Orientador: Prof. Dr. Marcos Duarte / Dissertação (mestrado) - Universidade Federal do ABC. Programa de Pós-Graduação em Engenharia Biomédica, 2016. / Atualmente os dispositivos de instrumentação biomecânica, sobretudo os sistemas de captura de movimento, são capazes de fornecer um grande número de dados. Seja para extrair informações para análise de diferenças entre grupos, para semi-automatização de procedimentos clínicos típicos de laboratórios de marcha ou para o desenvolvimento de sistemas inteligentes que fazem uso da informação do movimento, o reconhecimento de padrões de movimento é uma necessidade. Estes dados podem ser reduzidos a séries temporais, sendo assim estamos diante de um problema de mineração de dados de séries temporais. No entanto enquanto a maior parte das pesquisas se concentram em tarefas de mineração, o problema fundamental de como representar uma série temporal ainda não foi plenamente abordado até agora. Este projeto objetiva estudar representações simbólicas para séries temporais provenientes de dados de captura de movimentos, uma abordagem ainda muito pouco explorada. Propondo a elaboração de um algoritmo que realiza o mapeamento entre um conjunto de séries temporais e uma sequência de símbolos, levando em consideração informações do domínio da biomecânica e controle motor. Um conjunto de movimentos discretos foram convertidos em uma representação simbólica, a partir da qual foi realizado um agrupamento hierárquico e classificação em 10 atividades rotuladas, com uma exatidão de 84.72% e um padrão sequencial foi detectado no andar. Estes resultados foram obtidos em um tempo de processamento relativamente baixo e a partir de apenas 3 ângulos no plano sagital do membro inferior direito. Com este trabalho atingimos o objetivo proposto e estabelecemos uma representação simbólica do movimento, denominada palavras do movimento, com boa parte das características almejadas. É uma representação simples e prática, a partir da qual foi possível estabelecer uma métrica que quantifica a similaridade entre movimentos. / Currently biomechanical instrumentation devices, especially motion capture systems, are able to provide a large amount of data. Be it for extract information to compare differences among different groups, for semi-automation of typical clinical gait analysis procedures or for developing intelligent systems that make use of motion information, recognition of motion patterns is a need. These data can be reduced to time series, so we are facing a problem of mining time series data. However, while most of the research communities have concentrated on the mining tasks, the fundamental problem on how to represent a time series has not yet been fully addressed so far. This project aims to study symbolic representation for time series data from motion capture, an approach still not much explored. The development of the algorithm that performs the mapping between a set of time series and a sequence of symbols, taking into account information from the field of biomechanics and motor control. A set of discrete motions were converted to a symbolic representation, from which we performed a hierarchical clustering and classified in 10 labeled activities, with an accuracy of 84.72% and a sequential pattern was detected in gait. These results were obtained at a relatively low processing time and from only 3 angles in the sagittal plane of the right lower limb. We achieved our objective and established a symbolic representation of the movement, called motion words, with most of it desired characteristics. It is a simple and practical representation, from which it was possible establish a metric that quantifies the similarity between movements.

