Fachbereich Informatik
Refine
H-BRS Bibliography
- yes (62)
Departments, institutes and facilities
Document Type
- Master's Thesis (37)
- Bachelor Thesis (19)
- Report (4)
- Diploma Thesis (1)
- Study Thesis (1)
Year of publication
Keywords
- Emergency support system (2)
- Mobile sensors (2)
- Robotik (2)
- chemoCR (2)
- 0-1-Integer-Problem (1)
- 3D-Laserscanner (1)
- 3D-Punktwolke (1)
- 3D-Scanner (1)
- ASAG (1)
- Active Learning (1)
Die Matrix-Vektor-Multiplikation für dünn besetzte Matrizen (SpMV) stellt für weitreichende wissenschaftliche Anwendungen eine der Kernoperationen des High-Performance-Computing-Bereichs dar. Für die verteilte Berechnung mit immer beliebter werdenden hybriden Rechenclustern kommt dabei die Frage nach einer geeigneten Partitionierungsstrategie für die Verteilung von Daten und Berechnung auf. Diese Arbeit beschäftigt sich damit welchen Einfluss die Struktur der Matrix und die unterschiedlichen Prozessortypen auf die Leistung der SpMV haben und schlägt ein Modell vor, um für diese eine lastbalancierte Verteilung zu erreichen. Wesentliche Bestandteile sind dabei die Laufzeitvorhersage für aktuelle CPUs und GPUs basierend auf einem abgewandelten Roofline-Modell sowie die bewährte Methode der Graph-Partitionierung.
Das Fraunhofer-Institut für Intelligente Analyse- und Informationssysteme (IAIS) betreibt seit mehreren Jahren auf dem Campus Schloss Birlinghoven in Sankt Augustin angewandte Forschung in den Bereichen Multisensordatenanalyse und Datenvisualisierung.
Im Rahmen einer mehrjährigen Kooperation zwischen dem Fraunhofer-IAIS und der Wehrtechnischen Dienststelle 71 (WTD71) wurde das Seeraumüberwachungssystem iLEXX entwickelt. Es soll den Benutzer auf auffällige Situationen hinweisen und ihm kontextabhängig alle notwendigen Handlungsoptionen zur weiteren Aufklärung der Situation oder der Abwehr einer Bedrohung aufzeigen. Das iLEXX-System verarbeitet eine Vielzahl von Sensordaten und Ereignissen. Abhängig vom Szenario kommen hier mehrere tausend Updates pro Sekunde zusammen, die in Echtzeit vorverarbeitet und visualisiert werden müssen.
Estimation of Prediction Uncertainty for Semantic Scene Labeling Using Bayesian Approximation
(2018)
With the advancement in technology, autonomous and assisted driving are close to being reality. A key component of such systems is the understanding of the surrounding environment. This understanding about the environment can be attained by performing semantic labeling of the driving scenes. Existing deep learning based models have been developed over the years that outperform classical image processing algorithms for the task of semantic labeling. However, the existing models only produce semantic predictions and do not provide a measure of uncertainty about the predictions. Hence, this work focuses on developing a deep learning based semantic labeling model that can produce semantic predictions and their corresponding uncertainties. Autonomous driving needs a real-time operating model, however the Full Resolution Residual Network (FRRN) [4] architecture, which is found as the best performing architecture during literature search, is not able to satisfy this condition. Hence, a small network, similar to FRRN, has been developed and used in this work. Based on the work of [13], the developed network is then extended by adding dropout layers and the dropouts are used during testing to perform approximate Bayesian inference. The existing works on uncertainties, do not have quantitative metrics to evaluate the quality of uncertainties estimated by a model. Hence, the area under curve (AUC) of the receiver operating characteristic (ROC) curves is proposed and used as an evaluation metric in this work. Further, a comparative analysis about the influence of dropout layer position, drop probability and the number of samples, on the quality of uncertainty estimation is performed. Finally, based on the insights gained from the analysis, a model with optimal configuration of dropout is developed. It is then evaluated on the Cityscape dataset and shown to be outperforming the baseline model with an AUC-ROC of about 90%, while the latter having AUC-ROC of about 80%.
