Orientation – the process by which animals determine their position in an environment – can be accomplished by using the visually distinct properties of objects or surfaces, known as features (i.e., colour or pattern) or the relationship among objects and surfaces, known as geometry (i.e., wall length or angular information). Although features have been shown to facilitate the encoding of geometry, little is known as to whether restricting one’s viewpoint to include fewer features will still facilitate the encoding of geometry. During this experiment, men and women were trained to search near either an acute or an obtuse corner of a virtual parallelogram-shaped room that contained either three or four discrete and distinctive features. Participants were subsequently tested for their encoding of wall length and angles when the cues were presented in isolation, together, or in conflict. Results showed that the number of features present during training did not influence the encoding of geometry. However, the discrete and distinctive properties of the features overshadowed the encoding of angles by women as well as by participants who were trained with the obtuse corner. Although some groups of participants did not encode angular information when this was the only available geometric cue, all groups weighed angles more heavily than wall length when the cues provided conflicting information. This result suggests that one type of geometric cue (i.e., wall length) can facilitate the encoding of another (i.e., angles).

Analysis and extraction of features in video streaming refer to the process of identifying and extracting specific characteristics or patterns from a video stream that can be used for various purposes, such as object detection, recognition, and tracking, as well as video compression, indexing, and retrieval. The extracted features can be used for different purposes, depending on the specific needs and requirements of the application. End-to-end streaming latency refers to the delay between the time a video frame is captured and the time it is displayed on the user’s device. Analysis and extraction of features in video streaming can be used to measure end-to-end streaming latency by extracting specific characteristics or patterns from the video stream that indicate the start and end points of the video stream and the time stamps of each frame. In this work we propose a simple but effective way to measure the end-to-end streaming latency by using object detection and image-to-text conversion, both tasks based on the extraction of features of the underlying content. Shot boundary detection is the process of identifying the boundaries between shots in a video stream. Shot boundary detection is an important task in video processing, as it is used for various applications, such as video editing, indexing, retrieval, and summarization. Analysis and extraction of features in video streaming can be used for shot boundary detection by extracting specific characteristics or patterns from the video stream that indicate changes in the visual and audio content. Once these features are extracted, various techniques can be used to detect shot boundaries, such as thresholding, clustering, and machine learning algorithms. In this work, we analyze state-of-the-art deep learning algorithms for shot boundary detection tasks and datasets and propose several new models that improve the efficiency of the last state-of-the-art models meanwhile keeping or even improving the resulting metrics. Video temporal segmentation in scenes is the process of dividing a video stream into coherent temporal segments grouping all shots that are visually and semantically related to each other. In this work, we take advantage of the improvements done in the task of shot boundary detection to propose a foundational model in the task of segmenting the video into scenes, from a previous segmentation in shots. We propose a model based on visual similarity and we also contribute with a specific dataset for the task.

No description available.

Automatic speech recognition requires many hours of transcribed speech recordings in order for an acoustic model to be effectively trained. However, recording speech corpora is time-consuming and expensive, so such quantities of data exist only for a handful of languages — there are many languages for which little or no data exist. Given that there are acoustic similarities between different languages, it may be fruitful to use data from a well-supported source language for the task of training a recogniser in a target language with little training data. Since most languages do not share a common phonetic inventory, we propose an indirect way of transferring information from a source language model to a target language model. Tandem features, in which class-posteriors from a separate classifier are decorrelated and appended to conventional acoustic features, are used to do that. They have the advantage that the language used to train the classifier, typically a Multilayer Perceptron (MLP) need not be the same as the target language being recognised. Consistent with prior work, positive results are achieved for monolingual systems in a number of different languages. Furthermore, improvements are also shown for the cross-lingual case, in which the tandem features were generated using a classifier not trained for the target language. We examine factors which may predict the relative improvements brought about by tandem features for a given source and target pair. We examine some cross-corpus normalization issues that naturally arise in multilingual speech recognition and validate our solution in terms of recognition accuracy and a mutual information measure. The tandem classifier in work up to this point in the thesis has been a phoneme classifier. Articulatory features (AFs), represented here as a multi-stream, discrete, multivalued labelling of speech, can be used as an alternative task. The motivation for this is that since AFs are a set of physically grounded categories that are not language-specific they may be more suitable for cross-lingual transfer. Then, using either phoneme or AF classification as our MLP task, we look at training the MLP using data from more than one language — again we hypothesise that AF tandem will resulting greater improvements in accuracy. We also examine performance where only limited amounts of target language data are available, and see how our various tandem systems perform under those conditions.

