Abstract Feature‐based time‐varying volume visualization is combined with illustrative visualization to tell the story of how mesoscale ocean eddies form in the Gulf Stream and transport heat and nutrients across the ocean basin. The internal structure of these three‐dimensional eddies and the kinematics with which they move are critical to a full understanding of ocean eddies. In this work, we apply a feature‐based method to track instances of ocean eddies through the time steps of a high‐resolution multi‐decadal regional ocean model and generate a series of eddy paths which reflect the life cycle of individual eddy instances. Based on the computed metadata, several important geometric and physical properties of eddy are computed. Illustrative visualization techniques, including visual effectiveness enhancement, focus+context, and smart visibility, are combined with the extracted volume features to explore eddy characteristics at different levels. An evaluation by domain experts indicates that combining our feature‐based techniques with illustrative visualization techniques provides an insight into the role eddies play in ocean circulation. The domain experts expressed a preference for our methods over existing tools.

This study investigated how people comprehend three‐dimensional (3D) visualizations and what properties of such visualizations affect comprehension. Participants were asked to draw the face of a 3D visualization after it was cut in half. We videotaped the participants as they drew, erased, verbalized their thoughts, gestured, and moved about a two‐dimensional paper presentation of the 3D visualization. The videorecords were analyzed using a grounded theory approach to generate hypotheses related to comprehension difficulties and visualization properties. Our analysis of the results uncovered three properties that made problem solving more difficult for participants. These were: (a) cuts that were at an angle in relation to at least one plane of reference, (b) nonplanar properties of the features contained in the 3D visualizations including curved layers and v‐shaped layers, and (c) mixed combinations of layers. In contrast, (a) cutting planes that were perpendicular or parallel to the 3D visualization diagram's planes of reference, (b) internal features that were flat/planar, and (c) homogeneous layers were easier to comprehend. This research has direct implications for the generation and use of 3D information visualizations in that it suggests design features to include and avoid.

For large-scale simulations, the data sets are so massive that it is sometimes not feasible to view the data with basic visualization methods, let alone explore all time steps in detail. Automated tools are necessary for knowledge discovery, i.e., to help sift through the data and isolate specific time steps that can then be further explored. Scientists study patterns and interactions and want to know when and where interesting things happen. Activity detection, the detection of specific interactions of objects which span a limited duration of time, has been an active research area in the computer vision community. In this paper, we introduce activity detection to scientific simulations and show how it can be utilized in scientific visualization. We show how activity detection allows a scientist to model an activity and can then validate their hypothesis on the underlying processes. Three case studies are presented.

Abstract A sampled object representation (SOR) defines a graphical model using data obtained from a sampling process, which takes a collection of samples at discrete positions in space in order to capture certain geometrical and physical properties of one or more objects of interest. Examples of SORs include images, videos, volume datasets and point datasets. Unlike many commonly used data representations in computer graphics, SORs lack in geometrical, topological and semantic information, which is much needed for controlling deformation and animation. Hence it poses a significant scientific and technical challenge to develop deformation and animation methods that operate upon SORs. Such methods can enable computer graphics and computer animation to benefit enormously from the advances of digital imaging technology. In this state of the art report, we survey a wide range of techniques that have been developed for manipulating, deforming and animating SORs. We consider a collection of elementary operations for manipulating SORs, which can serve as building blocks of deformation and animation techniques. We examine a collection of techniques that are designed to transform the geometry shape of deformable objects in sampled representations and pay particular attention to their deployment in surgical simulation. We review a collection of techniques for animating digital characters in SORs, focusing on recent developments in volume animation.

Curve-skeletons are thinned 1D representations of 3D objects useful for many visualization tasks including virtual navigation, reduced-model formulation, visualization improvement, animation, etc. There are many algorithms in the literature describing extraction methodologies for different applications; however, it is unclear how general and robust they are. In this paper, we provide an overview of many curve-skeleton applications and compile a set of desired properties of such representations. We also give a taxonomy of methods and analyze the advantages and drawbacks of each class of algorithms.

