Refine
Year of publication
Document Type
- Article (386)
- Conference Proceeding (318)
- Part of a Book (71)
- Workingpaper / Report (34)
- Master's Thesis (7)
- Book (5)
- Doctoral Thesis (5)
- Collection (4)
- Patent (3)
- Preprint (3)
Language
- English (847) (remove)
Keywords
- VSVR (39)
- DOAJ (37)
- Virtual (TV) Studio (23)
- FHD (22)
- Forena (12)
- Poster (8)
- security (8)
- security requirement (7)
- DEAL (6)
- Lehre (6)
Department/institution
- Fachbereich - Medien (342)
- Fachbereich - Maschinenbau und Verfahrenstechnik (293)
- Creative Media Production and Entertainment Computing (142)
- Fachbereich - Sozial- & Kulturwissenschaften (90)
- Fachbereich - Elektro- & Informationstechnik (67)
- Sound and Vibration Engineering (42)
- Digitale Vernetzung und Informationssicherheit (30)
- Fachbereich - Architektur (22)
- Fachbereich - Wirtschaftswissenschaften (17)
- Intelligente Mensch-Technik-Interaktion (12)
Isogeometric multi-patch analyses for mixed thin shells in the framework of non-linear elasticity
(2021)
The “Journal of Virtual Reality and Broadcasting” is an open access E-journal covering advanced media technology for the integration of human computer interaction and modern information systems. The main focus is on the creation of synergies between such basic technologies as computer graphics and state-of-the-art broadcasting techniques.
The main goals are to publish research results in the field of Virtual Reality and Broadcasting, to provoke discussions, and to promote the exchange of ideas and information. Developments in the area have a direct effect on society, therefore social aspects will also be considered. As an interdisciplinary field Virtual Reality requires multilateral collaboration in order to enable new applications.
RB publishes articles consecutively and in electronic form only. All articles are peer-reviewed in a strict review process by at least three independent experts from the appropriate field of research and appear in the English language. The articles are organized in one volume per year with ten to twenty articles. Material that has been previously presented at conferences undergo a major revision and are extended and modified by the authors with at least 20% new material according to the Journal's policy for previously published articles. The Journal has been established in 2004.
Currently, the submission and publication of articles is free of charge. No author fees are applied.
This paper investigates the implementation of a human body model for the simulation of a jumping motion in Simulink. It contains details to get an insight into modern biomechanics without previous experience. In the simulation, a Hill-Type muscle is used to generate joint momentum with a phenomenologically realistic relationship between muscle length and force. The simulation published with this article is one of many solutions to design a humanoid simulation. The paper guides through the implemented build and explains its range of application.
Sound and music are well-studied aspects of the quality of experience in restaurants; the role of the room acoustical conditions, their influence on the visitors’ soundscape evaluation and their impact on the overall customer satisfaction in restaurants, however, has received less scientific attention. The present field study therefore investigated whether sound pressure level, reverberation time, and soundscape pleasantness can predict factors associated with overall restaurant quality. In total, 142 persons visiting 12 restaurants in Berlin rated relevant acoustical and non-acoustical factors associated with restaurant quality. Simultaneously, the A-weighted sound pressure level (LA,eq,15) was measured, and the reverberation time in the occupied state (T20,occ) was obtained by measurements performed in the unoccupied room and a subsequent calculation of the occupied condition according to DIN 18041. Results from linear mixed-effects models revealed that both the LA,eq,15 and T20,occ had a significant influence on soundscape pleasantness and eventfulness, whereby the effect of T20,occ was meditated by the LA,eq,15. Also, the LA,eq,15 as well as soundscape pleasantness were significant predictors of overall restaurant quality. A comprehensive structural equation model including both acoustical and non-acoustical factors, however, indicates that the effect of soundscape pleasantness on overall restaurant quality is mediated by the restaurant’s atmosphere. Our results support and extend previous findings which suggest that the acoustical design of restaurants involves a trade-off between comfort and liveliness, depending on the desired character of the place.
