Efficient visual representation and reconstruction from generalised curvature measures

Erhardt Barth, Terry Caelli, and Christoph Zetzsche
ABSTRACT: In this paper we consider how some 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 which are zero for non-elliptic regions. Our approach also gives a new perspective on encoding principles in biological vision.

(pdf  0.25 MB) (PostScript 8 MB)