Nowadays perception is still an up-to-date scienti fic issue on a mobile robot system. This thesis introduces an approach on how to recognize objects, namely numbers, using a digital camera on a Volksbot robot. The robot used in this thesis has been specifi cally designed for the SICK robot day. The development of the vision algorithm was done in two stages: the region of interest detection and the actual number recognition. Diff erent algorithms had been tested and evaluated and the Canny edge detector with contour finding has been proven to be the best choice for the region of interest detection and the Tesseract OCR engine was the best decision for number recognition. To integrate the vision component on an existing robot system, ROS was used. This thesis also discusses the integration of the EPOS motor controller into ROS.
Machine learning-based solutions are frequently adapted in several applications that require big data in operations. The performance of a model that is deployed into operations is subject to degradation due to unanticipated changes in the flow of input data. Hence, monitoring data drift becomes essential to maintain the model’s desired performance. Based on the conducted review of the literature on drift detection, statistical hypothesis testing enables to investigate whether incoming data is drifting from training data. Because Maximum Mean Discrepancy (MMD) and Kolmogorov-Smirnov (KS) have shown to be reliable distance measures between multivariate distributions in the literature review, both were selected from several existing techniques for experimentation. For the scope of this work, the image classification use case was experimented with using the Stream-51 dataset. Based on the results from different drift experiments, both MMD and KS showed high Area Under Curve values. However, KS exhibited faster performance than MMD with fewer false positives. Furthermore, the results showed that using the pre-trained ResNet-18 for feature extraction maintained the high performance of the experimented drift detectors. Furthermore, the results showed that the performance of the drift detectors highly depends on the sample sizes of the reference (training) data and the test data that flow into the pipeline’s monitor. Finally, the results also showed that if the test data is a mixture of drifting and non-drifting data, the performance of the drift detectors does not depend on how the drifting data are scattered with the non-drifting ones, but rather their amount in the test set
Die letzten zwei Jahrzehnte wurden durch das exponentielle Wachstum der zur Verfügung stehenden Daten geprägt. Täglich produzieren Menschen und Maschinen mehr und mehr Daten, die oftmals in verteilten Datenspeichern abgelegt werden. Anwendungsgebiete lassen sich beispielsweise in der Physik und Astronomie finden, wo immense Datenmengen von Teilchenbeschleunigern oder Satelliten erzeugt werden, die gespeichert und verarbeitet werden müssen. Aus diesen Datenmengen können weder vom Menschen direkt noch durch traditionelle Analysemethoden neue Erkenntnisse gewonnen werden. Zur Verarbeitung dieser Datenmassen sind parallele sowie verteilte Datenanalyseverfahren notwendig. [MTT18,NEKH+18]
Neural network based object detectors are able to automatize many difficult, tedious tasks. However, they are usually slow and/or require powerful hardware. One main reason is called Batch Normalization (BN) [1], which is an important method for building these detectors. Recent studies present a potential replacement called Self-normalizing Neural Network (SNN) [2], which at its core is a special activation function named Scaled Exponential Linear Unit (SELU). This replacement seems to have most of BNs benefits while requiring less computational power. Nonetheless, it is uncertain that SELU and neural network based detectors are compatible with one another. An evaluation of SELU incorporated networks would help clarify that uncertainty. Such evaluation is performed through series of tests on different neural networks. After the evaluation, it is concluded that, while indeed faster, SELU is still not as good as BN for building complex object detector networks.
High-dimensional and multi-variate data from dynamical systems such as turbulent flows and wind turbines can be analyzed with deep learning due to its capacity to learn representations in lower-dimensional manifolds. Two challenges of interest arise from data generated from these systems, namely, how to anticipate wind turbine failures and how to better understand air flow through car ventilation systems. There are deep neural network architectures that can project data into a lower-dimensional space with the goal of identifying and understanding patterns that are not distinguishable in the original dimensional space. Learning data representations in lower dimensions via non-linear mappings allows one to perform data compression, data clustering (for anomaly detection), data reconstruction and synthetic data generation.
