Browsing by Author "Weiskopf, Daniel (Prof. Dr.)"

Now showing 1 - 17 of 17

Open Access
3D visualization of multivariate data
(2012) Sanftmann, Harald; Weiskopf, Daniel (Prof. Dr.)
Nowadays large amounts of data are organized in tables, especially in relational databases where the rows store the data items to which multiple attributes are stored in the columns. Information stored this way, having multiple (more than two or three) attributes, can be treated as multivariate data. Therefore, visualization methods for multivariate data have a large application area and high potential utility. This thesis focuses on the application of 3D scatter plots for the visualization of multivariate data. When dealing with 3D, spatial perception needs to be exploited, by effectively using depth cues to convey spatial information to the user. To improve the presentation of individual 3D scatter plots, a technique is presented that applies illumination to them, thus using the shape-from-shading depth cue. To enable the analysis not only of 3D but of multivariate data, a novel technique is introduced that allows the navigation between 3D scatter plots. Inspecting the large number of 3D scatter plots that can be projected from a multivariate data set is very time consuming. The analysis of multivariate data can benefit from automatic machine learning approaches. A presented method uses decision trees to increase the speed a user can gain an understanding of the multivariate data at no extra cost. Stereopsis can also support the display of 3D scatter plots. Here an improved anaglyph rendering technique is presented, significantly reducing ghosting artifacts. The technique is not only applicable for information visualization, but for general rendering or to present stereoscopic image data. Some information visualization algorithms require high computation time. Many of these algorithms can be parallelized to run interactively. A framework that supports the parallelization on shared and distributed memory systems is presented.
Open Access
Adaptation of point- and line-based visualization
(2024) Rodrigues, Nils; Weiskopf, Daniel (Prof. Dr.)
Visualization plays an important role in the lives of various heterogeneous parts of society: from a voter looking for the latest results of an election, to statisticians examining a distribution, to analysts trying to make sense of multidimensional data sets. This thesis adapts existing point- and line-based visualization methods to improve knowledge gain. The included contributions address three research questions: How to scale unit visualization for 1D data? How to improve navigation between 2D visualizations of multivariate data? How to combine the advantages of multiple 2D views in a single static visualization for multivariate data? The first part of the thesis focuses on unit visualization of 1D data with dot plots. Compared to the previous state of the art, the developed visualizations fit a wider range of data and expand the number of potential users by requiring less prior knowledge for interpretation. They adapt the definition of dot plots to scale nonlinearly with sample count, accurately show value frequencies in high-dynamic-range data, reduce positional error in displayed data points, and enhance the perception of subtle nuances in the data while avoiding moiré effects. We provide evidence for claimed improvements through evaluation with computational metrics and a crowdsourced user study. The second part of the dissertation focuses on visualizing multivariate data with scatter plots and scatter plot matrices. First, we evaluate six animated transitions between plots of different 2D subspaces with respect to task performance for tracking individual points and interactions between clusters. The results of a quantitative study with 170 participants show that orthographic rotation animation performs best and should be adopted more widely. Next, we develop a novel concept for recommending views in scatter plot matrices. It provides user- and task-specific suggestions by focusing on the data of interest to the viewer. Together, animation and recommendation adapt scatter plots to improve the user's ability to analyze more complex data effectively. In the third part, we develop a new visualization technique that extends parallel coordinate plots to provide a static alternative to scatter plots with animated transitions. The approach does not require interaction to display data flow between 2D subspace clusters. A custom density-based rendering technique enables the visibility of individual lines and structures within highly overdrawn regions. Our technique can communicate fuzzy clustering results through binning and color mapping. Finally, we discuss the presented contributions with respect to the original main questions and show possible directions for future research.
Open Access
Animated surfaces in physically-based simulation
(2018) Huber, Markus; Weiskopf, Daniel (Prof. Dr.)
Physics-based animation has become a ubiquitous element in all application areas of computer animation, especially in the entertainment sector. Animation and feature films, video games, and advertisement contain visual effects using physically-based simulation that blend in seamlessly with animated or live-action productions. When simulating deformable materials and fluids, especially liquids, objects are usually represented by animated surfaces. The visual quality of these surfaces not only depends on the actual properties of the surface itself but also on its generation and relation to the underlying simulation. This thesis focuses on surfaces of cloth simulations and fluid simulations based on Smoothed Particle Hydrodynamics (SPH), and contributes to improving the creation of animations by specifying surface shapes, modeling contact of surfaces, and evaluating surface effects of fluids. In many applications, there is a reference given for a surface animation in terms of its shape. Matching a given reference with a simulation is a challenging task and similarity is often determined by visual inspection. The first part of this thesis presents a signature for cloth animations that captures characteristic shapes and their temporal evolution. It combines geometric features with physical properties to represent accurately the typical deformation behavior. The signature enables calculating similarities between animations and is applied to retrieve cloth animations from collections by example. Interactions between particle-based fluids and deformable objects are usually modeled by sampling the deformable objects with particles. When interacting with cloth, however, this would require resampling the surface at large planar deformations and the thickness of cloth would be bound to the particle size. This problem is addressed in this thesis by presenting a two-way coupling technique for cloth and fluids based on the simulation mesh of the textile. It allows robust contact handling and intuitive control of boundary conditions. Further, a solution for intersection-free fluid surface reconstruction at contact with thin flexible objects is presented. The visual quality of particle-based fluid animation highly depends on the properties of the reconstructed surface. An important aspect of the reconstruction method is that it accurately represents the underlying simulation. This thesis presents an evaluation of surfaces at interfaces of SPH simulations incorporating the connection to the simulation model. A typical approach in computer graphics is compared to surface reconstruction used in material sciences. The behavior of free surfaces in fluid animations is highly influenced by surface tension. This thesis presents an evaluation of three types of surface tension models in combination with different pressure force models for SPH to identify the individual characteristics of these models. Systematic tests using a set of benchmark scenes are performed to reveal strengths and weaknesses, and possible areas of applications.