Much of the visualization research has focused on improving the rendering quality and speed, and enhancing the perceptibility of features in the data. Recently, significant emphasis has been placed on focus+context (F+C) techniques (e.g., fisheye views and magnification lens) for data exploration in addition to viewing transformation and hierarchical navigation. However, most of the existing data exploration techniques rely on the manipulation of viewing attributes of the rendering system or optical attributes of the data objects, with users being passive viewers. In this paper, we propose a more active approach to data exploration, which attempts to mimic how we would explore data if we were able to hold it and interact with it in our hands. This involves allowing the users to physically or actively manipulate the geometry of a data object. While this approach has been traditionally used in applications, such as surgical simulation, where the original geometry of the data objects is well understood by the users, there are several challenges when this approach is generalized for applications, such as flow and information visualization, where there is no common perception as to the normal or natural geometry of a data object. We introduce a taxonomy and a set of transformations especially for illustrative deformation of general data exploration. We present combined geometric or optical illustration operators for focus+context visualization, and examine the best means for preventing the deformed context from being misperceived. We demonstrated the feasibility of this generalization with examples of flow, information and video visualization.

Splitting a volumetric object is a useful operation in volume graphics and its applications, but is not widely supported by existing systems for volume-based modeling and rendering. In this paper, we present an investigation into two main algorithmic approaches, namely, explicit and implicit splitting, for modeling and rendering splitting actions. We consider a generalized notion based on scalar fields, which encompasses discrete specifications (e.g., volume data sets) as well as procedural specifications (e.g., hypertextures) of volumetric objects. We examine the correctness, effectiveness, efficiency, and deficiencies of each approach in specifying and controlling a spatial and temporal specification of splitting. We propose methods for implementing these approaches and for overcoming their deficiencies. We present a modeling tool for creating specifications of splitting functions, and describe the use of volume scene graphs for facilitating direct rendering of volume splitting. We demonstrate the use of these approaches with examples of volume visualization, medical illustration, volume animation, and special effects.

In this paper, we explore the concept of a "library of brain images", which implies not only a repository of brain images, but also efficient search and retrieval mechanisms that are based on models derived from IR practice. As a preliminary study, we have worked with a collection of functional MRI brain images assembled in the study of several distinct cognitive tasks. We adapt several classical IR methods (inverted indexing, TFIDF and Latent Semantic Indexing(LSI)) to content-based brain image retrieval. Our results show that efficient and accurate retrieval of brain images is possible, and that representations motivated by the IR perspective are somewhat more effective than are methods based on retaining the full image information.

Video visualization is a newly emerged technology drawing the concepts and methodologies from volume and flow visualization, image and video processing, and vision science. It extracts meaningful information from a video data set and conveys the extracted information to users in appropriate visual representations. It is not intended to provide fully automatic solutions to the traditional problems in video processing, but aims at offering an approach that involves human in the loop of intelligent reasoning while removing the burden of viewing videos. In this paper, we examine the technical scope of this new technology and relate it to the technical advances in video processing. In particular, we highlight the areas where video visualization can offer more generic solutions with minimal effort for adjusting scene-specific parameters and hard-coding application-specific logic, as well as areas where video visualization can benefit from a large collection of theories and algorithms which have already been developed for video processing. We also briefly discuss the potential applicability of advances in video processing to the broader spectrum of time-varying data visualization

The output from the massively parallel scientific simulations is so voluminous and complex that advanced visualization technologies are necessary to interpret the calculated results. Even though visualization technology has progressed significantly in recent years, we are barely capable of visualizing and analyzing terascale data to its full extent, and petascale datasets are on the horizon. This workshop aims at addressing this pressing issue by fostering communication between visualization researchers and practitioners. The workshop attendees will be introduced to the latest and greatest research innovations in large data visualization and also help direct further research direction through an open discussion session.