In this work, we explore the potential that variational autoencoders (VAE) have to learn low-dimensional data representations in order to tackle the problems posed by the two dynamical systems mentioned above. A VAE is a neural network architecture that combines the mechanisms of the standard autoencoder and variational bayes. The goal here is to train a neural network to minimize a loss function defined by a reconstruction term together with a variational term defined as a Kulback-Leibler (KL) divergence.
The report discusses the results obtained for the two different data domains: wind turbine time series and turbulence data from computational fluid dynamics (CFD) simulations.
We report on the reconstruction, clustering and unsupervised anomaly detection of wind turbine multi-variate time series data using a variant of a VAE called Variational Recurrent Autoencoder (VRAE). We trained a VRAE to cluster normal and abnormal wind turbine series (two class problem) as well as normal and multiple abnormal series (multi-class problem). We found that the model is capable of distinguishing between normal and abnormal cases by reducing the dimensionality of the input data and projecting it to two dimensions using techniques such as Principal Component Analysis (PCA) and t-distributed stochastic neighbor embedding (t-SNE). A set of anomaly scoring methods is applied on top of these latent vectors in order to compute unsupervised clustering. We have achieved an accuracy of up to 96% with the KM eans + + algorithm.
We also report the data reconstruction and generation results of two dimensional turbulence slices corresponding to CFD simulation of a HVAC air duct. For this, we have trained a Convolutional Variational Autoencoder (CVAE). We have found that the model is capable of reconstructing laminar flows up to a certain degree of resolution as well generating synthetic turbulence data from the learned latent distribution.
The introduction of gestures as a supplementary input modality has become of increasing interest to human computer interaction design, especially for 3D computer environments. This thesis describes the concepts and development of a gesture recognition system based on the machine learning technique of Hidden Markov Models. Well-known from the field of speech recognition, this statistical method is employed in this thesis to represent and recognize predefined gestures. Within this work, gestures are defined as symbols, such as simple geometric shapes or Roman letters. They are extracted from a stream of three-dimensional optical tracking data which is resampled, reduced to 2D and quantized to be used as input to discrete Hidden Markov Models. A set of prerecorded training data is used to learn the parameters of the models and recognition is achieved by evaluating the trained models. The devised system was used to augment an existing virtual reality prototype application which serves as a demonstration and development platform for the VRGeo consortium. The consortium's goal is to investigate and utilize the benefits of virtual reality technology for the oil and gas industry. An isolated test of the system with seven gestures showed accuracies of up to 98.57% and the review from experts in the fields of virtual reality and geophysics was predominantly positive.
This master thesis describes a supervised approach to the detection and the identification of humans in TV-style video sequences. In still images and video sequences, humans appear in different poses and views, fully visible and partly occluded, with varying distances to the camera, at different places, under different illumination conditions, etc. This diversity in appearance makes the task of human detection and identification to a particularly challenging problem. A possible solution of this problem is interesting for a wide range of applications such as video surveillance and content-based image and video processing. In order to detect humans in views ranging from full to close-up view and in the presence of clutter and occlusion, they are modeled by an assembly of several upper body parts. For each body part, a detector is trained based on a Support Vector Machine and on densely sampled, SIFT-like feature points in a detection window. For a more robust human detection, localized body parts are assembled using a learned model for geometric relations based on Gaussians. For a flexible human identification, the outward appearance of humans is captured and learned using the Bag-of-Features approach and non-linear Support Vector Machines. Probabilistic votes for each body part are combined to improve classification results. The combined votes yield an identification accuracy of about 80% in our experiments on episodes of the TV series "Buffy the Vampire Slayer". The Bag-of-Features approach has been used in previous work mainly for object classification tasks. Our results show that this approach can also be applied to the identification of humans in video sequences. Despite the difficulty of the given problem, the overall results are good and encourage future work in this direction.