Atualmente, com a diversificação dos produtos demandados, os sistemas de manufatura têm se tornado cada vez mais complexos e de difícil operacionalização. Dificilmente o responsável pela manufatura tem toda a informação necessária para antever os detalhes necessários à sua execução. Por outro lado, as empresas não dispõem de tempo nem de recursos para testar alternativas de manufatura no sistema real. Com isso, a simulação baseada nas features, ou seja, na geometria da peça, se torna importante, pois retrata fielmente as características do produto a ser manufaturado, facilitando a gestão da manufatura. Esse trabalho foi estruturado a partir da revisão da literatura sobre os conceitos de sistemas de manufatura, tecnologia de features, simulação, realidade virtual e modelos de processos de software. O objetivo do trabalho foi desenvolver um ambiente a partir de uma nova abordagem para simulação da manufatura baseado nas features das peças com uma interface tridimensional baseada em realidade virtual. A partir dessas simulações, o usuário poderá tomar decisões baseadas em indicadores como: tempo total de manufatura, custo e tempo por recurso de fabricação e movimentação, tempo de espera em fila e taxa de utilização de cada recurso num sistema de manufatura específico sem alterar o cotidiano da empresa. O desenvolvimento do sistema foi executado dentro do paradigma da orientação a objetos e foi dividido em etapas: análise de requisitos, projeto, implementação, verificação e validação e documentação. Para mostrar que as funcionalidades do ambiente de simulação decorrentes da abordagem proposta no trabalho são aplicáveis em sistemas de manufatura reais foram propostas duas aplicações com suas soluções. / Nowadays with the diversification of products demanded, manufacturing systems have become increasingly complex and with difficult operation. Hardly the person responsible for the manufacturing area has all the information necessary to anticipate the details for their implementation. On the other hand, companies don\'t have the time or resources to test alternatives in the real manufacturing system. With this simulation based on the features, i.e., the part geometry becomes important because it represents faithfully the characteristics of the product being manufactured, making it easy the management of manufacturing. This work was structured from the literature review on the concepts of manufacturing systems, features technology simulation, virtual reality and software process models. The objective of this study was to develop an environment from a new approach for simulation of manufacturing based on features of the parts with a three-dimensional interface based on virtual reality. From these simulations, the user can make decisions based on indicators such as: total time manufacturing, costs and time for manufacturing and resource handling, time of queuing and rate of utilization resource in a manufacturing system without changing the daily life of the company. The development of the system was implemented within the paradigm of object orientation and was divided into steps: requiriments analysis, project, implementation, verification and validation and documentation. To show that the functions of the simulation environment due the approach proposed in this work are applicable to real manufacture systems have been proposed two applications with their solutions.

Atualmente, com a diversificação dos produtos demandados, os sistemas de manufatura têm se tornado cada vez mais complexos e de difícil operacionalização. Dificilmente o responsável pela manufatura tem toda a informação necessária para antever os detalhes necessários à sua execução. Por outro lado, as empresas não dispõem de tempo nem de recursos para testar alternativas de manufatura no sistema real. Com isso, a simulação baseada nas features, ou seja, na geometria da peça, se torna importante, pois retrata fielmente as características do produto a ser manufaturado, facilitando a gestão da manufatura. Esse trabalho foi estruturado a partir da revisão da literatura sobre os conceitos de sistemas de manufatura, tecnologia de features, simulação, realidade virtual e modelos de processos de software. O objetivo do trabalho foi desenvolver um ambiente a partir de uma nova abordagem para simulação da manufatura baseado nas features das peças com uma interface tridimensional baseada em realidade virtual. A partir dessas simulações, o usuário poderá tomar decisões baseadas em indicadores como: tempo total de manufatura, custo e tempo por recurso de fabricação e movimentação, tempo de espera em fila e taxa de utilização de cada recurso num sistema de manufatura específico sem alterar o cotidiano da empresa. O desenvolvimento do sistema foi executado dentro do paradigma da orientação a objetos e foi dividido em etapas: análise de requisitos, projeto, implementação, verificação e validação e documentação. Para mostrar que as funcionalidades do ambiente de simulação decorrentes da abordagem proposta no trabalho são aplicáveis em sistemas de manufatura reais foram propostas duas aplicações com suas soluções. / Nowadays with the diversification of products demanded, manufacturing systems have become increasingly complex and with difficult operation. Hardly the person responsible for the manufacturing area has all the information necessary to anticipate the details for their implementation. On the other hand, companies don\'t have the time or resources to test alternatives in the real manufacturing system. With this simulation based on the features, i.e., the part geometry becomes important because it represents faithfully the characteristics of the product being manufactured, making it easy the management of manufacturing. This work was structured from the literature review on the concepts of manufacturing systems, features technology simulation, virtual reality and software process models. The objective of this study was to develop an environment from a new approach for simulation of manufacturing based on features of the parts with a three-dimensional interface based on virtual reality. From these simulations, the user can make decisions based on indicators such as: total time manufacturing, costs and time for manufacturing and resource handling, time of queuing and rate of utilization resource in a manufacturing system without changing the daily life of the company. The development of the system was implemented within the paradigm of object orientation and was divided into steps: requiriments analysis, project, implementation, verification and validation and documentation. To show that the functions of the simulation environment due the approach proposed in this work are applicable to real manufacture systems have been proposed two applications with their solutions.

