Les méthodes de trajectographie conventionnelles par mesures d’angle supposent que la source est en mouvement rectiligne uniforme tandis que l’observateur est manœuvrant. Dans cette thèse, nous remettons en cause cette hypothèse en proposant un autre modèle de cinématique de la source : le mouvement circulaire uniforme. Nous prouvons qu’une telle trajectoire est observable à partir d’un observateur en mouvement rectiligne uniforme. Puis, nous étudions l’apport de mesures additionnelles de fréquence ou la faisabilité de la trajectographie par mesures de distances. Le cas d’une source en mouvement rectiligne uniforme et d’un observateur manœuvrant est étudié pour ce dernier type de mesures. Chaque cas donne lieu à une analyse de l’observabilité de la trajectoire de la source et à la mise au point de l’estimateur du maximum de vraisemblance. Nous montrons que ce dernier s’avère le plus souvent efficace. / The conventional bearings-only target motion analysis methods assume that the source is in constant velocity motion (constant speed and heading) while the observer maneuvers. In this thesis, we reassess this hypothesis and propose another model of the kinematics of the source: the constant turn motion (an arc of circle followed at constant speed). We prove that this kind of trajectory is observable by an observer in constant velocity motion. Then, we study the contribution of the addition of frequency measurements or the feasibility of target motion analysis methods that use range only measurements. The case of a source in constant velocity motion with a maneuvering observer is examined for this last type of measurements. Each case leads to an analysis of the observability of the trajectory of the source and to the development of the associated maximum likelihood estimator. We show that this estimator often appears to be efficient.

Suite à une lésion (ex: blessure médullaire, accident vasculaire cérébral) ou une maladie neurodégénérative (ex: maladie de Parkinson), le système nerveux central humain peut être sujet à de multiples déficiences sensori-motrices menant à des handicaps plus ou moins lourds au cours du temps.Face aux méthodes thérapeutiques classiques, la stimulation électrique fonctionnelle (SEF) des muscles préservés permet de restaurer le mouvement et de fournir une assistance afin d’améliorer la condition des personnes atteintes et de faciliter leur réadaptation fonctionnelle.De nombreuses problématiques intrinsèques à la complexité du système musculo-squelettique et aux contraintes technologiques rendent néanmoins difficile la démocratisation de solutions de stimulation électro-fonctionnelle en dépit d’avancées majeures dans le domaine.Visant à favoriser l’utilisabilité et l’adaptabilité de telles solutions, cette thèse s’appuie sur un réseau de capteurs/actionneurs génériques embarqués sur le sujet, afin d’utiliser la connaissance issue de l’observation et l’analyse du mouvement pathologique des membres inférieurs pour étudier et valider expérimentalement de nouvelles solutions de commande de la SEF à travers une approche orientée-patient. / The human central nervous system (CNS) can be subject to multiple dysfunctions. Potentially due to physical lesions (e.g.: spinal cord injuries, hemorrhagic or ischemic stroke) or to neurodegenerative disorders (e.g.: Parkinson’s disease), these deficiencies often result in major functional impairments throughout the years.As an alternative to usual therapeutic approaches, functional electrical stimulation (FES) of preserved muscles enables to assist individuals in executing functional movements in order to improve their daily life condition or to help enhancing rehabilitation process.Despite major technological advances in rehabilitation engineering, the complexity of the musculoskeletal system and the technological constraints associated have led to a very slow acceptance of neurorehabilitation technologies.To promote usability and adaptability, several approaches and algorithms were studied through this thesis and were experimentally validated in different clinical and pathological contexts, using low-cost wearable sensors combined to programmable stimulators to assess and control motion through a patient-centered approach.

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

As is seen from the work on gait recognition, there is a de-facto consensus about the silhouette of a person being the low-level representation of choice. It has been hypothesized that the performance degradation that is observed when one compares sequences taken on different surfaces, hence against different backgrounds, or when one considers outdoor sequences is due to the low silhouette quality and its variation. If only one can get better silhouettes the perfomance of gait recognition would be high. This thesis challenges that hypothesis. In the context of the HumanID Gait Challenge problem, we constructed a set of ground truth silhouttes over one gait cycles for 71 subjects, to test recognition across two conditions, shoe and surface. Using these, we show that the performance with ground truth silhouette is as good as that obtained by those obtained by a basic background subtraction algorithm. Therefore further research into ways to enhance silhouette extraction does not appear to be the most productive way to advance gait recognition. We also show, using the manually specified part level silhouettes, that most of the gait recognition power lies in the legs and the arms. The recognition power in various static gait recognition factors as extracted from a single view image, such as gait period, cadence, body size, height, leg size, and torso length, does not seem to be adequate. Using cummulative silhouette error images, we also suggest that gait actually changes when one changes walking surface; in particular the swing phase of the gait gets effected the most.

