We investigate the perceived visual complexity (VC) in data visualizations using objective image-based metrics. We collected VC scores through a large-scale crowdsourcing experiment involving 349 participants and 1,800 visualization images. We then examined how these scores align with 12 image-based metrics spanning pixel-based and statistic-information-theoretic (clutter), color, shape, and our two new object-based metrics (meaningful-color-count (MeC) and text-to-ink ratio (TiR)). Our results show that both low-level edges and high-level elements affect perceived VC in visualization images; the number of corners and distinct colors are robust metrics across visualizations. Second, feature congestion, a statistical information-theoretic metric capturing color and texture patterns, is the strongest predictor of perceived complexity in visualizations rich in the same continuous color/texture stimuli; edge density effectively explains VC in node-link diagrams. Additionally, we observe a bell-curve effect for texts: increasing TiR initially reduces complexity, reaching an optimal point, beyond which further text increases VC. Our quantification model is also interpretable-enabling metric-based explanations-grounded in the VisComplexity2K dataset, bridging computational metrics with human perceptual responses. The preregistration is available at osf.io/5xe8a. osf.io/bdet6 has the dataset and analysis code.

We present a data-domain sampling regime for quantifying CNNs' graphic perception behaviors. This regime lets us evaluate CNNs' ratio estimation ability in bar charts from three perspectives: sensitivity to training-test distribution discrepancies, stability to limited samples, and relative expertise to human observers. After analyzing 16 million trials from 800 CNN models and 6,825 trials from 113 human participants, we arrived at a simple and actionable conclusion: CNNs can outperform humans and their biases simply depend on the training-test distance. We show evidence of this simple, elegant behavior of the machines when they interpret visualization images. osf.io/gfqc3 provides registration, the code for our sampling regime, and experimental results.

We present experimental results to explore a form of bivariate glyphs for representing large-magnitude-range vectors. The glyphs meet two conditions: (1) two visual dimensions are separable; and (2) one of the two visual dimensions uses a categorical representation (e.g., a categorical colormap). We evaluate how much these two conditions determine the bivariate glyphs' effectiveness. The first experiment asks participants to perform three local tasks requiring reading no more than two glyphs. The second experiment scales up the search space in global tasks when participants must look at the entire scene of hundreds of vector glyphs to get an answer. Our results support that the first condition is necessary for local tasks when a few items are compared. But it is not enough for understanding a large amount of data. The second condition is necessary for perceiving global structures of examining very complex datasets. Participants' comments reveal that the categorical features in the bivariate glyphs trigger emergent optimal viewers' behaviors. This work contributes to perceptually accurate glyph representations for revealing patterns from large scientific results. We release source code, quantum physics data, training documents, participants' answers, and statistical analyses for reproducible science at https://osf.io/4xcf5/?view_only=94123139df9c4ac984a1e0df811cd580.

Node-link diagrams are the most popular form for graph visualization. Yet, salient information of a node-link diagram cannot be fully depicted by solely presenting the visualization. We propose to augment node-link diagrams by creating textual descriptions for interested information. We conduct an expert review and a user interview to identify six requirements of generated interpretations, including three requirements for connection extraction and three requirements for visual expression. Our solution, GraphDescriptor, generates textual descriptions with two stages: feature extraction and description generation. The first one identifies and extracts features of node-link diagrams, like node connections, visual designs, and types of graph layouts. The second stage creates a group of hierarchical sentences based on a pre-defined schema. To the best of our knowledge, our approach is the first attempt to generate textual descriptions automatically. Three use cases and the in-lab user study confirm the superiority of our approach.

Video summaries of scientific publications have gained more and more popularity over the last years, requiring many researchers to familiarize themselves with the tools and techniques of video production which can be an overwhelming task. This paper introduces a video structuring framework embedded into the authoring tool Pub2Vid. The tool supports users with the creation of their video outline and script, providing real video examples and recommendations based on the analysis of 40 publication summarization videos which were rated in a user study with 68 participants. Following a four-tier evaluation methodology, the application’s usability is assessed and improved via amateur and expert interviews, two rounds of usability tests and two case studies. It is shown that the tool and its recommendations are particularly useful for beginners due to the simple design and intuitive components as well as suggestions based on real video examples.

