- Data Visualization and Analytics
- Gaze Tracking and Assistive Technology
- Geographic Information Systems Studies
- Video Analysis and Summarization
- Data Management and Algorithms
- Tactile and Sensory Interactions
- Data-Driven Disease Surveillance
- Augmented Reality Applications
- Interactive and Immersive Displays
- Advanced Text Analysis Techniques
- 3D Shape Modeling and Analysis
- Advanced Clustering Algorithms Research
- Image and Video Quality Assessment
- Advanced Vision and Imaging
- Color Science and Applications
- Virtual Reality Applications and Impacts
- Advanced Image and Video Retrieval Techniques
- Semantic Web and Ontologies
- Public Relations and Crisis Communication
- Natural Language Processing Techniques
- Species Distribution and Climate Change
- Context-Aware Activity Recognition Systems
- Anomaly Detection Techniques and Applications
- Mobile Crowdsensing and Crowdsourcing
- Complex Network Analysis Techniques
University of Stuttgart
2017-2025
Abstract Eye tracking provides a unique perspective on the inherently visual discourse between visualisation systems and their users, has recently become sufficiently precise affordable to be integrated as regular input into workstations virtual or augmented reality headsets alike. As such, real‐time eye can now contribute significantly towards development of gaze‐aware visualisations that infer monitor users' needs actively support activities. To facilitate such we make three contributions....
We present an exploratory study on the accessibility of images in publications when viewed with color vision deficiencies (CVDs). The is based 1,710 sampled from a visualization dataset (VIS30K) over five years. simulated four CVDs each image. First, researchers (one CVD) identified existing issues and helpful aspects subset images. Based resulting labels, 200 crowdworkers provided 30,000 ratings CVD analyzed this data for correlations, clusters, trends, free text comments to gain first...
Abstract Adaptive visualization and interfaces pervade our everyday tasks to improve interaction from the point of view user performance experience. This approach allows using several inputs, whether physiological, behavioral, qualitative, or multimodal combinations, enhance interaction. Due multitude approaches, we outline current research trends inputs used adapt visualizations interfaces. Moreover, discuss methodological approaches in mixed reality, physiological computing, visual...
Visual analytics (VA) research provides helpful solutions for interactive visual data analysis when exploring large and complex datasets. Due to recent advances in eye tracking technology, promising opportunities arise extend these traditional VA approaches. Therefore, we discuss foundations support systems. We first review the structure range of typical Based on a widely used model, present five comprehensive examples that cover wide usage scenarios. Then, demonstrate model can be...
Visualization in virtual 3D environments can provide a natural way for users to explore data. Often, arm and short head movements are required interaction augmented reality, which be tiring strenuous though. In an effort toward more user-friendly interaction, we developed prototype that allows manipulate objects using combination of eye gaze external clicker device. Using this prototype, performed user study comparing four different input methods plus was preferred by most participants.
Conventional dot plots use a constant size and are typically applied to show the frequency distribution of small data sets. Unfortunately, they not designed for high dynamic range frequencies. We address this problem by introducing nonlinear plots. Adopting idea scaling from logarithmic bar charts, our allow dots varying so that columns with large number samples reduced in height. For construction these diagrams, we introduce an efficient two-way sweep algorithm leads dense symmetrical...
Gaze tracking in 3D has the potential to improve interaction with objects and visualizations augmented reality. However, previous research showed that subjective perception of distance varies between real virtual surroundings. We wanted determine whether objectively measured gaze depth through eye also exhibits differences entirely environments. To this end, we conducted an experiment (N = 25) which used Microsoft HoloLens a binocular add-on from Pupil Labs. Participants performed task...
Visualizing time series data with a spatial context is problem that appears more and often, since small lightweight GPS devices allow us to enrich the position information. One example visualization of energy output power plants. We present web-based application aims provide information about production specified region, along location The intended be used as solid basis for political discussions, nudging, story telling German transition renewables, called "Energiewende". It was therefore...
The analysis of eye-tracking data can be very useful when evaluating controlled user studies. To support the in a fast and easy fashion, we have developed web-based framework for visual inspection comparison scanpaths based on filtering fixations similarity measures. Concerning first part, introduce multiscale aggregation saccades spatial partitioning that reduces clutter overlaid without changing overall impression large-scale eye movements. technique abstracts individual allows an analyst...
We propose a new technique for visual analytics and annotation of long-term pervasive eye tracking data which combined analysis gaze egocentric video is necessary. Our approach enables two important tasks such hour-long videos from individual participants: (1) efficient (2) direct interpretation the results. Exemplary time spans can be selected by user are then used as query that initiates fuzzy search similar based on features. In an iterative refinement loop, interface provides suggestions...
