pPCA: calculate or modify a probablistic PCA based on...

In zarquon42b/RvtkStatismo: Integrating statismo and R using the vtkStandardMeshRepresenter

Description

calculate or modify a probablistic PCA based on 3D-coordinates

Usage

pPCA(x, align = FALSE, use.lm = NULL, deselect = FALSE, sigma = NULL,
  exVar = 1, scale = FALSE, representer = NULL)

UpdateModel(model, sigma = NULL, exVar = 1)

## S4 method for signature 'pPCA'
UpdateModel(model, sigma = NULL, exVar = 1)

Arguments

x	array of dimensions k x 3 x n, where k=number of coordinates and n=sample size - or a list containing triangular meshes.
align	logical: if TRUE, the data will be aligned first
use.lm	integer vector: specifies the indices of the points that are to be used in the constrained model
deselect	logical: if TRUE, use.lm references the missing coordinates instead of the present ones.
sigma	estimate of error variance (sensible is a value estimating coordinate error in terms of observer error)
exVar	numeric value with 0 < exVar <= 1 specifying the PCs to be included by their cumulative explained Variance
scale	logical: allow scaling in Procrustes fitting
representer	a triangular mesh, where the vertices correspond to the coordinates in array, leave NULL for pointclouds.
model	object of class pPCA

Value

returns a probabilistic PCA model as S4 class "pPCA" (see pPCA-class). UpdateModel is used to modify existing models by changing sigma and exVar.

References

Lüthi M, Albrecht T, Vetter T. 2009. Probabilistic modeling and visualization of the flexibility in morphable models. In: Mathematics of Surfaces XIII. Springer. p 251-264

Examples

require(Morpho)
data(boneData)
model <- pPCA(boneLM[,,])
## change parameters without recomputing Procrustes fit
model1 <- UpdateModel(model, sigma=1, exVar=0.8)

zarquon42b/RvtkStatismo documentation built on May 4, 2019, 9:09 p.m.