Face aux limites auxquelles doivent faire face les systèmes optoélectroniques (matériel lourd, champ de mesure limité), les capteurs inertiels constituent une alternative prometteuse pour la mesure du mouvement humain. Grâce aux dernières avancées techniques, notamment en termes de miniaturisation des capteurs, leur utilisation en ambulatoire c’est-à-dire de façon autonome et embarquée est devenue possible. Mais ces opérations de miniaturisation ne sont pas sans effet sur les performances de ces capteurs. En effet, une telle mesure est dégradée par différents types de perturbations (stochastiques et déterministes) qui sont alors propagées au cours du processus dit de fusion des données visant à estimer l'orientation des segments humains. Classiquement, cette opération est réalisée à l'aide d'un filtre de Kalman dont le rôle est justement d'estimer une grandeur à partir d'une mesure bruitée en la confrontant à un modèle d'évolution.Dans ce contexte, nous proposons diverses méthodologies dans le but d'accéder à une mesure suffisamment précise pour être exploitée dans le cadre de l'analyse du mouvement humain. La première partie de cette thèse se focalise sur les capteurs. Tout d'abord, nous étudions les bruits de mesure issus des capteurs inertiels, puis nous leur attribuons un modèle afin de les prendre en compte au sein du filtre de Kalman. Ensuite, nous analysons les procédures de calibrage et évaluons leurs effets réels sur la mesure afin d'émettre quelques propositions en termes de compromis performance/facilité de réalisation.Dans une seconde partie, nous nous consacrons à l'algorithme de fusion des données. Après avoir proposé un filtre de Kalman adapté à la mesure du mouvement humain, nous nous focalisons sur un problème récurrent à ce stade : l'identification des matrices de covariance dont le rôle est d'attribuer une caractérisation globale aux erreurs de mesure. Cette méthode, basée sur une confrontation de la mesure avec une référence issue d'un système optoélectronique, met en évidence la nécessité de traiter ce problème rigoureusement.Dans une troisième partie, nous commençons à aborder les problèmes liés à l'utilisation des capteurs inertiels pour la mesure du mouvement humain, notamment le calibrage anatomique et le positionnement des capteurs.En conclusion, les gains apportés par les diverses propositions avancées dans cette thèse sont évalués et discutés. / To face the limits of optoelectronic systems (heavy device, restricted measurement field), inertial sensors are a promising alternative for human motion analysis. Thanks to the latest technical advancements like sensor miniaturization, they can now work autonomously which makes possible to directly embed them on the human segments. But, as a counterpart of these developments, inertial sensor measurement still suffers from both stochastic and deterministic perturbations. The induced errors then propagate over the so-called fusion algorithm used to estimate human segment orientation. A common tool to perform such an operation is the Kalman filter that estimates unknown variables by correcting noisy measurements by the use of a dynamic model.With the aim of achieving a sufficiently accurate measurement to perform human motion analysis, various methodologies are proposed in the present work. The first part of this thesis focuses on the sensors. First, inertial sensor noises are studied and modeled in order to be integrated into the Kalman filter. Calibration processes as their effects over the measurement are for that purposed analyzed. Some recommendations are thus proposed to reach a compromise between calibration performance and complexity.In a second part, the data fusion algorithm is approached. A specific Kalman filter dedicated to human motion measurement is first proposed. Then, a recurrent problem is studied in details: the definition of the covariance matrix that represents a globalcharacterization of the measurement errors. Considering an optoelectronic system as a reference to compare inertial measurement, a method is proposed for this covariance matrix identification, which also highlights the need to address this problem rigorously.In a third part, we begin to address the use of inertial sensors for human motion analysis by focusing on models and IMU-to-segment calibration.To conclude, the benefits made by the proposed methodologies are evaluated and discussed.

De etiologia multifatorial, a síndrome dolorosa fêmoro-patelar (SDPF) acomete de 7 a 15% da população, em sua maioria jovens, adultos e ativos. Causa dor difusa anterior ou retropatelar e é exacerbada durante atividades funcionais, como subir e descer degraus, permanecer por um período prolongado sentado, agachado ou ajoelhado (COWAN, BENNELL e HODGES, 2002). Tanto o processo de avaliação quanto a evolução do tratamento baseiam-se, fundamentalmente, no relato de dor feito pelo paciente e na incapacidade funcional. Diagnosticar corretamente o distúrbio possibilitaria um tratamento mais adequado e permitiria o acompanhamento do quadro evolutivo do paciente ao longo do tratamento proposto. Desta maneira, propôs-se a análise de alguns parâmetros biomecânicos para melhor caracterização destes indivíduos. 