median.SO3: Compute the projected or intrinsic median estimate of the...
In heike/rotations: Tools for working with rotational data

Description Usage Arguments Value References See Also

The median-type estimators are defined as

\widetilde{\bm{S}}=\argmin_{\bm{S}\in SO(3)}∑_{i=1}^nd_D(\bm{R}_i,\bm{S})

. If the choice of distance metrid, d_D, is Riemannian then the estimator is called the intrinsic, and if the distance metric in Euclidean then it projected. The algorithm used in the intrinsic case is discussed in Hartley et al. (2011) and the projected case was written by the authors.

1 2	median.SO3(Rs, type = "projected", epsilon = 1e-05, maxIter = 2000, na.rm = FALSE)

`Rs`	A n\times 9 matrix where each row corresponds to a random rotation in matrix form
`type`	choice of geometry: intrinsic or projected
`epsilon`	the stopping rule for the iterative process
`maxIter`	maximum number of iterations to allow
`na.rm`	how to handle NAs

the median

Hartley R, Aftab K and Trumpf J (2011). "L1 rotation averaging using the Weiszfeld algorithm." In _2011 IEEE Conference on Computer Vision and Pattern Recognition (CVPR)_, pp. 3041-3048. IEEE.

mean.SO3

heike/rotations documentation built on May 17, 2019, 3:24 p.m.