This February 2022 issue of theIEEE Transactions on Visualization and Computer Graphics (TVCG)contains the proceedings of IEEE VIS 2021, held online on October 24-29, 2021, with General Chairs from Tulane University and Universidade de Sao Paulo. With IEEE VIS 2021, the conference series is in its 32nd year.

We present document domain randomization (DDR), the first successful transfer of convolutional neural networks (CNNs) trained only on graphically rendered pseudo-paper pages to real-world document segmentation. DDR renders pseudo-document pages by modeling randomized textual and non-textual contents of interest, with user-defined layout and font styles to support joint learning of fine-grained classes. We demonstrate competitive results using our DDR approach to extract nine document classes from the benchmark CS-150 and papers published in two domains, namely annual meetings of Association for Computational Linguistics (ACL) and IEEE Visualization (VIS). We compare DDR to conditions of style mismatch, fewer or more noisy samples that are more easily obtained in the real world. We show that high-fidelity semantic information is not necessary to label semantic classes but style mismatch between train and test can lower model accuracy. Using smaller training samples had a slightly detrimental effect. Finally, network models still achieved high test accuracy when correct labels are diluted towards confusing labels; this behavior hold across several classes.

Visualization, as a vibrant field for researchers, practitioners, and higher educational institutions, is growing and evolving very rapidly. Tremendous progress has been made since 1987, the year often cited as the beginning of data visualization as a distinct field. As such, the number of visualization resources and the demand for those resources are increasing at a very fast pace. We present a collection of open visualization resources for all those with an interest in interactive data visualization and visual analytics. Because the number of resources is so large, we focus on collections of resources, of which there are already very many ranging from literature collections to collections of practitioner resources. We develop a novel classification of visualization resource collections based on the resource type, e.g. literature-based, web-based, etc. The result is a helpful overview and details-on-demand of many useful resources. The collection offers a valuable jump-start for those seeking out data visualization resources from all backgrounds spanning from beginners such as students to teachers, practitioners, and researchers wishing to create their own advanced or novel visual designs.

We present the VIS30K dataset, a collection of 29,689 images that represents 30 years of figures and tables from each track of the IEEE Visualization conference series (Vis, SciVis, InfoVis, VAST). VIS30K's comprehensive coverage of the scientific literature in visualization not only reflects the progress of the field but also enables researchers to study the evolution of the state-of-the-art and to find relevant work based on graphical content. We describe the dataset and our semi-automatic collection process, which couples convolutional neural networks (CNN) with curation. Extracting figures and tables semi-automatically allows us to verify that no images are overlooked or extracted erroneously. To improve quality further, we engaged in a peer-search process for high-quality figures from early IEEE Visualization papers. With the resulting data, we also contribute VISImageNavigator (VIN, visimagenavigator.github.io), a web-based tool that facilitates searching and exploring VIS30K by author names, paper keywords, title and abstract, and years.

We design and evaluate a novel layout fine-tuning technique for node-link diagrams that facilitates exemplar-based adjustment of a group of substructures in batching mode. The key idea is to transfer user modifications on a local substructure to other substructures in the entire graph that are topologically similar to the exemplar. We first precompute a canonical representation for each substructure with node embedding techniques and then use it for on-the-fly substructure retrieval. We design and develop a light-weight interactive system to enable intuitive adjustment, modification transfer, and visual graph exploration. We also report some results of quantitative comparisons, three case studies, and a within-participant user study.