Open Access
Computational methods for SPH-based fluid animation
(2021) Reinhardt, Stefan; Weiskopf, Daniel (Prof. Dr.)
In der Computergrafik hat sich Smoothed Particle Hydrodynamics (SPH) zu einem der wichtigsten Ansätze für physikalisch basierte Fluidanimation entwickelt. Die ursprüngliche Formulierung wurde in den letzten Jahrzehnten von vielen Autoren weiterentwickelt. Dadurch ist SPH zu einer vielseitig einsetzbaren Technik geworden, welche in der Lage ist, ein breites Spektrum von Phänomenen zu modellieren. Ein Kernaspekt jüngerer Forschung ist die Entwicklung von Methoden zur Verbesserung der Effizienz, Genauigkeit und visuellen Qualität der ursprünglichen Formulierung. Trotz aller Fortschritte ist die Forschung zu SPH noch immer ein sehr aktives Gebiet. In dieser Arbeit werden Methoden zur Verbesserung der physikalisch basierten Animation von Fluiden mit SPH vorgestellt. Die vorgestellten Beiträge befassen sich mit verschiedenen Herausforderungen, welche sich aus der Simulation von Fluiden mit SPH ergeben. Nach einer ausführlichen Diskussion der Grundlagen zu SPH-basierter Fluidanimation wird ein Ansatz zum visuellen Debugging von SPH-Simulationen vorgestellt. Dieser ist so konzipiert, dass er die Entwicklung neuer Techniken für SPH-basierte Fluidsimulation unterstützt. Eine Anforderungsanalyse wird durchgeführt, um die speziellen Bedürfnisse systematisch zu erörtern und die Anwendung wird dementsprechend gestaltet. Im Weiteren wird ein besonderes Augenmerk auf den Diskretisierungsprozess des numerischen Modells gelegt. Zunächst wird ein asynchrones Zeitintegrationsverfahren vorgestellt. Durch die Verwendung individueller Zeitschrittweiten wird die Effizienz des Simulationsprozesses verbessert. Anschließend wird ein konsistentes Verfahren zur Korrektur des Glättungskerns präsentiert. Dieser Ansatz trägt dazu bei, die im Diskretisierungsprozess auftretenden Fehler zu reduzieren und damit die Genauigkeit des Simulationsmodells zu verbessern. Er basiert auf der bekannten Shepard-Interpolation, beseitigt aber Inkonsistenzen, welche bei der Anwendung der Shepard-Interpolation auf SPH entstehen. Zuletzt wird ein Ansatz vorgestellt, um feine Details auf eine animierte Oberfläche hinzuzufügen. Solch eine Oberfläche resultiert beispielsweise aus einer SPH-basierten Flüssigkeitsanimation. Diese auf der Oberfläche modellierten Effekte werden durch das Geschwindigkeitsfeld der Basissimulation getrieben. Mit der präsentierten Methode können feine Details sehr effizient simuliert werden und das Erscheinungsbild der Fluids wird verbessert. Die vorgestellten Methoden berücksichtigen die speziellen Anforderungen von Computergrafikanwendungen. Insbesondere wird auf Effizienz, Genauigkeit und visuelle Qualität geachtet. Die Effizienz der Simulation ist im Bereich der Computergrafik von besonderer Bedeutung. Typischerweise sind dort sehr viele Simulationsdurchläufe der gleichen Sequenz nötig, da die Animation eines Fluids nach und nach verfeinert wird. Die Verbesserung der Effizienz des Simulationsprozesses ermöglicht zusätzlich eine feinere räumliche Diskretisierung, womit der Realitätsgrad der Simulation weiter erhöht werden kann. Bei der physikalisch basierten Animation von Fluiden mit SPH hängt die Genauigkeit nicht nur von der Simulationsauflösung ab, sondern auch von der Genauigkeit der SPH-Approximation. Verbesserungen in diesem Punkt erlauben es die physikalischen Phänomene besser zu modellieren und tragen zu einer realistisch wirkenden Simulation bei. Visuelle Qualität ist bei Computergrafikanwendungen immer von besonderer Bedeutung, da ansprechende Visualisierungen meist das gewünschte Ergebnis sind.
