Definition A Gaussian Mixture Model (GMM) is a parametric probability density function represented as a weighted sum of Gaussian component densities. GMMs are commonly used as a parametric model of the probability distribution of continuous measurements or features in a biometric system, such as vocal-tract related spectral features in a speaker recognition system. GMM parameters are estimated ...

A novel variational algorithm is developed for video inpainting. Within a Bayesian framework, using standard maximum a posteriori to variational formulation rationale, we derive a minimum energy formulation for the estimation of a reconstructed sequence as well as motion recovery. From the EulerLagrange Equations, we propose a full multiresolution algorithm in order to compute a good local mini...

We develop in this paper a generic Bayesian framework for the joint estimation of motion and recovery of missing data in a damaged video sequence. Using standard maximum a posteriori to variational formulation rationale, we derive generic minimum energy formulations for the estimation of a reconstructed sequence as well as motion recovery. We instantiate these energy formulations and from their...

A simultaneous multi-frame super-resolution video reconstruction procedure, utilizing spatio-temporal smoothness constraints and motion estimator confidence parameters is proposed. The ill-posed inverse problem of reconstructing super-resolved imagery from the low resolution, degraded observations is formulated as a statistical inference problem and a Bayesian, maximum a-posteriori (MAP) approa...

A simultaneous multi-frame super-resolution video reconstruction procedure, utilizing spatio-temporal smoothness constraints and motion estimator confidence parameters is proposed. The ill-posed inverse problem of reconstructing super-resolved imagery from the low resolution, degraded observations is formulated as a statistical inference problem and a Bayesian, maximum a-posteriori (MAP) approa...

Bayesian estimation of the tail index of a heavy-tailed distribution is addressed when data are randomly right-censored. Maximum a posteriori and mean posterior estimators are constructed for various prior distributions of the tail index and their consistency and asymptotic normality are established. Finitesample properties of the proposed estimators are investigated via simulations. Tail index...

Arterial Spin Labeling (ASL) techniques potentially allow the absolute, non-invasive quantification of brain perfusion. This can be achieved either by fitting a kinetic model to the data acquired at a number of inversion times (TI) or by applying the model to a single TI assuming values for all other parameters. The accuracy of the model estimation strongly depends on the distribution of the TI...