EGPGV07: Eurographics Symposium on Parallel Graphics and Visualization

Permanent URI for this collection

https://diglib.eg.org/handle/10.2312/360

Browse

Now showing 1 - 13 of 13

Asynchronous BVH Construction for Ray Tracing Dynamic Scenes on Parallel Multi-Core Architectures
(The Eurographics Association, 2007) Ize, Thiago; Wald, Ingo; Parker, Steven G.; Jean M. Favre and Luis Paulo Santos and Dirk Reiners
Recent developments have produced several techniques for interactive ray tracing of dynamic scenes. In particular, bounding volume hierarchies (BVHs) are efficient acceleration structures that handle complex triangle distributions and can accommodate deformable scenes by updating (refitting) the bounding primitive without restructuring the entire tree. Unfortunately, updating only the bounding primitive can result in a degradation of the quality of the BVH, and in some scenes will result in a dramatic deterioration of rendering performance. The typical method to avoid this degradation is to rebuild the BVH when a heuristic determines the tree is no longer efficient, but this rebuild results in a disruption of interactive system response. We present a method that removes this gradual decline in performance while enabling consistently fast BVH performance. We accomplish this by asynchronously rebuilding the BVH concurrently with rendering and animation, allowing the BVH to be restructured within a handful of frames.
Direct Send Compositing for Parallel Sort-Last Rendering
(The Eurographics Association, 2007) Eilemann, Stefan; Pajarola, Renato; Jean M. Favre and Luis Paulo Santos and Dirk Reiners
In contrast to sort-first, sort-last parallel rendering has the distinct advantage that the task division for parallel geometry processing and rasterization is simple, and can easily be incorporated into most visualization systems. However, the efficient final depth-compositing for polygonal data, or alpha-blending for volume data of partial rendering results is the key to achieve scalability in sort-last parallel rendering. In this paper, we demonstrate the efficiency as well as flexibility of the direct send sort-last compositing algorithm, and compare it to existing approaches, both in a theoretical analysis and in an experimental setting.
Distributed Collaborative Data Analysis with Heterogeneous Visualisation Systems
(The Eurographics Association, 2007) Duessel, Thomas; Zilken, Herwig; Frings, W.; Eickermann, T.; Gerndt, A.; Wolter, Marc; Kuhlen, Torsten; Jean M. Favre and Luis Paulo Santos and Dirk Reiners
A system for the distributed, collaborative online visualisation in heterogeneous visualisation environments was developed and tested in the application project KoDaVis, which is part of the german optical network testbed VIOLA. The aim of KoDaVis is the visualisation of huge data sets from atmosphere research. The core of the presented distributed computer supported collaborative work system is a framework for the coupling of heterogeneous visualisation systems and the design and implementation of two distinct servers, one for the collaborative aspects and one for the direct remote access to centrally stored data. Interfaces to the VTK-based virtual reality visualisation system ViSTA and to the modular visualisation environment AVS/Express were implemented and tested. The successful coupling of these two different visualisation systems as well as the benefit of a fast optical network for parallel data access and for distributed collaboration could be demonstrated in a test setup.
Dynamic Regions of Interest for Interactive Flow Exploration
(The Eurographics Association, 2007) Wolter, Marc; Bischof, C.; Kuhlen, Torsten; Jean M. Favre and Luis Paulo Santos and Dirk Reiners
Virtual Reality (VR) provides a useful tool for understanding complex, unsteady flow phenomena. The user can directly interact with the data and therefore benefits from a spatial coherence of action and result. However, visualization in virtual environments imposes very high demands on interactivity in order to maintain this coherence. Exploration of large, unsteady datasets in VR requires efficient visualization or data reduction algorithms to produce results within acceptable waiting times. We propose a technique for reducing required data especially suited for direct interaction in virtual environments. We use a distributed system to parallely extract a dynamic region of interest (DROI) from the simulation data. This DROI is adapted according to the user s interaction behavior and allows for the analysis of local flow features. With this reduction we provide interactive extraction of local features from large, time-varying datasets.
Exploiting Parallelism in Physically-Based Simulations on Multi-Core Processor Architectures
(The Eurographics Association, 2007) Thomaszewski, Bernhard; Pabst, Simon; Blochinger, Wolfgang; Jean M. Favre and Luis Paulo Santos and Dirk Reiners