A robot (e.g. mobile manipulator) that interacts with its environment to perform its tasks, often faces situations in which it is unable to achieve its goals despite perfect functioning of its sensors and actuators. These situations occur when the behavior of the object(s) manipulated by the robot deviates from its expected course because of unforeseeable ircumstances. These deviations are experienced by the robot as unknown external faults. In this work we present an approach that increases reliability of mobile manipulators against the unknown external faults. This approach focuses on the actions of manipulators which involve releasing of an object. The proposed approach, which is triggered after detection of a fault, is formulated as a three-step scheme that takes a definition of a planning operator and an example simulation as its inputs. The planning operator corresponds to the action that fails because of the fault occurrence, whereas the example simulation shows the desired/expected behavior of the objects for the same action. In its first step, the scheme finds a description of the expected behavior of the objects in terms of logical atoms (i.e. description vocabulary). The description of the simulation is used by the second step to find limits of the parameters of the manipulated object. These parameters are the variables that define the releasing state of the object.
Using randomly chosen values of the parameters within these limits, this step creates different examples of the releasing state of the object. Each one of these examples is labelled as desired or undesired according to the behavior exhibited by the object (in the simulation), when the object is released in the state corresponded by the example. The description vocabulary is also used in labeling the examples autonomously. In the third step, an algorithm (i.e. N-Bins) uses the labelled examples to suggest the state for the object in which releasing it avoids the occurrence of unknown external faults.
The proposed N-Bins algorithm can also be used for binary classification problems. Therefore, in our experiments with the proposed approach we also test its prediction ability along with the analysis of the results of our approach. The results show that under the circumstances peculiar to our approach, N-Bins algorithm shows reasonable prediction accuracy where other state of the art classification algorithms fail to do so. Thus, N-Bins also extends the ability of a robot to predict the behavior of the object to avoid unknown external faults. In this work we use simulation environment OPENRave that uses physics engine ODE to simulate the dynamics of rigid bodies.
Intelligente Dialogsysteme – Chatbots – werden immer häufiger als virtuelle Ansprechpartner von Unternehmen und Institutionen eingesetzt. Auf Basis einer Wissensdatenbank können Chatbots einen größeren Anteil von Kundenanfragen automatisiert beantworten. Analog ist der Einsatz von Chatbots als digitaler Ansprechpartner öffentlicher Verwaltungen denkbar. Sie könnten Bürgern helfen, sich innerhalb der behördlichen Strukturen zu orientieren und Verwaltungsleistungen effizient und effektiv in Anspruch zu nehmen.
Diese Arbeit überprüft den Einsatz eines Chatbots in der öffentlichen Verwaltung hinsichtlich der entstehenden Kosten und des erwartbaren Nutzens. Auf Basis einer umfangreichen Literaturauswertung und der prototypischen Realisierung eines Chatbots für ein Stadtportal werden dabei Herausforderungen dieser Anwendungsdomäne herausgearbeitet, konkrete Funktionsweise und Implementierungsstrategien von Chatbots erörtert und einige Erfolgsfaktoren formuliert, die den Kern einer Handlungsempfehlung für Entscheidungsträger öffentlicher Verwaltungen bilden.
Volumen Rendering ist ein eigenes Thema der Computergrafik und wurde in den letzten Jahren fortlaufend optimiert. Neben verschiedenen Ansätzen, die in Software implementiert sind, gibt es auch einige spezielle Methoden, die die Grafikhardware geeignet nutzen. 2003 wurde ein erstes Paper von J. Krüger und R. Westermann veröffentlicht, in dem eine Hardwareimplementierung eines Raycasting Volumen Renderers gezeigt wurde, ein Ansatz, der bislang nicht geeignet in Hardware umgesetzt werden konnte. Die Vorteile von diesem Ansatz bestehen in zwei Beschleunigungstechniken, die entweder bei fast opaken Darstellungen der Datensätze oder bei Darstellungen mit wenig sichtbaren Daten ausgespielt werden können.
