Feature displacement interpolation
- 27 November 2002
- proceedings article
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- Vol. 3, 208-212
- https://doi.org/10.1109/icip.1998.999011
Abstract
Given a sparse set of feature matches, we want tocompute an interpolated dense displacement map. Theapplication may be stereo disparity computation, flowcomputation, or non-rigid medical registration. Alsoestimation of missing image data, may be phrasedin this framework. Since the features often are verysparse, the interpolation model becomes crucial. Weshow that a maximum likelihood estimation based onthe covariance properties (Kriging) show propertiesmore expedient than methods such...Keywords
This publication has 7 references indexed in Scilit:
- Normalized and differential convolutionPublished by Institute of Electrical and Electronics Engineers (IEEE) ,2002
- 4D shape-preserving modelling of bone growthPublished by Springer Nature ,1998
- Regression with Gaussian ProcessesPublished by Springer Nature ,1997
- Regularization, Scale-Space, and Edge Detection FiltersJournal of Mathematical Imaging and Vision, 1997
- Visual ReconstructionPublished by MIT Press ,1987
- From Images to SurfacesPublished by MIT Press ,1981
- Fractional Brownian Motions, Fractional Noises and ApplicationsSIAM Review, 1968