Understanding the Behavior of Belief Propagation

Probabilistic graphical models are a powerful concept for modeling high-dimensional distributions. Besides modeling distributions, probabilistic graphical models also provide an elegant framework for performing statistical inference; because of the high-dimensional nature, however, one must often use approximate methods for this purpose. Belief propagation performs approximate inference, is efficient, and looks back on a long success-story. Yet, in most cases, belief propagation lacks any performance and convergence guarantees. Many realistic problems are presented by graphical models with loops, however, in which case belief propagation is neither guaranteed to provide accurate estimates nor that it converges at all. This thesis investigates how the model parameters influence the performance of belief propagation. We are particularly interested in their influence on (i) the number of fixed points, (ii) the convergence properties, and (iii) the approximation quality. For this purpose, we take a different perspective ...

Christian Knoll — Graz University of Technology


Three dimensional shape modeling: segmentation, reconstruction and registration

Accounting for uncertainty in three-dimensional (3D) shapes is important in a large number of scientific and engineering areas, such as biometrics, biomedical imaging, and data mining. It is well known that 3D polar shaped objects can be represented by Fourier descriptors such as spherical harmonics and double Fourier series. However, the statistics of these spectral shape models have not been widely explored. This thesis studies several areas involved in 3D shape modeling, including random field models for statistical shape modeling, optimal shape filtering, parametric active contours for object segmentation and surface reconstruction. It also investigates multi-modal image registration with respect to tumor activity quantification. Spherical harmonic expansions over the unit sphere not only provide a low dimensional polarimetric parameterization of stochastic shape, but also correspond to the Karhunen-Lo´eve (K-L) expansion of any isotropic random field on the unit sphere. Spherical ...

Li, Jia — University of Michigan


Factor Graph Based Detection Schemes for Mobile Terrestrial DVB Systems with Long OFDM Blocks

This PhD dissertation analyzes the performance of second generation digital video broadcasting (DVB) systems in mobile terrestrial environments and proposes an iterative detection algorithm based on factor graphs (FG) to reduce the distortion caused by the time variation of the channel, providing error-free communication in very severe mobile conditions. The research work focuses on mobile scenarios where the intercarrier interference (ICI) is very high: high vehicular speeds when long orthogonal frequency-division multiplexing (OFDM) blocks are used. As a starting point, we provide the theoretical background on the main topics behind the transmission and reception of terrestrial digital television signals in mobile environments, long with a general overview of the main signal processing techniques included in last generation terrestrial DVB systems. The proposed FG-based detector design is then assessed over a simpli ed bit-interleaved coded modulation (BICM)-OFDM communication scheme for a ...

Ochandiano, Pello — University of Mondragon


Analysis of Message Passing Algorithms and Free Energy Approximations in Probabilistic Graphical Models

Probabilistic graphical models are a powerful concept for representing complex relations between components in systems with uncertain behavior. Stemming originally from statistical mechanics, their applications stretch across various fields such as computer vision, speech recognition, social network analysis, and many more. Often they allow for a compact formulation of practical challenges in terms of fundamental computational problems, that are summarized under the term probabilistic inference. Unfortunately, these problems (e.g., the computation of marginal probabilities and the partition function) are computationally intractable so that we need to approximate the solution. In this thesis we consider two different categories of deterministic approximation methods: message passing algorithms and (variational) free energy approximations. Specifically, we focus on the most important representatives from either class, which are loopy belief propagation on the one hand, and the Bethe free energy approximation on the other hand. Although ...

Leisenberger, Harald — Graz University of Technology


A Robust Face Recognition Algorithm for Real-World Applications

Face recognition is one of the most challenging problems of computer vision and pattern recognition. The difficulty in face recognition arises mainly from facial appearance variations caused by factors, such as expression, illumination, partial face occlusion, and time gap between training and testing data capture. Moreover, the performance of face recognition algorithms heavily depends on prior facial feature localization step. That is, face images need to be aligned very well before they are fed into a face recognition algorithm, which requires precise facial feature localization. This thesis addresses on solving these two main problems -facial appearance variations due to changes in expression, illumination, occlusion, time gap, and imprecise face alignment due to mislocalized facial features- in order to accomplish its goal of building a generic face recognition algorithm that can function reliably under real-world conditions. The proposed face recognition algorithm ...

