This thesis investigates on clinical applicability of a novel sensing technology in the areas of postural steadiness and stroke assessment. The mechanically simple Distributive Tactile Sensing approach is applied to extract motion information from flexible surfaces to identify parameters and disorders of human movement in real time. The thesis reports on the design, implementation and testing of smart platform devices which are developed for discrimination applications through the use of linear and non-linear data interpretation techniques and neural networks for pattern recognition. In the thesis mathematical models of elastic plates, based on finite element and finite difference methods, are developed and described. The models are used to identify constructive parameters of sensing devices by investigating sensitivity and accuracy of Distributive Tactile Sensing surfaces. Two experimental devices have been constructed for the investigation. These are a sensing floor platform for standing applications and a sensing chair for sitting applications. Using a linear approach, the sensing floor platform is developed to detect centre of pressure, an important parameter widely used in the assessment of postural steadiness. It is demonstrated that the locus of centre of pressure can be determined with an average deviation of 1.05mm from that of a commercialised force platform in a balance application test conducted with five healthy volunteers. This amounts to 0.4% of the sensor range. The sensing chair used neural networks for pattern recognition, to identify the level of motor impairment in people with stroke through performing functional reaching task while sitting. The clinical studies with six real stroke survivors have shown the robustness of the sensing technique to deal with a range of possible motion in the reaching task investigated. The work of this thesis demonstrates that the novel Distributive Tactile Sensing approach is suited to clinical and home applications as screening and rehabilitation systems. Mechanical simplicity is a merit of the approach and has potential to lead to versatile low-cost units.

Procedural human motion generation is still an open area of research. Most research into procedural human motion focus on two problem areas: the realism of the generated motion and the computation time required to generate the motion. Realism is a problem because humans are very adept at spotting the subtle nuances of human motion and so the computer generated motion tends to look mechanical. Computation time is a problem because the complexity of the motion generation algorithms results in lengthy processing times for greater levels of realism. The balancing human problem poses the question of how to procedurally generate, in real-time, realistic standing poses of an articulated human body. This report presents the balancing human algorithm that addresses both concerns: realism and computation time. Realism was addressed by integrating two existing algorithms. One algorithm addressed the physics of the human motion and the second addressed the prediction of the next pose in the animation sequence. Computation time was addressed by identifying techniques to simplify or constrain the algorithms so that the real-time goal can be met. The research methodology involved three tasks: developing and implementing the balancing human algorithm, devising a real-time simulation graphics engine, and then evaluating the algorithm with the engine. An object-oriented approach was used to model the balancing human as an articulated body consisting of systems of rigid-bodies connected together with joints. The attributes and operations of the object-oriented model were derived from existing published algorithms.

