intrinsic structure of