Tradutor da língua gestual portuguesa modelo de tradução bidireccional

O projeto tem como objetivo desenvolver e avaliar um modelo que facilita o acesso para pessoas surdas ou com deficiência auditiva, o acesso ao conteúdo digital - em particular o conteúdo educacional e objetos de aprendizagem – a criação de condições para uma maior inclusão social de surdos e deficientes auditivos. Pretende-se criar um modelo bidirecional, em que permite a pessoas com deficiências auditivas, possam se comunicar com outras pessoas, com a tradução da Língua Gestual Portuguesa (LGP) para a Língua Portuguesa (LP) e que outras pessoas não portadoras de qualquer deficiência auditiva possam por sua vez comunicar com os surdos ou deficientes auditivos através da tradução da LP para a LGP. Há um conjunto de técnicas que poderíamos nos apoiar para desenvolver o modelo e implementar a API de tradução da LGP em LP. Muitos estudos são feitos com base nos modelos escondidos de Markov (HMM) para efetuar o reconhecimento. Recentemente os estudos estão a caminhar para o uso de técnicas como o “Dynamic Time Warping” (DTW), que tem tido mais sucesso do que outras técnicas em termos de performance e de precisão. Neste projeto optamos por desenvolver a API e o Modelo, com base na técnica de aprendizagem Support Vector Machines (SVM) por ser uma técnica simples de implementar e com bons resultados demonstrados em reconhecimento de padrões. Os resultados obtidos utilizando esta técnica de aprendizagem foram bastante ótimos, como iremos descrever no decorrer do capítulo 4, mesmo sabendo que utilizamos dois dispositivos para capturar dados de descrição de cada gesto. Toda esta tese integra-se no âmbito do projeto científico/ investigação a decorrer no grupo de investigação GILT, sob a coordenação da professora Paula Escudeiro e suportado pela Fundação para Ciência e Tecnologia (FCT).

Kalman tracking linear predictor for vowel intelligibility enhancement on european portuguese HMM based speech synthesis

The recent developments on Hidden Markov Models (HMM) based speech synthesis showed that this is a promising technology fully capable of competing with other established techniques. However some issues still lack a solution. Several authors report an over-smoothing phenomenon on both time and frequencies which decreases naturalness and sometimes intelligibility. In this work we present a new vowel intelligibility enhancement algorithm that uses a discrete Kalman filter (DKF) for tracking frame based parameters. The inter-frame correlations are modelled by an autoregressive structure which provides an underlying time frame dependency and can improve time-frequency resolution. The system’s performance has been evaluated using objective and subjective tests and the proposed methodology has led to improved results.

Adaptive filtering for high quality HMM based speech synthesis

In this work an adaptive filtering scheme based on a dual Discrete Kalman Filtering (DKF) is proposed for Hidden Markov Model (HMM) based speech synthesis quality enhancement. The objective is to improve signal smoothness across HMMs and their related states and to reduce artifacts due to acoustic model's limitations. Both speech and artifacts are modelled by an autoregressive structure which provides an underlying time frame dependency and improves time-frequency resolution. Themodel parameters are arranged to obtain a combined state-space model and are also used to calculate instantaneous power spectral density estimates. The quality enhancement is performed by a dual discrete Kalman filter that simultaneously gives estimates for the models and the signals. The system's performance has been evaluated using mean opinion score tests and the proposed technique has led to improved results.

Human activity recognition for an intelligent knee orthosis

Dissertação para obtenção do Grau de Mestre em Engenharia Biomédica

Human activity recognition with accelerometry: novel time and frequency features

Human Activity Recognition systems require objective and reliable methods that can be used in the daily routine and must offer consistent results according with the performed activities. These systems are under development and offer objective and personalized support for several applications such as the healthcare area. This thesis aims to create a framework for human activities recognition based on accelerometry signals. Some new features and techniques inspired in the audio recognition methodology are introduced in this work, namely Log Scale Power Bandwidth and the Markov Models application. The Forward Feature Selection was adopted as the feature selection algorithm in order to improve the clustering performances and limit the computational demands. This method selects the most suitable set of features for activities recognition in accelerometry from a 423th dimensional feature vector. Several Machine Learning algorithms were applied to the used accelerometry databases – FCHA and PAMAP databases - and these showed promising results in activities recognition. The developed algorithm set constitutes a mighty contribution for the development of reliable evaluation methods of movement disorders for diagnosis and treatment applications.

