Abstract
Recently, methods based on local image features have shown promise for texture and object recognition tasks. This paper presents a large-scale evaluation of an approach that represents images as distributions (signatures or histograms) of features extracted from a sparse set of keypoint locations and learns a Support Vector Machine classifier with kernels based on two effective measures for comparing distributions, the Earth Mover’s Distance and the ÷2 distance. We first evaluate the performance of our approach with different keypoint detectors and descriptors, as well as different kernels and classifiers. We then conduct a comparative evaluation with several state-of-the-art recognition methods on 4 texture and 5 object databases. On most of these databases, our implementation exceeds the best reported results and achieves comparable performance on the rest. Finally, we investigate the influence of background correlations on recognition performance.
Original language | English |
---|---|
Title of host publication | Conference on Computer Vision and Pattern Recognition Workshop, 2006 |
Subtitle of host publication | (CVPRW'06) |
Editors | Cordelia Schmid, Stefano Soatto, Carlo Tomasi |
Publisher | IEEE |
Pages | 93-100 |
Number of pages | 8 |
ISBN (Print) | 0769526462 |
DOIs | |
Publication status | Published - 2006 |
Event | IEEE Computer Society, 2006 Conference on Computer Vision and Pattern Recognition (CVPRW'06) - New York, United States Duration: 17 Jun 2006 → 22 Jun 2006 |
Conference
Conference | IEEE Computer Society, 2006 Conference on Computer Vision and Pattern Recognition (CVPRW'06) |
---|---|
Country/Territory | United States |
City | New York |
Period | 17/06/06 → 22/06/06 |