Dissertations / Theses on the topic 'Motion template'

The use of conversational agents has increased across the world. From providing automated support for companies to being virtual psychologists they have moved from an academic curiosity to an application with real world relevance. While many researchers have focused on the content of the dialogue and synthetic speech to give the agents a voice, more recently animating these characters has become a topic of interest. An additional use for character animation technology is in the film and video game industry where having characters animated without needing to pay for expensive labour would save tremendous costs. When animating characters there are many aspects to consider, for example the way they walk. However, to truly assist with communication automated animation needs to duplicate the body language used when speaking. In particular conversational agents are often only an animation of the upper parts of the body, so head motion is one of the keys to a believable agent. While certain linguistic features are obvious, such as nodding to indicate agreement, research has shown that head motion also aids understanding of speech. Additionally head motion often contains emotional cues, prosodic information, and other paralinguistic information. In this thesis we will present our research into synthesising head motion using only recorded speech as input. During this research we collected a large dataset of head motion synchronised with speech, examined evaluation methodology, and developed a synthesis system. Our dataset is one of the larger ones available. From it we present some statistics about head motion in general. Including differences between read speech and story telling speech, and differences between speakers. From this we are able to draw some conclusions as to what type of source data will be the most interesting in head motion research, and if speaker-dependent models are needed for synthesis. In our examination of head motion evaluation methodology we introduce Forced Canonical Correlation Analysis (FCCA). FCCA shows the difference between head motion shaped noise and motion capture better than standard methods for objective evaluation used in the literature. We have shown that for subjective testing it is best practice to use a variation of MUltiple Stimuli with Hidden Reference and Anchor (MUSHRA) based testing, adapted for head motion. Through experimentation we have developed guidelines for the implementation of the test, and the constraints on the length. Finally we present a new system for head motion synthesis. We make use of simple templates of motion, automatically extracted from source data, that are warped to suit the speech features. Our system uses clustering to pick the small motion units, and a combined HMM and GMM based approach for determining the values of warping parameters at synthesis time. This results in highly natural looking motion that outperforms other state of the art systems. Our system requires minimal human intervention and produces believable motion. The key innovates were the new methods for segmenting head motion and creating a process similar to language modelling for synthesising head motion.

This thesis presents a novel lip-reading approach to classifying utterances from video data, without evaluating voice signals. This work addresses two important issues which are • the efficient representation of mouth movement for visual speech recognition • the temporal segmentation of utterances from video. The first part of the thesis describes a robust movement-based technique used to identify mouth movement patterns while uttering phonemes. This method temporally integrates the video data of each phoneme into a 2-D grayscale image named as a motion template (MT). This is a view-based approach that implicitly encodes the temporal component of an image sequence into a scalar-valued MT. The data size was reduced by extracting image descriptors such as Zernike moments (ZM) and discrete cosine transform (DCT) coefficients from MT. Support vector machine (SVM) and hidden Markov model (HMM) were used to classify the feature descriptors. A video speech corpus of 2800 utterances was collected for evaluating the efficacy of MT for lip-reading. The experimental results demonstrate the promising performance of MT in mouth movement representation. The advantages and limitations of MT for visual speech recognition were identified and validated through experiments. A comparison between ZM and DCT features indicates that th e accuracy of classification for both methods is very comparable when there is no relative motion between the camera and the mouth. Nevertheless, ZM is resilient to rotation of the camera and continues to give good results despite rotation but DCT is sensitive to rotation. DCT features are demonstrated to have better tolerance to image noise than ZM. The results also demonstrate a slight improvement of 5% using SVM as compared to HMM. The second part of this thesis describes a video-based, temporal segmentation framework to detect key frames corresponding to the start and stop of utterances from an image sequence, without using the acoustic signals. This segmentation technique integrates mouth movement and appearance information. The efficacy of this technique was tested through experimental evaluation and satisfactory performance was achieved. This segmentation method has been demonstrated to perform efficiently for utterances separated with short pauses. Potential applications for lip-reading technologies include human computer interface (HCI) for mobility-impaired users, defense applications that require voice-less communication, lip-reading mobile phones, in-vehicle systems, and improvement of speech-based computer control in noisy environments.

