No description available.

No description available.

Les systèmes multi-capteurs exploitent les complémentarités des différentes sources sensorielles. Par exemple, le capteur visuo-inertiel permet d'estimer la pose à haute fréquence et avec une grande précision. Les méthodes de vision mesurent la pose à basse fréquence mais limitent la dérive causée par l'intégration des données inertielles. Les centrales inertielles mesurent des incréments du déplacement à haute fréquence, ce que permet d'initialiser la vision et de compenser la perte momentanée de celle-ci. Cette thèse analyse deux aspects du problème. Premièrement, nous étudions les méthodes visuelles directes pour l'estimation de pose, et proposons une nouvelle technique basée sur la corrélation entre des images et la pondération des régions et des pixels, avec une optimisation inspirée de la méthode de Newton. Notre technique estime la pose même en présence des changements d'illumination extrêmes. Deuxièmement, nous étudions la fusion des données a partir de la théorie de la commande. Nos résultats principaux concernent le développement d'observateurs pour l'estimation de pose, biais IMU et l'autocalibrage. Nous analysons la dynamique de rotation d'un point de vue non linéaire, et fournissons des observateurs stables dans le groupe des matrices de rotation. Par ailleurs, nous analysons la dynamique de translation en tant que système linéaire variant dans le temps, et proposons des conditions d'observabilité uniforme. Les analyses d'observabilité nous permettent de démontrer la stabilité uniforme des observateurs proposés. La méthode visuelle et les observateurs sont testés et comparés aux méthodes classiques avec des simulations et de vraies données visuo-inertielles.

PURPOSE: Thermal ablation therapy is an emerging local cancer treatment to destroy cancer tissue using heat. However variations in blood flow and energy absorption rates make it extremely challenging to monitor thermal changes. Insufficient ablation may lead to recurrence of the cancer while excessive ablation may damage adjacent healthy tissues. Ultrasound could be a convenient and inexpensive imaging modality for real-time monitoring of the ablation. For the development and optimization of these methods, it is essential to have ground truth data and a reliable and quantitative validation technique before beginning clinical trials on humans. In this dissertation, my primary focus was to solve the image-to-physical space registration problem using stereotactic fiducials that provide accurate correlation of ultrasound and pathology (ground truth) images. METHOD: A previously developed validation test-bed prototype was evaluated using phantom experiments to identify the shortcomings and limitations. In order to develop an improved validation platform, a simulator was implemented for evaluating registration methods as well as different line fiducial structures. New fiducial line structures were proposed, and new methods were implemented to overcome the limitations of the old system. The new methods were then tested using simulation results and phantom studies. Phantom experiments were conducted to improve the visibility of fiducials, as well as the quality of acquired ultrasound and pathology image datasets. RESULTS: The new system outperforms the previous one in terms of accuracy, robustness, and simplicity. The new registration method is robust to missing fiducials. I also achieved complete fiducial visibility in all images. Enhancing the tissue fixation medium improved the ultrasound data quality. The quality of pathology images were improved by a new imaging method. Simulation results show improvement in pose recovery accuracy using my proposed fiducial structure. This was validated by phantom studies reducing spatial misalignment between the US and pathology image sets. CONCLUSION: A new generation of test-bed was developed that provides a reliable and quantitative validation technique for evaluating and optimizing ablation monitoring methods. / Thesis (Master, Computing) -- Queen's University, 2011-09-29 20:31:55.159

This thesis is concerned with the task of autonomous selection of objects to remove (unload) them from a pile in robotic manipulation systems. Applications such as the automation of logistics processes and service robots require an ability to autonomously manipulate objects in the environment. A collapse of a pile of objects due to an inappropriate choice of the object to be removed from the pile cannot be afforded for an autonomous robotic manipulation system. This dissertation presents an indepth analysis of the problem and proposes methods and algorithms to empower robotic manipulation systems to select a safe object from a pile elaborately and autonomously. The contributions presented in this thesis are three-fold. First, a set of algorithms is proposed for extracting a minimal set of high level symbolic relations, namely, gravitational act and support relations, of physical interactions between objects composing a pile. The symbolic relations, extracted by a geometrical reasoning method and a static equilibrium analysis can be readily used by AI paradigms to analyze the stability of a pile and reason about the safest set of objects to be removed. Considering the problem of undetected objects and the uncertainty in the estimated poses as they exist in realistic perception systems, a probabilistic approach is proposed to extract the support relations and to make a probabilistic decision about the set of safest objects using notions from machine learning and decision theory. Second, an efficient search based algorithm is proposed in an internal representation to automatically resolve the inter-penetrations between the shapes of objects due to errors in the poses estimated by an existing object detection module. Refining the poses by resolving the inter-penetrations results in a geometrically consistent model of the environment, and was found to reduce the overall pose error of the objects. This dissertation presents the concept of minimum translation search for object pose refinement and discusses a discrete search paradigm based on the concept of depth of penetration between two polyhedrons. Third, an application centric evaluation of ranging sensors for selecting a set of appropriate sensors for the task of object detection in the design process of a real-world robotics manipulation system is presented. The performance of the proposed algorithms are tested on data sets generated in simulation and from real-world scenarios.