We report empirical study results on the color encoding of ensemble scalar and orientation to visualize diffusion magnetic resonance imaging (DMRI) tubes. The experiment tested six scalar colormaps for average fractional anisotropy (FA) tasks (grayscale, blackbody, diverging, isoluminant-rainbow, extended-blackbody, and coolwarm) and four three-dimensional (3D) spherical colormaps for tract tracing tasks (uniform gray, absolute, eigenmaps, and Boy's surface embedding). We found that extended-blackbody, coolwarm, and blackbody remain the best three approaches for identifying ensemble average in 3D. Isoluminant-rainbow colormap led to the same ensemble mean accuracy as other colormaps. However, more than 50 percent of the answers consistently had higher estimates of the ensemble average, independent of the mean values. The number of hues, not luminance, influences ensemble estimates of mean values. For ensemble orientation-tracing tasks, we found that both Boy's surface embedding (greatest spatial resolution and contrast) and absolute colormaps (lowest spatial resolution and contrast) led to more accurate answers than the eigenmaps scheme (medium resolution and contrast), acting as the uncanny-valley phenomenon of visualization design in terms of accuracy. Absolute colormap broadly used in brain science is a good default spherical colormap. We could conclude from our study that human visual processing of a chunk of colors differs from that of single colors.

This paper develops a new Neyman–Pearson detection approach, to be called band-specified virtual dimensionality (BSVD), to estimating the number of bands required by band selection (BS),$n_{\mathrm {BS}}$, as well as finding desired bands at the same time. Its idea is derived from target-specified virtual dimensionality (TSVD) where targets under hypotheses as signal sources in TSVD are replaced with bands as signal sources and the test statistics derived for a Neyman–Pearson detector (NPD) is signal-to-noise ratio (SNR) that is used to derive orthogonal subspace projection (OSP) approach for hyperspectral image classification and dimensionality reduction. Accordingly, the resulting virtual dimensionality is referred to as OSP-based BSVD. Several benefits resulting from BSVD cannot be offered by the traditional BS methods. One is its direct approach to dealing with$n_{\mathrm {BS}}$. Another is no-search strategy needed for finding optimal bands. Instead, it uses NPD to determine and rank desired bands for band prioritization. Most importantly, it determines$n_{\mathrm {BS}}$and finds desired bands simultaneously and progressively.

BACKGROUND: A major challenge of high throughput transcriptome studies is presenting the data to researchers in an interpretable format. In many cases, the outputs of such studies are gene lists which are then examined for enriched biological concepts. One approach to help the researcher interpret large gene datasets is to associate genes and informative terms (iTerm) that are obtained from the biomedical literature using the eGIFT text-mining system. However, examining large lists of iTerm and gene pairs is a daunting task. RESULTS: We have developed WebGIVI, an interactive web-based visualization tool ( http://raven.anr.udel.edu/webgivi/ ) to explore gene:iTerm pairs. WebGIVI was built via Cytoscape and Data Driven Document JavaScript libraries and can be used to relate genes to iTerms and then visualize gene and iTerm pairs. WebGIVI can accept a gene list that is used to retrieve the gene symbols and corresponding iTerm list. This list can be submitted to visualize the gene iTerm pairs using two distinct methods: a Concept Map or a Cytoscape Network Map. In addition, WebGIVI also supports uploading and visualization of any two-column tab separated data. CONCLUSIONS: WebGIVI provides an interactive and integrated network graph of gene and iTerms that allows filtering, sorting, and grouping, which can aid biologists in developing hypothesis based on the input gene lists. In addition, WebGIVI can visualize hundreds of nodes and generate a high-resolution image that is important for most of research publications. The source code can be freely downloaded at https://github.com/sunliang3361/WebGIVI . The WebGIVI tutorial is available at http://raven.anr.udel.edu/webgivi/tutorial.php .

