A statistical learning/inference framework for color demosaicing is presented. We start with simplistic assumptions about color constancy, and recast color demosaicing as a blind linear inverse problem: color parameterizes the unknown kernel, while brightness takes on the role of a latent variable. An expectationmaximization algorithm naturally suggests itself for the estimation of them both. T...

In this project1, we first study the Gaussian-based hidden Markov random field (HMRF) model and its expectationmaximization (EM) algorithm. Then we generalize it to Gaussian mixture model-based hidden Markov random field. The algorithm is implemented in MATLAB. We also apply this algorithm to color image segmentation problems and 3D volume segmentation problems.

We investigate two approaches to automatic discrimination of similar languages: Expectationmaximization algorithm for estimating conditional probability P (word|language) and a series of byte level language models. The accuracy of these methods reached 86.6 % and 88.3 %, respectively, on set A of the DSL Shared task 2016 competition.

We investigate the application of expectationmaximization (EM) algorithms to the classical problem of multiple target tracking (MTT) for a known number of targets. Conventional algorithms, which deal with this problem, have a computational complexity that depends exponentially on the number of targets, and usually divide the problem into a localization stage and a tracking stage. The new algori...

This paper applies a new clustering approach for identifying and segmenting motion in image sequences. We estimate a matrix whose entries represent similarity probabilities between local motion estimates. We adopt a two step iterative algorithm which consists of a variant of the expectationmaximization algorithm for segmenting regions with similar motion. The proposed algorithm updates cluster ...

We propose a detailed model of evolution of exon-intron structure of eukaryotic genes that takes into account gene-specific intron gain and loss rates, branch-specific gain and loss coefficients, invariant sites incapable of intron gain, and rate variability of both gain and loss which is gamma-distributed across sites. We develop an expectationmaximization algorithm to estimate the parameters ...

When using Black-Scholes formula to price options, the key is the estimation of the stochastic return variance. In this paper we discuss an approach based on Bayes filters which combines the GARCH model and the implied volatilities. Empirical experiments demonstrate the better pricing accuracy of this approach. Furthermore, we show that we can re-estimate the parameters of the dynamics system u...

We prove a closed-form solution to second-order Tensor Voting (TV), and employ the resulting structure-aware tensors in ExpectationMaximization (EM). Our new algorithm, aptly called EM-TV, is an efficient and robust method for parameter estimation. Quantitative comparison shows that our method performs better than the conventional second-order TV and other representative techniques in parameter...

This paper discusses the supervised learning of morphology using stochastic transducers, trained using the ExpectationMaximization (EM) algorithm. Two approaches are presented: first, using the transducers directly to model the process, and secondly using them to define a similarity measure, related to the Fisher kernel method (Jaakkola and Haussler, 1998), and then using a Memory-Based Learnin...

Speech recognition performance degrades significantly when there is a mismatch between testing and training conditions. Linear transformation-based maximum-likelihood (ML) techniques have been proposed recently to tackle this problem. In this paper, we extend this approach to use nonlinear transformations. These are implemented by multilayer perceptrons (MLPs) which transform the Gaussian means...