The fundamental task in human motion analysis is the extraction or capture of human motion and the established industrial technique is marker-based human motion capture. However, marker-based systems, apart from being expensive, are obtrusive and require a complex, time-consuming experimental setup, resulting in increased user discomfort. As an alternative solution, research on markerless human motion analysis has increased in prominence. In this thesis, we present three human motion analysis algorithms performing markerless tracking and classification from multiple-view studio-based video sequences using particle swarm optimisation and charting, a subspace learning technique.In our first framework, we formulate, and perform, human motion tracking as a multi-dimensional non-linear optimisation problem, solved using particle swarm optimisation (PSO), a swarm-intelligence algorithm. PSO initialises automatically, does not need a sequence-specific motion model, functioning as a blackbox system, and recovers from tracking divergence through the use of a hierarchical search algorithm (HPSO). We compare experimentally HPSO with particle filter, annealed particle filter and partitioned sampling annealed particle filter, and report similar or better tracking performance. Additionally we report an extensive experimental study of HPSO over ranges of values of its parameters and propose an automatic-adaptive extension of HPSO called as adaptive particle swarm optimisation. Next, in line with recent interest in subspace tracking, where low-dimensional subspaces are learnt from motion models of actions, we perform tracking in a low-dimensional subspace obtained by learning motion models of common actions using charting, a nonlinear dimensionality reduction tool. Tracking takes place in the subspace using an efficient modified version of particle swarm optimisation. Moreover, we perform a fast and efficient pose evaluation by representing the observed image data, multi-view silhouettes, using vector-quantized shape contexts and learning the mapping from the action subspace to shape space using multi-variate relevance vector machines. Tracking results with various action sequences demonstrate the good accuracy and performance of our approach.Finally, we propose a human motion classification algorithm, using charting-based low-dimensional subspaces, to classify human action sub-sequences of varying lengths, or snippets of poses. Each query action is mapped to a single subspace space, learnt from multiple actions. Furthermore we present a system in which, instead of mapping multiple actions to a single subspace, each action is mapped separately to its action-specific subspace. We adopt a multi-layered subspace classification scheme with layered pruning and search. One of the search layers involves comparing the input snippet with a sequence of key-poses extracted from the subspace. Finally, we identify the minimum length of action snippet, of skeletal features, required for classification, using competing classification systems as the baseline. We test our classification component on HumanEva and CMU mocap datasets, achieving similar or better classification accuracy than various comparable systems. human motion and the established industrial technique is marker-based human motion capture. However, marker-based systems, apart from being expensive, are obtrusive and require a complex, time-consuming experimental setup, resulting in increased user discomfort. As an alternative solution, research on markerless human motion analysis has increased in prominence. In this thesis, we present three human motion analysis algorithms performing markerless tracking and clas- si?cation from multiple-view studio-based video sequences using particle swarm optimisation and charting, a subspace learning technique. In our ?rst framework, we formulate, and perform, human motion tracking as a multi-dimensional non-linear optimisation problem, solved using particle swarm optimisation (PSO), a swarm-intelligence algorithm. PSO initialises automat- ically, does not need a sequence-speci?c motion model, functioning as a black- box system, and recovers from temporary tracking divergence through the use of a powerful hierarchical search algorithm (HPSO). We compare experiment- ally HPSO with particle ?lter, annealed particle ?lter and partitioned sampling annealed particle ?lter, and report similar or better tracking performance. Addi- tionally we report an extensive experimental study of HPSO over ranges of values of its parameters and propose an automatic-adaptive extension of HPSO called as adaptive particle swarm optimisation. Next, in line with recent interest in subspace tracking, where low-dimensional subspaces are learnt from motion models of actions, we perform tracking in a low-dimensional subspace obtained by learning motion models of common actions using charting, a nonlinear dimensionality reduction tool. Tracking takes place in the subspace using an e?cient modi?ed version of particle swarm optimisa- tion. Moreover, we perform a fast and e?cient pose evaluation by representing the observed image data, multi-view silhouettes, using vector-quantized shape contexts and learning the mapping from the action subspace to shape space us- ing multi-variate relevance vector machines. Tracking results with various action sequences demonstrate the good accuracy and performance of our approach. Finally, we propose a human motion classi?cation algorithm, using charting-based low-dimensional subspaces, to classify human action sub-sequences of varying lengths, or snippets of poses. Each query action is mapped to a single subspace space, learnt from multiple actions. Furthermore we present a system in which, instead of mapping multiple actions to a single subspace, each action is mapped separately to its action-speci?c subspace. We adopt a multi-layered subspace classi?cation scheme with layered pruning and search. One of the search lay- ers involves comparing the input snippet with a sequence of key-poses extracted from the subspace. Finally, we identify the minimum length of action snippet, of skeletal features, required for accurate classi?cation, using competing classi?ca- tion systems as the baseline. We test our classi?cation component on HumanEva and CMU mocap datasets, achieving similar or better classi?cation accuracy than

Image based motion capture is a problem that has recently gained a lot of attention in the domain of understanding human motion in computer vision. The problem involves estimating the 3D configurations of a human body from a set of images and has applications that include human computer interaction, smart surveillance, video analysis and animation. This thesis takes a machine learning based approach to reconstructing 3D pose and motion from monocular images or video. It makes use of a collection of images and motion capture data to derive mathematical models that allow the recovery of full body configurations directly from image features. The approach is completely data-driven and avoids the use of a human body model. This makes the inference extremely fast. We formulate a class of regression based methods to distill a large training database of motion capture and image data into a compact model that generalizes to predicting pose from new images. The methods rely on using appropriately developed robust image descriptors, learning dynamical models of human motion, and kernelizing the input within a sparse regression framework. Firstly, it is shown how pose can effectively and efficiently be recovered from image silhouettes that are extracted using background subtraction. We exploit sparseness properties of the relevance vector machine for improved generalization and efficiency, and make use of a mixture of regressors for probabilistically handling ambiguities that are present in monocular silhouette based 3D reconstruction. The methods developed enable pose reconstruction from single images as well as tracking motion in video sequences. Secondly, the framework is extended to recover 3D pose from cluttered images by introducing a suitable image encoding that is resistant to changes in background. We show that non-negative matrix factorization can be used to suppress background features and allow the regression to selectively cue on features from the foreground human body. Finally, we study image encoding methods in a broader context and present a novel multi-level image encoding framework called ‘hyperfeatures' that proves to be effective for object recognition and image classification tasks.