Pulse oximetry is used to measure heart rate (HR) and arterial oxygen saturation (SpO2) from photoplethysmographic (PPG) waveforms. PPG waveforms are highly sensitive to motion artifact (MA), limiting the implementation of pulse oximetry in mobile physiological monitoring using wearable devices. Previous studies have shown that multichannel pulse oximetry can successfully acquire diverse signal information during simple, repetitive motion, thus leading to differences in motion tolerance across channels. In this study, we introduce a multichannel forehead-mounted pulse oximeter and investigate the performance of this novel sensor under a variety of intense motion artifacts. We have developed a multichannel template-matching algorithm that chooses the channel with the least amount of motion artifact to calculate HR and SpO2 every 2 seconds. We show that for a wide variety of random motion, channels respond differently to motion, and the multichannel estimate outperforms single channel estimates in terms of motion tolerance, signal quality, and HR and SpO2 error. Based on 31 data sets of PPG waveforms corrupted by random motion, the mean relative HR error was decreased by an average of 5.6 bpm when the multichannel-switching algorithm was compared to the worst performing channel. The percentage of HR measurements with absolute errors ≤ 5 bpm during motion increased by an average of 27.8 % when the multichannel-switching algorithm was compared to the worst performing channel. Similarly, the mean relative SpO2 error was decreased by an average of 4.3 % during motion when the multichannel-switching algorithm was compared to each individual channel. The percentage of SpO2 measurements with absolute error ≤ 3 % during motion increased by an average of 40.7 % when the multichannel-switching algorithm was compared to the worst performing channel. Implementation of this multichannel algorithm in a wearable device will decrease dropouts in HR and SpO2 measurements during motion. Additionally, the differences in motion frequency introduced across channels observed in this study shows precedence for future multichannel-based algorithms that make pulse oximetry measurements more robust during a greater variety of intense motion.

La reconstruction 3D monoculaire déformable est le problème général d'estimation de forme 3D d'un objet déformable à partir d'images 2D. Plusieurs scénarios ont émergé : le Shape-from-Template (SfT) et le Non-Rigid Structure-from-Motion (NRSfM) sont deux approches qui ont été grandement étudiées pour leur applicabilité. La première utilise une seule image qui montre un objet se déformant et un patron (une forme 3D texturée de l'objet dans une pose de référence). La seconde n'utilise pas de patron, mais utilise plusieurs images et estime la forme 3D dans chaque image. Les deux approches s'appuient sur le mouvement de points de correspondances entre les images et sur des a priori de déformations, restreignant ainsi leur utilisation à des surfaces texturées qui se déforment de manière lisse. Cette thèse fait avancer l'état de l'art du SfT et du NRSfM dans deux directions. La première est l'étude du SfT dans le cas de patrons 1D (c’est-à-dire des courbes comme des cordes et des câbles). La seconde direction est le développement d'algorithmes de SfT et de NRSfM qui exploitent plusieurs indices visuels et qui résolvent des cas réels et complexes non-résolus précédemment. Nous considérons des déformations isométriques et reconstruisons la partie extérieure de l'objet. Les contributions techniques et scientifiques de cette thèse sont divisées en quatre parties.La première partie de cette thèse étudie le SfT curvilinéaire, qui est le cas du patron curvilinéaire plongé dans un espace 2D ou 3D. Nous proposons une analyse théorique approfondie et des solutions pratiques pour le SfT curvilinéaire. Malgré son apparente simplicité, le SfT curvilinéaire s'est avéré être un problème complexe : il ne peut pas être résolu à l'aide de solutions locales non-holonomes d'une équation différentielle ordinaire et ne possède pas de solution unique, mais un nombre fini de solutions ambiguës. Une contribution technique majeure est un algorithme basé sur notre théorie, qui génère toutes les solutions ambiguës. La deuxième partie de cette thèse traite d'une limitation des méthodes de SfT : la reconstruction de plis. Cette limitation vient de la parcimonie de la contrainte de mouvement et de la régularisation. Nous proposons deux contributions qui s'appuient sur un cadre de minimisation d'énergie non-convexe. Tout d'abord, nous complétons la contrainte de mouvement avec une contrainte robuste de bord. Ensuite, nous modélisons implicitement les plis à l'aide d'une représentation dense de la surface basée maillage et d'une contrainte robuste de lissage qui désactive automatiquement le lissage de la courbure sans connaître a priori la position des plis.La troisième partie de cette thèse est dédiée à une autre limitation du SfT : la reconstruction de surfaces peu texturées. Cette limitation vient de la difficulté d'obtenir des correspondances (parcimonieuses ou denses) sur des surfaces peu texturées. Comme l'ombrage révèle les détails sur des surfaces peu texturées, nous proposons de combiner l'ombrage avec le SfT. Nous présentons deux contributions. La première est une initialisation en cascade qui estime séquentiellement la déformation de la surface, l'illumination de la scène, la réponse de la caméra et enfin les albédos de la surface à partir d'images monoculaires où la surface se déforme. La seconde est l'intégration de l'ombrage à notre précédent cadre de minimisation d'énergie afin de raffiner simultanément les paramètres photométriques et de déformation.La dernière partie de cette thèse relâche la connaissance du patron et aborde deux limitations du NRSfM : la reconstruction de surfaces peu texturées avec des plis. Une contribution majeure est l'extension du second cadre d'optimisation pour la reconstruction conjointe de la forme 3D de la surface sur toutes les images d'entrée et des albédos de la surface sans en connaître un patron
Monocular deformable 3D reconstruction is the general problem of recovering the 3D shape of a deformable object from monocular 2D images. Several scenarios have emerged: the Shape-from-Template (SfT) and the Non-Rigid Structure-from-Motion (NRSfM) are two approaches intensively studied for their practicability. The former uses a single image depicting the deforming object and a template (a textured 3D shape of this object in a reference pose). The latter does not use a template, but uses several images and recovers the 3D shape in each image. Both approaches rely on the motion of correspondences between the images and deformation priors, which restrict their use to well-textured surfaces which deform smoothly. This thesis advances the state-of-the-art in SfT and NRSfM in two main directions. The first direction is to study SfT for the case of 1D templates (i.e. curved, thin structures such as ropes and cables). The second direction is to develop algorithms in SfT and NRSfM that exploit multiple visual cues and can solve complex, real-world cases which were previously unsolved. We focus on isometric deformations and reconstruct the outer part of the object. The technical and scientific contributions of this thesis are divided into four parts. The first part of this thesis studies the case of a curvilinear template embedded in 2D or 3D space, referred to Curve SfT. We propose a thorough theoretical analysis and practical solutions for Curve SfT. Despite its apparent simplicity, Curve SfT appears to be a complex problem: it cannot be solved locally using exact non-holonomic partial differential equation and is only solvable up to a finite number of ambiguous solutions. A major technical contribution is a computational solution based on our theory, which generates all the ambiguous solutions.The second part of this thesis deals with a limitation of SfT methods: reconstructing creases. This is due to the sparsity of the motion constraint and regularization. We propose two contributions which rely on a non-convex energy minimization framework. First, we complement the motion constraint with a robust boundary contour constraint. Second, we implicitly model creases with a dense mesh-based surface representation and an associated robust smoothing constraint, which deactivates curvature smoothing automatically where needed, without knowing a priori the crease location. The third part of this thesis is dedicated to another limitation of SfT: reconstructing poorly-textured surfaces. This is due to correspondences which cannot be obtained so easily on poorly-textured surfaces (either sparse or dense). As shading reveals details on poorly-textured surfaces, we propose to combine shading and SfT. We have two contributions. The first is a cascaded initialization which estimates sequentially the surface's deformation, the scene illumination, the camera response and then the surface albedos from deformed monocular images. The second is to integrate shading to our previous energy minimization framework for simultaneously refining deformation and photometric parameters.The last part of this thesis relaxes the knowledge of the template and addresses two limitations of NRSfM: reconstructing poorly-textured surfaces with creases. Our major contribution is an extension of the second framework to recover jointly the 3D shapes of all input images and the surface albedos without any template