Ekenel, Hazim Kemal — University of Karlsruhe


Audiovisual Speech Synthesis Based on Hidden Markov Models

In this dissertation, new methods for audiovisual speech synthesis using Hidden Markov Models (HMMs) are presented and their properties are investigated. The problem of audiovisual speech synthesis is to computationally generate both audible speech as well as a matching facial animation or video (a “visual speech signal”) for any given input text. This results in “talking heads” that can read any text to a user, with applications ranging from virtual agents in human-computer interaction to characters in animated films and computer games. For recording and playback of facial motion, an optical marker-based facial motion capturing hardware system and 3D animation software are employed, which represent the state of the art in the animation industry. For modeling the acoustic and motion parameters of the synchronously recorded speech data, an existing HMM-based acoustic speech synthesis framework has been extended to the visual ...

Schabus, Dietmar — Graz University of Technology, Signal Processing and Speech Communication Laboratory


Channel Estimation Architectures for Mobile Reception in Emerging DVB Standards

Throughout this work, channel estimation techniques have been analyzed and proposed for moderate and very high mobility DVB (digital video broadcasting) receivers, focusing on the DVB-T2 (Digital Video Broadcasting - Terrestrial 2) framework and the forthcoming DVB-NGH (Digital Video Broadcasting - Next Generation Handheld) standard. Mobility support is one of the key features of these DVB specifications, which try to deal with the challenge of enabling HDTV (high definition television) delivery at high vehicular speed. In high-mobility scenarios, the channel response varies within an OFDM (orthogonal frequency-division multiplexing) block and the subcarriers are no longer orthogonal, which leads to the so-called ICI (inter-carrier interference), making the system performance drop severely. Therefore, in order to successfully decode the transmitted data, ICI-aware detectors are necessary and accurate CSI (channel state information), including the ICI terms, is required at the receiver. With the ...

Martínez, Lorena — University of Mondragon


Image Sequence Restoration Using Gibbs Distributions

This thesis addresses a number of issues concerned with the restoration of one type of image sequence namely archived black and white motion pictures. These are often a valuable historical record but because of the physical nature of the film they can suffer from a variety of degradations which reduce their usefulness. The main visual defects are ‘dirt and sparkle’ due to dust and dirt becoming attached to the film or abrasion removing the emulsion and ‘line scratches’ due to the film running against foreign bodies in the camera or projector. For an image restoration algorithm to be successful it must be based on a mathematical model of the image. A number of models have been proposed and here we explore the use of a general class of model known as Markov Random Fields (MRFs) based on Gibbs distributions by ...

Morris, Robin David — University of Cambridge


Bayesian Signal Processing Techniques for GNSS Receivers: from multipath mitigation to positioning

This dissertation deals with the design of satellite-based navigation receivers. The term Global Navigation Satellite Systems (GNSS) refers to those navigation systems based on a constellation of satellites, which emit ranging signals useful for positioning. Although the american GPS is probably the most popular, the european contribution (Galileo) will be operative soon. Other global and regional systems exist, all with the same objective: aid user's positioning. Initially, the thesis provides the state-of-the-art in GNSS: navigation signals structure and receiver architecture. The design of a GNSS receiver consists of a number of functional blocks. From the antenna to the fi nal position calculation, the design poses challenges in many research areas. Although the Radio Frequency chain of the receiver is commented in the thesis, the main objective of the dissertation is on the signal processing algorithms applied after signal digitation. These ...

Closas, Pau — Universitat Politecnica de Catalunya


Deep Learning Techniques for Visual Counting

The explosion of Deep Learning (DL) added a boost to the already rapidly developing field of Computer Vision to such a point that vision-based tasks are now parts of our everyday lives. Applications such as image classification, photo stylization, or face recognition are nowadays pervasive, as evidenced by the advent of modern systems trivially integrated into mobile applications. In this thesis, we investigated and enhanced the visual counting task, which automatically estimates the number of objects in still images or video frames. Recently, due to the growing interest in it, several Convolutional Neural Network (CNN)-based solutions have been suggested by the scientific community. These artificial neural networks, inspired by the organization of the animal visual cortex, provide a way to automatically learn effective representations from raw visual data and can be successfully employed to address typical challenges characterizing this task, ...