Judėjimas yra pagrindinis žmogaus veiklos komponentas. Buvo atlikta daug mėginimų atskleisti jo principus pasitelkiant fiziką bei dinamiką. Kartu kompiuterinė grafika ir robotų technika plėtoja šias pastangas, tačiau daug problemų lieka neišspręstų, netgi ir aprašant paprasčiausią atvejį: linijinį, tiesiaeigį, ritmišką ėjimą. Taigi netiesinių sistemų tyrimo tikslas yra surasti tvarką chaose; surasti įrodymų, kad nereguliari elgsena yra valdoma nedidelės deterministinių lygčių sistemos, pritaikant ją eksperimentiniams signalams laike. Tokio tyrimo sėkmės galima tikėtis, nustačius, kad šios sistemos būsenos kintamieji yra tvirtai suporuoti tarpusavyje. Chaotiškų sistemų tyrimų tikslas yra nustatyti dvi pagrindines jų savybes: dimensiją ir entropijos spektrą. Paprastai kalbant, dimensija yra dydis, parodantis diferencialinių lygčių skaičių, reikalingą aprašyti sistemai, o entropija yra dydis, parodantis informacijos apie sistemos būseną praradimą laiko bėgyje. Teigiama baigtinė entropija yra chaoso egzistavimo įrodymas. Šio darbo tikslas yra sukurti chaotiško signalo analizės sistemą, kuri leistų ištirti elementarius monotoniškus žmogaus rankos judesius dvimatėje plokštumoje. / Since we encounter many phenomena with irregular motion, e.g. the weather, turbulence, carbon resistor noise, chemical reactions and biological signals (human motion), we are tempted to investigate whether we could model the dynamics with nonlinear differential equations. Our aim is to find order within the chaos; to find evidence that the irregular behavior is governed by a small set of deterministic equations, using experimental time series. We might be successful in particular when the state variables of the system are strongly coupled. In this report, we will restrict ourselves to the determination of several properties that describe a chaotic system, including the dimension and entropy spectra. Loosely speaking, the dimension is a measure for the number of differential equations needed to describe the system, while the entropy is a measure for the loss of information about the state of the system in the course of time. Positive but finite entropy is a hall-mark of chaos. In this paper, we will describe few experiments that were performed on a portion of human motion data, and compare the results to theoretical model of system for signal analysis.

The performance of a classifier is affected by a number of factors including classifier type, the input features and the desired output. This thesis examines the impact of feature selection and classification problem division on classification accuracy and complexity. Proper feature selection can reduce classifier size and improve classifier performance by minimizing the impact of noisy, redundant and correlated features. Noisy features can cause false association between the features and the classifier output. Redundant and correlated features increase classifier complexity without adding additional information. Output selection or classification problem division describes the division of a large classification problem into a set of smaller problems. Problem division can improve accuracy by allocating more resources to more difficult class divisions and enabling the use of more specific feature sets for each sub-problem. The first part of this thesis presents two methods for creating feature-selected hierarchical classifiers. The feature-selected hierarchical classification method jointly optimizes the features and classification tree-design using genetic algorithms. The multi-modal binary tree (MBT) method performs the class division and feature selection sequentially and tolerates misclassifications in the higher nodes of the tree. This yields a piecewise separation for classes that cannot be fully separated with a single classifier. Experiments show that the accuracy of MBT is comparable to other multi-class extensions, but with lower test time. Furthermore, the accuracy of MBT is significantly higher on multi-modal data sets. The second part of this thesis focuses on input feature selection measures. A number of filter-based feature subset evaluation measures are evaluated with the goal of assessing their performance with respect to specific classifiers. Although there are many feature selection measures proposed in literature, it is unclear which feature selection measures are appropriate for use with different classifiers. Sixteen common filter-based measures are tested on 20 real and 20 artificial data sets, which are designed to probe for specific feature selection challenges. The strengths and weaknesses of each measure are discussed with respect to the specific feature selection challenges in the artificial data sets, correlation with classifier accuracy and their ability to identify known informative features. The results indicate that the best filter measure is classifier-specific. K-nearest neighbours classifiers work well with subset-based RELIEF, correlation feature selection or conditional mutual information maximization, whereas Fisher's interclass separability criterion and conditional mutual information maximization work better for support vector machines. Based on the results of the feature selection experiments, two new filter-based measures are proposed based on conditional mutual information maximization, which performs well but cannot identify dependent features in a set and does not include a check for correlated features. Both new measures explicitly check for dependent features and the second measure also includes a term to discount correlated features. Both measures correctly identify known informative features in the artificial data sets and correlate well with classifier accuracy. The final part of this thesis examines the use of feature selection for time-series data by using feature selection to determine important individual time windows or key frames in the series. Time-series feature selection is used with the MBT algorithm to create classification trees for time-series data. The feature selected MBT algorithm is tested on two human motion recognition tasks: full-body human motion recognition from joint angle data and hand gesture recognition from electromyography data. Results indicate that the feature selected MBT is able to achieve high classification accuracy on the time-series data while maintaining a short test time.