Orchard fruits and vegetable crops are vulnerable to wild birds and animals. These wild birds and animals can cause critical damage to the produce. Traditional methods of scaring away birds such as scarecrows are not long-term solutions but short-term solutions. This is a huge problem especially near areas like San Luis Obispo where there are vineyards. Bird damage can be as high as 50% for grapes being grown in vineyards. The total estimated revenue lost annually in the 10 counties in California due to bird and rodent damage to 22 selected crops ranged from $168 million to $504 million (in 2009 dollars). A more effective and permanent system needs to be put into place. Monitoring systems in agricultural settings could potentially provide a lot of data for image processing. Most current monitoring systems however don’t focus on image processing but instead really heavily on sensors. Just having sensors for certain systems work, but for birds, monitoring it is not an option because they are not domesticated like pigs, cows etc. in which most these agricultural monitoring systems work on. Birds can fly in and out of the area whereas domesticated animals can be confined to certain physical regions. The most crucial step in a smart scarecrow system would be how a threat would v be detected. Image processing methods can be effectively applied to detecting items in video footage. This paper will focus on bird detection and will analyze motion detection with image subtraction, bird detection with template matching, and bird detection with the Viola-Jones Algorithm. Of the methods considered, bird detection with the Viola-Jones Algorithm had the highest accuracy (87%) with a somewhat low false positive rate. This image processing step would ideally be incorporated with hardware (such as a microcontroller or FPGA, sensors, a camera etc.) to form a smart scarecrow system.

This thesis investigates ways in which a nanoscale production line may be built from synthetic DNA components. One property of a production line is motion, the coordinated movement of components, in this case strands of DNA, between specific locations. Another property is the ability to assemble a product, where smaller molecular building blocks are attached to D A and react when brought together by the DNA assembly line. An important fea- ture of either task is the ability of the mechanism to proceed with minimum user interaction: it is preferable that the assembly line be autonomous. The challenges and design principles of molecular machines working in nano- scale environments are first considered. Previous studies demonstrating the use of synthetic DNA not only as a self-assembling material to build nano- structures, but also to coordinate motion, are summarized. All DNA nano- machines that operate through the exchange of DNA strands are coordinated by toeholds. A 'split toehold', one that combines two smaller toeholds on distal sections of DNA held in proximity, is proposed as a way to allow a single cargo strand to interact with many different components. A molecular motor is then developed that transports a cargo between track locations. The fuel strands are hairpins, that carry instructions directing the cargo to the next anchorage. The switching of cargo direction in response to the chemical environment is also investigated. Two mechanisms that may allow the autonomous assembly of components are investigated, one of which is demonstrated using DNA-linked cleavable molecular building blocks. Further extensions to the mechanism are investi- gated, for example, the ability to use the DNA mechanism itself as a barcode containing information on the order of assembled ingredients.