BACKGROUND: Global scale brain research collaborations such as the ENIGMA (Enhancing Neuro Imaging Genetics through Meta Analysis) consortium are beginning to collect data in large quantity and to conduct meta-analyses using uniformed protocols. It becomes strategically important that the results can be communicated among brain scientists effectively. Traditional graphs and charts failed to convey the complex shapes of brain structures which are essential to the understanding of the result statistics from the analyses. These problems could be addressed using interactive visualization strategies that can link those statistics with brain structures in order to provide a better interface to understand brain research results. RESULTS: We present ENIGMA-Viewer, an interactive web-based visualization tool for brain scientists to compare statistics such as effect sizes from meta-analysis results on standardized ROIs (regions-of-interest) across multiple studies. The tool incorporates visualization design principles such as focus+context and visual data fusion to enable users to better understand the statistics on brain structures. To demonstrate the usability of the tool, three examples using recent research data are discussed via case studies. CONCLUSIONS: ENIGMA-Viewer supports presentations and communications of brain research results through effective visualization designs. By linking visualizations of both statistics and structures, users can gain more insights into the presented data that are otherwise difficult to obtain. ENIGMA-Viewer is an open-source tool, the source code and sample data are publicly accessible through the NITRC website ( http://www.nitrc.org/projects/enigmaviewer_20 ). The tool can also be directly accessed online ( http://enigma-viewer.org ).

We present the results of a comprehensive multi-pass analysis of visualization paper keywords supplied by authors for their papers published in the IEEE Visualization conference series (now called IEEE VIS) between 1990-2015. From this analysis we derived a set of visualization topics that we discuss in the context of the current taxonomy that is used to categorize papers and assign reviewers in the IEEE VIS reviewing process. We point out missing and overemphasized topics in the current taxonomy and start a discussion on the importance of establishing common visualization terminology. Our analysis of research topics in visualization can, thus, serve as a starting point to (a) help create a common vocabulary to improve communication among different visualization sub-groups, (b) facilitate the process of understanding differences and commonalities of the various research sub-fields in visualization, (c) provide an understanding of emerging new research trends, (d) facilitate the crucial step of finding the right reviewers for research submissions, and (e) it can eventually lead to a comprehensive taxonomy of visualization research. One additional tangible outcome of our work is an online query tool (http://keyvis.org/) that allows visualization researchers to easily browse the 3952 keywords used for IEEE VIS papers since 1990 to find related work or make informed keyword choices.

We have created and made available to all a dataset with information about every paper that has appeared at the IEEE Visualization (VIS) set of conferences: InfoVis, SciVis, VAST, and Vis. The information about each paper includes its title, abstract, authors, and citations to other papers in the conference series, among many other attributes. This article describes the motivation for creating the dataset, as well as our process of coalescing and cleaning the data, and a set of three visualizations we created to facilitate exploration of the data. This data is meant to be useful to the broad data visualization community to help understand the evolution of the field and as an example document collection for text data visualization research.

We designed and evaluated SplitVectors, a new vector field display approach to help scientists perform new discrimination tasks on large-magnitude-range scientific data shown in three-dimensional (3D) visualization environments. SplitVectors uses scientific notation to display vector magnitude, thus improving legibility. We present an empirical study comparing the SplitVectors approach with three other approaches - direct linear representation, logarithmic, and text display commonly used in scientific visualizations. Twenty participants performed three domain analysis tasks: reading numerical values (a discrimination task), finding the ratio between values (a discrimination task), and finding the larger of two vectors (a pattern detection task). Participants used both mono and stereo conditions. Our results suggest the following: (1) SplitVectors improve accuracy by about 10 times compared to linear mapping and by four times to logarithmic in discrimination tasks; (2) SplitVectors have no significant differences from the textual display approach, but reduce cluttering in the scene; (3) SplitVectors and textual display are less sensitive to data scale than linear and logarithmic approaches; (4) using logarithmic can be problematic as participants' confidence was as high as directly reading from the textual display, but their accuracy was poor; and (5) Stereoscopy improved performance, especially in more challenging discrimination tasks.

While the primary goal of visual analytics research is to improve the quality of insights and findings, a substantial amount of research in provenance has focused on the history of changes and advances throughout the analysis process. The term, provenance, has been used in a variety of ways to describe different types of records and histories related to visualization. The existing body of provenance research has grown to a point where the consolidation of design knowledge requires cross-referencing a variety of projects and studies spanning multiple domain areas. We present an organizational framework of the different types of provenance information and purposes for why they are desired in the field of visual analytics. Our organization is intended to serve as a framework to help researchers specify types of provenance and coordinate design knowledge across projects. We also discuss the relationships between these factors and the methods used to capture provenance information. In addition, our organization can be used to guide the selection of evaluation methodology and the comparison of study outcomes in provenance research.