Scatter plots are popular for displaying 2D data, but in practice, many data sets have more than two dimensions. For the analysis of such multivariate it is often necessary to switch between scatter different dimension pairs, e.g., a plot matrix (SPLOM). Alternative approaches include "grand tour" an overview entire set or creating artificial axes from dimensionality reduction (DR). A cross-cutting concern all techniques ability viewers find correspondence points views. Previous work...
Scatter plots are popular for displaying 2D data, but in practice, many data sets have more than two dimensions. For the analysis of such multivariate it is often necessary to switch between scatter different dimension pairs, e.g., a plot matrix (SPLOM). Alternative approaches include "grand tour" an overview entire set or creating artificial axes from dimensionality reduction (DR). A cross-cutting concern all techniques ability viewers find correspondence points views. Previous work...
We present a framework for rapid prototyping of pervasive visual deficiency simulation in the context graphical interfaces, virtual reality, and augmented reality. Our facilitates emulation various deficiencies wide range applications, which allows users with normal vision to experience combinations conditions such as myopia, hyperopia, presbyopia, cataract, nyctalopia, protanopia, deuteranopia, tritanopia, achromatopsia. provides an infrastructure encourage researchers evaluate...
We present a visual analytics approach to support the workload management process for z/OS mainframes at IBM. This typically requires analysis of records consisting 100 150 performance-related metrics, sampled over time. aim replacing previous spreadsheet-based workflow with an easier, faster, and more scalable one regarding measurement periods collected performance metrics. To achieve this goal, we collaborate developer embedded IBM in formative process. Based on that experience, discuss...
We propose a three-step concept and visual design for supporting the exploration of high-dimensional data in scatterplots through eye-tracking. First, we extract subsets underlying using existing classifications, automated clustering algorithms, or For latter, map gaze to dimensions scatterplot. Clusters points that have been focus viewers' are marked as clusters interest (eye-mind hypothesis). In second step, our extracts various properties from statistics scagnostics clusters. The third...
Word clouds are a good way to represent textual data with meta information. However, when it comes analyzing multiple sources, they difficult use. This is due their poor comparability. The proposed RadCloud merges standard into single one, while retaining the information about origin of words. Using separate as well this new visualization technique, tool for analysis created: MuWoC. resulting software then evaluated by using sources ranging from encyclopedia articles, over literature custom...
We investigate the problem of analyzing word frequencies in multiple text sources with aim to give an overview word-based similarities several texts as a starting point for further analysis. To reach this goal, we designed visual analytics approach composed typical stages and processes, combining algorithmic analysis, visualization techniques, human users their perceptual abilities, well interaction methods both data analysis component. By our first generate multivariate dataset where words...
We instrument Group Diagrams (GDs) to reduce clutter in sets of eye-tracking scanpaths. consist trajectory subsets that cover, or represent, the whole set trajectories with respect some distance measure and an adjustable threshold. The original GDs allow for application various measures. implement GD framework evaluate it on scanpaths were collected by a former user study public transit maps. find Fréchet is most appropriate get meaningful results, yet flexible enough cover outliers. discuss...
Abstract We instrument Group Diagrams (GDs) to reduce clutter in sets of eye-tracking scanpaths. consist trajectory subsets that cover, or represent, the whole set trajectories with respect some distance measure and an adjustable threshold. The original GDs allow for application various measures. implement GD framework evaluate it on scanpaths were collected by a former user study public transit maps. find Fréchet is most appropriate get meaningful results, yet flexible enough cover...
画面や広告の閲覧者の視線軌跡を分析することで,人々が何に関心を持っているのかを特定し,商品の推薦や広告デザインの改良などが可能となる.しかし,視線の分析・可視化の手法には複数の課題が残されている.本論文では,これらの課題のうち(1)複数の関心領域(AOI)を推移するパターンの可視化,(2)複数の視線軌跡の差分の可視化,の新たな手法として複数人の視線軌跡に含まれるパターンを可視化する手法を提案する.本手法ではまずWebページや画像などに階層型AOIを設定し,その結果を用いて視線軌跡を記号列化する.生成した記号列にはN-gramを適用し,AOI間の推移パターンを抽出する.最後にパターン抽出結果の一覧と特徴的な要素の形状を可視化する.実行結果として,Wikipediaのページとポスターの2種類の刺激を観察した際の視線軌跡を可視化した結果を紹介する.これらの結果から,テキスト・図表の間の視線の推移の傾向や,被験者間の差分を効率的に発見できることを示す.
We introduce relaxed dot plots as an improvement of nonlinear for unit visualization. Our produce more faithful data representations and reduce moiré effects. Their contour is based on a customized kernel frequency estimation to match the shape distribution underlying values. Previous layouts column-centric scaling diameters visualization high-dynamic-range with high peaks. provide mathematical approach convert that our smooth envelope shape. This formalism allows us use linear, root,...