A long-standing goal in computer graphics is to create and control realistic motion for virtual human characters. Despite the progress made over the last decade, it remains challenging to design a system that allows a random user to intuitively create and control life-like human motions. This dissertation focuses on exploring theory, algorithms and applications that enable novice users to quickly and easily create and control natural-looking motions, including both full-body movement and hand articulations, for human characters. More specifically, the goals of this research are: (1) to investigate generative statistical models and physics-based dynamic models to precisely predict how humans move and (2) to demonstrate the utility of our motion models in a wide range of applications including motion analysis, synthesis, editing and acquisition. We have developed two novel generative statistical models from prerecorded motion data and show their promising applications in real time motion editing, online motion control, offline animation design, and motion data processing. In addition, we have explored how to model subtle contact phenomena for dexterous hand grasping and manipulation using physics-based dynamic models. We show for the first time how to capture physically realistic hand manipulation data from ambiguous image data obtained by video cameras.

Στην παρούσα διπλωματική εργασία γίνεται μελέτη και ανάλυση της ανθρώπινης κίνησης με σκοπό την αναγνώριση και τον χαρακτηρισμό της. Στο κεφάλαιο 1 παρουσιάζεται το θεωρητικό υπόβαθρο, περιγράφονται εν συντομία τα μέρη της ανάλυσης μιας ολοκληρωμένης διαδικασίας και ορίζονται οι έννοιες οι οποίες θα χρησιμοποιηθούν στην συνέχεια. Στο κεφάλαιο 2 παρουσιάζονται τα μοντέλα και οι τεχνικές που χρησιμοποιούνται κυρίως για την αφαίρεση φόντου σε μια εικόνα και γίνεται υλοποίηση και εφαρμογή, ορισμένων από αυτών, σε βίντεο με συγκεκριμένα χαρακτηριστικά με στόχο την σύγκριση των αποτελεσμάτων. Στο κεφάλαιο 3 παρουσιάζονται οι τεχνικές, και οι κύριοι αντιπρόσωποι αυτών, που χρησιμοποιούνται για την αναγνώριση κινούμενης οντότητας εντός μιας ακολουθίας εικόνων. Στο κεφάλαιο 4 γίνεται υλοποίηση αλγόριθμων, σύμφωνα με τις τεχνικές που αναπτύχτηκαν στο κεφάλαιο 3, και εφαρμογής τους σε βίντεο ώστε να μελετήσουμε τα αποτελέσματα, επίσης παρουσιάζονται οι δυνατότητες του simulink και πως μπορούμε να το χρησιμοποιήσουμε ως εργαλείο για να πετύχουμε ίδια αποτελέσματα με αυτά από την συγγραφή κώδικα σε matlab. Στο τελευταίο κεφάλαιο παρουσιάζονται οι τεχνικές που έχουν χρησιμοποιηθεί στην διεθνή βιβλιογραφία για την αναγνώριση κίνησης και στην συνέχεια γίνεται ανάπτυξη αλγόριθμου που χρησιμοποιεί ως αναγνωριστικό χαρακτηριστικό το κέντρο μάζας της κινούμενης οντότητας και μέσω αυτού προσδιορίζεται η μορφή της κίνησης. / n this thesis study and analysis of human motion for the recognition and characterization of. Chapter 1 presents the theoretical background, outlines the parts of analysis of an integrated process and defines the concepts that will used then. Chapter 2 presents the models and techniques are mainly used to remove a background image and is implementation and enforcement, some of them, in video certain characteristics in order to compare the results. At Chapter 3 presents the techniques, and the main representatives of those who used to identify an entity within a moving sequence of images. Chapter 4 is implementing algorithms under the techniques being developed in Chapter 3, and their application to video To study the results also shows the potential of simulink and how we can use it as a tool to achieve same results with the ones writing code in matlab. In the last chapter presents the techniques used in international literature to identify traffic and then becomes growth algorithm used as an identifier attribute the center of mass the moving entity and this is determined by the shape of motion.