The thesis deals with head pose estimation in stereo data. The theoretical part provides the basis for understanding the geometry of the camera, its parameters and the method of calibration. The following describes the principles of stereo analysis and creating of disparity maps. In the research section, the methods used for head pose modelling are presented and an analysis of selected published articles is given. In the course of the master’s thesis, a system of two cameras for stereoscopic acquisition of motion of the head was designed and several measurements were carried out. The obtained data was prepared for creation of disparity maps and further processing. Based on the detection of facial features, in particular the inner and outer corners of the eyes and corners of the mouth, and their correspondences, a simple geometric model in shape of triangle was created to illustrate the inclination of the facial plane in space. By computing the angle of inclination in three axes, the current head pose is obtained. Motion is modelled by tracking detected points during video sequences.

Master’s thesis deals with object pose estimation using monocular camera. As an object is considered every rigid, shape fixed entity with strong edges, ideally textureless. Object position in this work is represented by transformation matrix, which describes object translation and rotation towards world coordinate system. First chapter is dedicated to explanation of theory of geometric transformations and intrinsic and extrinsic parameters of camera. This chapter also describes detection algorithm Chamfer Matching, which is used in this work. Second chapter describes all development tools used in this work. Third, fourth and fifth chapter are dedicated to practical realization of this works goal and achieved results. Last chapter describes created application, that realizes known object pose estimation in scene.

The aim of this master's thesis was to create a user interface that allows controlling the PC using voice commands, static poses and dynamic gestures. It contains an implementation of a natural user interface that emulates the functionality of the keyboard and the mouse. In this thesis for detecting and tracking the user was used the sensor Kinect. For implementation I have chosen platform Windows and the Kinect for Windows SDK with programming language C#. During the implementation and testing of the interface, sensor Kinect for Xbox 360 was used. The result of this master's thesis is a natural user interface that allows controlling desktop applications without using keyboard or mouse, and that is comfortable and user-friendly.

With the development of posibilities of image processing, stereoscopy, prices of web cameras and power of computers an opportunity to multiply an experience with working with 3D programs showed. From the picture from webcamera an estimation of a pose of user's head can be made. According to this pose a view on 3D scene can be changed. Then, when user moves his head, he will have a feeling as if monitor was a window through which one can see the scene behind. With the system which is the result of this project it will be possible to easily and cheaply add this kind of behaviour to any 3D application.

Not all categories are created equal in object recognition. Fine-grained visual categorization (FGVC) is a branch of visual object recognition that aims to distinguish subordinate categories within a basic-level category. Examples include classifying an image of a bird into specific species like "Western Gull" or "California Gull". Such subordinate categories exhibit characteristics like small inter-category variation and large intra-class variation, making distinguishing them extremely difficult. To address such challenges, an algorithm should be able to focus on object parts and be invariant to object pose. Like many other computer vision tasks, FGVC has witnessed phenomenal advancement following the resurgence of deep neural networks. However, the proposed deep models are usually treated as black boxes. Network interpretation and understanding aims to unveil the features learned by neural networks and explain the reason behind network decisions. It is not only a necessary component for building trust between humans and algorithms, but also an essential step towards continuous improvement in this field. This dissertation is a collection of papers that contribute to FGVC and neural network interpretation and understanding. Our first contribution is an algorithm named Pose and Appearance Integration for Recognizing Subcategories (PAIRS) which performs pose estimation and generates a unified object representation as the concatenation of pose-aligned region features. As the second contribution, we propose the task of semantic network interpretation. For filter interpretation, we represent the concepts a filter detects using an attribute probability density function. We propose the task of semantic attribution using textual summarization that generates an explanatory sentence consisting of the most important visual attributes for decision-making, as found by a general Bayesian inference algorithm. Pooling has been a key component in convolutional neural networks and is of special interest in FGVC. Our third contribution is an empirical and experimental study towards a thorough yet intuitive understanding and extensive benchmark of popular pooling approaches. Our fourth contribution is a novel LMPNet for weakly-supervised keypoint discovery. A novel leaky max pooling layer is proposed to explicitly encourages sparse feature maps to be learned. A learnable clustering layer is proposed to group the keypoint proposals into final keypoint predictions. 2020 marks the 10th year since the beginning of fine-grained visual categorization. It is of great importance to summarize the representative works in this domain. Our last contribution is a comprehensive survey of FGVC containing nearly 200 relevant papers that cover 7 common themes.