abstract: Visual object recognition has achieved great success with advancements in deep learning technologies. Notably, the existing recognition models have gained human-level performance on many of the recognition tasks. However, these models are data hungry, and their performance is constrained by the amount of training data. Inspired by the human ability to recognize object categories based on textual descriptions of objects and previous visual knowledge, the research community has extensively pursued the area of zero-shot learning. In this area of research, machine vision models are trained to recognize object categories that are not observed during the training process. Zero-shot learning models leverage textual information to transfer visual knowledge from seen object categories in order to recognize unseen object categories. Generative models have recently gained popularity as they synthesize unseen visual features and convert zero-shot learning into a classical supervised learning problem. These generative models are trained using seen classes and are expected to implicitly transfer the knowledge from seen to unseen classes. However, their performance is stymied by overfitting towards seen classes, which leads to substandard performance in generalized zero-shot learning. To address this concern, this dissertation proposes a novel generative model that leverages the semantic relationship between seen and unseen categories and explicitly performs knowledge transfer from seen categories to unseen categories. Experiments were conducted on several benchmark datasets to demonstrate the efficacy of the proposed model for both zero-shot learning and generalized zero-shot learning. The dissertation also provides a unique Student-Teacher based generative model for zero-shot learning and concludes with future research directions in this area. / Dissertation/Thesis / Masters Thesis Computer Science 2020

The neural pathways underlying the symptoms of Post Traumatic Stress Disorder (PTSD) have not been fully elucidated. Intrusive memories, persistent anxiety and other cognitive deficits have been attributed to maladaptive or otherwise aberrant processing in specific brain regions, including the hippocampus, amygdala and prefrontal cortex. Our laboratory has developed an animal model of PTSD which results in the enhancement of memory for a place associated with exposure to a predator, anxiety-like behavior, increased startle and impaired memory in a non-aversive memory task. To better understand how the interaction of the hippocampus and prefrontal cortex contribute to the different symptoms of the disorder, we investigated the transient inactivation of each structure during an intense stressor. Our results show that long-term contextual fear associations involve activity in both the hippocampus and the prefrontal cortex, but only the prefrontal cortex is involved in cued fear memories as well.

No description available.

This work deals with recognition of hand-drawn objects traced by children with mental disorders. The aim is to classify object’s geometrical primitives into classes so then can be plotted along with the idealized shape of the input object. Level of mental retardation is determined by the variance of the input (drawn) object from idealized shape of the object (artwork).

Diploma thesis deals with a robot vision and its application to the problem of manipulation of coincidentally placed objects. There is mentioned an overview of current principles of the most frequently used vision systems on the market. With regard to the required task to be solved, there are mentioned various possibilities of using basic softsensors during the recognition of different objects. The objective of this Diploma thesis is also programming and realization of a demonstration application applying knowledge of PLC programming, knowledge of expert programming KRL language (for KUKA robots), knowledge of designing scripts for smart camera in Spectation software and knowledge of network communication among all devices used in this case.

Les mains robotiques sont pour la plupart utilisées pour reproduire la dextérité humaine. Au delà des challenges mécaniques et de contrôle que ceci peut représenter, la connaissance de l’environnent avec lequel la main interagit est important pour assurer la dextérité. Donc, la reconnaissance tactile des objets est devenue une capacité importante pour les systèmes de manipulation. Dans ce thèse, on propose une méthode pour qu'une main robotique puisse comprendre la nature géométrique d'un objet que lui a été donné. En plus des données statique récupérées quand la main a saisie l'objet, le mouvements qu'elle fait pendant la saisi sont aussi exploitées. Comme première contribution, on propose les signatures de formes proprioceptives. Ceci est un descripteur qui est basé uniquement sur des données proprioceptives et qui est invariant à la taille et à la position de l'objet dans la main. Il contient l'information sur la forme globale de l'objet. Comme deuxième contribution, on propose un outil pour extraire l'information sur l'objet saisi en utilisant l'information dynamique générée pendant la saisie. Pour cela, les mouvements des doigts pendant le saisie sont interprétés en fonction de la stratégie de saisie utilisée. On présente une méthode pour faire la reconnaissance de la forme d'un objet de façon séquentielle. Pour cela, on utilise une collection des Forêt d'arbres décisionnels. Ceci permet de mettre a jour le modèle de reconnaissance quand des nouveaux objets doivent être reconnus. De cette façon, le temps du processus d’entraînement de l'algorithme est réduit. / Robotic anthropomorphic hands are mostly used to reproduce the human dexterity in manipulation. Beyond the mechanical and control challenges that this represents, perceptive knowledge of the environment with which the hand interacts is key to ensure that dexterity is achieved. In this sense, tactile object recognition has become an important asset for manipulation systems. Regardless of the advances in this domain, it continues to be a valid subject of research today. In this thesis, we propose a method to enable a robotic hand to quickly understand the geometrical nature of an object that has been handled by it. Aside from the static data obtained once the object has been fully grasped, the movements of the hand during the grasp execution will also be exploited. As a first contribution, we propose the proprioceptive shape signature. This descriptor, based solely on proprioceptive data, is invariant to the size and pose of the object within the hand and it contains information about the global shape of the object almost as soon as the grasp execution ends. As a second contribution, we propose a tool to extract information about the grasped object from the dynamic data generated during the grasp execution. For this, the movements of the fingers during the grasping process will be interpreted based on the grasp strategy. Finally, we present a method to perform sequential object shape identification based on a collection of random forests. This method allows to update the recognition model as new shapes are desired to be identified. Thus, the time-consuming process of training the model from scratch is avoided.

