VG05: Eurographics/IEEE VGTC Workshop on Volume Graphics 2005ISBN 3-905673-26-6https://diglib.eg.org:443/handle/10.2312/4722024-03-29T15:04:58Z2024-03-29T15:04:58ZGPU Accelerated Image Aligned SplattingNeophytou, NeophytosMueller, Klaushttps://diglib.eg.org:443/handle/10.2312/VG.VG05.197-2052022-03-28T09:54:54Z2005-01-01T00:00:00ZGPU Accelerated Image Aligned Splatting
Neophytou, Neophytos; Mueller, Klaus
Klaus Mueller and Thomas Ertl and Eduard Groeller
Splatting is a popular technique for volume rendering, where voxels are represented by Gaussian kernels, whose pre-integrated footprints are accumulated to form the image. Splatting has been mainly used to render pre-shaded volumes, which can result in significant blurring in zoomed views. This can be avoided in the image-aligned splatting scheme, where one accumulates kernel slices into equi-distant, parallel sheet buffers, followed by classification, shading, and compositing. In this work we attempt to evolve this algorithm to the next level: GPU based acceleration. First we describe the challenges that the highly parallel Gather architecture of modern GPUs poses to the Scatter based nature of a splatting algorithm. We then describe a number of strategies that exploit newly introduced features of the latest-generation hardware to address these limitations. Two crucial operations to boost the performance in image-aligned splatting are the early elimination of hidden splats and the skipping of empty buffer-space. We will describe mechanisms which take advantage of the early z-culling hardware facilities to accomplish both of these operations efficiently in hardware.
2005-01-01T00:00:00ZiSBVR: Isosurface-aided Hardware Acceleration Techniques for Slice-Based Volume RenderingXue, DaqingZhang, CaixiaCrawfis, Rogerhttps://diglib.eg.org:443/handle/10.2312/VG.VG05.207-2152022-03-28T09:54:39Z2005-01-01T00:00:00ZiSBVR: Isosurface-aided Hardware Acceleration Techniques for Slice-Based Volume Rendering
Xue, Daqing; Zhang, Caixia; Crawfis, Roger
Klaus Mueller and Thomas Ertl and Eduard Groeller
In this paper, we examine the performance of the early z-culling feature on current high-end commodity graphics cards and present an isosurface-aided hardware acceleration algorithm for slice-based volume rendering (iSBVR) to maximize its utilization. We analyze the computational models for early z-culling of the texture based volume rendering. We demonstrate that the performance improves with two to four times speedup against an original straightforward SBVR on an ATI 9800 pro display board. As volumetric shaders become increasingly complex, the advantages of fast z-culling will become even more pronounced.
2005-01-01T00:00:00ZA Simple and Flexible Volume Rendering Framework for Graphics-Hardware-based RaycastingStegmaier, SimonStrengert, MagnusKlein, ThomasErtl, Thomashttps://diglib.eg.org:443/handle/10.2312/VG.VG05.187-1952022-03-28T09:54:46Z2005-01-01T00:00:00ZA Simple and Flexible Volume Rendering Framework for Graphics-Hardware-based Raycasting
Stegmaier, Simon; Strengert, Magnus; Klein, Thomas; Ertl, Thomas
Klaus Mueller and Thomas Ertl and Eduard Groeller
In this work we present a flexible framework for GPU-based volume rendering. The framework is based on a single pass volume raycasting approach and is easily extensible in terms of new shader functionality. We demonstrate the flexibility of our system by means of a number of high-quality standard and non-standard volume rendering techniques. Our implementation shows a promising performance in a number of benchmarks while producing images of higher accuracy than obtained by standard pre-integrated slice-based volume rendering.
2005-01-01T00:00:00ZGPU-based Object-Order Ray-Casting for Large DatasetsHong, WeiQiu, FengKaufman, Ariehttps://diglib.eg.org:443/handle/10.2312/VG.VG05.177-1852022-03-28T09:54:35Z2005-01-01T00:00:00ZGPU-based Object-Order Ray-Casting for Large Datasets
Hong, Wei; Qiu, Feng; Kaufman, Arie
Klaus Mueller and Thomas Ertl and Eduard Groeller
We propose a GPU-based object-order ray-casting algorithm for the rendering of large volumetric datasets, such as the Visible Human CT datasets. A volumetric dataset is decomposed into small sub-volumes, which are then organized using a min-max octree structure. The small sub-volumes are stored in the leaf nodes of the min-max octree, which are also called cells. The cells are classified using a transfer function, and the visible cells are then loaded into the video memory or the AGP memory. The cells are sorted and projected onto the image plane front to back. The cell projection is implemented using a volumetric ray-casting algorithm on the GPU. In order to make the cell projection more efficient, we devise a propagation method to sort cells into layers. The cells within the same layer are projected at the same time. We demonstrate the efficiency of our algorithm using the Visible Human datasets and a segmented photographic brain dataset on commodity PCs.
2005-01-01T00:00:00Z