Improvement of prostheses requires knowledge of how the body adapts. A transradial prosthesis without a dynamic wrist component may cause awkward compensatory motion leading to fatigue, injury or rejection of the prosthesis. This work analyzed the movements of shoulder, elbow and torso during four tasks: drinking from a cup, opening a door, lifting a box and turning a steering wheel. The main purpose of this study was to determine if using a basic transradial prosthesis that lacks motion of the forearm and wrist would cause significant compensatory motion of the shoulder, elbow and torso during the tasks. The second purpose of the study was to determine if the location of added mass would affect compensatory movements during these tasks. A group of able-bodied participants were asked to complete the tasks, without and with a brace, simulating a basic transradial prosthesis to determine if bracing is an appropriate way to study prosthetic use. Transradial prosthesis wearers also completed the tasks without and with added mass at the elbow or at the wrist to determine if distribution of mass has an effect on the motions. Using a motion capture system movements of the shoulder, elbow and torso were analyzed. For the bilateral tasks, the degree of asymmetry (DoA) was calculated for each subject. Statistical analysis was completed within subject comparing the mass interventions and between subjects comparing the control, braced and prosthesis wearing groups. While opening a door and lifting a box, prosthesis users compensated predominantly by bending the torso sideways toward affected side. During the steering wheel task, amputees used more elbow flexion to accommodate for the lack of forearm rotation. While drinking from a cup, compensation occurred by bending the cervical spine, although this was not measured. Adding mass increased the joint forces and moments during the box lift. This research can be used for transradial prosthesis design improvements as well as improving methods of prosthesis fitting and therapeutic training by providing quantitative data of compensatory motion. The data from this study is being used to develop a model for an upper limb prosthesis.

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.

This paper reviews a number of aspects of visual motion analysis in biological systems from a computational perspective. We illustrate the kinds of insights that have been gained through computational studies and how these observations can be integrated with experimental studies from psychology and the neurosciences to understand the particular computations used by biological systems to analyze motion. The particular areas of motion analysis that we discuss include early motion detection and measurement, the optical flow computation, motion correspondence, the detection of motion discontinuities, and the recovery of three-dimensional structure from motion.

Perceptual studies suggest that the visual system uses the "rigidity" assumption to recover three dimensional structures from motion. Ullman (1984) recently proposed a computational scheme, the incremental rigidity scheme, which uses the rigidity assumptions to recover the structure of rigid and non-rigid objects in motion. The scheme assumes the input to be discrete positions of elements in motion, under orthographic projection. We present formulations of Ullmans' method that use velocity information and perspective projection in the recovery of structure. Theoretical and computer analyses show that the velocity based formulations provide a rough estimate of structure quickly, but are not robust over an extended time period. The stable long term recovery of structure requires disparate views of moving objects. Our analysis raises interesting questions regarding the recovery of structure from motion in the human visual system.

Analyzing human motion is important in a number of ways. An athlete constantly needs to evaluate minute details about his or her motion pattern. In physical rehabilitation, the doctor needs to evaluate how well a patient is rehabilitating from injuries. Some systems are being developed in order to identify people only based on their gait. Automatic interpretation of sign language is another area that has received much attention. While all these applications can be considered useful in some sense, the analysis of human motion can also be used for pure entertainment. For example, by filming a sport activity from one view, it is possible to create a 3D reconstruction of this motion, that can be rendered from a view where no camera was originally placed. Such a reconstruction system can be enjoyable for the TV audience. It can also be useful for the computer-game industry. This thesis presents ideas and new methods on how such reconstructions can be obtained. One of the main purposes of this thesis is to identify a number of qualitative constraints that strongly characterizes a certain class of motion. These qualitative constraints provide enough information about the class so that every motion satisfying the constraints will "look nice" and appear, according to a human observer, to belong to the class. Further, the constraints must not be too restrictive; a large variation within the class is necessary. It is shown how such qualitative constraints can be learned automatically from a small set of examples. Another topic that will be addressed concerns analysis of motion in terms of quality assessment as well as classification. It is shown that in many cases, 2D projections of a motion carries almost as much information about the motion as the original 3D representation. It is also shown that single-view reconstruction of 2D data for the purpose of analysis is generally not useful. Using these facts, a prototype of a "virtual coach" that is able to track and analyze image data of human action is developed. Potentials and limitations of such a system are discussed in the the thesis. / QC 20100601

Kinematic analysis of basketball shooting is evolving, however the effects of fatigue on free throw shooting have not been studied. Therefore the effects of fatigue on the kinematics of free throw shooting among elite male basketball players was assessed. Ten healthy male collegiate basketball players participated in the study. Resting and fatigue heart rates of the participants were measured. After a 15 minute warm-up period, markers were placed on seven locations on the shooting arm&rsquo / s side upper and lower extremities. The free throw shots were recorded with two digital cameras at a speed of 60 frames/s at a stereoscopic position. Data were analyzed with the photogrammetry technique. Each participant performed free throw shots (pre-fatigue condition) until the two successful and two unsuccessful shots were collected. Then participants completed a fatigue protocol, which included sprints and squat jumping, until reaching their volitional exhaustion and free throw shots were repeated (post-fatigue condition). The elbow, trunk, knee and ankle joint angles were measured. Successful and unsuccessful shots were compared for pre- and post-fatigue conditions. The results demonstrated that fatigue did not affect free throw shooting and there was no significant joint angle difference (p&gt / .05) between successful and unsuccessful shots (p&gt / .05). It was concluded that fatigue does not affect the kinematics of free throw shooting of healthy male collegiate basketball players and there were no differences in the kinematics of selected joint angles for successful and unsuccessful free throw shots.