Open Access
Computational visualization of scalar fields
(2014) Ament, Marco; Weiskopf, Daniel (Prof. Dr.)
Scalar fields play a fundamental role in many scientific disciplines and applications. The increasing computational power offers scientists and digital artists novel opportunities for complex simulations, measurements, and models that generate large amounts of data. In technical domains, it is important to understand the phenomena behind the data to advance research and development in the application domain. Visualization is an essential interface between the usually abstract numerical data and human operators who want to gain insight. In contrast, in visual media, scalar fields often describe complex materials and their realistic appearance is of highest interest by means of accurate rendering models and algorithms. Depending on the application focus, the different requirements on a visualization or rendering must be considered in the development of novel techniques. The first part of this thesis presents three novel optical models that account for the different goals of photorealistic rendering and scientific visualization of volumetric data. In the first case, an accurate description of light transport in the real world is essential for realistic image synthesis of natural phenomena. In particular, physically based rendering aims to produce predictive results for real material parameters. This thesis presents a physically based light transport equation for inhomogeneous participating media that exhibit a spatially varying index of refraction. In addition, an extended photon mapping algorithm is introduced that provides a solution of this optical model. In scientific volume visualization, spatial perception and interactive controllability of the visual representation are usually more important than physical accuracy, which offers researchers more flexibility in developing goal-oriented optical models. This thesis presents a novel illumination model that approximates multiple scattering of light in a finite spherical region to achieve advanced lighting effects like soft shadows and translucency. The main benefit of this contribution is an improved perception of volumetric features with full interactivity of all relevant parameters. Additionally, a novel model for mapping opacity to isosurfaces that have a small but finite extent is presented. Compared to physically based opacity, the presented approach offers improved control over occlusion and visibility of such interval volumes. In addition to the visual representation, the continuously growing data set sizes pose challenges with respect to performance and data scalability. In particular, fast graphics processing units (GPUs) play a central role for current and future developments in distributed rendering and computing. For volume visualization, this thesis presents a parallel algorithm that dynamically decomposes image space and distributes work load evenly among the nodes of a multi-GPU cluster. The presented technique facilitates illumination with volumetric shadows and achieves data scalability with respect to the combined GPU memory in the cluster domain. Distributed multi-GPU clusters become also increasingly important for solving compute-intense numerical problems. The second part of this thesis presents two novel algorithms for efficiently solving large systems of linear equations in multi-GPU environments. Depending on the driving application, linear systems exhibit different properties with respect to the solution set and choice of algorithm. Moreover, the special hardware characteristics of GPUs in combination with the rather slow data transfer rate over a network pose additional challenges for developing efficient methods. This thesis presents an algorithm, based on compressed sensing, for solving underdetermined linear systems for the volumetric reconstruction of astronomical nebulae from telescope images. The technique exploits the approximate symmetry of many nebulae combined with regularization and additional constraints to define a linear system that is solved with iterative forward and backward projections on a distributed GPU cluster. In this way, data scalability is achieved by combining the GPU memory of the entire cluster, which allows one to automatically reconstruct high-resolution models in reasonable time. Despite their high computational power, the fine grained parallelism of modern GPUs is problematic for certain types of numerical linear solvers. The conjugate gradient algorithm for symmetric and positive definite linear systems is one the most widely used solvers. Typically, the method is used in conjunction with preconditioning to accelerate convergence. However, traditional preconditioners are not suitable for efficient GPU processing. Therefore, a novel approach is introduced, specifically designed for the discrete Poisson equation, which plays a fundamental role in many applications. The presented approach builds on a sparse approximate inverse of the matrix to exploit the strengths of the GPU.
Open Access
Encoding high dynamic range and wide color gamut imagery
(2017) Fröhlich, Jan; Weiskopf, Daniel (Prof. Dr.)
