EuroVisPosters2022

Permanent URI for this collection

https://diglib.eg.org/handle/10.2312/2633198

Browse

Now showing 1 - 20 of 35

Accurate Molecular Atom Selection in VR
(The Eurographics Association, 2022) Molina, Elena; Vázquez, Pere-Pau; Krone, Michael; Lenti, Simone; Schmidt, Johanna
Accurate selection in cluttered scenes is complex because a high amount of precision is required. In Virtual Reality Environments, it is even worse, because it is more difficult for us to point a small object with our arms in the air. Not only our arms move slightly, but the button/trigger press reduces our weak stability. In this paper, we present two alternatives to the classical ray pointing intended to facilitate the selection of atoms in molecular environments. We have implemented and analyzed such techniques through an informal user study and found that they were highly appreciated by the users. This selection method could be interesting in other crowded environments beyond molecular visualization.
Interactive Visualization of Machine Learning Model Results Predicting Infection Risk
(The Eurographics Association, 2022) Schäfer, Steffen; Baumgartl, Tom; Wulff, Antje; Kuijper, Arjan; Marschollek, Michael; Scheithauer, Simone; von Landesberger, Tatiana; Krone, Michael; Lenti, Simone; Schmidt, Johanna
We present a novel visual-interactive interface to show results of a machine learning algorithm, which predicts the infection probability for patients in hospitals. The model result data is complex and needs to be presented in a clear and intuitive way to microbiology and infection control experts in hospitals. Our visual-interactive interface offers linked views which allow for detailed analysis of the model results. Feedback from microbiology and infection control experts showed that they were able to extract new insights regarding outbreaks and transmission pathways.
Validating Perception of Hyperspectral Textures in Virtual Reality Systems
(The Eurographics Association, 2022) Díaz-Barrancas, Francisco; Cwierz, Halina; Gil-Rodríguez, Raquel; Pardo, Pedro J.; Krone, Michael; Lenti, Simone; Schmidt, Johanna
Virtual reality (VR) environments are increasingly offering higher quality content. They use different computing techniques to improve the final user experience. In this work, we create different light sources and introduce hyperspectral textures for the object reflectance to boost the VR environment's quality. In addition, we perform a quantitative study to demonstrate that hyperspectral textures improve the final quality of the content in virtual reality systems.
Explorative Visual Analysis of Spatio-temporal Regions to Detect Hemodynamic Biomarker Candidates
(The Eurographics Association, 2022) Derstroff, Adrian; Leistikow, Simon; Nahardani, Ali; Ebrahimi, Mahyasadat; Hoerr, Verena; Linsen, Lars; Krone, Michael; Lenti, Simone; Schmidt, Johanna
Biomarkers are measurable biological properties that allow for distinguishing subjects of different cohorts such as healthy vs. diseased. In the context of diagnosing diseases of the cardiovascular system, researchers aim - among others - at detecting biomarkers in the form of spatio-temporal regions of blood flow obtained by medical imaging or of derived hemodynamical parameters. As the search space for such biomarkers in time-varying volumetric multi-field data is extremely large, we present an interactive visual exploration system to support the analysis of the potential of spatio-temporal regions to discriminate cohorts.
GDot-i: Interactive System for Dot Paintings of Graphs
(The Eurographics Association, 2022) Eades, Peter; Hong, Seok-Hee; McGrane, Martin; Meidiana, Amyra; Krone, Michael; Lenti, Simone; Schmidt, Johanna
This poster presents GDot-i, an interactive system visualizing graphs and networks as dot paintings, inspired by the dot painting style of Central Australia. We describe the implementation of GDot-i, a web-based interactive system, including the user interface and typical use cases.
Chord2DS: An Extension to Chord Diagram to Show Data Elements from Two Heterogeneous Data Sources
(The Eurographics Association, 2022) Humayoun, Shah Rukh; Brahmadevara, Likhitha; Krone, Michael; Lenti, Simone; Schmidt, Johanna
The standard Chord diagram, a radial layout, shows data elements in a circular fashion from one data source. In this paper, we propose an extension to the standard Chord diagram to show data elements from two heterogeneous data sources into one single diagram. The main Chord diagram is used for showing data elements and the relations between them from one data source, while we use an outer layer to show data elements from the second data source. The relationships between data elements from both data sources are shown through visual cues. The proposed solution uses space efficiently compared to using multiple diagrams in the scenarios of two heterogeneous data sources.