Ciampi Luca — University of Pisa


Signal Strength Based Localization and Path-loss Exponent Self-Estimation in Wireless Networks

Wireless communications and networking are gradually permeating our life and substantially influencing every corner of this world. Wireless devices, particularly those of small size, will take part in this trend more widely, efficiently, seamlessly and smartly. Techniques requiring only limited resources, especially in terms of hardware, are becoming more important and urgently needed. That is why we focus this thesis around analyzing wireless communications and networking based on signal strength (SS) measurements, since these are easy and convenient to gather. SS-based techniques can be incorporated into any device that is equipped with a wireless chip. More specifically, this thesis studies \textbf{SS-based localization} and \textbf{path-loss exponent (PLE) self-estimation}. Although these two research lines might seem unrelated, they are actually marching towards the same goal. The former can easily enable a very simple wireless chip to infer its location. But to solve ...

Hu, Yongchang — Delft University of Technology


Fire Detection Algorithms Using Multimodal Signal and Image Analysis

Dynamic textures are common in natural scenes. Examples of dynamic textures in video include fire, smoke, clouds, volatile organic compound (VOC) plumes in infra-red (IR) videos, trees in the wind, sea and ocean waves, etc. Researchers extensively studied 2-D textures and related problems in the fields of image processing and computer vision. On the other hand, there is very little research on dynamic texture detection in video. In this dissertation, signal and image processing methods developed for detection of a specific set of dynamic textures are presented. Signal and image processing methods are developed for the detection of flames and smoke in open and large spaces with a range of up to $30$m to the camera in visible-range (IR) video. Smoke is semi-transparent at the early stages of fire. Edges present in image frames with smoke start loosing their sharpness ...

Toreyin, Behcet Ugur — Bilkent University


Generalized Consistent Estimation in Arbitrarily High Dimensional Signal Processing

The theory of statistical signal processing finds a wide variety of applications in the fields of data communications, such as in channel estimation, equalization and symbol detection, and sensor array processing, as in beamforming, and radar systems. Indeed, a large number of these applications can be interpreted in terms of a parametric estimation problem, typically approached by a linear filtering operation acting upon a set of multidimensional observations. Moreover, in many cases, the underlying structure of the observable signals is linear in the parameter to be inferred. This dissertation is devoted to the design and evaluation of statistical signal processing methods under realistic implementation conditions encountered in practice. Traditional statistical signal processing techniques intrinsically provide a good performance under the availability of a particularly high number of observations of fixed dimension. Indeed, the original optimality conditions cannot be theoretically guaranteed ...

Rubio, Francisco — Universitat Politecnica de Catalunya


Cosparse regularization of physics-driven inverse problems

Inverse problems related to physical processes are of great importance in practically every field related to signal processing, such as tomography, acoustics, wireless communications, medical and radar imaging, to name only a few. At the same time, many of these problems are quite challenging due to their ill-posed nature. On the other hand, signals originating from physical phenomena are often governed by laws expressible through linear Partial Differential Equations (PDE), or equivalently, integral equations and the associated Green’s functions. In addition, these phenomena are usually induced by sparse singularities, appearing as sources or sinks of a vector field. In this thesis we primarily investigate the coupling of such physical laws with a prior assumption on the sparse origin of a physical process. This gives rise to a “dual” regularization concept, formulated either as sparse analysis (cosparse), yielded by a PDE ...

Kitić, Srđan — Université de Rennes 1


Bayesian Approaches in Image Source Seperation

In this thesis, a general solution to the component separation problem in images is introduced. Unlike most existing works, the spatial dependencies of images are modelled in the separation process with the use of Markov random fields (MRFs). In the MRFs model, Cauchy density is used for the gradient images. We provide a general Bayesian framework for the estimation of the parameters of this model. Due to the intractability of the problem we resort to numerical solutions for the joint maximization of the a posteriori distribution of the sources, the mixing matrix and the noise variances. For numerical solution, four different methods are proposed. In first method, the difficulty of working analytically with general Gibbs distributions of MRF is overcome by using an approximate density. In this approach, the Gibbs distribution is modelled by the product of directional Gaussians. The ...

Kayabol, Koray — Istanbul University

The current layout is optimized for mobile phones. Page previews, thumbnails, and full abstracts will remain hidden until the browser window grows in width.

The current layout is optimized for tablet devices. Page previews and some thumbnails will remain hidden until the browser window grows in width.