This Medical Image Computing and Computer Assisted Intervention (MICCAI) 2003 paper by Lefohn et al. describes a brain tumor segmentation study performed with a new GPU-based level-set solver. This paper demonstrates that the ability to interact with a level-set computation in real time enables users to quickly produce segmentations from MRI data that qualitatively and quantitatively compare favorably with expert hand-segmentations. (Interactive, GPU-Based Level Sets for 3D Brain Tumor Segmentation. Aaron E. Lefohn, Joshua E. Cates and Ross T. Whitaker. To appear at “Medical Image Computing and Computer Assisted Intervention,” (MICCAI) 2003.)

This paper describes a framework for the implementation of linear algebra operators on GPUs, providing the building blocks for the design of more complex numerical algorithms. The framework takes advantage of sparse and banded matrices in particular. The paper demonstrates the approach by implementing direct solvers for sparse matrices with application to multi-dimensional finite difference equations, i.e. the 2D wave equation and the incompressible Navier-Stokes equations. (Linear Algebra Operators for GPU Implementation of Numerical Algorithms. Jens Krüger and Rüdiger Westermann. To appear in the proceedings of SIGGRAPH 2003.)

This EGSR 2003 paper by Goodnight et al. demonstrates the implementation of a real-time tone mapping algorithm on programmable graphics hardware. This allows interactive applications to achieve higher levels of realism by rendering with physically based, unclamped lighting values and high dynamic range texture maps. (Interactive Time-Dependent Tone Mapping Using Programmable Graphics Hardware. In Proceedings of Eurographics Symposium on Rendering 2003.)

GPGPU now has it’s own WWW domain! We have registered gpgpu.org and gpgpu.com. We plan to add new features to GPGPU in the future. Possible additions include a discussion forum, benchmark and utility code repository, and more. To start, we’ve made the site searchable (use the search box on the right). We received much positive feedback during SIGGRAPH, Graphics Hardware, and NVIDIA-U last week. We appreciate all suggestions, contributions, and news items. Thanks!

This GH 2003 paper by Kenneth Moreland and Edward Angel describes how to use programmable GPUs to perform the fast Fourier transform (FFT). Their system that can synthesize an image by conventional means, perform the FFT, filter the image, and finally apply the inverse FFT in well under 1 second for a 512 by 512 image. (The FFT on a GPU. Kenneth Moreland and Edward Angel. Graphics Hardware 2003.)

This report by Coombe et al. describes a technique for computing radiosity, including an adaptive subdivision of the model, using graphics hardware. The technique uses floating point textures and fragment programs to perform progressive refinement using a novel implementation of hemicube radiosity on the GPU. (Radiosity on Graphics Hardware. Greg Coombe, Mark J. Harris, Anselmo Lastra. UNC TR03-020. June, 2003.)

Sh is a metaprogramming language for programmable GPUs, developed at the University of Waterloo. From the “About Sh” page: “A high-level language allows programming GPUs with familiar constructs and syntax, without worrying about the details of the hardware. Sh is such a high-level language. It offers the convenient syntax of C++ and takes the burden of register allocation and other low-level issues away from the programmer. This allows GPU programs to be written much quicker and makes porting such programs extremely simple.” Sh is an open-source project hosted on SourceForge http://libsh.sourceforge.net.

This paper by Li et al. proposes a method for reconstruction and rendering visual hulls using programmable graphics hardware. The visual hull is an approximate 3D geometry representation that is reconstructed from multiple silhouette images. Projective texture mapping and alpha channel modulation are exploited to reconstruct visual hulls implicitly in image-space. View-dependent textures can be applied in the same rendering pass. The performance of this approach is significantly faster than that of previously reported similar systems. (Hardware-accelerated Visual Hull Reconstruction and Rendering. In Proceedings of Graphics Interface 2003).

This paper from the upcoming book ShaderX 2 Programming by Ádám Moravánszky gives a detailed description of implementing dense matrix operations on programmable GPUs. Matrix multiplication is applied to solving linear systems of equations and the linear complementarity problem, which can in turn be used to simulate soft body and rigid body physics. The performance of the GPU implementation is compared to the SSE2 optimized ATLAS library running on the CPU. DirectX 9 pixel and vertex shader programs are provided. (Dense Matrix Algebra on the GPU. Ádám Moravánszky. To appear in ShaderX 2 Programming, Wordware, 2003.)

This article at ExtremeTech.com describes David Kirk’s vision of the future of 3D computer graphics and graphics hardware. The article links to the GPGPU website, and is based on a presentation by David Kirk that describes some recent GPGPU research. Because of this, GPGPU was mentioned in a Slashdot post and comments. The combination of links resulted in over 12,000 site visits in under 4 days (that’s over 25% of all visits to this page, ever.)