33 voluntárias - 11 com diagnóstico médico de dor fêmoro-patelar e 22 clinicamente normais - foram submetidas à avaliação por meio de eletromiografia de superfície das porções medial e lateral do quadríceps; análise tridimensional de movimento; e plataforma de força durante a subida de degraus. As voluntárias do grupo SDFP foram caracterizadas por um atraso no tempo de atividade máxima do músculo vasto medial em relação ao vasto lateral (4 ms), aumento da rotação externa do fêmur (3,14 graus) e menor variação global do torque (1,14 Nm), quando comparadas aos indivíduos do grupo controle. Os resultados obtidos complementam-se e ajudam a entender características de comportamento neuromotor em indivíduos com dor fêmoro-patelar. Conjuntamente, esses instrumentos apresentam grande potencial para a avaliação e classificação dos indivíduos com relação à síndrome. / From multifactorial etiology, patellofemoral pain syndrome (PFPS) affects 7 to 15% of the population, mostly young, adults and active people. It causes anterior or behind patella diffuse pain and is exacerbated during functional activities such as climbing and descending stairs or stay for a long time sitting, kneeling or crouching down (COWAN, BENNELL e HODGES, 2002). The evaluation process as much as the evolution of treatment is fundamentally based in the reporting of pain by the patient and the functional disability. Correct diagnostic of the disorder would allow a more appropriate treatment as well as to keep up with the patient evolution during the proposed treatment. Thus, it was proposed the biomechanical analysis of some parameters to better characterize these individuals. 33 female volunteers - 11 with medical diagnosis of patellofemoral pain and 22 clinically normal - were submitted to evaluation of surface electromyography of the medial and lateral portions of the quadriceps, three-dimensional motion analysis and force plate during the ascent of steps. The volunteers of the PFPS group were characterized by a delay in time of maximum activity of the vastus medialis muscle related to the vastus lateralis (4 ms), increased femoral external rotation (3.14 degrees) and less overall variation in the torque (1.14 Nm) when compared to individuals in the control group. The results complement each other and help to understand features of neuromotor behavior in individuals with patellofemoral pain. Together, these tools have great potential for assessment and classification of these individuals related to the syndrome.

CrossFit® has gained popularity as a functional training method that caters to individuals with varying levels of fitness, including those with little to no athletic background. Regardless, shoulders represent the highest injury rate in this sport, most frequently associated with pull-up variations. Therefore, ensuring safety during these movements is crucial. This thesis utilized optical motion tracking, force sensors, and Electromyography to measure nine experienced non-professional CrossFit® athletes performing three pull-up techniques used in CrossFit® (strict, kipping, and butterfly). Data were used in musculoskeletal modeling to assess and compare kinematics, kinetics, and muscle activations. Findings were then associated with common injury factors observed in overhead athletes. The novel dataset generated in this study demonstrated excellent repeatability and revealed significant differences between the techniques for most kinematic and kinetic variables, while variations in muscle activations were less systematic. Results indicated shoulder joint angles linked to impingement risk, with the kipping pull-up displaying more prominent risk factors. The butterfly pull-up exhibited extreme angular velocities and negative joint powers; however, direct associations with injuries could not be established. Notably, shoulder joint moments were high in all three techniques, suggesting significant muscle stresses. These results underscore the importance of individualized training load management to mitigate the risk of overuse injuries. Furthermore, the evidence presented in this study supports the importance of maintaining proper mechanics and developing strength before attempting kipping and butterfly pull-ups. The high joint moments, combined with severe positions and velocities observed during these techniques, highlight the necessity of prioritizing foundational strength and technique to minimize injury risks. / CrossFit® har blivit populärt som en funktionell träningsmetod som lockar individueller med olika fitnessnivåer, även de utan tidigare idrottserfarenhet. Trots detta utgör axelskada den högsta skadefrekvensen inom denna sport och är oftast förknippad med räckhäv varianter. Därför är det avgörande att garantera säkra övningarna. Denna studie använde optisk rörelsespårning, kraftsensorer och elektromyografi för att mäta nio tränade icke-professionella CrossFit® atleter utförande av tre olika räckhäv tekniker som används inom CrossFit® (strikta, kipping och butterfly). Data användes för muskuloskeletala modeller för att uppskatta och jämföra kinematik, kinetik och muskelaktivering. Resultaten förknippades med till vanliga skadefaktorer som observeras hos overhead-idrottare. Det nya datasetet som genererades i denna studie visade utmärkt repeterbarhet och avslöjade signifikanta skillnader mellan teknikerna för de flesta kinematiska och kinetiska variablerna, medan variationer i muskelaktiveringar var mindre systematiska. Resultaten visade på axelledens vinklar som var kopplade till inklämningsrisk, där kipping pull-up uppvisade mer framträdande riskfaktorer. Butterfly pull-up uppvisade extrema vinkelhastigheter och negativa ledkrafter; men direkta samband med skador har inte upprättats. Det är anmärkningsvärt att axelledsmomenten var höga i alla tre tekniker, vilket tyder på betydande muskelspänningar. Dessa resultat understryker vikten av individualiserad hantering av träningsbelastning för att minska risken för överbelastningsskador. Dessutom stöder de bevis som presenteras i den här studien vikten av att behålla rätt mekanik och utveckla styrka innan man försöker göra kipping och butterfly pull-ups. De höga ledmomenten i kombination med svåra positioner och hastigheter som observerades under dessa tekniker lyfter fram behovet av att prioritera grundläggande styrka och teknik för att minimera risk för skador.