In this paper we describe a GPU-based technique for creating illustrative visualization through interactive manipulation of volumetric models. It is partly inspired by medical illustrations, where it is common to depict cuts and deformation in order to provide a better understanding of anatomical and biological structures or surgical processes, and partly motivated by the need for a real-time solution that supports the specification and visualization of such illustrative manipulation. We propose two new feature-aligned techniques, namely surface alignment and segment alignment, and compare them with the axis-aligned techniques which was reported in previous work on volume manipulation. We also present a mechanism for defining features using texture volumes, and methods for computing correct normals for the deformed volume in respect to different alignments. We describe a GPU-based implementation to achieve real-time performance of the techniques and a collection of manipulation operators including peelers, retractors, pliers and dilators which are adaptations of the metaphors and tools used in surgical procedures and medical illustrations. Our approach is directly applicable in medical and biological illustration, and we demonstrate how it works as an interactive tool for focus+context visualization, as well as a generic technique for volume graphics.

We present a 3D matching framework based on a many-to-many matching algorithm that works with skeletal representations of 3D volumetric objects. We demonstrate the performance of this approach on a large database of 3D objects containing more than 1000 exemplars. The method is especially suited to matching objects with distinct part structure and is invariant to part articulation. Skeletal matching has an intuitive quality that helps in defining the search and visualizing the results. In particular, the matching algorithm produces a direct correspondence between two skeletons and their parts, which can be used for registration and juxtaposition.

Curve-skeletons are a 1D subset of the medial surface of a 3D object and are useful for many visualization tasks including virtual navigation, reduced-model formulation, visualization improvement, mesh repair, animation, etc. There are many algorithms in the literature describing extraction methodologies for different applications; however, it is unclear how general and robust they are. In this paper, we provide an overview of many curve-skeleton applications and compile a set of desired properties of such representations. We also give a taxonomy of methods and analyze the advantages and drawbacks of each class of algorithms.

In this paper, we describe a methodology and implementation for interactive dataset traversal using motion-controlled transfer functions. Dataset traversal here refers lo the process of translating a transfer function along a specific path. In scientific visualization, it is often necessary to manipulate transfer functions in order to visualize datasets more effectively. This manipulation of transfer functions is usually performed globally, i.e., a new transfer function is applied to the entire dataset. Our approach allows one to locally manipulate transfer functions while controling its movement along a traversal path. The method we propose allows the user to select a traversal path within the dataset, based on the shape of the volumetric model and manipulate a transfer function along this path. Examples of dataset traversal include the animation of transfer functions along a pre-defined path, the simulation of flow in vascular structures, and the visualization of convoluted shapes. For example, this type of traversal is often used in medical illustration to highlight flow in blood vessels. We present an interactive implementation of our method using graphics hardware, based on the decomposition of the volume. We show examples of our approach using a variety of volumetric datasets, and we also demonstrate that with our novel decomposition, the rendering process is faster.

Little is known about the cognitive abilities which influence the comprehension of scientific and information visualizations and what properties of the visualization affect comprehension. Our goal in this paper is to understand what makes visualizations difficult. We address this goal by examining the spatial ability differences in a diverse population selected for spatial ability variance. For example, how is, spatial ability related to visualization comprehension? What makes a particular visualization difficult or time intensive for specific groups of subjects? In this paper, we present the results of an experiment designed to answer these questions. Fifty-six subjects were tested on a basic visualization task and given standard paper tests of spatial abilities. An equal number of males and females were recruited in this study in order to increase spatial ability variance. Our results show that high spatial ability is correlated with accuracy on our three-dimensional visualization test, but not with time. High spatial ability subjects also had less difficulty with object complexity and the hidden properties of an object.

Most data used in the study of seafloor hydrothermal plumes consists of sonar (acoustic) scans and sensor readings. Visual data captures only a portion of the sonar data range due to the prohibitive cost and physical infeasibility of taking sufficient lighting and video equipment to such extreme depths. However, visual images are available from research dives and from the recent IMAX movie, volcanoes of the deep sea. In this application paper, we apply existing lighting models with forward scattering and light attenuation to the 3D sonar data in order to mimic the visual images available. These generated images are compared to existing visual images. This can help the geoscientists understand the relationship between these different data modalities and elucidate some of the mechanisms used to capture the data.

