This workshop will first focus on an entry-level GPGPU introduction by Robert Strzodka, followed by a discussion of hardware-efficient PDE solvers with applications in Image Processing. After a brief review of the current state-of-the-art of FEM codes on the CPU side, algorithmic design patterns for efficient GPU implementations will be experimentally evaluated. (Note: talks will be given in German.) http://www.mathematik.uni-dortmund.de/~goeddeke/workshop/index.html
Excerpted from GPU Gems 2, “The GeForce 6 Series GPU Architecture” (Chapter 30) describes the architecture of the GeForce 6 Series family of GPUs, including details on the overall system architecture, vertex processor, fragment processor, and various other features. (Emmett Kilgariff and Randima Fernando. “The GeForce 6 Series GPU Architecture”, in GPU Gems 2, Addison-Wesley 2005.)
gDEBugger, an OpenGL API debugger, traces application activity on top of the OpenGL API, letting programmers see what is happening within the graphic system implementation. The new V1.3 is capable of independently enabling and disabling graphics pipeline stages to pinpoint the location of graphics performance bottlenecks. These new gDebugger features include: Ignore all OpenGL Draw Commands; Force Single-Pixel View Port; and Force 2×2 Stub Textures. By observing application performance using the heads-up FPS display when turning off individual stages, programmers can determine the exact location of bottlenecks. (www.gremedy.com)
Ne@tware Player 2005 is a shader player. It supports video over a model with real-time visual special effects using programmable HLSL shaders. Its multithreaded media engine supports media mixing, media codecs, and Shader Model 3.0. Developers can also design and test their own HLSL shaders and FX effects in Ne@tware Player 2005.
Efficient and visually compelling reproduction of effects due to multiple scattering in participating media remains one of the most difficult tasks in computer graphics. Although several fast techniques were recently developed, most of them work only for special types of media (for example, uniform or sufficiently dense) or require extensive precomputation. In this paper we present a lighting model for the general case of inhomogeneous medium and demonstrate its implementation on programmable graphics hardware. It is capable of producing high quality imagery at interactive frame rates with only mild assumptions about medium scattering properties and a moderate amount of simple precomputation. (A Lighting Model for General Participating Media. Kyle Hegeman, Michael Ashikhmin and Simon Premoze. Accepted for publication. Proceedings of ACM SIGGRAPH Symposium on Interactive 3D Graphics and Games, April 2005.)
This thesis by Robert Strzodka describes the design of robust quantized schemes and their hardware efficient implementation on data-stream-based architectures for PDE-based image processing. The focus lies on enhancing both performance and accuracy by an efficient use of appropriate hardware resources. Quantized schemes which, despite roundoff errors, preserve the qualitative behavior of the continuous models are constructed, and examined on different GPUs, a FPGA and a reconfigurable array processor. The pros and cons of the hardware designs and the memory gap problem are discussed in detail. (Hardware Efficient PDE Solvers in Quantized Image Processing. Robert Strzodka. PhD thesis, University of Duisburg-Essen, 2004.)
This paper by Rao et al. at UNC Charlotte describes an algorithm to track human limbs at interactive rates without using markers. 3d point cloud data is derived from a modified visual hull algorithm. This data is fed into a particle filtering algorithm that runs on the GPU. The tracking system runs at interactive rates. (Interactive marker-less tracking of human limbs. Rao S., Hodges L.F to be submitted to Transactions on Visualization and Computer Graphics.)
This paper by Jansen et al. describes how to utilize current commodity graphics hardware to perform Fourier volume rendering directly on the GPU. The paper presents a novel implementation of the Fast Fourier Transform: This Split-Stream-FFT maps the recursive structure of the FFT to the GPU in an efficient way. Additionally, high-quality resampling within the frequency domain is discussed. The implementation enables visualization of large volumetric data sets at interactive frame rates on a mid-range computer system. (Fourier Volume Rendering on the GPU Using a Split-Stream FFT)
A second GPU Gems 2 sample chapter, Streaming Architectures and Technology Trends (Chapter 29), by John Owens is now available. The first sample chapter Per-Pixel Displacement with Distance Functions (Chapter 8), was released last week.
A Special Issue of the Elsevier Journal “Simulation Practice and Theory” about Programmable Graphics Hardware is planned for 2005. Authors of papers which explore simulation studies and algorithms utilizing graphics hardware are invited to participate in the special issue. Please see the Call for Papers for more information.