The purpose of this thesis was to investigate how motion analysis can be used as service to reduce the rate of upper-body musculoskeletal disorders (MSDs) in forestry, and propose a design based on the identified requirements. The project was carried out in collaboration with a Power Tools and Accessories Manufacturer (PTAM). Work-related upper-limb musculoskeletal disorders cost the European Union about 0.5-2% of its Gross Domestic Product yearly, cause immeasurable human suffering, and strain societal resources, not to mention the effect on the finances of the firm employing the afflicted individual. Forestry is the most prolific industry in terms musculoskeletal disorders, with a prevalence rate of circa 6 recorded MSDs per 100 workers and year. MSDs can be prevented, if individuals are aware of the risk. However, the traditional ways of creating awareness are clearly not working, considering the high rates of MSDs. Self-tracking technologies are therefore proposed as a new, more effective, way of increasing risk awareness among forestry workers. By increasing awareness, exposure and subsequent risk can be reduced. This thesis was initiated by a literature study, user observations and an interview study at PTAM, followed by a technical evaluation and synthesis of the ergonomic parameters. Early concepts were developed, tested with users, and reworked according to their feedback. Some relevant ethical dimensions of this innovation were also considered as the technology has great potential for both positive and negative influence, which can have profound effects on the users and affect their willingness to engage in self-tracking. The proposed solution is called the Ergonomic Risk Assessment service (ERA). The ERA is a conceptual service that uses motion analysis to assess the risk of developing musculoskeletal disorders caused by hazardous working postures. The ERA is composed of two parts: the tracking unit and the data analysis. The tracking unit is a sports shirt with integrated Bluetooth and IMU sensors which gathers motion data that is used in the automated analysis to assess exposure and risk. The assessments are communicated back to the user via real-time indications and long-term overviews. As a service, the ERA has potential to generate considerate value and have a significant impact on the social and economic sustainability for nations, firms, and individuals, if designed and implemented in an ethically responsible manner. / Syftet med den här rapporten var att undersöka hur rörelseanalys kunde användas som en tjänst för att minska förekomsten av muskuloskeletala besvär (MSDs) i överkroppen bland skogsarbetare, samt att föreslå en design av ett sådant system baserat på de identifierade behoven och kraven. Projektet utfördes i samarbete med PTAM, ett företag som tillverkar motordrivna verktyg och tillbehör. Arbetsrelaterade muskuloskeletala besvär i överkroppen kostar årligen EU ca 0.5–2% av dess BNP, orsakar ett omätbart lidande och belastar samhällets resurser, för att inte nämna dess inverkan på ekonomin bland de drabbade individernas arbetsgivare. Bland de olika industrierna är skogsbruk värst drabbat av MSDs, med en prevalens på ca 6 anmälda MSDs per 100 anställda och år. MSDs kan förebyggas, om individerna är medvetna om risken. De traditionella sätten att skapa riskmedvetenhet fungerar uppenbarligen inte, med tanke på den höga förekomsten av besvär. Därför föreslås teknik för självövervakning som ett nytt och mer effektivt sätta att öka riskmedvetandet bland skogsarbetare. Genom att öka medvetenheten kan exponeringen och risken att utveckla muskuloskeletala besvär minskas Den här masteruppsatsen inleddes av en litteraturstudie, användarobservationer, och en intervjustudie på PTAM, som följdes av en teknisk utvärdering och en syntes av de ergonomiska parametrarna. Preliminära koncept utvecklades, testades med användare, och anpassades efter deras återkoppling. Några relevanta etiska dimensioner av den här innovationen beaktades, då tekniken har potential till positiv så väl som negativ inverkan, vilket kan ha djupgående effekter på användarna och påverka deras villighet att använda sig av självövervakningssystemet. Den föreslagna lösningen kallas för ERA (Ergonomic Risk Assessment service). ERA är en konceptuell tjänst som använder sig av rörelseanalys för att uppskatta risken att utveckla muskuloskeletala besvär orsakade av skadliga arbetsställningar. Lösningen består av två delar: spårningsanordningen och dataanalysen. Spårningsanordningen består av Bluetooth- och tröghetssensorer vilka samlar rörelsedata, som i sin tur används i analysen för att utvärdera exponering och risk. Utvärderingen kommuniceras sedan till användaren via realtidsindikationer samt långtidsöversikter. Som en tjänst har ERA potential att generera avsevärt värde och ha en betydande påverkan på den sociala och ekonomiska hållbarheten för nationer, företag, och individer, och det designas och implementeras på ett etiskt, ansvarsfullt vis.