In dieser Dissertation wird ein szenischer Bewegtbilddatensatz mit erweitertem Dynamikumfang (High Dynamic Range, HDR) und großem Farbumfang (Wide Color Gamut, WCG) eingeführt und es werden Modelle zur Kodierung von HDR und WCG Bildern vorgestellt. Die objektive und visuelle Evaluation neuer HDR und WCG Bildverarbeitungsalgorithmen, Kompressionsverfahren und Bildwiedergabegeräte erfordert einen Referenzdatensatz hoher Qualität. Daher wird ein neuer HDR- und WCG-Video-Datensatz mit einem Dynamikumfang von bis zu 18 fotografischen Blenden eingeführt. Er enthält inszenierte und dokumentarische Szenen. Die einzelnen Szenen sind konzipiert um eine Herausforderung für Tone Mapping Operatoren, Gamut Mapping Algorithmen, Kompressionscodecs und HDR und WCG Bildanzeigegeräte darzustellen. Die Szenen sind mit professionellem Licht, Maske und Filmausstattung aufgenommen. Um einen cinematischen Bildeindruck zu erhalten, werden digitale Filmkameras mit ‘Super-35 mm’ Sensorgröße verwendet. Der zusätzliche Informationsgehalt von HDR- und WCG-Videosignalen erfordert im Vergleich zu Signalen mit herkömmlichem Dynamikumfang eine neue und effizientere Signalkodierung. Ein Farbraum für HDR und WCG Video sollte nicht nur effizient quantisieren, sondern wegen der unterschiedlichen Monitoreigenschaften auf der Empfängerseite auch für die Dynamik- und Farbumfangsanpassung geeignet sein. Bisher wurden Methoden für die Quantisierung von HDR Luminanzsignalen vorgeschlagen. Es fehlt jedoch noch ein entsprechendes Modell für Farbdifferenzsignale. Es werden daher zwei neue Farbräume eingeführt, die sich sowohl für die effiziente Kodierung von HDR und WCG Signalen als auch für die Dynamik- und Farbumfangsanpassung eignen. Diese Farbräume werden mit existierenden HDR und WCG Farbsignalkodierungen des aktuellen Stands der Technik verglichen. Die vorgestellten Kodierungsschemata erlauben es, HDR- und WCG-Video mittels drei Farbkanälen mit 12 Bits tonaler Auflösung zu quantisieren, ohne dass Quantisierungsartefakte sichtbar werden. Während die Speicherung und Übertragung von HDR und WCG Video mit 12-Bit Farbtiefe pro Kanal angestrebt wird, unterstützen aktuell verbreitete Dateiformate, Videoschnittstellen und Kompressionscodecs oft nur niedrigere Bittiefen. Um diese existierende Infrastruktur für die HDR Videoübertragung und -speicherung nutzen zu können, wird ein neues bildinhaltsabhängiges Quantisierungsschema eingeführt. Diese Quantisierungsmethode nutzt Bildeigenschaften wie Rauschen und Textur um die benötigte tonale Auflösung für die visuell verlustlose Quantisierung zu schätzen. Die vorgestellte Methode erlaubt es HDR Video mit einer Bittiefe von 10 Bits ohne sichtbare Unterschiede zum Original zu quantisieren und kommt mit weniger Rechenkraft im Vergleich zu aktuellen HDR Bilddifferenzmetriken aus.
Open Access
Enhancing fluid animation with fine detail
(2021) Morgenroth, Dieter; Weiskopf, Daniel (Prof. Dr.)
Wasser oder allgemein Flüssigkeiten sind beliebte Bestandteile für Actionszenen in Filmen. Deshalb ist im Bereich der Computergrafik für visuelle Effekte (VFX) die Simulation und das Rendern von Flüssigkeiten eine häufig benötigte Fähigkeit. Die Flüssigkeitssimulationen sind dabei meist groß angelegt und decken Längenskalen im Meter- bis Kilometerbereich ab. Es gibt aber physikalische Effekte von Flüssigkeiten, die sich im kleinen Maßstab abspielen, aber eine auffällige optische Wirkung auch im Großen haben können. Die Berechnung dieser kleinskaligen Effekte benötigt eine sehr hohe Auflösung und ist deshalb bei der Simulation großer Szenen aus Zeit- und Kostengründen oft nicht möglich. Diese Dissertation diskutiert Strategien, um Fluidsimulationen mit kleinskaligen physikalischen Effekten zu ergänzen. Der erste Teil dieser Dissertation beschreibt eine VFX-Produktionspipeline für Flüssigkeitssimulationen, die die neuen Möglichkeiten von Cloud-Computing- Angeboten ausnutzt, um die Auflösung dank erweiterter Rechenleistung zu erhöhen und so mehr Details zu erreichen. Dabei wurde mittels einer Client/ Serverarchitektur und Remote-Procedure-Calls ein System aufgebaut, das interaktives Arbeiten an einer Simulationsszene in einer lokalen Applikation ermöglicht, in dem die aufwendigen Berechnungen der Simulation auf einem ausgelagerten Rechner in der Cloud stattfinden. Dabei wurde auch auf spezielle GPU-Hardware zurückgegriffen. Ein weiterer Beitrag im Rahmen dieses Systems ist die Einführung von “Blind Particles”, bei deren Verwendung es möglich wird, unnötige Partikel aus Datensätzen zu löschen, ohne das visuelle Ergebnis zu beeinflussen. Dadurch kann Bandbreite und Renderzeit gespart werden. Der zweite Teil der Dissertation stellt eine direkte Raytracing-Methode für implizit beschriebene Flüssigkeitsoberflächen vor, die die Kappilareffekte an den Grenzflächen zu Festkörpern berücksichtigt. Das Verfahren verwendet die analytische Lösung der Meniskusform an der Fluidgrenzfläche, um den Effekt der Oberflächenspannung zwischen Wasseroberfläche und Festkörper zu erzielen. Das Verfahren erzeugt korrekte Kontaktwinkel an den Rändern, ohne dass eine rechenintensive Simulation erforderlich ist. Zur Renderzeit kombiniert es die analytische Lösung für einen kleinskaligen Effekt mit der numerischen Lösung einer großskaligen Simulation. Das Verfahren garantiert den richtigen Kontaktwinkel und liefert in bestimmten Szenarien die richtige Lösung über die gesamte Grenzfläche; selbst in allgemeinen Szenarien liefert es plausible Ergebnisse. Im letzten Teil wird ein Verfahren vorgestellt, um Fluidströmungen auf sich entwickelnden Oberflächen zu simulieren, z.B. einen Ölfilm auf einer Wasseroberfläche. Bei einer animierten Oberfläche (z.B. extrahiert aus einer partikelbasierten Fluidsimulation) im dreidimensionalen Raum wird eine zweite Simulation auf der Eingabeoberfläche hinzugefügt. Im Allgemeinen wird eine partielle Differentialgleichung auf einer Level-Set-Oberfläche gelöst. Es werden Kopplungsstrategien zwischen Eingabeeigenschaften und der Simulation eingeführt, und Masse- und Impulserhaltung wird zu bestehenden Methoden hinzugefügt. Auf diese Weise werden hochauflösende 2D-Simulationen auf groben Eingabeflächen effizient berechnet.
