UNIGE document Chapitre d'actes
previous document  unige:47706  next document
add to browser collection
Title

Denoising with infinite mixture of Gaussians

Authors
Published in Proceedings of the 13th European Signal Processing Conference, EUSIPCO 2005. Antalya (Turkey) - Sep 4-8 - . 2005, p. 1-4
Abstract We show in this paper how an Infinite Mixture of Gaussians model can be used to estimate/denoise non-Gaussian data with local linear estimators based on the Wiener filter. The decomposition of the data in Gaussian components is straightforwardly computed with the Gaussian Transform, previously derived in [2]. The estimation is based on a two-step procedure, the first step consisting in variance estimation, and the second step in data estimation through Wiener filtering. We propose new generic variance estimators based on the Infinite Gaussian Mixture prior such as the cumulative estimator or the local-global estimator, as well as more classical Bayesian estimators. Results are presented in terms of distortion for the case of Generalized Gaussian data.
Keywords EquationsEstimationGaussian distributionTransformsVectorsWiener filters
Full text
Structures
Research groups Computer Vision and Multimedia Laboratory
Multimodal Interaction Group
Stochastic Information Processing Group
Citation
(ISO format)
ALECU, Teodor, VOLOSHYNOVSKYY, Svyatoslav, PUN, Thierry. Denoising with infinite mixture of Gaussians. In: Proceedings of the 13th European Signal Processing Conference, EUSIPCO 2005. Antalya (Turkey). [s.l.] : [s.n.], 2005. p. 1-4. https://archive-ouverte.unige.ch/unige:47706

85 hits

26 downloads

Update

Deposited on : 2015-03-06

Export document
Format :
Citation style :