As multi-core processor systems become more and more widespread, the demand for designing efficient parallel algorithms propagates also into the field of computer graphics. This is especially true for the physically-based simulation, which is notorious for expensive numerical methods. In this paper we explore possibilities for accelerating these algorithms on modern multi-core architectures. As an application we focus on physically-based cloth simulation. In this context, two distinct problems can be identified: the physical model and the collision handling stage both bearing potential bottlenecks for the simulation. From the parallelization point of view these two components are substantially different. The physical model can be treated efficiently using static problem decomposition. The collision handling problem, however, requires a different approach, due to its dynamically changing structure. We address this problem using multi-threaded programming with fully dynamic task decomposition. Furthermore, we propose a new task splitting approach based on a robust work estimate. The associated data is derived from temporal coherence. Altogether, the combination of different parallelization techniques leads to a concise and yet versatile framework for highly efficient physical simulation.
Hybrid CPU-GPU Unstructured Meshes Parallel Volume Rendering on PC Clusters
(The Eurographics Association, 2007) Juliachs, Manuel; Carrard, Thierry; Nomine, Jean-Philippe; Jean M. Favre and Luis Paulo Santos and Dirk Reiners
Large-scale numerical simulation produces datasets with ever-growing size and complexity. In particular, unstructured meshes are encountered in many applications. Volume rendering provides a way to efficiently analyze such datasets. Recent advances in graphics hardware have enabled the implementation of efficient unstructured volume rendering algorithms on the GPU. However, GPU architecture limitations make these methods difficultly amenable to a parallel implementation, which is necessary to render very large datasets at interactive speeds and high resolutions. Many previous parallel approaches have focused on softwarebased algorithms. In this paper, we present a hybrid object-space/image-space CPU-GPU distributed parallel volume rendering method, taking advantage of the flexibility afforded by the CPU, including SIMD processing capabilities, and using GPUs to perform repetitive tasks like depth-sorting and compositing. We present the impact of the different phases on the overall rendering time as a function of node number.
Interactive Iso-Surface Ray Tracing of Massive Volumetric Data Sets
(The Eurographics Association, 2007) Friedrich, Heiko; Wald, Ingo; Guenther, J.; Marmitt, G.; Slusallek, Phillip; Jean M. Favre and Luis Paulo Santos and Dirk Reiners
The visualization of iso-surfaces from gridded volume data is an important tool in many scientific applications. Today, it is possible to ray trace high-quality iso-surfaces at interactive frame rates even on commodity PCs. However, current algorithms fail if the data set exceeds a certain size either because they are not designed for outof- core data sets or the loading times are too high because there is too much overhead involved in the out-of-core (OOC) techniques. We propose a kD-tree based OOC data structure that allows to ray trace iso-surfaces of large volumetric data sets of many giga bytes at interactive frame rates on a single PC. A LOD technique is used to bridge loading times of data that is fetched asynchronously in the background. Using this framework we are able to ray trace iso-surfaces between 2 and 4 fps on a single dual-core Opteron PC at 640×480 resolution and an in-core memory footprint that is only a fraction of the entire data size.
Interactive Particle Visualization with Advanced Shading Models using Lazy Evaluation
(The Eurographics Association, 2007) Gribble, Christiaan P.; Parker, Steven G.; Jean M. Favre and Luis Paulo Santos and Dirk Reiners
Particle-based simulation methods are used to model a wide range of complex phenomena and to solve timedependent problems of various scales. Effective visualizations of the resulting state will communicate subtle changes in the three-dimensional structure, spatial organization, and qualitative trends within a simulation as it evolves. We describe a visualization process targeting upcoming, highly parallel multicore desktop systems that enables interactive navigation and exploration of large particle datasets rendered with illumination effects from advanced shading models. These expensive illumination effects are evaluated lazily by decoupling interactive display and high quality rendering. We explore the performance characteristics of this approach and demonstrate its effectiveness using several large particle datasets.
iRun: Interactive Rendering of Large Unstructured Grids