Open Access
Interactive visual analysis of vector fields
(2013) Bachthaler, Sven; Weiskopf, Daniel (Prof. Dr.)
Visualization is a very active research area due to several reasons. For years, data sets have been getting larger and more complex, increasing the difficulty of handling this data. Furthermore, in technical application areas, visualization is an essential part of the engineering process. These developments drive the need for improvements of all aspects of scientific visualization, as well as the integration of information visualization techniques. This thesis focuses on the development of visualization and analysis techniques for different types of vector fields - vector fields representing the flow of air or water, but also magnetic fields and vector fields derived computationally from scalar fields. The different techniques that were developed to handle such fields are organized in three parts: the first part presents methods that visualize vector fields in dense manner. The second part discusses methods that rely on topological approaches - the complexity of the visualization is reduced by concentrating on features of the data. In the third and final part, continuous scatterplots are introduced, which are designed to analyze correlations in multivariate data sets. In the first part, the goal is to show as much information as possible and using every available pixel of the viewport to do so. However, one of the challenges of dense visualization methods is to maintain interactivity for high resolution visualizations. A cluster environment is used here to offer increased rendering performance and memory size for large and complex data sets. Additionally, an animation-based approach is presented that allows one to decouple the line-like patterns of LIC from the direction of animation. This decoupling is desirable since perception research suggests that LIC-based techniques combined with animation are non-optimal for local motion detection of the human visual system. The second part focuses on topological methods to filter the data and hence, reduce the complexity of the resulting visualization. For time-dependent vector fields, Lagrangian coherent structures are used to visualize space-time manifolds that represent the topology of these fields. Furthermore, the dynamic of such fields is visualized directly on these space-time manifolds, allowing us to quantify the hyperbolicity close to the topological skeleton. In addition, another technique is presented in the second part that allows one to visualize the topology of magnetic fields based on dipoles. Here, traditional topological methods are non-optimal, hence, an alternative topology is developed that visualizes the existence and magnitude of magnetic flux between dipoles. In the final part, the mathematical basis and several computational approaches are presented to compute continuous scatterplots. These plots are designed to work with data sets defined on a continuous domain, which is typical for scientific visualization data. In contrast to traditional scatterplots, they visualize the density in the data domain, instead of merely plotting data attached at discrete sampling positions. The additional computational approaches are an improvement of the original approach in terms of flexibility - they allow a trade-off between output quality and rendering performance, as well as the use of generic interpolation methods.
Open Access
Matrix methods in visualization
(2024) Krake, Tim; Weiskopf, Daniel (Prof. Dr.)
The theory of matrices has a long history that began over 4000 years ago. It took a while until matrices were studied systematically in the context of linear algebra. While these results from the 18th and 19th century were mainly characterized by theoretical thoughts, the modern use of matrices is usually linked to computational aspects. This aspect made the theory of matrices extremely useful for applied sciences, such as computer graphics and visualization, and paved the way for innovative matrix methods. The overall goal of this thesis is to integrate such matrix methods into the field of data analysis and visualization, where emphasis is placed on matrix decompositions. In this context, the following four concepts are addressed: the examination of linear structures and matrix formulations, the utilization of matrix formulations and matrix methods, the customization of matrix methods for visualization, and the augmentation of visualization techniques. These four conceptual steps characterize a sequential process that is used throughout the chapters of this thesis. With a main focus on data-driven methods that reveal time evolutionary and statistical patterns, the contents of the chapters refer to different fields of application. Chapter 2 demonstrates applications of Dynamic Mode Decomposition in the context of visual computing, and Chapter 3 addresses the challenges of uncertainty propagation and visualization. In contrast, Chapters 4 and 5 present methods in the context of structural analysis (solid mechanics) and smoothed particle hydrodynamics (fluid mechanics). The overall content of this thesis demonstrates the versatile, effective use of matrices for visual computing.
