Abstract | ||
---|---|---|
We present a nonparametric probability density estimation model. The classical Parzen window approach builds a spherical Gaussian density around every input sample. Our method has a first stage where hard neighborhoods are determined for every sample. Then soft clusters are considered to merge the information coming from several hard neighborhoods. Our proposal estimates the local principal directions to yield a specific Gaussian mixture component for each soft cluster. This leads to outperform other proposals where local parameter selection is not allowed and/or there are no smoothing strategies, like the manifold Parzen windows. |
Year | DOI | Venue |
---|---|---|
2008 | 10.1016/j.patrec.2008.07.010 | Pattern Recognition Letters |
Keywords | DocType | Volume |
classical parzen window approach,soft cluster,hard neighborhood,nonparametric modeling,soft clustering,probability density estimation,manifold parzen windows,local parameter selection,parzen windows.,specific gaussian mixture component,nonparametric probability density function,nonparametric probability density estimation,spherical gaussian density,parzen window,hard neighbourhood,local principal direction,input sample,probability density function,probability density | Journal | 29 |
Issue | ISSN | ISBN |
16 | Pattern Recognition Letters | 3-540-74689-7 |
Citations | PageRank | References |
5 | 0.55 | 14 |
Authors | ||
4 |
Name | Order | Citations | PageRank |
---|---|---|---|
Ezequiel López-Rubio | 1 | 323 | 39.73 |
Juan Miguel Ortiz-de-lazcano-lobato | 2 | 68 | 11.59 |
Domingo López-Rodríguez | 3 | 55 | 9.24 |
María Del Carmen Vargas-gonzález | 4 | 8 | 2.90 |