Knowledge is our most valuable asset! Due to the increasing fluctuation caused by demographic change, the change in society and the industry, companies are increasingly confronted with knowledge management. Therefore, companies have problems with carrying out projects effectively. There are several models, methods and tools to develop an effective concept for knowledge management. Can the process capability of projects be ensured despite high fluctuation? What appropriate methods and tools are available to manage knowledge? These and more questions are clarified in the course of this work and presented in a conceptual proposal.
Land of High Hopes
(2019)
We are living in the era of sophisticated, state-of-the-art technology, artificial intelligence, accelerating data sciences at ultimate speed and sheer unlimited possibilities. We are able to shoot unmanned rockets equipped with vehicles into the orbit. We can afford to spend fortunes on research and development. However, and that leaves me bloodcurdling and furious, we haven´t managed yet to settle at least some of the most pressing global humanitarian needs: access to food, water, electricity, education and healthcare. So how can we be proud of any achievements? What does make us proud anyway?! ...
In this paper we introduce a system for tracking persons walking or standing on a large planar surface and for using the acquired data to easily configure position based interactions for virtual studio productions. The tracking component of the system, radarTRACK, is based on a laser scanner device capable of delivering interaction points on a large configurable plane. By using the device on the floor it is possible to use the delivered data to detect feet positions and derive the position and orientation of one or more users in real time. The second component of the system, named OscCalibrator, allows for the easy creation of multidimensional linear mappings between input and output parameters and the routing of OSC messages within a single modular design environment. We demonstrate the use of our system to flexibly create position-based interactions in a virtual studio environment.
Large-Scale Storage Technologies in the Context of future German Transmission Grid Congestions
(2012)
The study “lidA – leben in der Arbeit. German cohort study on work, age and health” was a two-wave panel study funded by the Federal Ministry of Education and Research and examines the relationship between work, health and work participation among aging employees from an interdisciplinary perspective (www.lida-studie.de). The lidA study was carried out by an interdisciplinary consortium consisting of the Institute for Safety Technology at Bergische University Wuppertal, the Institute for Employment Research (IAB), Nuremberg, the Institute of Social Medicine and Health Economics at Otto-von-Guericke-University Magdeburg, the Institute of the History, Philosophy and Ethics of Medicine at Ulm University, infas Institute for Applied Social Sciences, Bonn and the Federal Institute for Occupational Safety and Health (BAuA), Berlin, as an associated partner. lidA contains a representative sample of persons belonging to the 1959 and 1965 birth cohorts and employed on the reference date for sampling, 31 December 2009. They were interviewed personally (CAPI) in 2011 (wave 1; N=6,585) and 2014 (wave 2; N=4,244). The data contain information about work and current employment relationship characteristics, work demands, workloads, subjective health indicators, socioeconomic status, household context and allow both cross-sectional and longitudinal analysis. This dataset is now available as a Scientific Use File for the scientific community via the Research Data Centre of the Federal Employment Agency at the Institute for Employment Research. The report contains information on data collection, data editing, anonymisation and the description of the variables of the two waves.
The presentation of virtual environments in real time has always been a demanding task. Specially designed graphics hardware is necessary to deal with the large amounts of data these applications typically produce. For several years the chipsets that were used allowed only simple lighting models and fixed algorithms. But recent development has produced new graphics processing units (GPUs) that are much faster and more programmable than their predecessors. This paper presents an approach to take advantage of these new features. It uses a video texture as part of the lighting calculations for the passenger compartment of a virtual train and was run on the GPU of a recent PC graphics card. The task was to map the varying illumination of a filmed landscape onto the virtual objects and also onto another video texture (showing two passengers), thereby enhancing the realism of the scene.