We designed and evaluated SplitVector, a new vector field display approach to help scientists perform new discrimination tasks on scientific data shown in virtual environments (VEs). Our empirical study compared the SplitVector approach with three other approaches of direct linear representation, log, and text display common in information-rich VEs or IRVEs. Our results suggest the following: (1) SplitVectors improve the accuracy by about 10 times compared to the linear mapping and by 4 times to log in discrimination tasks; (2) SplitVectors lead to no significant differences from the IRVE text display approach, yet reduce the clutter; and (3) SplitVector improved task performance in both mono and stereoscopy conditions.

BACKGROUND: High-throughput methods are generating biological data on a vast scale. In many instances, genomic, transcriptomic, and proteomic data must be interpreted in the context of signaling and metabolic pathways to yield testable hypotheses. Since humans can interpret visual information rapidly, a means for interactive visual exploration that lets biologists interpret such data in a comprehensive and exploratory manner would be invaluable. However, humans have limited memory capacity. Current visualization tools have limited viewing and manipulation capabilities to address complex data analysis problems, and visual exploratory tools are needed to reduce the high mental workload imposed on biologists. RESULTS: We present PathRings, a new interactive web-based, scalable biological pathway visualization tool for biologists to explore and interpret biological pathways. PathRings integrates metabolic and signaling pathways from Reactome in a single compound graph visualization, and uses color to highlight genes and pathways affected by input data. Pathways are available for multiple species and analysis of user-defined species or input is also possible. PathRings permits an overview of the impact of gene expression data on all pathways to facilitate visual pattern finding. Detailed pathways information can be opened in new visualizations while maintaining the overview, that form a visual exploration provenance. A dynamic multi-view bubbles interface is designed to support biologists' analytical tasks by letting users construct incremental views that further reflect biologists' analytical process. This approach decomposes complex tasks into simpler ones and automates multi-view management. CONCLUSIONS: PathRings has been designed to accommodate interactive visual analysis of experimental data in the context of pathways defined by Reactome. Our new approach to interface design can effectively support comparative tasks over substantially larger collection than existing tools. The dynamic interaction among multi-view dataset visualization improves the data exploration. PathRings is available free at http://raven.anr.udel.edu/~sunliang/PathRings and the source code is hosted on Github: https://github.com/ivcl/PathRings .

We present an assessment of the state and historic development of evaluation practices as reported in papers published at the IEEE Visualization conference. Our goal is to reflect on a meta-level about evaluation in our community through a systematic understanding of the characteristics and goals of presented evaluations. For this purpose we conducted a systematic review of ten years of evaluations in the published papers using and extending a coding scheme previously established by Lam et al. [2012]. The results of our review include an overview of the most common evaluation goals in the community, how they evolved over time, and how they contrast or align to those of the IEEE Information Visualization conference. In particular, we found that evaluations specific to assessing resulting images and algorithm performance are the most prevalent (with consistently 80-90% of all papers since 1997). However, especially over the last six years there is a steady increase in evaluation methods that include participants, either by evaluating their performances and subjective feedback or by evaluating their work practices and their improved analysis and reasoning capabilities using visual tools. Up to 2010, this trend in the IEEE Visualization conference was much more pronounced than in the IEEE Information Visualization conference which only showed an increasing percentage of evaluation through user performance and experience testing. Since 2011, however, also papers in IEEE Information Visualization show such an increase of evaluations of work practices and analysis as well as reasoning using visual tools. Further, we found that generally the studies reporting requirements analyses and domain-specific work practices are too informally reported which hinders cross-comparison and lowers external validity.