Thesis (M.S.)--West Virginia University, 1999.
Title from document title page. Document formatted into pages; contains xii, 122 p. : ill. (some col.) Includes abstract. Includes bibliographical references (p. 119-122).

Current methodologies for recognizing similar protein motifs are predominantly based upon establishing a homology relationship between the sequences. These methods are widely exploited to annotate new genomes and assign putative functions to new genes. However they are usually based on sequence data alone. More recent approaches have incorporated structural data into methods to improve the predictions compared to just sequence based methods alone. So far these approaches have not been widely exploited in bioinformatics for identifying common, small motifs. A test system was examined containing such a degenerate but short, repeating motif, the tetratricopeptide repeat (TPR). Sequence analysis was done to assess the effectiveness of common search tools for finding TPR motifs. These methods included BLAST, PSI-BLAST and Hidden Markov Models and found the latter to be easily the most effective search strategy. Further sequence analysis of the TPR motif was carried out to demonstrate the extent to which TPRs with similar sequences are related in functional terms. In addition a full structural analysis was also performed. The results of the sequence and structural analysis of the TPR allowed structural information to be obtained and structurally conserved features in TPRs comprising conserved interacting residues pair positions were revealed. Comparative models were built and evaluated for all annotated TPR sequences with unknown structures to assess their compatibility with the TPR motif structure. From these and other models the interaction energy of structurally adjacent residues pairs has been calculated. These models were generated by mutating residues in key conserved positions to all possible amino acid combinations. The energy is then evaluated for each of these 20x20 pair combinations. This energy is then integrated into sequence based methods such as Hidden Markov Models with the aim of improving TPR prediction. An improvement in search sensitivity and specificity is demonstrated which should allow improved identification and annotation of this motif in sequence databases.

With the improvement of multimedia technologies such as broadband-enabled HDTV, video on demand and internet TV, the computer and the TV are merging to become a single device. Moreover the previously cited technologies as well as DVD or Blu-ray can provide menu navigation and interactive content. The growing interest in video conferencing led to the integration of the webcam in different devices such as laptop, cell phones and even the TV set. Our approach is to directly use an embedded webcam to remotely control a TV set using hand gestures. Using specific gestures, a user is able to control the TV. A dedicated interface can then be used to select a TV channel, adjust volume or browse videos from an online streaming server. This approach leads to several challenges. The first is the use of a simple webcam which leads to a vision based system. From the single webcam, we need to recognize the hand and identify its gesture or trajectory. A TV set is usually installed in a living room which implies constraints such as a potentially moving background and luminance change. These issues will be further discussed as well as the methods developed to resolve them. Video browsing is one example of the use of gesture recognition. To illustrate another application, we developed a simple game controlled by hand gestures. The emergence of 3D TVs is allowing the development of 3D video conferencing. Therefore we also consider the use of a stereo camera to recognize hand gesture.

Thesis (Ph. D.)--West Virginia University, 2001.
Title from document title page. Document formatted into pages; contains xviii, 130 p. : ill. (some col.). Vita. Includes abstract. Includes bibliographical references (p. 98-104).

Thesis (Ph. D.)--West Virginia University, 2003.
Title from document title page. Document formatted into pages; contains xii, 156 p. : ill. (some col.) + MPEG video files. Includes MPEG video files. Includes abstract. Includes bibliographical references.

Thesis (Ph. D.)--West Virginia University, 2000.
Title from document title page. Document formatted into pages; contains v, 277 p. Includes abstract. Includes bibliographical references (p. 268-277).

Au moyen de plusieurs mots (tabernacle, temple et Jérusalem céleste), tels qu'ils apparaissent dans l'histoire sainte de la bible, des conceptions différentes d'une demeure de dieu, d'un centre de culte sont recherchées à travers des images et des textes du moyen âge, (neuvième au douzième siècle). L'étude porte sur le rôle que joue cette iconographie par rapport aux textes bibliques
Research has been carried out into various concepts of three different words, tabernacle, temple and celestial Jerusalem, such as they appear in the holy history of the bible. They represent all three various concepts of the dwelling of god and the place of cult. Our research concerns the images (iconography) and their texts of the middle ages (ninth to twelfth century)