We describe a novel method for searching and comparing 3D objects. The method encodes the geometric and topological information in the form of a skeletal graph and uses graph matching techniques to match the skeletons and to compare them. The skeletal graphs can be manually annotated to refine or restructure the search. This helps in choosing between a topological similarity and a geometric (shape) similarity. A feature of skeletal matching is the ability to perform part-matching, and its inherent intuitiveness, which helps in defining the search and in visualizing the results. Also, the matching results, which are presented in a per-node basis can be used for driving a number of registration algorithms, most of which require a good initial guess to perform registration. We also describe a visualization tool to aid in the selection and specification of the matched objects.

Interactive multimedia walkthrough applications are useful tools for visualizing complex areas. These environments permit navigation through a virtual space based on intuitive actions like "go forward" or "go left". The space is generally constructed using computer graphics models and enhanced with video, still images, and sound. While video is usually incorporated into these models, it is played as taken, and generally as a dependent media, i.e. the navigation controls do not control the video even if the video is a "real" walkthrough of the virtual space. Integrating the real-world media and the computer graphics model by registering both within a common virtual reality framework would allow navigation in the computer graphics model to control video of the corresponding location in the real world and vice versa. This would cause the computer graphics model to be regenerated based on the camera position in the real world video. Additionally, by registering into the common framework, new media data of any type can be added into the presentation. This paper presents a multimedia data structure capable of supporting these operations. A prototypical mobile video capture device is discussed that can record the required video media and process it for inclusion in the framework described.

Applications of visualization techniques that facilitate comparison of simulation and field datasets of seafloor hydrothermal plumes are demonstrated in order to explore and confirm theories of plume behavior. In comparing these datasets, there is no one-to-one correspondence. We show the comparison by performing quantitative capturing of large scale observable features. The comparisons are needed not only to improve the relevance of the simulations to the field observations, but also to enable real time adjustment of shipboard data collection systems. Our approach for comparing simulation and field datasets is to use skeletonization and centerline representation. Features representing plumes are skeletonized. Skeleton points are used to construct a centerline and to quantify plume properties on planes normal to the centerline. These skeleton points are further used to construct an idealized cone representing a plume isosurface. The difference between the plume feature and the cone is identified as protrusions of turbulent eddies. Comparison of the simulation and field data sets through these abstractions illustrates how these abstractions characterize a plume.

Real-time visualization of large volume data sets demands high-performance computation, pushing the storage, processing and data communication requirements to the limits of current technology. General-purpose parallel processors have been used to visualize moderate-size data sets at interactive frame rates; however, the cost and size of these supercomputers inhibits the widespread use for real-time visualization. This paper surveys several special-purpose architectures that seek to render volumes at interactive rates. These specialized visualization accelerators have cost, performance and size advantages over parallel processors. All architectures implement ray casting using parallel and pipelined hardware. We introduce a new metric that normalizes performance to compare these architectures. The architectures included in this survey are VOGUE, VIRIM, Array-Based Ray Casting, EM-Cube and VIZARD II. We also discuss future applications of special-purpose accelerators.

Visualization and quantification methods are being developed to analyze our acoustic images of thermal plumes containing metallic mineral particles that discharge from hot springs on the deep seafloor. The acoustic images record intensity of backscattering from the particulate matter suspended in the plumes. The visualization methods extract, classify, visualize, measure and track reconstructions of the plumes, depicted by isointensity surfaces as 3D volume objects and 2D slices. The parameters measured, including plume volume, cross sectional area, centerline location (trajectory), surface area and isosurfaces at percentages of maximum backscatter intensity, are being used to derive elements of plume behavior including expansion with height, dilution, and mechanisms of entrainment of surrounding seawater. Our aim is to compare the observational data with predictions of plume theory to test and advance models of the behavior of hydrothermal plumes through the use of multiple representations.

3D time-varying unstructured and structured data sets are difficult to visualize and analyze because of the immense amount of data involved. These data sets contain many evolving amorphous regions, and standard visualization techniques provide no facilities to aid the scientist to follow regions of interest. In this paper, we present a basic framework for the visualization of time-varying data sets, and a new algorithm and data structure to track volume features in unstructured scalar data sets. The algorithm and data structure are general and can be used for structured, curvilinear, adaptive and hybrid grids as well. The features tracked can be any type of connected regions. Examples are shown from ongoing research.