We present results from a user study of task performance on streamtube visualizations, such as those used in three-dimensional (3D) vector and tensor field visualizations. This study used a tensor field sampled from a full-brain diffusion tensor magnetic resonance imaging (DTI) dataset. The independent variables include illumination model (global illumination and OpenGL-style local illumination), texture (with and without), motion (with and without), and task. The three spatial analysis tasks are: (1) a depth-judgment task: determining which of two marked tubes is closer to the user's viewpoint, (2) a visual-tracing task: marking the endpoint of a tube, and (3) a contact-judgment task: analyzing tube-sphere penetration. Our results indicate that global illumination did not improve task completion time for the tasks we measured. Global illumination reduced the errors in participants' answers over local OpenGLstyle rendering for the visual-tracing task only when motion was present. Motion contributed to spatial understanding for all tasks, but at the cost of longer task completion time. A high-frequency texture pattern led to longer task completion times and higher error rates. These results can help in the design of lighting model, such as flow or diffusion-tensor field visualizations and identify situations when the lighting is more efficient and accurate.

We report the impact of display characteristics (stereo and size) on task performance in diffusion magnetic resonance imaging (DMRI) in a user study with 12 participants. The hypotheses were that (1) adding stereo and increasing display size would improve task accuracy and reduce completion time, and (2) the greater the complexity of a spatial task, the greater the benefits of an improved display. Thus we expected to see greater performance gains when detailed visual reasoning was required. Participants used dense streamtube visualizations to perform five representative tasks: (1) determine the higher average fractional anisotropy (FA) values between two regions, (2) find the endpoints of fiber tracts, (3) name a bundle, (4) mark a brain lesion, and (5) judge if tracts belong to the same bundle. Contrary to our hypotheses, we found the task completion time was not improved by the use of the larger display and that performance accuracy was hurt rather than helped by the introduction of stereo in our study with dense DMRI data. Bigger was not always better. Thus cautious should be taken when selecting displays for scientific visualization applications. We explored the results further using the body-scale unit and subjective size and stereo experiences.

Our long-term observations from working with bat biologists reveal that they often have to switch between multiple working environments, e.g., they use matlab to conduct analysis then port the results into a visualization system for confirmation. If more analysis is needed, they switch back to matlab to make changes. Too often, the separation between analysis and visualization caused by current systems can interrupt the analytical thinking process.

In this paper we introduce an effective method to construct a global spatio-temporal representation for action recognition. This representation is inspired by the fact that human actions can be treated as 3D shapes induced by the silhouettes in the space-time volume. We estimate the silhouettes which contain detailed shape information of the action, and present an efficient sampling method to extract interest points along the silhouettes. The local interest point is represented by a spatio-temporal descriptor based on 2D DAISY. Our global space-time representation is the integration of these local descriptors in an order along the silhouette. In this manner, we not only utilize the static shape information, but also the spatial-temporal cue. We have obtained impressive results on publicly available action datasets.

In a user study comparing four visualization methods for three-dimensional vector data, participants used visualizations from each method to perform five simple but representative tasks: 1) determining whether a given point was a critical point, 2) determining the type of a critical point, 3) determining whether an integral curve would advect through two points, 4) determining whether swirling movement is present at a point, and 5) determining whether the vector field is moving faster at one point than another. The visualization methods were line and tube representations of integral curves with both monoscopic and stereoscopic viewing. While participants reported a preference for stereo lines, quantitative results showed performance among the tasks varied by method. Users performed all tasks better with methods that: 1) gave a clear representation with no perceived occlusion, 2) clearly visualized curve speed and direction information, and 3) provided fewer rich 3D cues (e.g., shading, polygonal arrows, overlap cues, and surface textures). These results provide quantitative support for anecdotal evidence on visualization methods. The tasks and testing framework also give a basis for comparing other visualization methods, for creating more effective methods, and for defining additional tasks to explore further the tradeoffs among the methods.

Interaction with medical volume data has often been difficult, due to the large memory and computational power required. By taking advantage of current high-end graphics hardware, we have developed a volumetric virtual environment that provides the ability to help people interact with the volumetric Visible Human data set. The application enables the user to explore the interior of a virtual human body in a natural and intuitive way.