Des motifs nanométriques d’oxydes métalliques (TiO2, Al2O3, ZrO2) étaient synthétisés avec une technique bottom-up. Un mélange de précurseurs moléculaires et de micelles préformées des copolymères à block est déposé sur des substrats tels que la silice, l’or ou l’ITO par voie de trempage. Des monocouches de micelles entourées par les espèces inorganiques sont deposées à des concentrations et des vitesses de dépôt faibles. Calcination élimine les copolymères et des motifs (perforations rondes, canaux, anneaux,…) rigides d’oxyde métallique sont obtenus. La présence de monocouches de perforations d’un diamètre entre 10 et 70nm ainsi que l’accessibilité de la surface du substrat a travers ces perforations a été confirme avec plusieurs méthodes de caractérisation. Les dessins du type nanocrateres sont bifunctionels, peuvent être functionalisé sélectivement et ont une bonne stabilité mécanique, thermique et chimique ce qui fait que ces matériaux sont intéressants pour plusieurs applications
Metal oxide nanopatterns (TiO2, Al2O3, ZrO2) were synthesised with a bottom up technique that combines the method of chemical solution deposition and the self-organisation properties of block copolymer micelles on a substrate surface through Evaporation Induced Micelle Packing (EIMP). Molecular precursors were mixed with micelles in EtOH/THF/H2O or EtOH/THF and the solutions were dip coated onto various substrates such as silicon wafers, gold or ITO. High dilution and low withdrawal speed lead to the deposition of micelle monolayers surrounded by inorganic precursors. A calcination step eliminates the block copolymer and rigid metal oxide nanopatterns with various motifs like circular perforations (nanocraters), channels or rings are obtained. The size and kind of the motifs can be controlled by adjusting several crucial parameters during solution preparation like the kind and size of the used block copolymer in combination with the used solvent composition, concentration and conditioning. High kinetics of evaporation during the dip coating further allow ordering of nanocrater perforations in hexagonal patterns. The synthesised patterns were characterised by ellipsometry, AFM, FEG-SEM, GISAXS, XPS, cyclic voltametry and contact angle measurements. The presence of monolayers of perforations with diameters of 10-70nm and accessibility of the substrate surface through the perforations was confirmed. Nanocrater patterns show inherent bifunctionality and the substrate surface and the pattern can be selectively functionalised. In addition, nanocrater patterns are mechanically, chemically and thermally stable and are therefore interesting materials for various kinds of applications

Hepatitis delta virus (HDV) is the smallest known mammalian RNA virus, containing a genome of ~ 1700 nt. Replication of HDV is extremely dependent on the host transcription machinery. Previous studies indicated that RNA polymerase II (RNAPII) directly binds to and forms an active preinitiation complex on the right terminal stem-loop fragment (R199G) of HDV genomic RNA, and that the polypyrimidine tract-binding protein-associated splicing factor (PSF) directly binds to the same region. Further studies demonstrated that PSF also binds to the carboxyl-terminal domain (CTD) of RNAP II. In my thesis, co-immunoprecipitation assays were performed to show that PSF stimulates the interaction of RNAPII with R199G. Results of co-immunoprecipitation experiments also suggest that both the RNA recognition motif 2 (RRM2) and N-terminal proline-rich region (PRR) of PSF are required for the interaction between PSF and RNAPII, while the two RNA recognition motifs (RRM1 and RRM2) might be required for the interaction of PSF with R199G. Furthermore, in vitro run-off transcription assays suggest that PSF facilitates the HDV RNA transcription from the R199G template. Together, the above experiments suggest that PSF might act as a transcription factor for the RNAPII transcription of HDV RNA by linking the CTD of RNAPII and the HDV RNA promoter. My experiments provide a better understanding of the mechanism of HDV RNA-dependent transcription by RNAP II.