Load Matching and Grid Interaction of Net Zero Energy Buildings. EuroSun Conference 2010, Graz
(2010)
“Net Zero Energy Building” has become a prominent wording to describe the synergy of energy efficient building and renewable energy utilization to reach a balanced energy budget over a yearly cycle. Taking into account the energy exchange with a grid infrastructure overcomes the limitations of seasonal energy storage on-site. Even though the wording “Net Zero Energy Building” focuses on the annual energy balance, large differences may occur between solution sets in the amount of grid interaction needed to reach the goal. The paper reports on the analysis of example buildings concerning the load matching and grid interaction. Indices to describe both issues are proposed and foreseen as part of a harmonized definition framework. The work is part of subtask A of the IEA SHCP Task40/ECBCS Annex 52: “Towards Net Zero Energy Solar Buildings”.
The study examined whether the extent of negative body-related thoughts and emotions evoked by looking in a mirror can be changed by cognitive-behavioural body image therapy (BIT) and whether any remaining cognitive and affective responses dropped faster after the treatment. Females with eating disorders (n = 17) looked in a mirror for 40 minutes under standardised conditions before and after BIT. A control group (CG) (n = 24) was also exposed to their own bodies twice. Cognitive and affective reactions were assessed four times during the session. Hierarchical Linear Modelling revealed that among eating-disordered patients, the entrance levels of negative thoughts and feelings were lower at post compared to pre-BIT, whereas they remained stable in the CG. The extent of reduction of the remaining body image distress was not changed after the treatment. BIT seems to be a potent treatment of negative emotions and cognitions in an in vivo situation.
Magneto-hydrothermal triple-convection in a W-shaped porous cavity containing oxytactic bacteria
(2022)
Bioconvective heat and mass transport phenomena have recently been the subject of interest in diverse fields of applications pertaining to the motion of fluids and their thermophysical properties. The transport processes in a system involving triple convective phenomena, irregular geometry, and boundary conditions constitute a complex phenomenon. This work aims to explore the mixed thermo-bioconvection of magnetically susceptible fluid containing copper nanoparticles and oxytactic bacteria in a novel W-shaped porous cavity. The buoyant convention is generated due to the isothermal heating at the wavy bottom wall, whereas the mixed convection is induced due to the shearing motion of the top-cooled sliding wall. Furthermore, the bioconvection is induced due to the manifestation of oxytactic bacteria or organisms. The inclined sidewalls are insulated. The geometry is packed with water based Cu nanoparticle mixed porous structure, which is subjected to a magnetizing field acted horizontally. The complex transport equations are transformed into nondimensional forms, which are then computed using the finite volume-based developed code. The coupled triple-convective flow physics are explored for a wide range of involved controlling parameters, which could provide helpful insight to the system designer for its proper operation. The shape of geometry can be considered one of the important parameters to control the heat and mass transport phenomena. In general, the influence of amplitude (δ) is more compared to the waviness number (m) of the undulations. The magnitude of heat (Nu) and mass (Sh) transfer rate for the W-shaped cavity is high compared to conventional square and trapezoidal-shaped cavities. The output of the analysis could be very helpful for the designer for modeling devices operating on nanotechnology-based bioconvection, microbial fuel cells, and others.
Virtual (tv) studios gain much more acceptance through improvements in computer graphics and camera tracking. Still commercial studios cannot have full interaction between actors and virtual scene because actors data are not completely digital available as well as the feedback for actors is still not sufficient. Markerless full body tracking might revolutionize virtual studio technology as it allows better interaction between real and virtual world. This article reports about using a markerless actor tracking in a virtual studio with a tracking volume of nearly 40 cubic meter enabling up to three actors within the green box. The tracking is used for resolving the occlusion between virtual objects and actors so that the Tenderer can output automatically a mask for virtual objects in the foreground in case the actor is behind. It is also used for triggering functions scripted within the Tenderer engine, which are attached to virtual objects, starting any kind of action (e.g., animation). Last but not least the system is used for controlling avatars within the virtual set. All tracking and rendering is done within a studio frame rate of 50 Hz with about 3 frames delay. The markerless actor tracking within virtual studios is evaluated by experts using an interview approach. The statistical evaluation is based on a questionnaire.