<p>Pervasive cameras are generating videos at an unprecedented pace, making videos the new frontier of big data. As the processors, e.g., CPU/GPU, become increasingly powerful, the cloud and edge nodes can generate useful insights from colossal video data. However, as the research in computer vision (CV) develops vigorously, the system area has been a blind spot in CV research. With colossal video data generated from cameras every day and limited compute resource budgets, how to design software systems to generate insights from video data efficiently?</p><p><br></p><p>Designing cost-efficient video analytics software systems is challenged by the expensive computation of vision operators, the colossal data volume, and the precious wireless bandwidth of surveillance cameras. To address above challenges, three software systems are proposed in this thesis. For the first system, we present VStore, a data store that supports fast, resource-efficient analytics over large archival videos. VStore manages video ingestion, storage, retrieval, and consumption and controls video formats through backward derivation of configuration: in the opposite direction along the video data path, VStore passes the video quantity and quality expected by analytics backward to retrieval, to storage, and to ingestion. VStore derives an optimal set of video formats, optimizes for different resources in a progressive manner, and runs queries as fast as 362x of video realtime. For the second system, we present a camera/cloud runtime called DIVA that supports querying cold videos distributed on low-cost wireless cameras. DIVA is built upon a novel zero-streaming paradigm: to save wireless bandwidth, when capturing video frames, a camera builds sparse yet accurate landmark frames without uploading any video data; when executing a query, a camera processes frames in multiple passes with increasingly more expensive operators. On diverse queries over 15 videos, DIVA runs at more than 100x realtime and outperforms competitive alternatives remarkably. For the third system, we present Clique, a practical object re-identification (ReID) engine that builds upon two unconventional techniques. First, Clique assesses target occurrences by clustering unreliable object features extracted by ReID algorithms, with each cluster representing the general impression of a distinct object to be matched against the input. Second, to search across camera videos, Clique samples cameras to maximize the spatiotemporal coverage and incrementally adds cameras for processing on demand. Through evaluation on 25 hours of traffic videos from 25 cameras, Clique reaches a high recall at 5 of 0.87 across 70 queries and runs at 830x of video realtime in achieving high accuracy.</p>

A laser radar can be used to generate 3D data about the terrain in a very high resolution. The development of new support technologies to analyze these data is critical to the effective and efficient use of these data in decision support systems, due to the large amounts of data that are generated. Adequate technology in this regard is currently not available and development of new methods and algorithms to this end are important goals of this work. A semi-qualitative data structure for terrain surface modelling has been developed. A categorization and triangulation process has also been developed to substitute the high resolution 3D model for this data structure. The qualitative part of the structure can be used for detection and recognition of terrain features. The quantitative part of the structure is, together with the qualitative part, used for visualization of the terrain surface. Substituting the 3D model for the semi-qualitative structures means that a data reduction is performed. A number of algorithms for detection and recognition of different terrain objects have been developed. The algorithms use the qualitative part of the previously developed semi-qualitative data structure as input. The taken approach is based on matching of symbols and syntactic pattern recognition. Results regarding the accuracy of the implemented algorithms for detection and recognition of terrain objects are visualized. A further important goal has been to develop a methodology for determining driveability using 3D-data and other geographic data. These data must be fused with vehicle data to determine the properties of the terrain context of our operations with respect to driveability. This fusion process is therefore called context fusion. The recognized terrain objects are used together with map data in this method. The uncertainty associated with the imprecision of the data has been taken into account as well. / <p>Report code: LiU-Tek-Lic-2008:29.</p>

Although sensor planning in computer vision has been a subject of research for over two decades, a vast majority of the research seems to concentrate on two particular applications in a rather limited context of laboratory and industrial workbenches, namely 3D object reconstruction and robotic arm manipulation. Recently, increasing interest is engaged in research to come up with solutions that provide wide-area autonomous surveillance systems for object characterization and situation awareness, which involves portable, wireless, and/or Internet connected radar, digital video, and/or infrared sensors. The prominent research problems associated with multisensor integration for wide-area surveillance are modality selection, sensor planning, data fusion, and data exchange (communication) among multiple sensors. Thus, the requirements and constraints to be addressed include far-field view, wide coverage, high resolution, cooperative sensors, adaptive sensing modalities, dynamic objects, and uncontrolled environments. This article summarizes a new survey and analysis conducted in light of these challenging requirements and constraints. It involves techniques and strategies from work done in the areas of sensor fusion, sensor networks, smart sensing, Geographic Information Systems (GIS), photogrammetry, and other intelligent systems where finding optimal solutions to the placement and deployment of multimodal sensors covering a wide area is important. While techniques covered in this survey are applicable to many wide-area environments such as traffic monitoring, airport terminal surveillance, parking lot surveillance, etc., our examples will be drawn mainly from such applications as harbor security and long-range face recognition.

One of the main goals in computer vision is to achieve a human-like understanding of images. Nevertheless, image understanding has been mainly studied in the 2D image frame, so more information is needed to relate them to the 3D world. With the emergence of 3D sensors (e.g. the Microsoft Kinect), which provide depth along with color information, the task of propagating 2D knowledge into 3D becomes more attainable and enables interaction between a machine (e.g. robot) and its environment. This dissertation focuses on three aspects of indoor 3D scene understanding: (1) 2D-driven 3D object detection for single frame scenes with inherent 2D information, (2) 3D object instance segmentation for 3D reconstructed scenes, and (3) using room and floor orientation for automatic labeling of indoor scenes that could be used for self-supervised object segmentation. These methods allow capturing of physical extents of 3D objects, such as their sizes and actual locations within a scene.