Texture Segmentation is one of the most challenging areas of computer vision. One reason for this difficulty is the huge variety and variability of textures occurring in real world, making it very difficult to quantitatively study textures. One of the key tools used for texture segmentation is local invariant descriptors. Texture consists of textons, the basic building block of textures, that may vary by small nuisances like illumination variation, deformations, and noise. Local invariant descriptors are robust to these nuisances making them beneficial for texture segmentation. However, grouping dense descriptors directly for segmentation presents a problem: existing descriptors aggregate data from neighborhoods that may contain different textured regions, making descriptors from these neighborhoods difficult to group, leading to significant errors in segmentation. This work addresses this issue by proposing dense local descriptors, called Shape-Tailored Features, which are tailored to an arbitrarily shaped region, aggregating data only within the region of interest. Since the segmentation, i.e., the regions, are not known a-priori, we propose a joint problem for Shape-Tailored Features and the regions. We present a framework based on variational methods. Extensive experiments on a new large texture dataset, which we introduce, show that the joint approach with Shape-Tailored Features leads to better segmentations over the non-joint non Shape-Tailored approach, and the method out-performs existing state-of-the-art.

This study describes the gross anatomical, histological and surface morphological features of the oropharynx and proximal oesophagus of the emu in order to address the scarcity of information on this region in this commercially important bird. Heads obtained from birds at slaughter (and a younger and older bird from emergency farm slaughter) were used for this study and described using basic gross anatomical and histological techniques, supplemented by scanning electron microscopy. The findings of the study were compared with the relevant literature. The oral and pharyngeal cavities could not be morphologically separated and formed a single cavity. This cavity was dorso-ventrally flattened and clearly divided, both on the floor and the roof, into rostral pigmented and caudal non-pigmented parts. The non-pigmented floor housed the tongue and laryngeal mound which had a wide glottis and no papillae. The choana was triangular-shaped, with a small caudo-lateral fold on either side, and was situated in the nonpigmented part of the roof. Caudal to the choana were two rounded pharyngeal folds with a pitted ventral surface. A small bilateral projection from the caudo-lateral edge consisted mainly of diffuse lymphoid tissue. The pharyngeal folds contained numerous large simple branched tubular mucus-secreting glands as well as large accumulations of lymphoid tissue. The pigmented regions of the roof and floor were aglandular and lined by a keratinised stratified squamous epithelium which, particularly in the roof, contained numerous Herbst corpuscles in the underlying connective tissue. SEM revealed the surface to be composed of sheets of desquamating flattened polygonal cells. The non-pigmented regions were glandular and lined by a non-keratinised stratified squamous epithelium. Surface cells displayed a pattern of microplicae or microvilli while individual surface cells were seen to desquamate. The connective tissue housed small, simple tubular and large, simple branched tubular mucus-secreting glands, Herbst corpuscles (only absent from the pharyngeal folds and proximal oesophagus), lymphoid tissue, blood vessels and nerves. The glands of the upper digestive tract were polystomatic and named as follows according to their location: Caudal intermandibular, lingual, crico-arytenoid, oral angular, caudal palatine, pharyngeal and oesophageal. The openings of the glands to the surface were seen on SEM as variably sized holes on the surface, some being obscured by mucus secretions from the underlying glands. Taste receptors were sparse and present only in the caudal non-pigmented oropharyngeal floor, tongue root and proximal oesophagus. Accumulations of lymphoid tissue were identified at the junction between the two regions of the roof, and in the non-pigmented roof, the non-pigmented floor, tongue ventrum, root and frenulum, proximal oesophagus and pharyngeal folds. The consistent dense accumulation of lymphoid tissue in the pharyngeal folds constituted pharyngeal tonsils (Lymphonoduli pharyngeales). The lymphoid tissue of the non-pigmented floor was visible macroscopically as round raised nodules. Specific, unnamed larger lymphoid tissue aggregations were located at the junction of the tongue ventrum and frenulum and in the small folds lateral to the choana. Surface morphology, as seen by SEM, revealed a pattern of microridges on the surface cells of the keratinised areas, whereas the surface cells of the non-keratinised areas displayed microplicae, microvilli and cilia. Microvilli and cilia were associated with the gland openings and ducts. The proximal oesophagus was a cylindrical tube with a longitudinally folded mucosa and displayed the typical tissue layers described in birds. The mucosa was formed by a nonkeratinised stratified