Multi-Resolution Image Morphing

Manfred Kopp, Werner Purgathofer
Multi-Resolution Image Morphing
TR-186-2-97-20, October 1997 [ paper]
Content:

Information

Abstract

The proposed technique extends conventional image morphing algorithms by offering additional control over the processing of the individual multi-resolutions that are used in the wavelet representation when moving from the source to the destination image. This gives the animator the possibility to introduce a disparity between the morphing of global and detail features that can lead to interesting new effects that are not achievable with traditional methods. The calculation of an intermediate image in conventional image morphing algorithms is done in three steps: firstly the determination of feature correspondence and of an intermediate feature geometry, secondly the warping of the source and the destination images to intermediate images that match the feature geometry, and finally the blending of the two warped images. The proposed algorithm alters the third step by making the blending dependent on the multi-resolution levels using a new specialized 2D wavelet transformation.

Additional Files and Images

Additional files:
paper
paper




BibTeX

Download BibTeX-Entry
@techreport{Kopp-1997-MRM,
  title =      "Multi-Resolution Image Morphing",
  author =     "Manfred Kopp and Werner Purgathofer",
  year =       "1997",
  abstract =   "The proposed technique extends conventional image morphing
               algorithms by offering additional control over the
               processing of the individual multi-resolutions that are used
               in the wavelet representation when moving from the source to
               the destination image. This gives the animator the
               possibility to introduce a disparity between the morphing of
               global and detail features that can lead to interesting new
               effects that are not achievable with traditional methods.
               The calculation of an intermediate image in conventional
               image morphing algorithms is done in three steps: firstly
               the determination of feature correspondence and of an
               intermediate feature geometry, secondly the warping of the
               source and the destination images to intermediate images
               that match the feature geometry, and finally the blending of
               the two warped images. The proposed algorithm alters the
               third step by making the blending dependent on the
               multi-resolution levels using a new specialized 2D wavelet
               transformation.",
  address =    "Favoritenstrasse 9-11/186, A-1040 Vienna, Austria",
  institution = "Institute of Computer Graphics and Algorithms, Vienna
               University of Technology",
  note =       "human contact: technical-report@cg.tuwien.ac.at",
  month =      oct,
  number =     "TR-186-2-97-20",
  keywords =   "Animation, Image Morphing, Wavelets",
  URL =        "http://www.cg.tuwien.ac.at/research/publications/1997/Kopp-1997-MRM/",
}