Market and simulation analysis of PVT applications for the determination of new PVT test procedures
(2014)
Marktdaten
(2018)
Die in der Schriftenreihe Trading verwendeten Forschungsdaten beinhalten historische Kursdaten, geliefert von Lenz+Partner AG (Deutschland), für eine Auswahl an Aktien. Die Datei „FEBRDUSA_1“ zeigt diese Auswahl in Form der deutschen Wertpapierkennnummer (Spalte „Titel“), des Handelszeitraums (Spalten „Beginn“ und „Ende“) und des Aktiennamens. Kursdaten sind in dieser Datei nicht enthalten, sondern liegen in einer umfangreichen Datenbank vor.
Die Marktdaten umfassen 867 Aktien und sind das Ergebnis eines im Jahr 2015 durchgeführten Auswahlprozesses mit dem Ziel, Datenqualität und Handelbarkeit der Aktien zu verbessern. Es wurden folgende Bedingungen gefordert:
- Die Aktien werden an der Frankfurter Börse gehandelt.
- Sie sind in einem der deutschen Aktienindizes DAX, MDAX, TECDAX, SDAX, HDAX, CDAX, Technology All Share, Prime All Share und GEX oder in den amerikanischen Aktienindizes S&P 500 oder Nasdaq 100 gelistet.
- Die Kursdaten der in Frankfurt gehandelten Aktien enden nicht vor dem Jahr 2014.
- Der (unbereinigte) Eröffnungskurs Ende 2013 beträgt mindestens 1 EUR. Diese Bedingung soll dazu beitragen, das Handeln von „penny stocks“ zu verhindern.
Die vorliegende Marktdefinition resultierte in 867 Aktien, die frühestens am 11. Dezember 2014 enden, sodass eine Auswertung der Marktdaten mit vollständigen Kurswerten bis zu diesem Zeitpunkt möglich ist. Der Markt ist groß genug, um interessante Kaufbedingungen mit einer ausreichend großen Anzahl an Kaufkandidaten zu untersuchen.
High dynamic range environments maps based on still images or video streams are used for computer animation or interactive systems. The task of realistic light setup of scenes using captured environment maps might be eased as well as the visual quality improves. In this article, we discuss the light setting problem for virtual studio (tv) layout and system become more complex to handle this new feature of studio light capturing. The analysis of system requirements identifies the technical challenges.
Men afflicted with hypogonadism have reduced fat-free mass and in some cases a loss in muscle strength. We hypothesize that muscle elasticity of the lower extremities differs between patients with hypogonadism and healthy controls. The shear modulus of the soleus muscle was measured non-invasively using MR elastography (MRE) at 0%, 5%, 10%, 15%, and 20% of the subject’s maximum applied force. The results from this study showed statistical differences between the patients and controls, even though there were no statistical differences in applied force.
This dissertation, written for the course Master of European Social Policy Analysis – National University of Ireland Maynooth - , shows the context or so-called framework within which the medical social work service in two different welfare states: Ireland and Germany operates. This means the societal context as a whole, the health system in general (macro-level) and the hospital setting (mezzo-level) in particular. The idea of this publication is to give a short overview of the historic development of medical social work (micro-level) in Ireland and Germany and some background information about what medical social work is. The time frame examined starts from the 1980s up to present. The main questions to be answered are: What are the interdependencies between the macro-, mezzo- and micro-level? Have changes in the health system in general and the hospitals in particular influenced medical social work and if so how? What are the similarities and differences in the two countries?