Our understanding of the complexity of the brain is limited by the data we can collect and analyze. Because of experimental limitations and a desire for greater detail, most investigations focus on just one aspect of the brain. For example, brain function can be studied at many levels of abstraction including, but not limited to, gene expression, protein interactions, anatomical regions, neuronal connectivity, synaptic plasticity, and the electrical activity of neurons. By focusing on each of these levels, neuroscience has built up a detailed picture of how the brain works, but each level is understood mostly in isolation from the others. It is likely that interaction between all these levels is just as important. Therefore, a key hypothesis is that functional units spanning multiple levels of biological organization exist in the brain. This project attempted to combine neuronal circuitry analysis with functional proteomics and anatomical regions of the brain to explore this hypothesis, and took an evolutionary view of the results obtained. During the process we had to solve a number of technical challenges as the tools to undertake this type of research did not exist. Two informatics challenges for this research were to develop ways to analyze neurobiological data, such as brain protein expression patterns, to extract useful information, and how to share and present this data in a way that is fast and easy for anyone to access. This project contributes towards a more wholistic understanding of the fruit fly brain in three ways. Firstly, a screen was conducted to record the expression of proteins in the brain of the fruit fly, Drosophila melanogaster. Protein expression patterns in the fruit fly brain were recorded from 535 protein trap lines using confocal microscopy. A total of 884 3D images were annotated and made available on an easy to use website database, BrainTrap, available at fruitfly.inf.ed.ac.uk/braintrap. The website allows 3D images of the protein expression to be viewed interactively in the web browser, and an ontology-based search tool allows users to search for protein expression patterns in specific areas of interest. Different expression patterns mapped to a common template can be viewed simultaneously in multiple colours. This data bridges the gap between anatomical and biomolecular levels of understanding. Secondly, protein trap expression patterns were used to investigate the properties of the fruit fly brain. Thousands of protein-protein interactions have been recorded by methods such as yeast two-hybrid, however many of these protein pairs do not express in the same regions of the fruit fly brain. Using 535 protein expression patterns it was possible to rule out 149 protein-protein interactions. Also, protein expression patterns registered against a common template brain were used to produce new anatomical breakdowns of the fruit fly brain. Clustering techniques were able to naturally segment brain regions based only on the protein expression data. This is just one example of how, by combining proteomics with anatomy, we were able to learn more about both levels of understanding. Results are analysed further in combination with networks such as genetic homology networks, and connectivity networks. We show how the wealth of biological and neuroscience data now available in public databases can be combined with the Brain- Trap data to reveal similarities between areas of the fruit fly and mammalian brain. The BrainTrap data also informs us on the process of evolution and we show that genes found in fruit fly, yeast and mouse are more likely to be generally expressed throughout the brain, whereas genes found only in fruit fly and mouse, but not yeast, are more likely to have a specific expression pattern in the fruit fly brain. Thus, by combining data from multiple sources we can gain further insight into the complexity of the brain. Neural connectivity data is also analyzed and a new technique for enhanced motifs is developed for the combined analysis of connectivity data with other information such as neuron type data and potentially protein expression data. Thirdly, I investigated techniques for imaging the protein trap lines at higher resolution using electron microscopy (EM) and developed new informatics techniques for the automated analysis of neural connectivity data collected from serial section transmission electron microscopy (ssTEM). Measurement of the connectivity between neurons requires high resolution imaging techniques, such as electron microscopy, and images produced by this method are currently annotated manually to produce very detailed maps of cell morphology and connectivity. This is an extremely time consuming process and the volume of tissue and number of neurons that can be reconstructed is severely limited by the annotation step. I developed a set of computer vision algorithms to improve the alignment between consecutive images, and to perform partial annotation automatically by detecting membrane, synapses and mitochondria present in the images. Accuracy of the automatic annotation was evaluated on a small dataset and 96% of membrane could be identified at the cost of 13% false positives. This research demonstrates that informatics technology can help us to automatically analyze biological images and bring together genetic, anatomical, and connectivity data in a meaningful way. This combination of multiple data sources reveals more detail about each individual level of understanding, and gives us a more wholistic view of the fruit fly brain.

COORDENAÇÃO DE APERFEIÇOAMENTO DO PESSOAL DE ENSINO SUPERIOR
A pesquisa exegética sempre indicou um alto índice de referências a diversos textos vétero-testamentários no livro do Apocalipse. Dentre eles, destacam-se os textos de Ezequiel. A presença de textos do Antigo Testamento no Novo Testamento vem sendo analisada nos últimos anos sob a perspectiva da intertextualidade. A presente tese tem por escopo analisar Ap 22,1-5 e Ez 47,1-12 focalizando de modo particular o motivo da fonte de água que surge do Templo ou do trono de Deus. A perspectiva intertextual permitirá perceber melhor o modo como Ap 22,1-5 trabalhou este motivo de Ez 47,1-12. Outros textos véterotestamentários, apontados ao longo da pesquisa, contribuirão para verificar a relevância de Ez 47,1-12 na formulação de Ap 22,1-5.
The exegetical research has always indicated a high quantity of references to Old Testament texts present in the Book of Apocalypse. Among them the text of Ezekiel is specially relevant. The presence of Old Testament texts in the New Testament has been analyzed in the last years under the perspective of intertextuality. This dissertation has its focus on Apocalypse 22,1-5 and Ezekiel 47,1-12, specially the reason for the source of water which comes from the Temple or the throne of God. The intertextual perspective will allow a better understanding of the manner in which Apocalypse 22,1-5 worked this aspect from Ezekiel 47,1-12. Other Old Testament texts, signaled during the research, will contribute for the importance of Ezekiel 47,1-12 in the formulation of Apocalypse 22,1-5.

