gradientWd {LOMAR}

gradientWd

Description

Gradient of the objective function with respect to rotation and translation parameters

Usage

gradientWd(Tr, X, Y, CX, CY, w1 = NULL, w2 = NULL, S = NULL)

Arguments

Tr

Transformation vector as translation vector + rotation (angle in 2d, quaternion in 3d))

X

matrix of means of first GMM (i.e. reference point set)

Y

matrix of means of second GMM (i.e. moving point set)

CX

array of covariance matrices of first GMM such that X[i,] has covariance matrix C1[,,i]

CY

array of covariance matrices of second GMM such that Y[i,] has covariance matrix C2[,,i]

w1

(optional) vector of mixture weights of first GMM.

w2

(optional) vector of mixture weights of second GMM.

S

(optional) array of pre-computed sqrtm(sqrtm(CX[,,i]) %*% CY[,,j] %*% sqrtm(CX[,,i]))

Value

gradient vector