Recent advances in supercomputing capabilities pose a multi-faceted data retrieval challenge to the exploration and visualisation of the obtained results: the bandwidth between visualisation devices and the high-perfo...
详细信息
Collaborative exploration of scientific data sets across large high-resolution displays requires both high visual detail as well as low-latency transfer of image data (oftentimes inducing the need to trade one for the...
详细信息
Collaborative exploration of scientific data sets across large high-resolution displays requires both high visual detail as well as low-latency transfer of image data (oftentimes inducing the need to trade one for the other). In this work, we present a system that dynamically adapts the encoding quality in such systems in a way that reduces the required bandwidth without impacting the details perceived by one or more observers. Humans perceive sharp, colourful details, in the small foveal region around the centre of the field of view, while information in the periphery is perceived blurred and colourless. We account for this by tracking the gaze of observers, and respectively adapting the quality parameter of each macroblock used by the H.264 encoder, considering the so-called visual acuity fall-off. This allows to substantially reduce the required bandwidth with barely noticeable changes in visual quality, which is crucial for collaborative analysis across display walls at different locations. We demonstrate the reduced overall required bandwidth and the high quality inside the foveated regions using particle rendering and parallel coordinates.
Satisfaction surveys are an important measurement tool in fields such as market research or human resources management. Serious studies consist of numerous questions and contain answers from large population samples. ...
详细信息
ISBN:
(纸本)0780381548
Satisfaction surveys are an important measurement tool in fields such as market research or human resources management. Serious studies consist of numerous questions and contain answers from large population samples. Aggregation on both sides, the questions asked as well as the answers received, turns the multidimensional problem into a complex system of interleaved hierarchies. Traditional ways of presenting the results are limited to one-dimensional charts and cross-tables. We developed a visualization method called the parallel Coordinate Tree that combines multidimensional analysis with a tree structure representation. Distortion-oriented focus+context techniques are used to facilitate interaction with the visualization. In this paper we present a design study of a commercial application that we built, using this method to analyze and communicate results from large-scale customer satisfaction surveys.
The contour tree is a tool for understanding the topological structure of a scalar field. Recent work has built efficient contour tree algorithms for shared memory parallel computation, driven by the need to analyze l...
详细信息
ISBN:
(纸本)9781728184685
The contour tree is a tool for understanding the topological structure of a scalar field. Recent work has built efficient contour tree algorithms for shared memory parallel computation, driven by the need to analyze largedata sets in situ while the simulation is running. Unfortunately, methods for using the contour tree for practical data analysis are still primarily serial, including single isocontour extraction, branch decomposition and simplification. We report dataparallel methods for these tasks using a data structure called the hyperstructure and a general purpose approach called a hypersweep. We implement and integrate these methods with a Cinema database that stores features as depth images and with a web server that reconstructs the features for direct visualization.
Volumetric grids have recently been used by many recent works for representing complex scenes implicitly. A volumetric neural representation can be several orders of magnitude smaller in size while still preserving mo...
详细信息
ISBN:
(数字)9781665491563
ISBN:
(纸本)9781665491563
Volumetric grids have recently been used by many recent works for representing complex scenes implicitly. A volumetric neural representation can be several orders of magnitude smaller in size while still preserving most of high-frequency details. However, most volumes used in large-scale in situ visualization and analysis are partitioned and generated directly in parallel. Therefore, a compatible technique to create volumetric neural representations for these situations is much needed. In this project, we explore the possibility of constructing and optimizing such a representation for large-scale distributed volumes. We present our preliminary results in this poster. We also outline our plans to integrate our techniques with existing in situ visualization and analysis pipelines.
larger, higher resolution displays can be used to increase the scalability of information visualizations. But just how much can scalability increase using larger displays before hitting human perceptual or cognitive l...