Human motion analysis is one of the major problems in computer vision research. It deals with the study of the motion of human body in video data from different aspects, ranging from the tracking of body parts and reconstruction of 3D human body configuration, to higher level of interpretation of human action and activities in image sequences. When human motion is observed through video camera, it is perspectively distorted and may appear totally different from different viewpoints. Therefore it is highly challenging to establish correct relationships between human motions across video sequences with different camera settings. In this work, we investigate the geometric invariance in the motion of human body, which is critical to accurately understand human motion in video data regardless of variations in camera parameters and viewpoints. In human action analysis, the representation of human action is a very important issue, and it usually determines the nature of the solutions, including their limits in resolving the problem. Unlike existing research that study human motion as a whole 2D/3D object or a sequence of postures, we study human motion as a sequence of body pose transitions. We also decompose a human body pose further into a number of body point triplets, and break down a pose transition into the transition of a set of body point triplets. In this way the study of complex non-rigid motion of human body is reduced to that of the motion of rigid body point triplets, i.e. a collection of planes in motion. As a result, projective geometry and linear algebra can be applied to explore the geometric invariance in human motion. Based on this formulation, we have discovered the fundamental ratio invariant and the eigenvalue equality invariant in human motion. We also propose solutions based on these geometric invariants to the problems of view-invariant recognition of human postures and actions, as well as analysis of human motion styles. These invariants and their applicability have been validated by experimental results supporting that their effectiveness in understanding human motion with various camera parameters and viewpoints.

L'épaule est souvent affectée par des troubles musculo-squelettiques. Toutefois, leur évaluation est limitée à des mesures qualitatives qui nuisent à la spécificité et justesse du diagnostic. L'analyse de mouvement tridimensionnel pourrait complémenter le traitement conventionnel à l'aide de mesures quantitatives fonctionnelles. L'interaction entre les articulations de l'épaule est estimée par le rythme scapulo-huméral, mais la variabilité prononcée qu'il affiche nuit à son utilisation clinique. Ainsi, l'objectif général de cette thèse était de réduire la variabilité de la mesure du rythme scapulo-huméral. L'effet de la méthode de calcul du rythme scapulo-huméral et des conditions d'exécution