La période située entre les deux grandes vagues de destructions iconoclastes dans l’Angleterre du XVIe et du XVIIe siècle est aussi celle où se développe une nouvelle poétique dite « métaphysique », caractérisée par un usage frappant du conceit. Ce travail, consacré à l’œuvre de George Herbert, se propose d’étudier le glissement de propriétés habituellement dévolues à l’image religieuse vers la poésie. On montrera comment la rhétorique et l’imaginaire iconomaques informent le projet littéraire et religieux d’un auteur qui cherche à convertir et purifier la poésie. Là où l’iconoclasme anglais passé et à venir extériorise une crise profonde de la notion même d’image, George Herbert cherche à la reconstruire intérieurement par une poétique efficace de la discordia concors dont le modèle est la théologie de l’Incarnation. Si Herbert rompt, comme tant d’autres poètes à la même époque, avec certaines pratiques de l’image poétique, c’est afin de mieux figurer le divin. Au-delà de sa reconquête de la ressemblance par le transport de la métaphore, c’est aussi grâce à une modalité visuelle de la lettre que Herbert parvient à suggérer l’ineffable et la transcendance divine
The period of time which stretches between the two phases of violent iconoclastic destruction in Early Modern England also corresponds to the development of a new « metaphysical » poetics characteristically known for its striking conceits. The purpose of this study, which is devoted to George Herbert, will be to show how his poetry can become endowed with some of the properties usually ascribed to the religious image. The rhetorics of iconoclasm shape Herbert’s attempt to convert and purify poetry. Whereas past and future iconoclastic destruction testifies to a deeper crisis of the notion of image, George Herbert seeks to recreate it internally through a poetics of discordia concors whose model is the Incarnation. By distancing himself, as many other contemporary poets, from certain long-standing approaches to figurative language, he reclaims for poetry the power to body forth the divine. Such a recovery of resemblance is not only made possible through the « figure of transport », the metaphor, but aslo through an iconic use of writing. Thus Herbert is able to suggest the unspeakable and the transcendance of God

Dans notre corpus d’étude : La Présentation de Marie au Temple, Les Miracles de Nostre Dame par personnages et Le Mystère de la Passion d’Arnoul Gréban, le personnage de Marie se présente à l’intersection des langues, des genres et des lieux où se réinvente le théâtre. Ces aspects hétérogènes se dissolvent dans la mesure où le personnage de Marie s’ancre dans la logique de la foi chrétienne : au sein de la Vierge fut incarnée la Parole de Dieu qui, sous sa forme humaine, en la personne du Fils, parla aux hommes ; telle est l’affirmation fondamentale du christianisme qui souligne l’importance du Verbe. Néanmoins, les écrits néotestamentaires qui rapportent la parole divine sont discrets sur Marie, tandis que les jeux à sujet religieux donnent à voir et à entendre un personnage caractérisé par un langage théâtral. En écho avec les écrits scripturaires, le personnage de Notre Dame est composé à partir des écrits pieux en latin et en langue vernaculaire, des textes littéraires et rhétoriques, et sous l’influence des réalités sociales. Son personnage est en effet à la confluence de ces courants qu’il transcende et, à leur interstice, il modèle une nouvelle façon d’exprimer la foi à laquelle la forme dramatique donne une dimension collective. Appuyée sur l’analyse matérielle et poétique des manuscrits, cette thèse s’attache à montrer les dynamiques qui donnent vie au personnage de Marie et ses répercussions sur la composition du texte dramatique, ainsi que sur les communautés qu’il engendre
The dramatis personae of Mary in La Présentation de Marie au Temple, Les Miracles de Nostre Dame par personnages and Le Mystère de la Passion d’Arnoul Gréban is at the intersection of languages, poetical genres, and places where drama is performed. Her character is inspired by the Christian faith in the Incarnation: within the Virgin was incarnated the Word of God who, in his human form, in the person of the Son, spoke to men. The fundamental affirmation of Christianity underlines the importance of the Word. Nevertheless, the New Testament is discreet about Mary, while the religious plays show the mother of Lord characterized by a self-expression, a mainly lyrical one. Inspired by texts in Latin and in vernacular, from literary and rhetorical texts, and under the influence of social realities, the character of Mary is at the confluence of these currents that she transcends and, at their interstice, models a new way of expressing faith, to which the dramatic form gives a collective dimension. Based on the material and poetical analysis of the manuscripts, this thesis aims to show the dynamics that give life to Mary in theatre and the repercussions of her character on the poetry of the dramatic text, as well as on the communities it engenders

碩士
國立交通大學
資訊科學與工程研究所
100
This thesis introduces a bi-prediction scheme based on a joint application of template and block motion compensations. Since the template motion is decode-side inferable, this scheme needs only a motion overhead as that of uni-directional prediction. Two predictors derived from the template and block matchings are weighted in a pixel-adaptive manner using OBMC. From an analytical aspect, we provide an optimal design of window function in a parametric overlapped block motion compensation (OBMC) framework to further improve the efficiency of inter-frame prediction. In view of the tradeoff between the performance and the complexity, we discussed the impacts on the prediction efficiency of the proposed scheme when the number of template shapes and the number of motion hypotheses are changed. Also, a fast template search is provided, which greatly reduces the complexity at the decoder side. As compared with HM3.0, the proposed scheme achieves an average BD-rate saving of 2.2%, with a minimum of 1.1% and a maximum of 4.1%. The encoding time increases 46% while the decoding time increases moderately in 33%.

