Name
Title | ||
|---|---|---|
Multiscale colour texture retrieval using the geodesic distance between multivariate generalized Gaussian models. |
Abstract | ||
|---|---|---|
This contribution concerns the retrieval of colour tex- ture. The interband correlation structure is considered by modeling the heavy-tailed image wavelet histograms with a multivariate generalized Gaussian. As a similar- ity measure we propose to use the Rao geodesic distance, which, in contrast to the Kullback-Leibler divergence, ex- ists in a closed form for any fixed value of the shape pa- rameter of the distribution. We apply this in several re- trieval experiments. The modeling of the interband cor- relation significantly increases retrieval rates, while the geodesic distance is shown to outperform the Kullback- Leibler divergence. A multivariate Laplace distribution yields better results than a Gaussian, indicating the po- tential of a model with variable shape parameter together with the geodesic distance. |
Year | DOI | Venue |
|---|---|---|
2008 | 10.1109/ICIP.2008.4711718 | ICIP |
Keywords | Field | DocType |
laplace transforms,laplace distribution,shape parameter,gaussian processes,heavy tail,image texture,shape,databases,generalized gaussian distribution,image retrieval,geodesic distance,correlation,indexing terms,kullback leibler divergence,gaussian distribution | Pattern recognition,Laplace distribution,Similarity measure,Image texture,Gaussian,Shape parameter,Artificial intelligence,Gaussian process,Geodesic,Kullback–Leibler divergence,Mathematics | Conference |
ISSN | ISBN | Citations |
1522-4880 E-ISBN : 978-1-4244-1764-3 | 978-1-4244-1764-3 | 23 |
PageRank | References | Authors |
1.58 | 3 | 3 |
Name | Order | Citations | PageRank |
|---|---|---|---|
| Geert Verdoolaege | 1 | 119 | 9.23 |
| Steve De Backer | 2 | 200 | 15.14 |
| Paul Scheunders | 3 | 1190 | 102.87 |