du mouvement (rotation axiale du bras, charge, vitesse, activité musculaire) ont été testées. La cinématique des articulations de l'épaule a été calculé par chaîne cinématique et filtre de Kalman étendu sur des sujets sains avec un système optoélectronique. La méthode usuelle de calcul du rythme scapulo-huméral extrait les angles d'élévation gléno-humérale et de rotation latérale scapulo-thoracique. Puisque ces angles ne sont pas co-planaires au thorax, leur somme ne correspond pas à l'angle d'élévation du bras. Une nouvelle approche de contribution articulaire incluant toutes les rotations de chaque articulation est proposée et comparée à la méthode usuelle. La méthode usuelle surestimait systématiquement la contribution gléno-humérale par rapport à la méthode proposée. Ce nouveau calcul du rythme scapulo-huméral permet une évaluation fonctionnelle dynamique de l'épaule et réduit la variabilité inter-sujets. La comparaison d'exercices de réadaptation du supra-épineux contrastant la rotation axiale du bras a été réalisée, ainsi que l'effet d'ajouter une charge externe. L'exercice «full-can» augmentait le rythme scapulo-huméral et la contribution gléno-humérale ce qui concorde avec la fonction du supra-épineux. Au contraire, l'exercice «empty-can» augmentait la contribution scapulo-thoracique ce qui est associé à une compensation pour éviter la contribution gléno-humérale. L'utilisation de charge externe lors de la réadaptation du supra-épineux semble justifiée par un rythme scapulo-huméral similaire et une élévation gléno-humérale supérieure. Le mouvement de l'épaule est souvent mesuré ou évalué en condition statique ou dynamique et passive ou active. Cependant, l'effet de ces conditions sur la coordination articulaire demeure incertain. La comparaison des ces conditions révélait des différences significatives qui montrent l'importance de considérer les conditions de mouvement pour l'acquisition ou la comparaison des données. / The shoulder is often afflicted by musculo-skeletal diseases. However, shoulder evaluation is limited to qualitative measures affecting the specificity and accuracy of diagnosis. Three-dimensional motion analysis could complement conventional treatment with quantitative functional measures. The interaction between shoulder joints is estimated by the scapulo-humeral rhythm, but its pronounced variability of measurement ampers appropriate clinical use. Hence, the main objective of the thesis was to reduce the variability in the measure of the scapulo-humeral rhythm. The effect of the calculation method and conditions of motion (arm axial rotation, load, velocity, muscular activity) were tested on healthy subjects. Shoulder joints kinematics were obtained with an optoelectronic system using a kinematical chain and an extended Kalman filter. The common method of scapulo-humeral rhythm calculation