Generic system for human-computer gesture interaction: applications on sign language recognition and robotic soccer refereeing

Hand gestures are a powerful way for human communication, with lots of potential applications in the area of human computer interaction. Vision-based hand gesture recognition techniques have many proven advantages compared with traditional devices, giving users a simpler and more natural way to communicate with electronic devices. This work proposes a generic system architecture based in computer vision and machine learning, able to be used with any interface for human-computer interaction. The proposed solution is mainly composed of three modules: a pre-processing and hand segmentation module, a static gesture interface module and a dynamic gesture interface module. The experiments showed that the core of visionbased interaction systems could be the same for all applications and thus facilitate the implementation. For hand posture recognition, a SVM (Support Vector Machine) model was trained and used, able to achieve a final accuracy of 99.4%. For dynamic gestures, an HMM (Hidden Markov Model) model was trained for each gesture that the system could recognize with a final average accuracy of 93.7%. The proposed solution as the advantage of being generic enough with the trained models able to work in real-time, allowing its application in a wide range of human-machine applications. To validate the proposed framework two applications were implemented. The first one is a real-time system able to interpret the Portuguese Sign Language. The second one is an online system able to help a robotic soccer game referee judge a game in real time.

Spatio-temporal modelling of environmental data

Programa Doutoral em Matemática e Aplicações.

Modeling U.S. Inflation Dynamics: A Bayesian Nonparametric Approach

This paper uses an infinite hidden Markov model (IIHMM) to analyze U.S. inflation dynamics with a particular focus on the persistence of inflation. The IHMM is a Bayesian nonparametric approach to modeling structural breaks. It allows for an unknown number of breakpoints and is a flexible and attractive alternative to existing methods. We found a clear structural break during the recent financial crisis. Prior to that, inflation persistence was high and fairly constant.

Fusion of Multi-Atlas Segmentations with Spatial Distribution Modeling

In recent years, multi-atlas fusion methods have gainedsignificant attention in medical image segmentation. Inthis paper, we propose a general Markov Random Field(MRF) based framework that can perform edge-preservingsmoothing of the labels at the time of fusing the labelsitself. More specifically, we formulate the label fusionproblem with MRF-based neighborhood priors, as an energyminimization problem containing a unary data term and apairwise smoothness term. We present how the existingfusion methods like majority voting, global weightedvoting and local weighted voting methods can be reframedto profit from the proposed framework, for generatingmore accurate segmentations as well as more contiguoussegmentations by getting rid of holes and islands. Theproposed framework is evaluated for segmenting lymphnodes in 3D head and neck CT images. A comparison ofvarious fusion algorithms is also presented.

The InterPro database, an integrated documentation resource for protein families, domains and functional sites.

Signature databases are vital tools for identifying distant relationships in novel sequences and hence for inferring protein function. InterPro is an integrated documentation resource for protein families, domains and functional sites, which amalgamates the efforts of the PROSITE, PRINTS, Pfam and ProDom database projects. Each InterPro entry includes a functional description, annotation, literature references and links back to the relevant member database(s). Release 2.0 of InterPro (October 2000) contains over 3000 entries, representing families, domains, repeats and sites of post-translational modification encoded by a total of 6804 different regular expressions, profiles, fingerprints and Hidden Markov Models. Each InterPro entry lists all the matches against SWISS-PROT and TrEMBL (more than 1,000,000 hits from 462,500 proteins in SWISS-PROT and TrEMBL). The database is accessible for text- and sequence-based searches at http://www.ebi.ac.uk/interpro/. Questions can be emailed to interhelp@ebi.ac.uk.

Sound Texture Modelling

In the PhD thesis “Sound Texture Modeling” we deal with statistical modelling or textural sounds like water, wind, rain, etc. For synthesis and classification. Our initial model is based on a wavelet tree signal decomposition and the modeling of the resulting sequence by means of a parametric probabilistic model, that can be situated within the family of models trainable via expectation maximization (hidden Markov tree model ). Our model is able to capture key characteristics of the source textures (water, rain, fire, applause, crowd chatter ), and faithfully reproduces some of the sound classes. In terms of a more general taxonomy of natural events proposed by Graver, we worked on models for natural event classification and segmentation. While the event labels comprise physical interactions between materials that do not have textural propierties in their enterity, those segmentation models can help in identifying textural portions of an audio recording useful for analysis and resynthesis. Following our work on concatenative synthesis of musical instruments, we have developed a pattern-based synthesis system, that allows to sonically explore a database of units by means of their representation in a perceptual feature space. Concatenative syntyhesis with “molecules” built from sparse atomic representations also allows capture low-level correlations in perceptual audio features, while facilitating the manipulation of textural sounds based on their physical and perceptual properties. We have approached the problem of sound texture modelling for synthesis from different directions, namely a low-level signal-theoretic point of view through a wavelet transform, and a more high-level point of view driven by perceptual audio features in the concatenative synthesis setting. The developed framework provides unified approach to the high-quality resynthesis of natural texture sounds. Our research is embedded within the Metaverse 1 European project (2008-2011), where our models are contributting as low level building blocks within a semi-automated soundscape generation system.

