pt2dvar: pt2dvar
In evoshape/deformetrics: R functions for shape analysis

Description Usage Arguments Value Author(s) Examples

This function performs the Levi Civita Riemannian Parallel Transport according to explicit formulation presented in Varano et al (accepted) [Varano V, Gabriele S, Teresi L, Dryden I L, Puddu P E, Torromeo C, Piras P. Accepted. The TPS Direct Transport: a new method for transporting deformations in the Size-and-shape Space. International Journal of Computer Vision].

1 2	pt2dvar(array, group, doopa = F, tol = 1e-06, CR = NULL, locs = NULL, sss = T)

`array`	array: a kx2xn array of shapes
`group`	character: group structure
`doopa`	logical: if TRUE Ordinary Procrustes Analysis is performed on the workhorse function. It is not supposed to be set at TRUE as OPAs are already performed outside that step.
`tol`	numeric: tolerance parameter for rotation
`CR`	matrix:common Reference to which intra-group deformations are applied
`locs`	array: Local references (one for each level of group) from which intra-group deformations are computed
`sss`	logical: if TRUE analyses are performed on the Size and Shape Space. It is not supposed to be set at FALSE

final list: a list with transported shapes

Paolo Piras

## Not run:  
data(case1)
groupsimul<-factor(rep(1:5,each=21))
plottraj(procSym(case1,CSinit=T)$PCscores[,1:2],group,asp=1)######## a classic approach does not recover the cycle
ptcase1<-pt2dvar(case1,groupsimul,CR=procSym(case1[,,firstsfac(groupsimul)],CSinit=F,reflect=F,scale=F,pcAlign=F)$mshape,locs=case1[,,firstsfac(groupsimul)])
plottraj(procSym(ptcase1$final,CSinit=T)$PCscores[,1:2],groupsimul,asp=1) ####  with this kind Parallel Transport the cycle is not perfectly recovered as with the new approach of Direct Transport (Varano et al [accepted]). 

## End(Not run)