extracts the angles of gleno-humeral elevation and scapulo-thoracic lateral rotation. Since these angles are not co-planar, their sum does not correspond to the arm elevation angle. A novel approach of joint contribution including all the rotations of each joint was developed and compared to the common method. The common method systematically overestimated gleno-humeral contribution in comparison to the proposed method. The new scapulo-humeral rhythm calculation allows a dynamic functional evaluation of the shoulder and reduces inter-subject variability. Shoulder coordination was compared on typical supraspinatus rehabilitation exercises contrasting in arm axial rotation, and the effect of adding external load was tested. The ''full-can'' exercise increased the scapulo-humeral rhythm and gleno-humeral contribution, which correspond to supraspinatus function. In contrast, the ''empty-can'' exercise increased scapulo-thoracic contribution, which is associated to compensation to avoid gleno-humeral contribution. The use of external load in supraspinatus rehabilitation seems justified by a similar scapulo-humeral rhythm and superior gleno-humeral elevation. Shoulder motion is often measured or evaluated in static or dynamic and passive or active conditions. However, the effect of these conditions on shoulder coordination remains uncertain. The comparison of these conditions revealed significant differences on shoulder kinematics showing the importance to consider conditions of motion for acquisition or comparison of data.

❆❜str❛❝t ❚✐t❧❡✿ ❆♥❛❧②s✐s ♦❢ ♠✉s❝❧❡ ❛❝t✐✈✐t② ❞✉r✐♥❣ st❡❡r✐♥❣ ✇❤❡❡❧ ♠♦✈❡♠❡♥t ❖❜❥❡❝t✐✈❡s✿ ❚❤❡ ❛✐♠ ♦❢ t❤✐s t❤❡s✐s ✐s t♦ ♣r♦❝❡ss ❛♥ ❛✉t♦♠♦t✐✈❡ ❡r❣♦♥♦♠② ✐ss✉❡✳ ❋✉t❤❡r♠♦r❡✱ t♦ ❛♥❛❧②s❡ ✉♣♣❡r ❡①tr❡♠✐t② ♠♦✈❡♠❡♥t ❛♥❞ ❡❧❡❝tr✐❝❛❧ ❛❝t✐✈✐t② ♦❢ s♣❡✲ ❝✐✜❡❞ s❤♦✉❧❞❡r ♠✉s❝❧❡s ❞✉r✐♥❣ st❡❡r✐♥❣ ✇❤❡❡❧ ♠♦✈❡♠❡♥t ✇✐t❤ ❛ ❞✐✛❡r❡♥t ❤❛♥❞ ♣❧❛❝❡♠❡♥t✳ ❆♥♦t❤❡r ❛✐♠ ✐s t♦ ❝♦♠♣❛r❡ ♠❡❛s✉rr❡❞ ❡❧❡❝tr♦♠②♦❣r❛♣❤✐❝ ❞❛t❛ ✇✐t❤ ❛♥ ♦♣t✐♠❛❧ r❛♥❣❡ ♦❢ ♠✉s❝❧❡ ❛❝t✐✈✐t② ❞✉r✐♥❣ r❡♣❡t✐t✐✈❡ t❛s❦s✳ ▼❡t❤♦❞s✿ ❚❤❡ t❤❡s✐s ❤❛s ❛ ❝❤❛r❛❝t❡r ♦❢ ❛ ♣✐❧♦t st✉❞②✳ ❚❡♦r❡t✐❝❛❧ ❜❛s✐s ❤❛s ❜❡❡♥ ♦❜t❛✐♥❡❞ ❜② ❛ r❡✈✐❡✇ ♦❢ ❧✐t❡r❛t✉r❡ ❞❡❛❧ ✇✐t❤ s✐♠✐❧❛r q✉❡st✐♦♥s✳ ❊①♣❡r✐♠❡♥t❛❧ ♣❛rt ❤❛s ❜❡❡♥ ❝♦♥❝❡r♥❡❞ ✇✐t❤ ❡❧❡❝tr✐❝❛❧ ❛❝t✐✈✐t② ♦❢ s♣❡❝✐✜❡❞ s❤♦✉❧❞❡r ♠✉s❝❧❡s ❛♥❞ ♠♦t✐✲ ♦♥ ❛♥❛❧②s✐s ♦❢ ✉♣♣❡r ❡①tr❡♠✐t② ❞✉r✐♥❣ ❝♦♥tr❛❧❛t❡r❛❧ ❛♥❞ ✐♣s✐❧❛t❡r❛❧ st❡❡r✐♥❣ ✇❤❡❡❧ ♠♦✈❡♠❡♥t ✇✐t❤ ❛ ❞✐✛❡r❡♥t ❤❛♥❞ ♣❧❛❝❡♠❡♥t✳ ❖♥❧② t✇♦ s✉❜❥❡❝ts ❤❛s ❜❡❡♥ ❝❤♦s❡♥ ❞✉❡ t♦ ❛ t❡❝❤♥✐❝❛❧ ❞✐✣❝✉❧t② ♦❢ ❡①♣❡r✐♠❡♥t✳ ❚❤❡ ❡❧❡❝tr✐❝❛❧ ❛❝t✐✈✐t② ♦❢ ♠✉s❝❧❡s ❤❛s ❜❡❡♥ ❛♥❛❧②s❡❞ ✇✐t❤ s✉r❢❛❝❡ ❡❧❡❝tr♦♠②♦❣r❛♣❤②✳ ❚❤❡r❡ ❤❛s ❜❡❡♥ ♠❡❛s✉r❡❞ ❢♦❧❧♦✇✐♥❣ ♠✉s❝❧❡s✿ ♠✳ ❞❡❧t♦✐❞❡✉s ♣❛rs ❛♥t❡r✐♦r✱ ♠✳ ❞❡❧t♦✐❞❡✉s ♣❛rs ♠❡❞✐❛❧✐s✱ ♠✳ ♣❡❝t♦r❛❧✐s ♠❛❥♦r ♣❛rs ❝❧❛✈✐❝✉❧❛r✐s ❛ ♠✳ tr✐❝❡♣s ❜r❛❝❤✐✐ ❝❛♣✉t ❧♦♥❣✉♠✱ ♠✳ ♣❡❝t♦r❛❧✐s ♠❛❥♦r ♣❛rs st❡r♥♦❝♦st❛❧✐s✱ ♠✳ ❜✐❝❡♣s ❜r❛❝❤✐✐✱ ♠✳ ❞❡❧t♦✐❞❡✉s ♣❛rs ♣♦st❡r✐♦r✱ ♠✳ ✐♥❢r❛s♣✐✲ ♥❛t✉s ❛ ♠✳ tr✐❝❡♣s ❜r❛❝❤✐✐ ❝❛♣✉t ❧❛t❡r❛❧❡✳ ❚❤❡ ❡❧❡❝tr✐❝❛❧ ❛❝t✐✈✐t② ♦❢...