Open Access
Particle tracing methods for visualization and computer graphics
(2008) Schafhitzel, Tobias; Weiskopf, Daniel (Prof. Dr.)
This thesis discusses the broad variety of particle tracing algorithms with focus on flow visualization. Starting with a general overview of the basics of visualization and computer graphics, mathematics, and fluid dynamics, a number of methods using particle tracing for flow visualization and computer graphics are proposed. The first part of this thesis considers mostly texture-based techniques that are implemented on the graphics processing unit (GPU) in order to provide an interactive dense representation of 3D flow fields. This part considers particle tracing methods that can be applied on general vector fields and includes texture based visualization in volumes as well as on surfaces. Furthermore, it is described how particle tracing can be used for extracting flow structures, like path surfaces, of the given vector field. The second part of this thesis considers particle tracing on derived vector fields for flow visualization. Therefore, first a feature extraction criterion is applied on a fluid flow field. In most cases this results in a scalar field serving as base for the particle tracing methods. Here, it is shown how higher order derivatives of scalar fields can be used to extract flow features like 1D vortex core lines or 2D shear sheets. The extracted structures are further processed in terms of feature tracking. The third part generalizes particle tracing for arbitrary applications in visualization and computer graphics. Here, the particles' path either might be defined by the perspective of the human eye or by a force field that influences the particles' motion by considering second order ordinary differential equations. All three parts clarify the importance of particle tracing methods for a wide range of applications in flow visualization and computer graphics by various examples. Furthermore, it is shown how the flexibility of this method strongly depends on the underlying vector field, and how those vector fields can be generated in order to solve problems that go beyond traditional particle tracing in fluid flow fields.
Open Access
Uncertainty-aware visualization techniques
(2021) Schulz, Christoph; Weiskopf, Daniel (Prof. Dr.)
Nearly all information is uncertainty-afflicted. Whether and how we present this uncertainty can have a major impact on how our audience perceives such information. Still, uncertainty is rarely visualized and communicated. One difficulty is that we tend to interpret illustrations as truthful. For example, it is difficult to understand that a drawn point’s presence, absence, and location may not convey its full information. Similarly, it may be challenging to classify a point within a probability distribution. One must learn how to interpret uncertainty-afflicted information. Accordingly, this thesis addresses three research questions: How can we identify and reason about uncertainty? What are approaches to modeling flow of uncertainty through the visualization pipeline? Which methods are suitable for harnessing uncertainty? The first chapter is concerned with sources of uncertainty. Then, approaches to model uncertainty using descriptive statistics and unsupervised learning are discussed. Also, a model for validation and evaluation of visualization methods is proposed. Further, methods for visualizing uncertainty-afflicted networks, trees, point data, sequences, and time series are presented. The focus lies on modeling, propagation, and visualization of uncertainty. As encodings of uncertainty, we propose wave-like splines and sampling-based transparency. As an overarching approach to adapt existing visualization methods for uncertain information, we identify the layout process (the placement of objects). The main difficulty is that these objects are not simple points but distribution functions or convex hulls. We also develop two stippling-based methods for rendering that utilize the ability of the human visual system to cope with uncertainty. Finally, I provide insight into possible directions for future research.
Open Access
Video visual analytics
(2013) Höferlin, Markus Johannes; Weiskopf, Daniel (Prof. Dr.)
The amount of video data recorded world-wide is tremendously growing and has already reached hardly manageable dimensions. It originates from a wide range of application areas, such as surveillance, sports analysis, scientific video analysis, surgery documentation, and entertainment, and its analysis represents one of the challenges in computer science. The vast amount of video data renders manual analysis by watching the video data impractical. However, automatic evaluation of video material is not reliable enough, especially when it comes to semantic abstraction from the video signal. In this thesis, the visual analytics methodology is applied to the video domain to combine the complementary strengths of human cognition and machine processing. After depicting the challenges of scalable video analysis, a video visual analytics pipeline is proposed that relies on stream processing for scalability. The proposed video visual analytics pipeline consists of six stages that are processed successively--data stream selection, manipulation, feature extraction, filtering, relevance measure, and visualization--before the results are presented to the human analysts. The human analysts can interact and modify each of these stages iteratively. To support sense-making, the human analysts can directly integrate and organize reasoning artifacts into a reasoning sandbox. For the video visual analytics pipeline, various methods for the different stages are introduced that address data scalability, task scalability, and situational awareness. This work focuses mainly on the filtering and visualization stages, but provides reviews and discussions of techniques for the other stages as well. In the filtering stage, four interaction guidelines--easy-to-use filter definition, confidence-incorporated filter definition, decision-guided filter definition, and filter feedback--are defined and applied to formulate filters by properties, by sketch, or by example. Due to the suitability of trajectories for filtering, a configurable similarity metric for trajectories is introduced that allows combining different facets (features) with different similarity measures. Besides a survey on video visualization methods, the thesis contributes to the visualization stage by methods for fast-forward video visualization and hierarchical video exploration (the interactive schematic summaries). The VideoPerpetuoGram is extended and applied to different domains (video surveillance and snooker skill training), and an example of video visualization that solely depends on extracted features from video (the layered TimeRadarTrees) is discussed. Moreover, two sonification approaches with the purpose to improve situational awareness are introduced.