Diese Arbeit zeigt und erläutert, neben der theoretischen Einführung in das Thema, die Implementierung eines interaktiven raycasting-basierten Volumen Renderers auf neuester Grafikhardware mit Hilfe von Shaderprogrammen. Wesentliche Schritte folgen der Veröffentlichung von J. Krüger und R. Westermann, welche aber viele Details und Problemstellen verschweigt. Die Ergebnisse werden mit einem 3D-Textur Volumenrenderverfahren verglichen, wobei durch charakteristische Testdatensätze die beiden Beschleunigungstechniken des Raycasters untersucht werden. Weil beide Techniken bei fast allen Datensätzen eine Beschleunigung des Rendervorgangs hervorrufen sollten, werden die erzielten Ergebnisse miteinander verglichen und kritisch besprochen, um zu beurteilen, ob das hier implementierte Verfahren schneller als das bisher oft verwendete 3D-Texturverfahren ist.
Das WebDAV-Protokoll (Web-based Distributed Authoring and Versioning) ermöglicht die Bearbeitung und Verwaltung von Dateien auf einem Web-Server. Aus technischer Sicht ist WebDAV eine Erweiterung des HTTP-Protokolls. Durch die rasche Zunahme und den steigenden Verbreitungsgrad von WebDAV-basierten Anwendungen, wie etwa Dokumentenmanagementsystemen, steigen auch die Anforderungen an deren Zuverlässigkeit. Die voll umfassende Unterstützung von Transaktionen, d.h. die Zusammenfassung einer Menge von Verarbeitungsschritten zu einer logischen Einheit, würde hierzu einen wichtigen Beitrag leisten. Die für Transaktionen geforderten Eigenschaften, die gleichzeitig auch deren Hauptvorteile darstellen, werden durch das bekannte Akronym ACID beschrieben, welches für Atomarität (atomicity), Konsistenz (consistency), Isoliertheit (isolation) und Dauerhaftigkeit (durability) steht. Zurzeit unterstützt das WebDAV-Protokoll allerdings nur die Punkte Konsistenz und Dauerhaftigkeit, eine komplette und vor allem standardkonforme Unterstützung der ACID-Eigenschaften von Transaktionen ist nicht gegeben. Im Rahmen dieser Arbeit wurde nun ein Transaktionsmodell für den WebDAVStandard erarbeitet. Das Modell ermöglicht es, eine Menge von Dateioperationen transaktionsbasiert durchzuführen. Das Modell unterstützt dabei zur Sicherstellung der Serialisierbarkeit sowohl optimistische als auch pessimistische Verfahren. Die Unterstützung des optimistischen Verfahrens wurde dabei von der IETF (Internet Engineering Task Force) als zulässiges und sinnvolles Vorgehen zur Realisierung von Transaktionen mittels WebDAV bestätigt. Für die pessimistischen Verfahren wurde im Rahmen dieser Arbeit aufgezeigt, wie die bestehenden Konzepte des WebDAV-Standards erweitert werden müssen, um dies ebenfalls umsetzen zu können. Um die getroffene Entwurfsentscheidung zu verifizieren, wurde eine prototypische Implementierung des Modells vorgenommen. Hierbei wurde, nach einer entsprechenden Evaluierung und Bewertung, die optimistische Nebenläufigkeitskontrolle umgesetzt. Clientseitig setzt die Implementierung auf der Jackrabit-Library auf, die serverseitige Implementierung verwendet als Grundlage den WebDAV-Server von Subversion.
The recent explosion of available audio-visual media is the new challenge for information retrieval research. Audio speech recognition systems translate spoken content to the text domain. There is a need for searching and indexing this data which possesses no logical structure. One possible way to structure it on a high level of abstraction is by finding topic boundaries. Two unsupervised topic segmentation methods were evaluated with real-world data in the course of this work. The first one, TSF, models topic shifts as fluctuations in the similarity function of the transcript. The second one, LCSeg, approaches topic changes as places with the least overlapping lexical chains. Only LCSeg performed close to a similar real-world corpus. Other reported results could not be outperformed. Topic analysis based on the repeated word usage models renders topic changes more ambiguous than expected. This issue has more impact on the segmentation quality than the state-of-the-art ASR word error rate. It could be concluded that it is advisable to develop topic segmentation algorithms with real-world data to avoid potential biases to artificial data. Unlike evaluated approaches based on word usage analysis, methods operating with local contexts can be expected to perform better through emulation of semantic dependencies.