The name of the work: Shape changes in the axial system during respiration Goal of the work: the main objective is to perform a case study of brass parameters and changes the shape of the trunk during the ventilation maneuver, prepare the correct methodology of research and to conduct separate measurements. The thesis will be divided into two parts, the first will contain the theoretical background to the subject, in the second part will then be described by the research, where I measured the change in the volume of the lungs in relation to time, assess the impact of implemented ventilation maneuver on the individual probands and examine the respiratory dynamics in selected ventilation maneuver. Methods: in the framework of the fulfilment of the tasks of this work have been carried out experiments with the use of 3D kinematic record with Qualisys and spirometric measurement. The experiment involved three probands aged 20 to 40 specific common features. Each of the proband performed during the measurement of quiet breathing, maximum inhalation and exhalation and ventilation maneuver. Measurements were carried out in one day under standard conditions, measurements made on the day of probands quiet breathing, maximum inhalation and exhalation and ventilation maneuver (5 min total). Data records were...

L'objectif de ce travail est de développer des méthodes avancées d'apprentissage pour l’analyse et l'interprétation automatique du mouvement humain à partir de sources d'information diverses, telles que les images, les vidéos, les cartes de profondeur, les données de type “MoCap” (capture de mouvement), les signaux audio et les données issues de capteurs inertiels. A cet effet, nous proposons plusieurs modèles neuronaux et des algorithmes d’entrainement associés pour l’apprentissage supervisé et semi-supervisé de caractéristiques. Nous proposons des approches de modélisation des dépendances temporelles, et nous montrons leur efficacité sur un ensemble de tâches fondamentales, comprenant la détection, la classification, l’estimation de paramètres et la vérification des utilisateurs (la biométrie). En explorant différentes stratégies de fusion, nous montrons que la fusion des modalités à plusieurs échelles spatiales et temporelles conduit à une augmentation significative des taux de reconnaissance, ce qui permet au modèle de compenser les erreurs des classifieurs individuels et le bruit dans les différents canaux. En outre, la technique proposée assure la robustesse du classifieur face à la perte éventuelle d’un ou de plusieurs canaux. Dans un deuxième temps nous abordons le problème de l’estimation de la posture de la main en présentant une nouvelle méthode de régression à partir d’images de profondeur. Dernièrement, dans le cadre d’un projet séparé (mais lié thématiquement), nous explorons des modèles temporels pour l'authentification automatique des utilisateurs de smartphones à partir de leurs habitudes de tenir, de bouger et de déplacer leurs téléphones. Dans ce contexte, les données sont acquises par des capteurs inertiels embraqués dans les appareils mobiles. / The research goal of this work is to develop learning methods advancing automatic analysis and interpreting of human motion from different perspectives and based on various sources of information, such as images, video, depth, mocap data, audio and inertial sensors. For this purpose, we propose a several deep neural models and associated training algorithms for supervised classification and semi-supervised feature learning, as well as modelling of temporal dependencies, and show their efficiency on a set of fundamental tasks, including detection, classification, parameter estimation and user verification. First, we present a method for human action and gesture spotting and classification based on multi-scale and multi-modal deep learning from visual signals (such as video, depth and mocap data). Key to our technique is a training strategy which exploits, first, careful initialization of individual modalities and, second, gradual fusion involving random dropping of separate channels (dubbed ModDrop) for learning cross-modality correlations while preserving uniqueness of each modality-specific representation. Moving forward, from 1 to N mapping to continuous evaluation of gesture parameters, we address the problem of hand pose estimation and present a new method for regression on depth images, based on semi-supervised learning using convolutional deep neural networks, where raw depth data is fused with an intermediate representation in the form of a segmentation of the hand into parts. In separate but related work, we explore convolutional temporal models for human authentication based on their motion patterns. In this project, the data is captured by inertial sensors (such as accelerometers and gyroscopes) built in mobile devices. We propose an optimized shift-invariant dense convolutional mechanism and incorporate the discriminatively-trained dynamic features in a probabilistic generative framework taking into account temporal characteristics. Our results demonstrate, that human kinematics convey important information about user identity and can serve as a valuable component of multi-modal authentication systems.