O presente estudo teve como objetivos primeiramente classificar indivíduos em diferentes padrões por meio de inteligência artificial, caracterizando-os com base na composição corporal, aptidão física, assimetria do córtex pré frontal (CPF), perfil comportamental relacionado ao exercício e respostas psicofisiológicas antes, durante e depois de um teste incremental máximo (TIM), após realizadas essas classificações, comparar entre esses mesmos padrões suas respectivas respostas motivacionais, afetivas, de ativação e percepção subjetiva de esforço (PSE) antes, durante e depois de um TIM, avaliando hierarquicamente o quanto suas características (composição corporal, aptidão física, perfil comportamental e assimetria do CPF) influenciaram nas respostas de cada momento. Participaram do estudo 76 sujeitos de ambos os sexos com idade de 20 a 29 anos, realizando uma única visita laboratorial, cuja mesma consistiu em uma avaliação física, avaliação do perfil comportamental relacionado ao exercício por meio de questionários, coleta de eletroencefalograma (EEG) e a realização de um TIM na esteira. Os participantes foram posteriormente alocados em grupos por seus respectivos padrões de similaridades definidos por análise de Self-Organized Maps (SOM), e na sequência foram comparadas as respostas psicofisiológicas de afeto, PSE e ativação entre os grupos durante 8 momentos (antes, durante e após) do TIM, juntamente do poder de influência de 29 covariáveis em cada momento. Foram identificados 4 padrões (grupos) pelo SOM, sendo que o Grupo 1 (G1) possuía melhor composição corporal, aptidão física, e um perfil comportamental de maior afinidade ao exercício, enquanto os demais apresentaram características gradativamente opostas, consequentemente, esses grupos também diferiram nas respostas psicofisiológicas mediante o exercício, sendo que, o G1 apresentou maior motivação (f = 1.999, p < .05), afeto (f = 3.459, p < .05) e menor PSE (f = 3.030, p = .05) comparado aos outros grupos. Todas as respostas psicofisiológicas foram influenciadas pelas covariáveis, porém, a hierarquia de qual covariável influencia mais ou menos muda a cada momento do TIM. Concluindo, pôde-se identificar 4 padrões de indivíduos cujos mesmos diferem nas respectivas respostas psicofisiológicas nos períodos antes, durante e depois de um TIM. Entretanto, apesar das respostas psicofisiológicas mediante ao TIM sofrerem influências pelas variáveis de composição corporal, aptidão física, perfil comportamental e assimetria do CPF, a ordem hierárquica do poder de influência dessas mesmas variáveis diferem a cada momento do teste / The aim of the present study was to classify individuals into different patterns using artificial intelligence, characterizing them based on body composition, physical fitness, prefrontal cortex (CPF) asymmetry, exercise-related behavioral profile, and psychophysiological responses before, during and after a maximal incremental test (TIM), after performing these classifications, the aim was to compare between these same patterns their respective motivational, affective and arousal responses and rating perceived effort (PSE) before, during and after a TIM, evaluating hierarchically how much their respective characteristics (body composition, physical fitness, exercise-related behavioral profile, and CPF asymmetry) influenced the responses in each TIM moment. The sample composed by 76 subjects of both sexes, aged 20 to 29 years, performed a single laboratory visit which consisted of body composition measurements, evaluation of exercise-related behavioral profile through questionnaires, rest electroencephalogram (EEG) measurement and performing a TIM on the treadmill. Participants were later allocated into groups by their respective patterns of similarities defined by the Self Organized Maps (SOM) analysis, and the psychophysiological responses of motivation, affect, PSE and arousal between groups were compared in 8 moments (before, during and after) of TIM, together with the power of influence of 29 covariates at each moment. Four patterns (groups) were identified by SOM, and Group 1 (G1) had better body composition, physical fitness, and exercise-related behavioral profile with greater affinity to exercise, while other groups presented characteristics gradually opposite, consequently, these groups also differed in psychophysiological responses to exercise, being that G1 showed greater motivation (f = 1.999, p < .05) , afect (f = 3.459, p < .05) and lower PSE (f = 3.030, p = .05) compared to other groups. All the psychophysiological responses were influenced by the covariates, however, the hierarchy of which covariate influences more or less these responses changes at each moment of the TIM. In conclusion, it was possible to identify 4 patterns of individuals whose differ in their respective psychophysiological responses before, during and after a TIM. However, although psychophysiological responses to TIM be influenced by variables of body composition, physical fitness, exercise-related behavioral profile and CPF asymmetry, the hierarchical order of influence power of these same variables differ at each moment of test

This thesis describes a set of software applications for real-time gesturally con- trolled interactions with music and sound. The applications for each system are varied but related, addressing unsolved problems in the field of audio and music technology. The three systems presented in this work capture 3D human motion with spatial sensors and map position data from the sensors onto sonic parameters. Two different spatial sensors are used interchangeably to perform motion capture: the radiodrum and the Xbox Kinect. The first two systems are aimed at creating immersive virtually-augmented environments. The first application uses human ges- ture to move sounds spatially in a 3D surround sound by physically modelling the movement of sound in a space. The second application is a gesturally controlled self- organized music browser in which songs are clustered based on auditory similarity. The third application is specifically aimed at extending musical performance through the development of a digitally augmented vibraphone. Each of these applications is presented with related work, theoretical and technical details for implementation, and discussions of future work. / Graduate

No description available.

El uso de cámaras de video para obtener información de la zona costera se ha extendido en los últimos años, debido al alcance espacial y temporal y el bajo costo del equipo. Con esto se han desarrollado técnicas para realizar mediciones en las imágenes. Estas requieren supervisión, lo que limita el número de imágenes que pueden ser analizadas. En este trabajo se desarrolló una metodología para clasificar imágenes digitales según las características morfodinámicas observadas y reducir el conjunto de imágenes a unos prototipos representativos de todo el conjunto. Esta clasificación se realiza por medio de mapas auto-organizativos. Los mapas auto organizativos son redes neuronales competitivas no supervisadas diseñadas para trabajar en espacios multidimensionales. La metodología desarrollada se aplico en dos playas; la primera, macromareal, se estudio la evolución morfodinámica según el modelo de Wright y Short 1984. En la segunda, micromareal, se estudio, la evolución de la línea de costa. / The use of video cameras to study the coastal zone has been developed in the last years, due to the spatial and temporal possibilities. A large number of techniques to obtain measurements of images were developed too. Most of these techniques require the human supervision, this limit the number of images to be analyzed. In this Thesis, a methodology to classify digital images based on the morphodynamic and reduce the number of images to a few prototypes is proposed. The classification is done by means of Sel-Organizative Maps (SOMs). The SOMs are a competitive and unsupervised neural networks, specifically designed to deal with high dimensional data The methodology proposed was applied to two beaches. In the first one, macrotidal, the morphodynamic evolution was studied using the Wright and Short Model 1984. In the last one, a microtidal beach, the coastline evolution was studied.