A Caltech Library Service

Deformable kernels for early vision

Perona, Pietro (1991) Deformable kernels for early vision. In: 1991 IEEE Computer Society Conference on Computer Vision Patern Recognition (CVPR). IEEE Computer Society Press , Los Alamitos, CA, pp. 222-227. ISBN 0-8186-2148-6.

PDF - Published Version
See Usage Policy.


Use this Persistent URL to link to this item:


Early vision algorithms often have a first stage of linear filtering that 'extracts' from the image information at multiple scales of resolution and multiple orientations. A common difficulty in the design and implementation of such schemes is that one feels compelled to discretize coarsely the space of scales and orientations in order to reduce computation and storage costs. This discretization produces anisotropies due to a loss of traslation-, rotation- scaling- invariance that makes early vision algorithms less precise and more difficult to design. This need not be so: one can compute and store efficiently the response of families of linear filters defined on a continuum of orientations and scales. A technique is presented that allows (1) to compute the best approximation of a given family using linear combinations of a small number of 'basis' functions; (2) to describe all finite-dimensional families, i.e. the families of filters for which a finite dimensional representation is possible with no error. The technique is general and can be applied to generating filters in arbitrary dimensions. Experimental results are presented that demonstrate the applicability of the technique to generating multi-orientation multi-scale 20 edge-detection kernels. The implementation issues are also discussed.

Item Type:Book Section
Related URLs:
URLURL TypeDescription
Perona, Pietro0000-0002-7583-5809
Additional Information:© 1991 IEEE. Date of Current Version: 06 August 2002. I am grateful to Massimo Porrati, Alberto Grunbaum, David Donoho, Charles Desoer and Richard Murray for helpful conversations. The simulations in this work have been carried out using Paul Kube's "viz" image-manipulation package, Paul also gave allot of assistance. Part of this work was carried out at the International Computer Science Institute at Berkeley. This work was partially sponsored by U.S. Army Research Office grant number DAAL 03-86-K-0171 (Center for Intelligent Control Systems).
Funding AgencyGrant Number
Army Research Office (ARO)DAAL 03-86-K-0171
Other Numbering System:
Other Numbering System NameOther Numbering System ID
INSPEC Accession Number4132429
Record Number:CaltechAUTHORS:20120420-071031778
Persistent URL:
Official Citation:Perona, P.; , "Deformable kernels for early vision," Computer Vision and Pattern Recognition, 1991. Proceedings CVPR '91., IEEE Computer Society Conference on , vol., no., pp.222-227, 3-6 Jun 1991 doi: 10.1109/CVPR.1991.139691 URL:
Usage Policy:No commercial reproduction, distribution, display or performance rights in this work are provided.
ID Code:30208
Deposited By: Ruth Sustaita
Deposited On:20 Apr 2012 16:15
Last Modified:09 Nov 2021 19:39

Repository Staff Only: item control page