Recognition with local features: The kernel recipe

Christian Wallraven, Barbara Caputo, Arnulf Graf

Research output: Chapter in Book/Report/Conference proceedingConference contribution

257 Citations (Scopus)

Abstract

Recent developments in computer vision have shown thai local features can provide efficient representations suitable for robust object recognition. Support Vector Machines have been established as powerful learning algorithms with good generalization capabilities. In this paper, we combine these two approaches and propose a general kernel method for recognition with local features. We show that the proposed kernel satisfies the Mercer condition and that it is suitable for many established local feature frameworks. Large-scale recognition results are presented on three different databases, which demonstrate that SVMs with the proposed kernel perform better than standard matching techniques on local features. In addition, experiments on noisy and occluded images show that local feature representations significantly outperform global approaches.

Original languageEnglish
Title of host publicationProceedings of the IEEE International Conference on Computer Vision
Pages257-264
Number of pages8
Volume1
Publication statusPublished - 2003 Dec 2
Externally publishedYes
EventProceedings: Ninth IEEE International Conference on Computer Vision - Nice, France
Duration: 2003 Oct 132003 Oct 16

Other

OtherProceedings: Ninth IEEE International Conference on Computer Vision
CountryFrance
CityNice
Period03/10/1303/10/16

    Fingerprint

ASJC Scopus subject areas

  • Software
  • Computer Vision and Pattern Recognition

Cite this

Wallraven, C., Caputo, B., & Graf, A. (2003). Recognition with local features: The kernel recipe. In Proceedings of the IEEE International Conference on Computer Vision (Vol. 1, pp. 257-264)