Open Access
Vision-based methods for evaluating visualizations
(2018) Netzel, Rudolf; Weiskopf, Daniel (Prof. Dr.)
Open Access
Visual analytics of eye-tracking and video data
(2018) Kurzhals, Kuno; Weiskopf, Daniel (Prof. Dr.)
Eye tracking, i.e., the detection of gaze points, becomes increasingly popular in numerous research areas as a means to investigate perceptual and cognitive processes. In comparison to other evaluation methods, eye tracking provides insights into the distribution of attention and sequential viewing behavior, which are essential for many research questions. For visualization research, such insights help assess a visualization design and identify potential flaws. Gaze data coupled with a visual stimulus poses a complex analysis problem that is approached by statistical and visual methods. Statistical methods are often limited to hypothesis-driven evaluation and modeling of processes. Visualization is applied to confirm statistical results and for exploratory data analysis to form new hypotheses. Surveying the state of the art of visualizations for eye tracking shows a deficiency of appropriate methods, particularly for dynamic stimuli (e.g., videos). Video visualization and visual analytics provide methods that can be adapted to perform the required analysis processes. The automatic processing of video and gaze data is combined with interactive visualizations to provide an overview of the data, support efficient browsing, detect interesting events, and annotate important parts of the data. The techniques developed for this thesis focus on the analysis of videos from remote and from mobile eye tracking. The discussed remote eye-tracking scenarios consist of one video that is investigated by multiple participants. Mobile eye tracking comprises scenarios in which participants wear glasses with a built-in device to record their gaze. Both types of scenarios pose individual challenges that have to be addressed for an effective analysis. In general, the comparison of gaze behavior between participants plays an important role to detect common behavior and outliers. This thesis addresses the topic of eye tracking and visualization bidirectionally: Eye tracking is applied in user studies to evaluate visualization techniques beyond established performance measures and questionnaires. The current application of eye tracking in visualization research is surveyed. Further, it is discussed how existing methodology can be extended to incorporate eye tracking for future analysis scenarios. Vice versa, a set of new visualization techniques for data from remote and mobile eye-tracking devices are introduced that support the analysis of gaze behavior in general. Here, techniques for raw data and for data with annotations are introduced, as well as approaches to perform the tedious annotation process more efficiently.
Open Access
Visualization for integrated simulation systems
(2015) Hlawatsch, Marcel; Weiskopf, Daniel (Prof. Dr.)
Today, a large part of science as well as many applications in industry require the usage of simulation technology. Several key elements are relevant for an expedient use of modern simulation technology. It is important that the methods for analyzing the simulation outcome are suitable for the increasing complexity of simulations and the generated data. Larger or more complex data is analyzed visually in many cases. Therefore, visualization is essential for the work with simulation technology. Furthermore, an integrated simulation system providing simulation components as well as visualization components allows a more efficient work with simulation technology. Finally, uncertainties can be introduced at different stages in the simulation and analysis process. Correct interpretation of simulation results must consider these uncertainties. The theme of this thesis is to improve the work with simulation technology by employing novel visualization approaches and techniques. Each of these approaches targets one or more of these key elements. Since the key role of visualization in this context lies in displaying the simulation results, novel methods for visualizing different classes of data are introduced. It is discussed how scalar values covering a large value range can be visualized. Furthermore, several methods for time-dependent vector fields are presented. Another approach deals with the visualization of coherent structures in symmetric second-order tensor fields Effective analysis of simulation results is important, but not the only aspect that can be supported with visualization. Considering integrated simulation systems, such a complex software environment is typically built in a modular way. Workflows are a common way to control modular software and define the connections between the respective modules and their execution. This thesis presents two methods for visualizing the evolution of such workflows. They can be used to recapitulate previous sessions or analyze user behavior. This can help continue previous work or improve the respective software. Finally, three different types of uncertainty are discussed in this thesis and methods for handling them are presented: Uncertainty in the simulation model can be handled with a computational steering approach that allows the interactive change of simulation parameters. A glyph-based visualization method is introduced for time-dependent vector fields with uncertainty. Finally, the handling of uncertainty in user interaction is demonstrated in the context of flow visualization.
