Efficient visual representation and reconstruction from generalized curvature measures
by , ,
Abstract:
This paper considers how basic geometrical properties like curvature, rigidity, and possible embeddings can be related to efficient image encoding and the statistical concept of redundancy. In particular, the redundancy of planar and parabolic patches of images as surfaces is revealed by reconstructing the original image from curvature measures that are zero for non-elliptic regions. This approach also gives a new perspective on encoding principles in biological vision.
Download PDF
PDF URL: http://dx.doi.org/10.1117/12.49977
Reference:
Efficient visual representation and reconstruction from generalized curvature measures (Erhardt Barth, Terry M. Caelli, Christoph Zetzsche), In Geometric Methods in Computer Vision (Baba C. Vemuri, ed.), SPIE-Intl Soc Optical Eng, volume 1570, 1991.
Bibtex Entry:
@InProceedings{Barth1991,
  author    = {Erhardt Barth and Terry M. Caelli and Christoph Zetzsche},
  title     = {Efficient visual representation and reconstruction from generalized curvature measures},
  booktitle = {Geometric Methods in Computer Vision},
  year      = {1991},
  editor    = {Baba C. Vemuri},
  volume    = {1570},
  pages     = {86-95},
  month     = {sep},
  publisher = {{SPIE}-Intl Soc Optical Eng},
  abstract  = {This paper considers how basic geometrical properties like curvature, rigidity, and possible embeddings can be related to efficient image encoding and the statistical concept of redundancy. In particular, the redundancy of planar and parabolic patches of images as surfaces is revealed by reconstructing the original image from curvature measures that are zero for non-elliptic regions. This approach also gives a new perspective on encoding principles in biological vision.},
  doi       = {10.1117/12.49977},
  url       = {10.1117/12.49977">http://dx.doi.org/10.1117/12.49977},
}