Visualizing 3D time-varying fluid datasets is difficult because of the immense amount of data to be processed and understood. These datasets contain many evolving amorphous regions, and it is difficult to observe patterns and visually follow regions of interest. In this paper, we present a technique which isolates and tracks full-volume representations of regions of interest from 3D regular and curvilinear computational fluid dynamics datasets. Connected voxel regions ("features") are extracted from each time step and matched to features in subsequent time steps. Spatial overlap is used to determine the matching. The features from each time step are stored in octree forests to speed up the matching process. Once the features have been identified and tracked, the properties of the features and their evolutionary history can be computed. This information can be used to enhance isosurface visualization and volume rendering by color coding individual regions. We demonstrate the algorithm on four 3D time-varying simulations from ongoing research in computational fluid dynamics and show how tracking can significantly improve and facilitate the processing of massive datasets.

3D time varying datasets are difficult to visualize and analyze because of the immense amount of data involved. This is especially true when the datasets are turbulent with many evolving amorphous regions, as it is difficult to observe patterns and follow regions of interest. We present our volume based feature tracking algorithm and discuss how it can be used to help visualize and analyze large time varying datasets. We also address efficiency issues in dealing with massive time varying datasets.

We present a conceptual framework and a process model for feature extraction and iconic visualization. The features are regions of interest extracted from a dataset. They are represented by attribute sets, which play a key role in the visualization process. These attribute sets are mapped to icons, or symbolic parametric objects, for visualization. The features provide a compact abstraction of the original data, and the icons are a natural way to visualize them. We present generic techniques to extract features and to calculate attribute sets, and describe a simple but powerful modeling language which was developed to create icons and to link the attributes to the icon parameters. We present illustrative examples of iconic visualization created with the techniques described, showing the effectiveness of this approach.

Volume visualization is used heavily to view simulated or collected data sets in such applications as medical imaging, computational fluid dynamics, and climate modeling. However, software and low-cost hardware implementations of visualization algorithms do not have sufficient performance for interactive viewing. This paper discusses methods for low-cost, hardware acceleration of volume visualization using a PC-hosted FPGA board. Our methods focus on volume rendering approaches, since these techniques are widely used and are computationally expensive; our primary method uses a template-based, ray-casting algorithm. This hardware implementation is substantially faster than a software-only version running on the host PC.

Presents a conceptual framework and a process model for feature extraction and iconic visualization. Feature extraction is viewed as a process of data abstraction, which can proceed in multiple stages, and corresponding data abstraction levels. The features are represented by attribute sets, which play a key role in the visualization process. Icons are symbolic parametric objects, designed as visual representations of features. The attributes are mapped to the parameters (or degrees of freedom) of an icon. We describe some generic techniques to generate attribute sets, such as volume integrals and medial axis transforms. A simple but powerful modeling language was developed to create icons, and to link the attributes to the icon parameters. We present illustrative examples of iconic visualization created with the techniques described, showing the effectiveness of this approach.

A brief example of the visualization and quantification of a complex fluid interaction is presented in order to give one a feeling for the difficulty of dealing with geometrical and topological questions in three dimensions and time. To obtain a quantitative understanding, visiometric techniques, including thresholding, object isolation, ellipsoid fitting, abstraction, vector field line generation and data juxtaposition, were used.>

A method is presented for juxtaposing 4D space-time vector fields, of which one contains a source variable and the other the response field. Thresholding, ellipsoid fitting, and vortex line generation are used to reduce the amount of information and help analyze the relationship between two 3D vector variables evolving in time. The technique helps to highlight the topological relationship between the two in an effort to understand the causal connection. These concepts are applied to on-going research in evolving fluid dynamics problems.>

Geometric computations, like all numerical procedures, are extremely prone to roundoff error. However, virtually none of the numerical analysis literature directly applies to geometric calculations. Even for line intersection, the most basic geometric operation, there is no robust and efficient algorithm. Compounding the difficulties, many geometric algorithms perform iterations of calculations reusing previously computed data. In this paper, we explore some of the main issues in geometric computations and the methods that have been proposed to handle roundoff errors. In particular, we focus on one method and apply it to a general iterative intersection problem. Our initial results seem promising and will hopefully lead to robust solutions for more complex problems of computational geometry.