GPGPU |
General-Purpose Computation Using Graphics Hardware
|
IntroductionGPGPU stands for General-Purpose computation on GPUs. With the increasing programmability of commodity graphics processing units (GPUs), these chips are capable of performing more than the specific graphics computations for which they were designed. They are now capable coprocessors, and their high speed makes them useful for a variety of applications. The goal of this page is to catalog the current and historical use of GPUs for general-purpose computation.
|
Dissertation: Ray Tracing on a Stream Processor This dissertation by Tim Purcell of Stanford University discusses several topics relevant to GPGPU including a stream processor abstraction for GPUs, and GPU-based ray tracing and photon mapping algorithms. Much of this work has been reported on GPGPU before, but the description of the ray tracing work in particular is expanded and updated from previous papers with details about the Radeon 9700 ray tracer demonstrated at Siggraph 2002. Included on the web page are links to the dissertation defense talk slides and movies of the various demos. (Ray Tracing on a Stream Processor, Timothy J. Purcell, Ph.D. Dissertation, March 2004.)
Posted: 15 Mar 2004 [GPGPU /Advanced Rendering/Global Illumination] # Generalized Distance Transforms and Skeletons in Graphics Hardware This paper presents the computation of the feature distance transform (FDT) with an arbitrary distance function and the corresponding 2D skeleton or Voronoi diagram in DX8 or DX9 graphics hardware. Together with a direct, non-iterative distance measurement, the FDT enables us to compute a simply connected pixel-exact skeleton, which can be continously pruned with a regularization parameter. Real-time distance minimization in the zooming process delivers even sub-pixel accuracy. There are no restrictions on the distance function and additive and multiplicative weights can be applied. An adaptive tiling scheme delivers a ten-fold performance increase over a software or simple hardware implementation. (Generalized Distance Transforms and Skeletons in Graphics Hardware. Robert Strzodka and Alexandru Telea in Proceedings VisSym'04, to appear, 2004)
Posted: 15 Mar 2004 [GPGPU /Advanced Rendering/Image-Based Modeling & Rendering] # This paper presents an implementation of the Generalized Hough Transform (GHT) in DX8 graphics hardware. Given the 3D geometry of an object, the GHT is used to determine its pose, scale and position in an uncalibrated image. Without any a-priori knowledge about the image many different poses and scales must be tested. The implementation achieves a considerable speedup by increasing the operation count in favor of a data stream processing of the otherwise irregular memory access pattern of the GHT. The additional operations are used to regularize the problem, decreasing the number of the required candidate poses. (A Graphics Hardware Implementation of the Generalized Hough Transform for fast Object Recognition, Scale, and 3D Pose Detection. Robert Strzodka, Ivo Ihrke and Marcus Magnor in Proceedings ICIAP 2003, pp. 188-193, 2003.)
Posted: 15 Mar 2004 [GPGPU /Advanced Rendering/Image-Based Modeling & Rendering] # Streaming Geometric Optimization using Graphics Hardware This paper proposes algorithms for computing extent measures and approximate representations of a stream of points in R2 or R3. In particular, we study the problems of computing various extent measures (for example diameter, width, smallest enclosing rectangle, and smallest enclosing disk) and of approximating a set of points by a circle or a line. We show that these problems can be solved efficiently using graphics hardware even in the streaming model.
(P. Agarwal, S. Krishnan, N. Mustafa and S. Venkatasubramanian. Streaming Geometric Optimization Using Graphics Hardware. Proc. 11th European Symposium on Algorithms, Sep 2003.)
Posted: 15 Mar 2004 [GPGPU /Computational Geometry] # During the Advanced OpenGL Tutorial at the 2004 Game Developers Conference in San Jose, California, Mark Harris of NVIDIA will give a short talk on GPGPU for games. The OpenGL tutorial will be held Tuesday, March 23 from 10am until 6pm. Slides for this talk, "GPGPU : Beyond Graphics", as well as other talks from the OpenGL Tutorial are available at this link.
Posted: 15 Mar 2004 [GPGPU /Miscellaneous/Courses] # |
Categories
|