详细信息
larger, higher resolution displays can be used to increase the scalability of information visualizations. But just how much can scalability increase using larger displays before hitting human perceptual or cognitive limits? Are the same visualization techniques that are. good on a single monitor also the techniques that are best when they are scaled up using large, high-resolution displays? To answer these questions we performed a controlled experiment on user performance time, accuracy, and subjective workload when scaling up data quantity with different space-time-attribute visualizations using a large, tiled display. Twelve college students used small multiples, embedded bar matrices, and embedded time-series graphs either on a 2 megapixel (Mp) display or with data scaled up using a 32 Mp tiled display. Participants performed various overview and detail tasks on geospatially-referenced multidimensional time-series data. Results showed that current designs are perceptually scalable because they result in a decrease in task completion time when normalized per number of data attributes along with no decrease in accuracy. It appears that, for the visualizations selected for this study, the relative comparison between designs is generally consistent between display sizes. However, results also suggest that encoding is more important on a smaller display while spatial grouping is more important on a larger display. Some suggestions for designers are provided based on our experience designing visualizations for large displays.
We present a new parallel algorithm for probabilistic graphical model optimization. The algorithm relies on data-parallel primitives (DPPs), which provide portable performance over hardware architecture. We evaluate r...
详细信息
ISBN:
(纸本)9781538668733
We present a new parallel algorithm for probabilistic graphical model optimization. The algorithm relies on data-parallel primitives (DPPs), which provide portable performance over hardware architecture. We evaluate results on CPUs and GPUs for an image segmentation problem. Compared to a serial baseline, we observe runtime speedups of up to 13X (CPU) and 44X (GPU). We also compare our performance to a reference, OpenMP-based algorithm, and find speedups of up to 7X (CPU).
We propose a multidimensional visualization technique, named Viz3D, that creates a 3D representation of n-dimensional data that may be interactively manipulated by users to handle visual cluttering and object occlusio...
详细信息
ISBN:
(纸本)0769522270
We propose a multidimensional visualization technique, named Viz3D, that creates a 3D representation of n-dimensional data that may be interactively manipulated by users to handle visual cluttering and object occlusion. The projection performed in Viz3D is comparable in quality with the 3D projections obtained with well-known dimensionality reduction techniques, at a lower complexity cost. nile a 3D projection conveys more information, giving the user more control of the visual representation and an additional dimension, as compared to 2D, visual cluttering and object occlusion are still a problem in handling large multidimensional data sets. To produce more effective visualizations, two strategies are introduced Dimensionality is handled with a similarity clustering of attributes prior to projection. data set size is handled with a new strategy of visualizing data densities, rather than individual data records. Both the direct and density Viz3D visualizations provide the basis for a user driven visual clustering approach applicable to high-dimensional data sets that is very simple, intuitive and effective.
The proceedings contain 8 papers. The topics discussed include: rainbow: a rendering-aware index for high-quality spatial scatterplots with result-size budgets & profiling and visualizing GPU memory access and cac...
ISBN:
(纸本)9783038681755
The proceedings contain 8 papers. The topics discussed include: rainbow: a rendering-aware index for high-quality spatial scatterplots with result-size budgets & profiling and visualizing GPU memory access and cache behavior of ray tracers;iterative discrete element solver for efficient snow simulation;massively parallellarge scale inundation modelling & a flexible data streaming design for interactive visualization of large-scale volume data;automatic in situ camera placement for isosurfaces of large-scale scientific simulations;design and evaluation of a GPU streaming framework for visualizing time-varying AMR data;and GraphWaGu: GPU powered large scale graph layout computation and rendering for the web.
The dominant paradigm for searching and browsing largedata stores is text-based: presenting a scrollable list of search results in response to textual search term input. While this works well for the Web, there is op...
详细信息
The dominant paradigm for searching and browsing largedata stores is text-based: presenting a scrollable list of search results in response to textual search term input. While this works well for the Web, there is opportunity for improvement in the domain of personal information stores, which tend to have more heterogeneous data and richer metadata. In this paper, we introduce FacetMap, an interactive, query-driven visualization, generalizable to a wide range of metadata-rich data stores. FacetMap uses a visual metaphor for both input (selection of metadata facets as filters) and output. Results of a user study provide insight into tradeoffs between FacetMap's graphical approach and the traditional text-oriented approach.
暂无评论