碩士
國立臺灣大學
資訊工程學研究所
100
In this thesis, we propose a system to estimate head poses only using depth information in real-time. Two methods are developed. First, assuming that a head can be approximated by a bounding box, we find the best fitted plane for the frontal face by the least square error method. Thus, the normal vector of this plane represents the head orientation. Second, an optimization method based on 3D model fitting is developed. We iteratively minimize the distance between source and target point clouds of a user''s head. This method is more robust and the results are more precise. Both of the proposed methods give fully real-time responses (30fps) without needing the GPU speedup. We adopt a commodity depth sensor named Microsoft Kinect as well as Asus Xtion, and use the depth image as the only input so that our system will not be affected by illumination variations. The simplicity of this acquisition device comes at the cost of frequent noises in the acquired data. We demonstrate that 6 degree of freedom real-time head motion tracking in 3D space can be achieved with noisy depth data.

碩士
國立臺灣科技大學
資訊工程系
99
According to the record of National Police Agency, Ministry of The Interior, 2010 general situation of Offense of Illegal Gun Ammunition and Sword, the police seized 1,575 illegal guns, 31,760 bullets. 209 gun-related criminals within 84 murder, 76 robbery and hold-up. We can find that it becomes a serious problem in Taiwan and reveals the demands in detecting and preventing gun-violence. Gun-man detection has been discussed for so many years, if we could detect gun-man efficiently, it will not only help the society security, but also ensure the safety of safeguard. The characteristic of gun is no gesture, small, rectangle shape.No gesture means we could only use its shape for recognizing, but the characteristic of “small and rectangle shape” happens a lot in general background, then we can found that the hard portion of gun-man recognition is result in its lack of recognizable feature, This thesis proposes an intelligent surveillance system could identify pedestrian with holding a gun or not. Using the special shape feature of gun-holding-hand, motion image information and RGB pattern matching to determine is there anyone holding a gun in the image. The experimental results show that the system could detect human and gun precisely with input image resolution 640 x 480 in 17.8 fps.

碩士
國立交通大學
資訊科學與工程研究所
98
Abstract—Template matching prediction (TMP), which estimates the motion for a target block by using its surrounding pixels, has been observed to perform efficiently in inter-frame coding. In this paper, we expose, from a more theoretical viewpoint, the factors that determine the prediction efficiency of TMP. It is shown that the motion estimate found by template matching tends to be the motion associated with the template centroid and that TMP consistently outperforms SKIP prediction, but hardly competes with block motion compensation (BMC) unless both the motion and intensity fields are less random or have high spatial correlation. We also demonstrate how template and block motion estimates can jointly be applied in a parametric overlapped block motion compensation (OBMC) framework to further improve temporal prediction. Preliminary results show that combining TMP with OBMC can yield 2-16% reductions in mean-square prediction error, as compared with the single use of OBMC. The gain is even higher (18%) when the performance is compared with that of the standard BMC.

碩士
國立中央大學
土木工程研究所
94
In recent years, a topic of considerable research interest in the engineering seismology is to predict a suitable ground motion for seismic evaluation of a structure or earthquake risk assessment of a designated site. This study uses the discrete Fourier transform method to predict seismic ground motion of a recording site. The accelerograms at the recording site are divided into four sites according to which the location of epicenter, and for each site the suitable distributions of probability density function are selected to model Fourier spectrum and group delay time with the associated parameters being regressed as function of earthquake magnitude. Thus, with the given magnitude, predicted accelerograms can be generated using the inverse Fourier transform. The earthquake records with magnitudes equal to or greater than 5.0 from the Jr-Nan Temple Station in Taipei are selected for this study. For the Jr-Nan Temple Station, the following conclusions can be drawn：（1）the Fourier spectrum of records can be modeled as a lognormal probability density function；（2）separated frequency range into ten intervals group delay time of records can be modeled as a normal probability density function in each interval；（3）if more records with the epicenters located within small region are available, the accuracy of the predicted motion originated from that region can be enhanced；（4）if two regions have the same numbers of epicenter, the accuracy of the predicted ground motion originated from the region with epicenters spread over a small area will be better.

No
The minerals involved in the formation of metazoan skeletons principally comprise glassy silica, calcium phosphate or carbonate. Because of their ancient heritage, glass sponges (Hexactinellida) may shed light on fundamental questions such as molecular evolution, the unique chemistry and formation of the first skeletal silica-based structures, and the origin of multicellular animals. We have studied anchoring spicules from the metre-long stalk of the glass rope sponge (Hyalonema sieboldi; Porifera, Class Hexactinellida), which are remarkable for their size, durability, flexibility and optical properties. Using slow-alkali etching of biosilica, we isolated the organic fraction, which was revealed to be dominated by a hydroxylated fibrillar collagen that contains an unusual [Gly-3Hyp-4Hyp] motif. We speculate that this motif is predisposed for silica precipitation, and provides a novel template for biosilicification in nature.