Methanol Synthesis with Steel Mill Gases: Performance Investigations in an On‐Site Technical Center
(2021)
Method for producing a fermentation product: Angemeldet durch DIREVO Industrial Biotechnology GmbH
(2017)
Prioritätsdaten:
EP15191039A·2015-10-22; US201562245054P·2015-10-22
Bibliographische IPC:
C12P7/06; C12P7/14;
A method of determining the distance of an object from an automated vehicle based on images taken by a monocular image acquiring device. The object is recognized with an object-class by means of an image processing system. Respective position data are determined from the images using a pinhole camera model based on the object-class. Position data indicating in world coordinates the position of a reference point of the object with respect to the plane of the road is used with a scaling factor of the pinhole camera model estimated by means of a Bayes estimator using the position data as observations and under the assumption that the reference point of the object is located on the plane of the road with a predefined probability. The distance of the object from the automated vehicle is calculated from the estimated scaling factor using the pinhole camera model.
The task of the Center for Language Research is to provide content-based English language instruction for students of computer science and engineering. As such, we find ourselves at the confluence of many of the streams currently running through the English Language Teaching profession, including English for Science and Technology (EST), English for Academic Purposes (EAP), English for Specific Purposes (ESP), Computer-assisted language learning (CALL), content-based instruction, and multimedia applications in foreign language pedagogy. This paper describes our initial attempts to construct a number of World Wide Web pages where students will be able to study EST, EAP, and computer science topics on their own in a multimedia environment.
We report about several experiments on applying mixed reality technology in the context of accessing collective memories from atomic bombs, Holocaust and Second World War. We discuss the impact of Virtual Reality, Augmented Virtuality and Augmented Reality for specific memorial locations. We show how to use a virtual studio for demonstrating an augmented reality application for a specific location in a remote session within a video conference. Augmented Virtuality is used to recreate the local environment, thus providing a context and helping the participants recollect emotions related to a certain place. This technique demonstrates the advantages of using virtual (VR) and augmented (AR) reality environments for rapid prototyping and pitching project ideas in a live remote setting.
The article discusses the question of “How to convey the experience in a virtual environment to third parties?” and explains the different technical implementations which can be used for live streaming and recording of a mixed reality experience. The real-world applications of our approach include education, entertainment, e- sports, tutorials, and cinematic trailers, which can benefit from our research by finding a suitable solution for their needs. We explain and outline our Mixed Reality systems as well as discuss the experience of recorded demonstrations of different VR applications, including the need for calibrated camera lens parameters based on realtime encoder values.
Progressive web applications (PWAs) seem to be the next big thing in the mobile landscape. These are web applications with “super-powers” which are meant to provide the kinds of user experiences previously only native mobile applications could. Therefore they are able to match native applications in their capabilities but still thrive on the reachability of the web. They achieve this by implementing certain characteristics which originate from innovative standards and sophisticated best practices. Conveniently, as it is the web, PWAs are working seamlessly cross-platform. One of the closest things to a PWA so far may be an application created with the NativeScript framework. Such an application is also developed with web technologies yet able to use any native interface directly. This is made possible by using a designated runtime for mediating between JavaScript and the mobile system. Eventually, NativeScript is able to offer a high degree nativity yet also of convenient abstraction during development. This paper is set out to deliver further insight into both approaches by contrasting them based on the characteristics of PWAs. For this purpose, these characteristics are elaborated and adjusted to be applicable to native mobile applications. Hence a reasonable basis for what to expect from a native application is derived. Then NativeScript is assessed on this basis by the means of transferable concepts and technologies as well as a prototypic mobile application. Finally, an informed discussion and conclusion is performed based on the results. The comprehensive characteristics of PWAs resolve previous shortcomings with well thought out concepts and new technologies. These are transferable to native mobile applications to a far extent and may also be put into practice with NativeScript. The framework’s approach may be very well called ingenious, however, in the long run it might fall short of the innovative concept of PWAs. Having said this, it is still able to serve as a practical measure for creating appealing user experiences for multiple systems with arguably little effort. Due to large overlap a transformation between the two application types is a realistic option. Either way, web technologies are far from what they used to be and it is exciting to see how the mobile and web landscape will evolve in the future.