epithelium which on SEM showed minimal surface desquamation. The lamina propria contained numerous simple tubular mucus-secreting glands which sometimes branched and occasional diffuse lymphoid tissue aggregations. The gland openings to the surface were seen on SEM as small and large dark holes. The muscularis mucosae was very prominent and was a longitudinal smooth muscle layer separating the mucosa from the submucosa. The tunica muscularis was composed of a thicker inner circular and a thinner outer longitudinal smooth muscle layer surrounded by the outer loose connective tissue forming the tunica adventitia. The emu tongue was divided into a body and a root. The body was triangular, dorso-ventrally flattened, pigmented and displayed caudally directed lingual papillae on both the lateral and caudal margins. The root, a more conspicuous structure in comparison to other ratites, was triangular, with a raised bulbous component folding over the rostral part of the laryngeal fissure. The lingual skeleton was formed by the triangular-shaped paraglossum (hyaline cartilage), forming the core of the tongue body, and the rostral projection of the basihyale, ventral to the paraglossum. Following the general trend in ratites, the emu tongue was greatly reduced in comparison to the bill length and specifically adapted for swallowing during the cranioinertial method of feeding employed by palaeognaths. The tongue was invested by a non-keratinised stratified squamous epithelium. The glands in the connective tissue formed the bulk of the parenchyma and were composed of both small simple tubular and large simple branched tubular mucus-secreting glands similar to those seen in the oropharynx. The lingual glands were grouped as follows: dorsal and rostro-ventral (large glands), caudo-ventral and radical (large and small glands) and frenular (small glands). The large glands were visible macroscopically as doughnut-shaped structures. Melanocytes were absent from the tongue ventrum and occasionally from the tongue root. Lymphoid tissue was absent from the tongue dorsum. Herbst corpuscles were present in the tongue body and root and generally closely associated with the large mucus-secreting glands. The surface morphology varied in the different regions of the tongue. The dorsal and rostro-ventral tongue body showed individual desquamating cells and large gland openings only, the caudo-lateral ventrum showed less desquamation and both large and small openings. The mid-ventral aspect had an undulating uneven appearance with round raised cells on the surface which were densely packed with microvilli. Very large, large and small openings were present in this region and ciliated cells occurred in the vicinity of gland openings. This study presented various unique findings regarding the morphology of the emu oropharynx compared to other ratites. Although the sense of taste has been confirmed in many avian species, this study presented the first evidence of taste in the emu and ratites in general and suggests the possibility of taste being previously overlooked in the other birds studied (ostrich and greater rhea). The tongue root of the emu was clearly defined and is unique in structure and possible function amongst the ratites and other birds. Previously unmentioned functions of the emu tongue revealed by this study include: touch (Herbst corpuscles), taste (taste bud), lubrication and mechanical protection (mucus-secreting glands), immunological (lymphoid tissue) and digestive (swallowing). It was also noted that the various structures and organs of the oropharynx revealed important and often interesting differences between the emu and the other ratites documented. The prominent serrations of the rostral mandibular tomia of the emu also appear to be unique amongst ratites. The presence and wide distribution of Herbst corpuscles within the emu oropharynx and tongue show these areas to be highly sensitive to touch. The caudo-lateral projections of the pharyngeal folds effectively formed pharyngeal tonsils, a feature not apparent in other ratites. Despite the differences noted between the emu and other ratites it was possible to discern a common pattern of structures and features, with their modifications, both within and forming the oropharynx in this group of birds. / Dissertation (MSc)--University of Pretoria, 2009. / Anatomy and Physiology / unrestricted

No

No / Gender classification has multiple applications including, but not limited to, face perception, age, ethnicity and identity analysis, video surveillance and smart human computer interaction. The majority of computer based gender classification algorithms analyse the appearance of facial features predominantly based on the texture of the static image of the face. In this paper, we propose a novel algorithm for gender classification using the smile dynamics without resorting to the use of any facial texture information. Our experiments suggest that this method has great potential for finding indicators of gender dimorphism. Our approach was tested on two databases, namely the CK+ and the MUG, consisting of a total of 80 subjects. As a result, using the KNN algorithm along with 10-fold cross validation, we achieve an accurate classification rate of 80% for gender simply based on the dynamics of a person's smile.