Advanced Search

Evaluation of sets of oriented and non-oriented receptive fields as local descriptors

Research and Teaching Output of the MIT Community

Show simple item record Yokono, Jerry Jun en_US Poggio, Tomaso en_US 2004-10-20T21:05:24Z 2004-10-20T21:05:24Z 2004-03-24 en_US
dc.identifier.other AIM-2004-007 en_US
dc.identifier.other CBCL-237 en_US
dc.description.abstract Local descriptors are increasingly used for the task of object recognition because of their perceived robustness with respect to occlusions and to global geometrical deformations. We propose a performance criterion for a local descriptor based on the tradeoff between selectivity and invariance. In this paper, we evaluate several local descriptors with respect to selectivity and invariance. The descriptors that we evaluated are Gaussian derivatives up to the third order, gray image patches, and Laplacian-based descriptors with either three scales or one scale filters. We compare selectivity and invariance to several affine changes such as rotation, scale, brightness, and viewpoint. Comparisons have been made keeping the dimensionality of the descriptors roughly constant. The overall results indicate a good performance by the descriptor based on a set of oriented Gaussian filters. It is interesting that oriented receptive fields similar to the Gaussian derivatives as well as receptive fields similar to the Laplacian are found in primate visual cortex. en_US
dc.format.extent 20 p. en_US
dc.format.extent 3426196 bytes
dc.format.extent 1925439 bytes
dc.format.mimetype application/postscript
dc.format.mimetype application/pdf
dc.language.iso en_US
dc.relation.ispartofseries AIM-2004-007 en_US
dc.relation.ispartofseries CBCL-237 en_US
dc.subject AI en_US
dc.subject local descriptor en_US
dc.subject steerable filter en_US
dc.subject Gaussian derivatives en_US
dc.subject selectivity en_US
dc.subject invariance en_US
dc.title Evaluation of sets of oriented and non-oriented receptive fields as local descriptors en_US

Files in this item

Name Size Format Description 3.267Mb Postscript
AIM-2004-007.pdf 1.836Mb PDF

This item appears in the following Collection(s)

Show simple item record