Visualization Challenges of Variant Interpretation in Multiscale NGS Data
(The Eurographics Association, 2022) Ståhlbom, Emilia; Molin, Jesper; Lundström, Claes; Ynnerman, Anders; Krone, Michael; Lenti, Simone; Schmidt, Johanna
There is currently a movement in health care towards precision medicine, where genomics often is the central diagnostic component for tailoring the treatment to the individual patient. We here present results from a domain characterization effort to pinpoint problems and possibilities for visualization of genomics data in the clinical workflow, with analysis of copy number variants as an example task. Five distinct characteristics have been identified. Clinical genomics data is inherently multiscale, riddled with artifacts and uncertainty, and many findings have unknown significance, so it is a challenging visual analytics domain. Moreover, as in other clinical domains, high efficiency is key. This characterization will form the basis for follow-on visualization prototyping.
A Case Study on Implementing Screen Reader Accessibility in Dynamic Visualizations
(The Eurographics Association, 2022) Costa, Rita; Malveiro, Beatriz; Palmeiro, João; Bizarro, Pedro; Krone, Michael; Lenti, Simone; Schmidt, Johanna
Millions of people worldwide work in jobs where assessing dynamic data presented visually to them is a key part of their tasks. Since the data is only represented in a visual format, these occupations are out of reach for visually impaired people, making them unable to review hundreds of information-heavy cases per day and determine outcomes for each one in just a couple of minutes. In this work, we aim to shrink that gap by detailing the implementation of screen reader accessibility features to real-world visualizations used by fraud detection analysts. We propose a set of features that should be validated with users and, if proved to be useful, transformed into guidelines for creating these types of accessible charts.
Automatic Segmentation of Tooth Images: Optimization of Multi-parameter Image Processing Workflow
(The Eurographics Association, 2022) Bressan Fogalli, Giovani; Line, Sérgio Roberto Peres; Baum, Daniel; Krone, Michael; Lenti, Simone; Schmidt, Johanna
The development of specific algorithms in image processing are usually related to dataset characteristics. Those characteristics will influence the number of instructions required to solve a problem. Normally, the more complex a set of instructions is, the more parameters need to be set. Dealing with such degrees of freedom, sometimes leading to subjective decision making, is time-consuming and frequently leads to errors or sub-optimal results of the developed model. Here, we deal with a model for segmentation of masks of tooth images containing a pattern of bands called Hunter-Schreger Bands (HSB). They appear on tooth surface when lit from the side. This segmentation process is only one step of a pipeline whose overall goal is human biometric identification to be used, e.g., in forensics. The segmentation algorithm, which exploits the anisotropy of the image, uses several parameters and choosing the optimal combination of them is challenging. The aim of this work was to utilize visual data analysis tools to optimize the chosen parameters and to understand their influence on the performance of the algorithm. Our results reveal that a slightly better combination of parameter values can be found starting from the experimentally determined initial parameters. This approach can be repeatedly performed to achieve even better parameterizations. To more deeply understand the influence of the parameters on the final result, more sophisticated visual interaction tools will be explored in future work.
Visualizing the Evolution of Multi-agent Game-playing Behaviors
(The Eurographics Association, 2022) Agarwal, Shivam; Latif, Shahid; Rothweiler, Aristide; Beck, Fabian; Krone, Michael; Lenti, Simone; Schmidt, Johanna
Analyzing the training evolution of AI agents in a multi-agent environment helps to understand changes in learned behaviors, as well as the sequence in which they are learned. We train an existing Pommerman team from scratch and, at regular intervals, let it battle against another top-performing team. We define thirteen game-specific behaviors and compute their occurrences in 600 matches. To investigate the evolution of these behaviors, we propose a visualization approach and showcase its usefulness in an application example.