Mobile virtual archery
(2013)
A two-component system for control and monitoring of solar power towers with molten salt receivers is proposed. The control component consists of a model predictive control applica-tion (MPC) with a flexible objective function and on-line tunable weights, which runs on a In-dustrial PC and uses a reduced order dynamic model of the receiver’s thermal and flow dy-namics. The second component consists of a service-life monitoring unit, which estimates the service-life consumption of the absorber tubes depending on the current mode of operation based on thermal stresses and creep fatigue in the high temperature regime. The calculation of stresses is done based on a detailed finite element study, in which a digital twin of the re-ceiver was developed. By parallelising the model solver, the estimation of service-life con-sumption became capable of real-time operation. The system has been implemented at a test facility in Jülich, Germany, and awaits field experiments. In this paper, the modeling and archi-tecture are presented along simulation results, which were validated on a hardware-in-the-loop test bench. The MPC showed good disturbance rejection while respecting process variable constraints during the simulation studies.
We derive a nonlinear model of the steam cycle of a solar power plant, estimate its unknown parameters with measured data, and design a linear model predictive controller based on the resulting model. Simulations show that the linear model predictive controller is able to regulate the electrical power of the nonlinear steam process of the plant to a given reference trajectory very well. Moreover, it results in very good disturbance rejection.
Music listening has become a highly individualized activity with smartphones and music streaming services providing listeners with absolute freedom to listen to any kind of music in any situation. Until now, little has been written about the processes underlying the selection of music in daily life. The present study aimed to disentangle some of the complex processes among the listener, situation, and functions of music listening involved in music selection. Utilizing the experience sampling method, data were collected from 119 participants using a smartphone application. For 10 consecutive days, participants received 14 prompts using stratified-random sampling throughout the day and reported on their music-listening behavior. Statistical learning procedures on multilevel regression models and multilevel structural equation modeling were used to determine the most important predictors and analyze mediation processes between person, situation, functions of listening, and music selection. Results revealed that the features of music selected in daily life were predominantly determined by situational characteristics, whereas consistent individual differences were of minor importance. Functions of music listening were found to act as a mediator between characteristics of the situation and music-selection behavior. We further observed several significant random effects, which indicated that individuals differed in how situational variables affected their music selection behavior. Our findings suggest a need to shift the focus of music-listening research from individual differences to situational influences, including potential person-situation interactions.
Training autonomous vehicles requires lots of driving sequences in all situations. Collecting and labeling these drive scenes is a very time-consuming and expensive process. Currently, it is not possible to reuse these drive scenes with different optical properties, because there exists no numerically efficient model for the transfer function of the optical system. We present a numerical model for the point spread function (PSF) of an optical system that can efficiently model both experimental measurements and lens design simulations of the PSF. The numerical basis for this model is a nonlinear regression of the PSF with an artificial neural network. The novelty lies in the portability and the parameterization of this model. We present a lens measurement series, yielding a numerical function for the PSF that depends only on the parameters defocus, field, and azimuth. By convolving existing images and videos with this PSF, we generate images as if seen through the measured lens. The methodology applies to any optical scenario, but we focus on the context of autonomous driving, where the quality of the detection algorithms depends directly on the optical quality of the used camera system. With this model, it is possible to reuse existing recordings, with the potential to avoid millions of test drive miles. The parameterization of the optical model allows for a method to validate the functional and safety limits of camera-based advanced driver assistance systems based on the real, measured lens actually used in the product.
Developing security-critical systems is difficult, and there are many well-known examples of vulnerabilities exploited in practice. In fact, there has recently been a lot of work on methods, techniques, and tools to improve this situation already at the system specification and design. However, security-critical systems are increasingly long-living and undergo evolution throughout their lifetime. Therefore, a secure software development approach that supports maintaining the needed levels of security even through later software evolution is highly desirable. In this chapter, we recall the UMLsec approach to model-based security and discuss on tools and techniques to model and verify evolution of UMLsec models.