Brain tissue segmentation of fetal MR images

We present a segmentation method for fetal brain tissuesof T2w MR images, based on the well known ExpectationMaximization Markov Random Field (EM- MRF) scheme. Ourmain contribution is an intensity model composed of 7Gaussian distribution designed to deal with the largeintensity variability of fetal brain tissues. The secondmain contribution is a 3-steps MRF model that introducesboth local spatial and anatomical priors given by acortical distance map. Preliminary results on 4 subjectsare presented and evaluated in comparison to manualsegmentations showing that our methodology cansuccessfully be applied to such data, dealing with largeintensity variability within brain tissues and partialvolume (PV).

Weighted Shape-based Averaging with Neighborhood Prior Model for Multiple Atlas Fusion-based Medical Image Segmentation

In medical imaging, merging automated segmentations obtained from multiple atlases has become a standard practice for improving the accuracy. In this letter, we propose two new fusion methods: "Global Weighted Shape-Based Averaging" (GWSBA) and "Local Weighted Shape-Based Averaging" (LWSBA). These methods extend the well known Shape-Based Averaging (SBA) by additionally incorporating the similarity information between the reference (i.e., atlas) images and the target image to be segmented. We also propose a new spatially-varying similarity-weighted neighborhood prior model, and an edge-preserving smoothness term that can be used with many of the existing fusion methods. We first present our new Markov Random Field (MRF) based fusion framework that models the above mentioned information. The proposed methods are evaluated in the context of segmentation of lymph nodes in the head and neck 3D CT images, and they resulted in more accurate segmentations compared to the existing SBA.

Untangling the evolution of Rab G proteins: implications of a comprehensive genomic analysis.

BACKGROUND: Membrane-bound organelles are a defining feature of eukaryotic cells, and play a central role in most of their fundamental processes. The Rab G proteins are the single largest family of proteins that participate in the traffic between organelles, with 66 Rabs encoded in the human genome. Rabs direct the organelle-specific recruitment of vesicle tethering factors, motor proteins, and regulators of membrane traffic. Each organelle or vesicle class is typically associated with one or more Rab, with the Rabs present in a particular cell reflecting that cell's complement of organelles and trafficking routes. RESULTS: Through iterative use of hidden Markov models and tree building, we classified Rabs across the eukaryotic kingdom to provide the most comprehensive view of Rab evolution obtained to date. A strikingly large repertoire of at least 20 Rabs appears to have been present in the last eukaryotic common ancestor (LECA), consistent with the 'complexity early' view of eukaryotic evolution. We were able to place these Rabs into six supergroups, giving a deep view into eukaryotic prehistory. CONCLUSIONS: Tracing the fate of the LECA Rabs revealed extensive losses with many extant eukaryotes having fewer Rabs, and none having the full complement. We found that other Rabs have expanded and diversified, including a large expansion at the dawn of metazoans, which could be followed to provide an account of the evolutionary history of all human Rabs. Some Rab changes could be correlated with differences in cellular organization, and the relative lack of variation in other families of membrane-traffic proteins suggests that it is the changes in Rabs that primarily underlies the variation in organelles between species and cell types.

Brain Surface Segmentation of Magnetic Resonance Images of the Fetus

In this work we present a method for the image analysisof Magnetic Resonance Imaging (MRI) of fetuses. Our goalis to segment the brain surface from multiple volumes(axial, coronal and sagittal acquisitions) of a fetus. Tothis end we propose a two-step approach: first, a FiniteGaussian Mixture Model (FGMM) will segment the image into3 classes: brain, non-brain and mixture voxels. Second, aMarkov Random Field scheme will be applied tore-distribute mixture voxels into either brain ornon-brain tissue. Our main contributions are an adaptedenergy computation and an extended neighborhood frommultiple volumes in the MRF step. Preliminary results onfour fetuses of different gestational ages will be shown.