Open Access
Visualization techniques for group structures in graphs
(2015) Vehlow, Corinna; Weiskopf, Daniel (Prof. Dr.)
Graph visualization plays a key role analyzing relations between objects. With increasing size of the graph, it becomes difficult to understand global and local structures of the graph. Grouping objects of the graph based on their attributes or relations helps reveal global structures. Visualizing these group structures together with the graph topology can highlight central objects and reveal outliers. The ability of a visualization to help detecting these features becomes more difficult for groups that overlap or change over time. In many applications, groups cannot be interpreted as disjoint sets of objects. In fact, objects are often involved in several groups, sometimes even to different extent. With the existing types of overlapping groups, further analysis tasks arise that need to be considered for the visualization. In addition, real-world scenarios are not static but change over time and so do relations among objects. With the graph topology changing over time, the group structure changes as well. The challenge for visualizations of dynamic groups in dynamic graphs is to facilitate the analysis of group-related features not only for individual points in time but over time, showing group evolution events. This thesis presents visualization techniques for group structures in graphs that address these challenges: overlap and time dependency. As a basis, a survey of the state of the art in visualizing group structures in graphs is presented. The first part of this thesis is dedicated to the visualization of overlapping groups in static graphs, where different types of overlaps are considered. With each technique, the complexity of the groups increases. First, a visual analytics system for crisp overlapping groups in multivariate graphs is presented. This system integrates interactive filtering of large and dense networks with groupbased layouts of the resulting subnetworks and a technique to compare those subnetworks. Second, a technique that visualizes fuzzy overlapping groups in a graph based on layout strategies and further visual mappings is presented. This technique facilitates the investigation of fuzzy group memberships at different levels of detail based on a hierarchical aggregation model. In contrast to these techniques, the third visualization technique shows groups based on multivariate edge attributes rather than vertex attributes or the topology of the graph. In particular, edge-edge relations are visualized as curves that are directly integrated into the node-link diagram representing the object-relation structure. The second part is dedicated to visualization techniques for dynamic groups in dynamic graphs. Again, the complexity of the group structure rises from the first technique addressing flat groups to the second technique addressing more complex hierarchical groups. Within both techniques, the evolution of groups is encoded using a flow metaphor. The first technique visualizes the partially aggregated graphs by node-link diagrams, whereas the second technique is based on an extended adjacency matrix representation that encodes the hierarchical structure of vertices as well as changes in the graph topology. All presented techniques visualize the group structure integrated with the graph topology in a single image. Finally, the use of all techniques is demonstrated for real data sets from biology, one of the main application domains of group structures in graphs.
Open Access
Visualization techniques for parallel coordinates
(2013) Heinrich, Julian; Weiskopf, Daniel (Prof. Dr.)
Visualization plays a key role in knowledge discovery, visual data exploration, and visual analytics. Static images are an effective tool for visual communication, summarization, and pattern extraction in large and complex datasets. Only together with human-computer-interaction techniques, visual interfaces enable an analyst to explore large information spaces and to drive the whole analytical reasoning process. Scatterplots and parallel coordinates are well-recognized visualization techniques that are commonly employed for statistics (both explorative and descriptive) and data-mining, but are also gaining importance for scientific visualization. While scatterplots are restricted to the display of at most three dimensions due to the orthogonal layout of coordinate axes, a parallel arrangement allows for the visualization of multiple attributes of a dataset. Although both techniques rely on projections of higher-dimensional geometry and are related by a point–line duality, parallel coordinates enjoy great popularity for the visualization and analysis of multivariate data. Despite their popularity, parallel coordinates are subject to a number of limitations that remain to be solved. For large datasets, the potentially high amount of overlapping lines may hinder the observer from visually extracting meaningful patterns. Encoding observations with polylines make it difficult to follow lines over all dimensions, as they lose visual continuation across the axes. Clusters cannot be represented by the geometry of lines, and the order of axes has a high impact on the patterns exhibited by parallel coordinates. This thesis presents visualization techniques for parallel coordinates that address these limitations. As a foundation, an extensive review of the state of the art of parallel coordinates will be given. Based on the point–line duality, the existing model of continuous scatterplots is adapted to parallel coordinates for the visualization of data defined on continuous domains. To speed up computation and obtain interactive frame rates, a scalable and progressive rendering algorithm is introduced that further allows for arbitrary reconstruction and interpolation schemes. A curve-bundling model for parallel coordinates is evaluated with a user study showing that bundling is effective for cluster visualization based on geometric cues while being equally capable of revealing correlations between neighboring axes. To address the axis-order problem, a graph-based approach is presented that allows for the visualization of all pairwise relations in a matrix layout without redundancy. Finally, the use of parallel coordinates is demonstrated for real datasets from computational fluid dynamics, motion capturing, bioinformatics, and systems biology.