Context Specific Visualizations on Smartwatches
(The Eurographics Association, 2022) Islam, Alaul; Blascheck, Tanja; Isenberg, Petra; Krone, Michael; Lenti, Simone; Schmidt, Johanna
We present an analysis of the results of a full-day context-specific ideation exercise for smartwatch visualizations. Participants of the exercise created 34 sketches during a sightseeing activity. Our analysis of these sketches showed where visualizations could be applied and shown, what information needs they could target, and how data could be represented in the sightseeing context.
Parameter Sensitivity and Uncertainty Visualization in DTI
(The Eurographics Association, 2022) Siddiqui, Faizan; Höllt, Thomas; Vilanova, Anna; Krone, Michael; Lenti, Simone; Schmidt, Johanna
Diffusion Tensor Imaging is a powerful technique that provides a unique insight into the complex structure of the brain's white matter. However, several sources of uncertainty limit its widespread use. Data and modeling errors arise due to acquisition noise and modeling transformations. Moreover, the sensitivities of the user-defined parameters and region definitions are not usually evaluated, a small change in these parameters can add large variations in the results. Without showing these uncertainties any visualization of DTI data can potentially be misleading. In our work, we develop a visual analytic tool that provides insight into the accumulated uncertainty in the visualization pipeline. The primary goal of this project is to develop an efficient visualization strategy that will assist the end-user in making critical decisions and make fiber tracking analysis less cumbersome and more reliable, a crucial step towards adoption in the neurosurgical workflow.
Digital Twins of Smart Farms
(The Eurographics Association, 2022) Zhao, Yuhang; Jiang, Zheyu; Pang, Shanchen; Lv, Zhihan; Krone, Michael; Lenti, Simone; Schmidt, Johanna
In recent years, the development of Digital Twins has made rapid progress, and Digital Twins has gradually begun to combine various fields and applied to the current digitalization of the physical world. Digital Twins can play an important role in agriculture. Digital Twins can fully improve the yield and income of crop products and solve the problems of food security. In this paper, the development prospect of Digital Twins in agriculture is discussed.
A Mental Workload Estimation for Visualization Evaluation Using EEG Data and NASA-TLX
(The Eurographics Association, 2022) Yim, Soobin; Yoon, Chanyoung; Yoo, Sangbong; Jang, Yun; Krone, Michael; Lenti, Simone; Schmidt, Johanna
Mental workload is a cognitive effort felt by users while solving tasks, and good visualizations tend to induce a low mental workload. For better visualizations, various visualization techniques have been evaluated through quantitative methods that compare the response accuracy and performance time for completing visualization tasks. However, accuracy and time do not always represent the mental workload of a subject. Since quantitative approaches do not fully mirror mental workload, questionnaires and biosignals have been employed to measure mental workload in visualization assessments. The electroencephalogram (EEG) as biosignal is one of the indicators frequently utilized to measure mental workload. Since everyone judges and senses differently, EEG signals and mental workload differ from person to person. In this paper, we propose a mental workload personalized estimation model with EEG data specialized for each individual to evaluate visualizations. We use scatter plot, bar, line, and map visualizations and collect NASA-TLX scores as mental workload and EEG data. NASA-TLX and EEG data as training data are used for the mental workload estimation model.
Exploration and Analysis of Image-base Simulation Ensembles
(The Eurographics Association, 2022) Dahshan, Mai; Turton, Terece L.; Polys, Nicholas; Krone, Michael; Lenti, Simone; Schmidt, Johanna
Scientists run simulation ensembles to study the behavior of a phenomenon using varying initial conditions or input parameters. However, the I/O bottlenecks hinder performing large-scale multidimensional simulations. In situ visualization approaches address the variability of I/O performance by processing output data during simulation time and saving predetermined visualizations in image databases. This poster proposes a visual analytics approach to exploring and analyzing image-based simulation ensembles, taking advantage of semantic interaction, feature extraction, and deep learning techniques. Our approach uses deep learning and local feature techniques to learn image features and pass them along with the input parameters to the visualization pipeline for in-depth exploration and analysis of parameter and ensemble spaces simultaneously.
