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.
The upcoming GPU Gems 2 book is now in press and the first copies will be available at GDC 2005 in San Francisco. Several pieces of sample material from GPU Gems 2 have been released. These include a Visual Table of Contents (shown at right, which will be inside the front cover), the foreword (by Tim Sweeney, Epic Games), the preface, and the contributor biographies.
A first sample chapter, Per-Pixel Displacement Mapping with Distance Functions, has also been released. (GPU Gems 2 Homepage.)
The forums at GPGPU.org have proven quite popular. To keep them up to date with the latest web technology, we’ve added RSS syndication to the forums. This way, you can follow GPGPU discussions through any RSS-enabled “feed reader” (such as NetNewsWire, Feed Demon, or Firefox). Simply specify the URL http://www.gpgpu.org/forums/rss.php”. GPGPU.org has always had RSS available for the main news page.
We received news simultaneously from the developers of two new GPU ray tracers. Both projects are graduate-level thesis projects. One, called GPU-RT, is developed by Martin Christen and supports .3DS format meshes, multiple materials, and implements acceleration data structures. GPU-RT runs on NVIDIA GeForce 6 Series GPUs under D3D/HLSL and OpenGL/GLSL, and is available on SourceForge.net. The other project, “Ray Tracing on Programmable Graphics Hardware”, is by Filip Karlsson and Carl Johan Ljungstedt of Chalmers University of Technology. The thesis describes, among other things, how proximity clouds can be used to accelerate ray tracing on the GPU. (1. GPU-RT, Diploma Thesis by Martin Christen. 2. “Ray Tracing on Programmable Graphics Hardware”, Masters Thesis by Filip Karlsson and Carl Johan Ljungstedt.)
(Update 29 Jan 2005: Problems Solved.)
To our faithful readers and new visitors, please bear with us as we work out system problems after recent upgrades at our web host, ibiblio.org. We thank you for your patience. (While you wait: “Looks like somebody tried to cram-a-lam a swiss cake roll into the disk drive.”)
Suresh Venkatasubramanian will be teaching a GPGPU class at the University of Pennsylvania in Spring 2005. The class, titled “GPU Programming and Architecture” will focus on the stream programming abstraction of the GPU, and will cover the basic tools and techniques for designing and implementing algorithms for general purpose computations on the GPU. (UPenn GPGPU Course)