This paper presents a Bayesian framework for multi-cue 3D object tracking of deformable objects. The proposed spatio-temporal object representation involves a set of distinct linear subspace models or Dynamic Point Distribution Models (DPDMs), which can deal with both continuous and discontinuous appearance changes; the representation is learned fully automatically from training data. The representation is enriched with texture information by means of intensity histograms, which are compared using the Bhattacharyya coe#cient. Direct 3D measurement is furthermore provided by a stereo system.

... without addressing the so-called measurement-origin uncertainty. Part I and Part II deal with target motion models. Part III covers measurement models and associated techniques. Part IV is concerned with tracking techniques that are based on decisions regarding target maneuvers. This part surveys the multiple-model methods---the use of multiple models (and filters) simultaneously---which is the prevailing approach to maneuvering target tracking in the recent years. The survey is presented in a structured way, centered around three generations of algorithms: autonomous, cooperating, and variable structure. It emphasizes on the underpinning of each algorithm and covers various issues in algorithm design, application, and performance.

Abstract. We investigate the robustness of nonlinear filtering for continuous time finite state Markov chains, observed in white noise, with respect to misspecification of the model parameters. It is shown that the distance between the optimal filter and that with incorrect model parameters converges to zero uniformly over the infinite time interval as the misspecified model converges to the true model, provided the signal obeys a mixing condition. The filtering error is controlled through the exponential decay of the derivative of the nonlinear filter with respect to its initial condition. We allow simultaneously for misspecification of the initial condition, of the transition rates of the signal, and of the observation function. The first two cases are treated by relatively elementary means, while the latter case requires the use of Skorokhod integrals and tools of anticipative stochastic calculus. 1.

The concept of rainflow cycles is often used in fatigue of materials for analysing load processes, which in most realistic cases should be modelled stochastically. Methods are developed for computation of the rainflow matrix for random loads that are changing properties over time due to changes of the system dynamics. For a random vehicle load the change of properties could reflect dioeerent driving conditions. Mathematically, the random load is modelled by a switching process with Markov regime, i.e. the random load changes properties according to a hidden (not observed) Markov chain. An algorithm is developed for a switching process where each part of the load is modelled by a Markov chain. As only the local extremes are of importance for rainflow analysis, another approach is to model the sequence of turning points by a Markov chain. The main result of this paper is an algorithm for computation of the rainflow matrix for a switching process where each part is described by a Markov c...

Motivated by Hubert's segmentation procedure [16, 17], we discuss the application of hidden Markov models (HMM) to the segmentation of hydrological and enviromental time series. We use a HMM algorithm which segments time series of several hundred terms in a few seconds and is computationally feasible for even longer time series. The segmentation algorithm computes the Maximum Likelihood segmentation by use of an expectation / maximization iteration. We rigorously prove algorithm convergence and use numerical experiments, involving temperature and river discharge time series, to show that the algorithm usually converges to the globally optimal segmentation. The relation of the proposed algorithm to Hubert's segmentation procedure is also discussed.

Abstract not found

We consider the process dYt = ut dt + dWt, where u is a process not necessarily adapted to F Y (the filtration generated by the process Y) and W is a Brownian motion. We obtain a general representation for the likelihood ratio of the law of the Y process relative to Brownian measure. This representation involves only one basic filter (expectation of u conditional on observed process Y). This generalizes the result of Kailath and Zakai [Ann. Math. Statist. 42 (1971) 130–140] where it is assumed that the process u is adapted to F Y. In particular, we consider the model in which u is a functional of Y and of a random element X which is independent of the Brownian motion W. For example, X could be a diffusion or a Markov chain. This result can be applied to the estimation of an unknown multidimensional parameter θ appearing in the dynamics of the process u based on continuous observation of Y on the time interval [0,T]. For a specific hidden diffusion financial model in which u is an unobserved mean-reverting diffusion, we give an explicit form for the likelihood function of θ. For this model we also develop a computationally explicit E–M algorithm for the estimation of θ. In contrast to the likelihood ratio, the algorithm involves evaluation of a number of filtered integrals in addition to the basic filter. 1. Introduction. Let (�,F,P), {Ft,t ≤ T

We have developed a method to partition a set of data into clusters by use of Hidden Markov Models. Given a number of clusters, each of which is represented by one Hidden Markov Model, an iterative procedure finds the combination of cluster models and an assignment of data points to cluster models which maximizes the joint likelihood of the clustering.

This research aims at creating a framework to analyze the performance of iterative algorithms in distributed environments. The parallelization of certain iterative algorithms is indeed a crucial issue for the efficient solution of large or complex optimization problems. Diverse implementation techniques for such parallelizations have become popular. They are examined here with a view to understanding their impact on the algorithm behavior in a distributed environment. Several theoretical results concerning the sufficient conditions for, and speed of, convergence for parallel iterative algorithms are available. However, there is a gap between those results and what is relevant to the user at the application level. In particular, an estimate of the algorithm execution time is often desirable. The performance characterization presented in this dissertation follows a stochastic approach partially based on a Markov process. It addresses different characteristics of the algorithmic execution...

Cataloguing-in-Publication entry