Augmented Reality (AR) findet heutzutage sehr viele Anwendungsbereiche. Durch die Überlagerung von virtuellen Informationen mit der realen Umgebung eignet sich diese Technologie besonders für die Unterstützung der Benutzer bei technischen Wartungs- oder Reparaturvorgängen. Damit die virtuellen Daten korrekt mit der realen Welt überlagert werden, müssen Position und Orientierung der Kamera durch ein Trackingverfahren ermittelt werden. In dieser Arbeit wurde für diesen Zweck ein markerloses, modellbasiertes Trackingsystem implementiert. Während einer Initialisierungs-Phase wird die Kamerapose mithilfe von kalibrierten Referenzbildern, sogenannten Keyframes, bestimmt. In einer darauffolgenden Tracking-Phase wird das zu trackende Objekt weiterverfolgt. Evaluiert wurde das System an dem 1:1 Trainingsmodell des biologischen Forschungslabors Biolab, welches von der Europäischen Weltraumorganisation ESA zur Verfügung gestellt wurde.
Today publications are digitally available which enables researchers to search the text and often also the content of tables. On the contrary, images cannot be searched which is not a problem for most fields, but in chemistry most of the information are contained in images, especially structure diagrams. Next to the "normal" chemical structures, which represent exactly one molecule, there also exist generic structures, so called Markush structures. These contain variable parts and additional textual information which enable them to represent several molecules at once. This can vary between just a few and up to thousands or even millions. This ability lead to a spread of Markush structures in patents, because it enables patents to protect entire families of molecules at once. Next to the prevention of an enumeration of all structures it also has the advantage that, if a Markush structure is used in a patent, it is much harder to determine whether a specific structure is protected by it or not. To solve the question about the protection of a structure, it is necessary to search the patents. Appropriate databases for this task already do exist, but are filled manually. An automatic processing does not yet exist. In this project a Markush structure reconstruction prototype is developed which is able to reconstruct bitmaps including Markush structures (meaning a depiction of the structure and a text part describing the generic parts) into a digital format and save them in the newly developed context-free grammar based file format extSMILES. This format is searchable due to its context-free grammar based design. To be able to develop a Markush structure reconstruction prototype, an in depth analysis of the concept of Markush structures and their requirements for a reconstruction process was performed. Thereby it is stated, that the common connection table concept of the existing file formats is not able to store Markush structures. Especially challenging are conditions for most of the formats. Thus, a context-free grammar based file format is developed, which extends the SMILES format. This extSMILES called format assures the searchability of the results by its context-free grammar based concept, and is able to store all information contained in Markush structures. In addition it is generic, extendable and easily understandable. The developed prototype for the Markush structure reconstruction uses extSMILES as output format and is based on the chemical structure recognition tool chemoCR and the Unstructured Information Management Architecture UIMA. For chemoCR modules are developed which enable it to recognize and assemble Markush structures as well as to return the reconstruction result in extSMILES. For UIMA on the other hand, a pipeline is developed, which is able to analyse and translate the input text files to extSMILES. The results of both tools then are combined and presented in chemoCR. An evaluation of the prototype is performed on a representative set of twelve structures of interest and low image quality which contain all typical Markush elements. Trivial structures containing only one R-group are not evaluated. Due to the challenging nature of the images, no Markush structure could be correctly reconstructed. But by regarding the assumption, that R-group definitions which are described by natural language are excluded from the task, and under the condition that the core structure reconstruction is improved, the rate of success can be increased to 58.4%.
In dieser Arbeit wird eine von P. Ahlrichs und B. Dünweg entwickelte Methode [Ahlrichs und Dünweg, 1998] zur Simulation von Polymeren in Flüssigkeiten auf dem Cell-Prozessor implementiert. Dabei soll der Frage nachgegangen werden, wie performant der Cell-Processor in der Lage ist diese Simulation zu berechnen.