(The Eurographics Association, 2007) Vo, Huy T.; Callahan, Steven P.; Smith, Nathan; Silva, Claudio T.; Martin, William; Owen, David; Weinstein, David; Jean M. Favre and Luis Paulo Santos and Dirk Reiners
We present iRun, a system for interactively volume rendering large unstructured grids on commodity PCs. Rendering arbitrarily large datasets has been an active area of research for many years. However, the techniques required for polygonal data do not directly apply to the more complex problem of unstructured grids. In this paper, we describe the data structures and algorithms necessary to store large datasets on disk, keep an active portion of the dataset in main memory, and render visible regions to one or more displays. Our system leverages a combination of out-of-core data management, distributed rendering, hardware-accelerated volume rendering, and dynamic level-of-detail. On a commodity PC, our system can preprocess a dataset consisting of about 36 million tetrahedra in about an hour and can render it interactively with one or more PCs.
Multiresolution Visualization of Massive Models on a Large Spatial 3D Display
(The Eurographics Association, 2007) Bettio, Fabio; Gobbetti, Enrico; Pintore, Giovanni; Marton, Fabio; Jean M. Favre and Luis Paulo Santos and Dirk Reiners
We report on a cluster parallel multiresolution rendering system driving a spatial 3D display able to give multiple freely moving naked-eye viewers the illusion of seeing virtual objects floating at fixed physical locations situated in a human scale working volume. The efficiency of this approach is demonstrated by an application supporting interactive manipulation of colored highly tessellated models on a large (1.6x0.9 meters) 50Mpixel display that allows for a room-size working space.
Parallel Reflective Symmetry Transformation for Volume Data
(The Eurographics Association, 2007) Hong, Yuan; Shen, Han-Wei; Jean M. Favre and Luis Paulo Santos and Dirk Reiners
Many volume data possess symmetric features that can be clearly observed, for example, those existed in diffusion tensor image data sets. The exploitations of symmetries for volume data sets, however, are relatively limited due to the prohibitive computational cost of detecting the symmetries. In this paper we present an efficient parallel algorithm for symmetry computation in volume data represented by regular grids. Optimization is achieved by converting the raw data into a hierarchical tree-like structure.We design a novel algorithm to partition the tree and distribute the data among processors to minimize the data dependency at run time. The computed symmetries are useful for several volume data applications, including POF minimal opacity selection, transfer function generation and slice position selection.
Scalable Sort-First Parallel Direct Volume Rendering with Dynamic Load Balancing
(The Eurographics Association, 2007) Moloney, Brendan; Weiskopf, Daniel; Moeller, Torsten; Strengert, Magnus; Jean M. Favre and Luis Paulo Santos and Dirk Reiners
We describe a sort-first algorithm for parallel direct volume rendering on GPUs, with the intent of high scalability in regards to both performance and data set size. We explore three novel techniques for estimating the computation time for rendering each pixel, so that we can guarantee a good load balancing regardless of the level of frame-to frame coherence. A bricking technique is used to subdivide the object space, thus allowing each rendering node to load only the bricks of data that are needed to render their respective portions of the image space. This enables us to render data sets larger than an individual GPU's texture memory. We cull bricks that do not contribute to the final image in order to reduce the data that is loaded and provide a coarse method of empty space leaping. We introduce a novel method of eliminating the overhead of generating vertices for the proxy geometry of each brick, by creating a single template of vertices that are used to render all bricks of the same size. Finally, detailed performance measurements document the various aspects of our algorithm.
Work Stealing for Time-constrained Octree Exploration: Application to Real-time 3D Modeling
(The Eurographics Association, 2007) Soares, Luciano; Menier, Clement; Raffin, Bruno; Roch, Jean-Louis; Jean M. Favre and Luis Paulo Santos and Dirk Reiners
This paper introduces a dynamic work balancing algorithm, based on work stealing, for time-constrained parallel octree carving. The performance of the algorithm is proved and confirmed by experimental results where the algorithm is applied to a real-time 3D modeling from multiple video streams. Compared to classical work stealing, the proposed algorithm enforces a relaxed width first octree carving that enables to stop computations at anytime while ensuring a balanced carving.

Browse

Browsing EGPGV07: Eurographics Symposium on Parallel Graphics and Visualization by Title

Results Per Page

Sort Options