The vast majority of products and processes in industry and academia require human interaction. Thus, digital human models (DHMs) are becoming critical for improved designs, injury prevention, and a better understanding of human behavior. Although many capabilities in the DHM field continue to mature, there are still many opportunities for improvement, especially with respect to posture- and motion-prediction. Thus, this thesis investigates the use of artificial neural network (ANN) for improving predictive capabilities and for better understanding how and why human behave the way they do. With respect to motion prediction, one of the most challenging opportunities for improvement concerns computation speed. Especially, when considering dynamic motion prediction, the underlying optimization problems can be large and computationally complex. Even though the current optimization-based tools for predicting human posture are relatively fast and accurate and thus do not require as much improvement, posture prediction in general is a more tractable problem than motion prediction and can provide a test bead that can shed light on potential issues with motion prediction. Thus, we investigate the use of ANN with posture prediction in order to discover potential issues. In addition, directly using ANN with posture prediction provides a preliminary step towards using ANN to predict the most appropriate combination of performance measures (PMs) - what drives human behavior. The PMs, which are the cost functions that are minimized in the posture prediction problem, are typically selected manually depending on the task. This is perhaps the most significant impediment when using posture prediction. How does the user know which PMs should be used? Neural networks provide tools for solving this problem. This thesis hypothesizes that the ANN can be trained to predict human motion quickly and accurately, to predict human posture (while considering external forces), and to determine the most appropriate combination of PM(s) for posture prediction. Such capabilities will in turn provide a new tool for studying human behavior. Based on initial experimentation, the general regression neural network (GRNN) was found to be the most effective type of ANN for DHM applications. A semi-automated methodology was developed to ease network construction, training and testing processes, and network parameters. This in turn facilitates use with DHM applications. With regards to motion prediction, use of ANN was successful. The results showed that the calculation time was reduced from 1 to 40 minutes, to a fraction of a second without reducing accuracy. With regards to posture prediction, ANN was again found to be effective. However, potential issues with certain motion-prediction tasks were discovered and shed light on necessary future development with ANNs. Finally, a decision engine was developed using GRNN for automatically selecting four human PMs, and was shown to be very effective. In order to train this new approach, a novel optimization formulation was used to extract PM weights from pre-existing motion-capture data. Eventually, this work will lead to automatically and realistically driving predictive DHMs in a general virtual environment.

In this thesis, we present a new class of object trackers that are based ona boosted Multiple Instance Learning (MIL) algorithm to track an object in a video sequence. We show how the scope of such trackers can be expanded to the tracking of articulated movements by humans that frequently<br>result in large frame-to-frame variations in the appearance of what needs to be tracked. To deal with the problems caused by such variations, we present a component-based MIL (CMIL) algorithm with boosted learning. The components are the output of an image segmentation algorithm and give the boosted MIL the additional degrees of freedom that it needs in order to deal with the large frame-to-frame variations associated with articulated movements. Furthermore we explored two enhancements of the basic CMIL tracking algorithm. The first is based on an extended definition of positive learning samples for CMIL tracking. This extended definition can filter out false-positive learning samples in order to increase the robustness of CMIL tracking. The second enhancement is based on a combined motion prediction framework with the basic CMIL tracking for resolving issues arising from large and rapid translational human movements. The need for appropriate motion transition can be satisfied by probabilistic modeling of motion. Experimental results show that the proposed approaches yield robust tracking performances in various tracking environments, such as articulate human movements as well as ground human movements observed from aerial vehicles.