Zur Simulation der Polymere wird eine Molekular-Dynamik Simulation genutzt. Die Monomere der Polymerketten werden durch ein Kugel-Feder Modell gekoppelt. Die einzelnen Monomere der Polymere werden als einfache Punktteilchen betrachtet. Dies ermöglicht eine Interaktion der Monomere, unabhängig von deren Zeit- und Längenskalen, mit der Flüssigkeitssimulation durch Reibung. Die Flüssigkeit wird in dieser Arbeit durch die Lattice-Boltzmann-Methode simuliert.
This thesis proposes a multi-label classification approach using the Multimodal Transformer (MulT) [80] to perform multi-modal emotion categorization on a dataset of oral histories archived at the Haus der Geschichte (HdG). Prior uni-modal emotion classification experiments conducted on the novel HdG dataset provided less than satisfactory results. They uncovered issues such as class imbalance, ambiguities in emotion perception between annotators, and lack of representative training data to perform transfer learning [28]. Hence, the objectives of this thesis were to achieve better results by performing a multi-modal fusion and resolving the problems arising from class imbalance and annotator-induced bias in emotion perception. A further objective was to assess the quality of the novel HdG dataset and benchmark the results using SOTA techniques. Through a literature survey on the challenges, models, and datasets related to multi-modal emotion recognition, we created a methodology utilizing the MulT along with a multi-label classification approach. This approach produced a considerable improvement in the overall emotion recognition by obtaining an average AUC of 0.74 and Balanced-accuracy of 0.70 on the HdG dataset, which is comparable to state-of-the-art (SOTA) results on other datasets. In this manner, we were also able to benchmark the novel HdG dataset as well as introduce a novel multi-annotator learning approach to understand each annotator’s relative strengths and weaknesses for emotion perception. Our evaluation results highlight the potential benefits of the novel multi-annotator learning approach in improving overall performance by resolving the problems arising from annotator-induced bias and variation in the perception of emotions. Complementing these results, we performed a further qualitative analysis of the HdG annotations with a psychologist to study the ambiguities found in the annotations. We conclude that the ambiguities in annotations may have resulted from a combination of several socio-psychological factors and systemic issues associated with the process of creating these annotations. As these problems are also present in most multi-modal emotion recognition datasets, we conclude that the domain could benefit from a set of annotation guidelines to create standardized datasets.
This project focuses on object detection in dense volume data. There are several types of dense volume data, namely Computed Tomography (CT) scan, Positron Emission Tomography (PET), Magnetic Resonance Imaging (MRI). This work focuses on CT scans. CT scans are not limited to the medical domain; they are also used in industries. CT scans are used in airport baggage screening, assembly lines, and the object detection systems in these places should be able to detect objects fast. One of the ways to address the issue of computational complexity and make the object detection systems fast is to use low-resolution images. Low-resolution CT scanning is fast. The entire process of scanning and detection can be made faster by using low-resolution images. Even in the medical domain, to reduce the rad iation dose, the exposure time of the patient should be reduced. The exposure time of patients could be reduced by allowing low-resolution CT scans. Hence it is essential to find out which object detection model has better accuracy as well as speed at low-resolution CT scans. However, the existing approaches did not provide details about how the model would perform when the resolution of CT scans is varied. Hence in this project, the goal is to analyze the impact of varying resolution of CT scans on both the speed and accuracy of the model. Three object detection models, namely RetinaNet, YOLOv3, and YOLOv5, were trained at various resolutions. Among the three models, it was found that YOLOv5 has the best mAP and f1 score at multiple resolutions on the DeepLesion dataset. RetinaNet model h as the least inference time on the DeepLesion dataset. From the experiments, it could be asserted that sacrificing mean average precision (mAP) to improve inference time by reducing resolution is feasible.
Objektrelationale Datenbanken und Rough Sets für die Analyse von Contextualized Attention Metadata
(2009)
Robots integrated into a social environment with humans need the ability to locate persons in their surrounding area. This is also the case for the WelcomeBot which is developed at the Fraunhofer Institute IAIS. In the future, the robot should follow persons in the buildings and guide them to certain areas. Therefore, it needs the capability to detect and track a person in the environment. In this master thesis, an approach for fast and reliable tracking of a person via a mobile robotic platform is presented. Based on the investigation of different methods and sensors, a laser scanner and a camera are selected as the primary two sensors.