GitHub - heikowagner/Nonparametric-Registration-to-Low-Dimensional-Function-Spaces: Registration aims to decompose amplitude and phase variation of samples of curves. Phase variation is captured by warping functions which monotonically trans- form the domains. Resulting registered curves should then only exhibit amplitude variation. Most existing methods assume that all sample functions exhibit a typical sequence of shape features like peaks or valleys, and registration focuses on aligning these features. In this paper we adopt a more general point of view which goes be- yond feature alignment. We present a registration method where warping functions are defined in such a way that the resulting registered curves span a low dimensional linear function space. The approach may be used as a tool for analyzing any type of functional data satisfying a structural regularity condition called “bounded shape variation”. Problems of identifiability are discussed in detail, and connections to es- tablished registration procedures are analyzed. The method is applied to real and simulated data.

Branches Tags

Name		Name	Last commit message	Last commit date
Latest commit History 1 Commit
base		base
simulations		simulations
Figure_1_plus_kmean_comparison.R		Figure_1_plus_kmean_comparison.R
Figure_2_K1_K2_comparison.r		Figure_2_K1_K2_comparison.r
figure2c.r		figure2c.r
simulation5.r		simulation5.r

About

Registration aims to decompose amplitude and phase variation of samples of curves. Phase variation is captured by warping functions which monotonically trans- form the domains. Resulting registered curves should then only exhibit amplitude variation. Most existing methods assume that all sample functions exhibit a typical sequence of shape featu…