A Design Space for Explainable Ranking and Ranking Models
(The Eurographics Association, 2022) Hazwan, Ibrahim Al; Schmid, Jenny; Sachdeva, Madhav; Bernard, Jürgen; Krone, Michael; Lenti, Simone; Schmidt, Johanna
Item ranking systems support users in multi-criteria decision-making tasks. Users need to trust rankings and ranking algorithms to reflect user preferences nicely while avoiding systematic errors and biases. However, today only few approaches help end users, model developers, and analysts to explain rankings. We report on the study of explanation approaches from the perspectives of recommender systems, explainable AI, and visualization research and propose the first cross-domain design space for explainers of item rankings. In addition, we leverage the descriptive power of the design space to characterize a) existing explainers and b) three main user groups involved in ranking explanation tasks. The generative power of the design space is a means for future designers and developers to create more target-oriented solutions in this only weakly exploited space.
Interactive Attribution-based Explanations for Image Segmentation
(The Eurographics Association, 2022) Humer, Christina; Elharty, Mohamed; Hinterreiter, Andreas; Streit, Marc; Krone, Michael; Lenti, Simone; Schmidt, Johanna
Explanations of deep neural networks (DNNs) give users a better understanding of the inner workings and generalizability of a network. While the majority of research focuses on explanations for classification networks, in this work we focus on explainability for image segmentation networks. As a first contribution, we introduce a lightweight framework that allows generalizing certain attribution-based explanations, originally developed for classification networks, to also work for segmentation networks. The second contribution is a web-based tool that utilizes this framework and allows users to interactively explore segmentation networks. We demonstrate the approach using a self-trained mushroom segmentation network.
Visualizing Prediction Provenance in Regression Random Forests
(The Eurographics Association, 2022) Médoc, Nicolas; Ciorna, Vasile; Petry, Frank; Ghoniem, Mohammad; Krone, Michael; Lenti, Simone; Schmidt, Johanna
Random forest models are widely used in many application domains due to their performance and the fact that their constituent decision trees carry clear decision rules. Yet, the provenance of the predictions made by an entire forest is complex to grasp, which motivates application domain experts to adopt black-box testing strategies. We propose in this paper a coordinated multiple view system allowing to shed more light on prediction provenance, uncertainty and error in terms of bias and variance at the global model scale or at the local scale of decision paths and individual instances.
Scientific Convergence and Divergence in Visualization and Visual Analytics
(The Eurographics Association, 2022) He, Jiangen; Krone, Michael; Lenti, Simone; Schmidt, Johanna
We present preliminary results of a visualization tool designed to visualize scientific evolution by using scientific publication data, especially convergence-divergence processes. It aims to increase the efficiency and accuracy of our understanding of scientific knowledge in a certain field with limited domain knowledge. We visualized 2,435 papers published in IEEE VIS and EuroVis to demonstrate the tool and provide a big picture of the scientific evolution in the visualization community.
Visualizing Similarities between American Rap-Artists
(The Eurographics Association, 2022) Meinecke, Christofer; Schebera, Jeremias; Eschrich, Jakob; Wiegreffe, Daniel; Krone, Michael; Lenti, Simone; Schmidt, Johanna
Rap music is one of the biggest music genres in the world today. Since the early days of rap music, references not only to pop culture but also to other rap artists have been an integral part of the lyrics' artistry. In addition, rap musicians reference each other by adopting fragments of lyrics, for example, to give credit. This kind of text reuse can be used to create connections between individual artists. Due to the large amount of lyrics, only automated detection methods can efficiently detect similarities between the songs and the artists. Here, we present a visualization system for analyzing rap music lyrics. We also trained a network tailored specifically for rap lyrics to compute similarities in lyrics. Here a video of the prototype can be seen.

Browse

Browsing EuroVisPosters2022 by Issue Date

Results Per Page

Sort Options