While developing for multiple platforms at once seems like a convenient solution, there are several challenges arising when trying to abstract the entire mobile development. This paper is meant to evaluate current cross-platform development for mobile applications. The background for its necessity, its conceptual approach and the problems to face when developing cross-platform were determined and explained in detail. Afterwards, certain solutions were evaluated against the former insights. Based on the results, an informed discussion and conclusion was performed. The mobile environment consist of only two big players by now, Android and iOS. These operating systems differ in architecture, design and consequently in the way applications are developed for each of them. Therefore, high demands are made towards cross-platform solutions. Tools which allow for the creation of applications for multiple platforms at once have to match native applications in regard to user experience and performance. At the same time they need to be able to optimize developments with the goal of being cost efficient. Apache Cordova, Xamarin and NativeScript were selected for evaluation in regard to their ability to meet these requirements. Cordova acts as the comparison group of cross-platform tools. It is the big player in the field and there are reasons for this. However, aspiring solutions with higher nativity and ambitious approaches are emerging. Xamarin and NativeScript deliver top quality results while offering loosely coupled developments. Therefore it is possible to develop high quality applications and still benefit from the advantages of platform-independent solutions. As a consequence mobile development is about to change in the foreseeable future. More sophisticated approaches may lead to a higher number of developments done cross-platform, and rightfully so.
Men afflicted with chronic obstructive pulmonary disease (COPD) show signs of muscle dysfunction, such as increased muscle fatigue and acidosis during exercise. We hypothesize that muscle elasticity of the lower extremities differs between patients with COPD and age-related controls. The shear modulus of the soleus muscle was measured non-invasively using MR elastography (MRE) at 0%, 5%, 10%, 15%, and 20% of the subject’s maximum applied force. The results from this study showed statistical differences between the patients and controls, even though there were no statistical differences in the maximum applied force.
Listening to music played by small MP3-players is almost as ubiquituous as talking over mobile phones. In this paper we analyse interaction elements and usabilty of current MP3-players. To determine the usability, an empirical study based on a website that presented three simulated MP3-players was performed with over 1000 participants.
The results show that, despite most marketing claims, some interaction devices are not as intuitive and really suitable for the task as they schould be. Also we found some surprising results concerning erroneous usage of navigation keys.
Mass spectrometry data is one of the key sources of information in many workflows in medicine and across the life sciences. Mass fragmentation spectra are generally considered to be characteristic signatures of the chemical compound they originate from, yet the chemical structure itself usually cannot be easily deduced from the spectrum. Often, spectral similarity measures are used as a proxy for structural similarity but this approach is strongly limited by a generally poor correlation between both metrics. Here, we propose MS2DeepScore: a novel Siamese neural network to predict the structural similarity between two chemical structures solely based on their MS/MS fragmentation spectra. Using a cleaned dataset of > 100,000 mass spectra of about 15,000 unique known compounds, we trained MS2DeepScore to predict structural similarity scores for spectrum pairs with high accuracy. In addition, sampling different model varieties through Monte-Carlo Dropout is used to further improve the predictions and assess the model's prediction uncertainty. On 3600 spectra of 500 unseen compounds, MS2DeepScore is able to identify highly-reliable structural matches and to predict Tanimoto scores for pairs of molecules based on their fragment spectra with a root mean squared error of about 0.15. Furthermore, the prediction uncertainty estimate can be used to select a subset of predictions with a root mean squared error of about 0.1. Furthermore, we demonstrate that MS2DeepScore outperforms classical spectral similarity measures in retrieving chemically related compound pairs from large mass spectral datasets, thereby illustrating its potential for spectral library matching. Finally, MS2DeepScore can also be used to create chemically meaningful mass spectral embeddings that could be used to cluster large numbers of spectra. Added to the recently introduced unsupervised Spec2Vec metric, we believe that machine learning-supported mass spectral similarity measures have great potential for a range of metabolomics data processing pipelines.
