Feature displacement interpolation

Given a sparse set of feature matches, we want to compute an interpolated dense displacement map. The application may be stereo disparity computation, flowcomputation, or non-rigid medical registration. Also estimation of missing image data, may be phrased in this framework. Since the features often are verysparse, the interpolation model becomes crucial. We show that a maximum likelihood estimation based on the covariance properties (Kriging) show propertiesmore expedient than methods such as Gaussian interpolation or Tikhonov regularizations, also including scale-selection. The computational complexitiesare identical. We apply the maximum likelihood interpolation to growth analysis of the mandibular bone. Here, the features used are the crest-lines of the objectsurface.