Abstract The modulation-transfer function (MTF) is a fundamental optical metric to measure the optical quality of an imaging system. In the automotive industry it is used to qualify camera systems for ADAS/AD. Each modern ADAS/AD system includes evaluation algorithms for environment perception and decision making that are based on AI/ML methods and neural networks. The performance of these AI algorithms is measured by established metrics like Average Precision (AP) or precision-recall-curves. In this article we research the robustness of the link between the optical quality metric and the AI performance metric. A series of numerical experiments were performed with object detection and instance segmentation algorithms (cars, pedestrians) evaluated on image databases with varying optical quality. We demonstrate with these that for strong optical aberrations a distinct performance loss is apparent, but that for subtle optical quality differences – as might arise during production tolerances – this link does not exhibit a satisfactory correlation. This calls into question how reliable the current industry practice is where a produced camera is tested end-of-line (EOL) with the MTF, and fixed MTF thresholds are used to qualify the performance of the camera-under-test.
Multi-level hp-finite cell method for embedded interface problems with application in biomechanics
(2018)
This work presents a numerical discretization technique for solving 3-dimensional material interface problems involving complex geometry without conforming mesh generation. The finite cell method (FCM), which is a high-order fictitious domain approach, is used for the numerical approximation of the solution without a boundary-conforming mesh. Weak discontinuities at material interfaces are resolved by using separate FCM meshes for each material sub-domain and weakly enforcing the interface conditions between the different meshes. Additionally, a recently developed hierarchical hp-refinement scheme is used to locally refine the FCM meshes to resolve singularities and local solution features at the interfaces. Thereby, higher convergence rates are achievable for nonsmooth problems. A series of numerical experiments with 2- and 3-dimensional benchmark problems is presented, showing that the proposed hp-refinement scheme in conjunction with the weak enforcement of the interface conditions leads to a significant improvement of the convergence rates, even in the presence of singularities. Finally, the proposed technique is applied to simulate a vertebra-implant model. The application showcases the method's potential as an accurate simulation tool for biomechanical problems involving complex geometry, and it demonstrates its flexibility in dealing with different types of geometric description.
The interpretation process of complex data sets makes the integration of effective interaction techniques crucial. Recent work in the field of human-computer interaction has shown that there is strong evidence that multimodal user interaction, i.e. the integration of various input modalities and interaction techniques into one comprehensive user interface, can improve human performance when interacting with complex data sets. However, it is still unclear which factors make these user interfaces superior to unimodal user interfaces. The contribution of this work is an analytical comparison of a multimodal and a unimodal user interface for a scientific visualization application. We show that multimodal user interaction with simultaneously integrated speech and gesture input improves user performance regarding efficiency and ease of use.
Music consumption and distribution has changed drastically in the last twenty years due to digitalization. Instead of buying music products, like CDs or even downloaded digital music files, today music is predominantly distributed via online music streaming. Therefore, it is mandatory for the music industry as well as researchers to determine which factors are affecting consumers intention to use music streaming services. This study aims to investigate drivers of Behavioral Intention (BI) to use a music streaming service. Particularly, aspects of social preferences in the context of music streaming will be explored. Therefore, a research model is created, based on the Value-based Adoption Model (VAM) (Kim et al., 2007) with an extension to take social factors into account. An online questionnaire with 102 participants was conducted to test the research model and the corresponding hypotheses. In order to evaluate reliable indications, some adjustments needed to be made resulting in a corrected research model. The final independent variables of the corrected model are Perceived Value (PV) (which is composed of Perceived Usefulness (PU) and Perceived Enjoyment (PE)), Perceived Ease of Use (PEU), Perceived Price Value (PPV), Facilitating Conditions (FC), Perceived Fairness (PF), Social Connectivity (SC), and Peer-Effects (PFX). The empirical results show that Perceived Value (PV), Perceived Price Value (PPV), and Perceived Fairness (PF) are determinants of consumers Behavioral Intention (BI) to use a music streaming service.