CSE 690: GPGPU Lecture 11: Projects, Papers

1
CSE 690 GPGPULecture 11 Projects, Papers

• Klaus Mueller
• Computer Science, Stony Brook University

2
General Graphics
3
Raytracing

• Pioneering paper on this topic (2002)
• Ray Tracing on Programmable Graphics Hardware
  by Purcell et al.
• Raytracing on a Stream Processor, Purcells PhD
  dissertation at Stanford

4
Raytracing

• GPU-Based Non-Linear Raytracing by Weiskopf et
  al.
• should also present the original paper by
  Groeller and the paper by Pharr Rendering
  Complex Scenes with Memory-Coherent Ray Tracing
• simulates rays undergoing forces in dynamical
  systems, astrophysics, black holes

5
Sub-Surface Scattering

• GPU Algorithms for Radiosity and Subsurface
  Scattering by Carr et al.
• also present original paper A practical model
  for subsurface light transport by Jensen at al.
• compare with CPU approach in Interactive
  Subsurface Scattering for Translucent Meshes by
  Hao et al.

6
Radiosity

• Radiosity on Graphics Hardware by Coombe et al.
  
• should also present Kellers Instant Radiosity
  paper
• and The Ray-Engine by Carr and Hart
• and Real-Time Global Illumination on GPU by
  Nijasure et al.

7
Photon Mapping

• A technique to achieve global illumination
• Photon Mapping on Programmable Graphics
  Hardware by Purcell et al.
• new paper Simulating Photon Mapping for
  Real-time Applications by Larsen et al.
• should also present the original photon mapping
  paper by Jensen et al.

8
Volume Rendering
9
Standard Volume Rendering

• Acceleration Techniques for GPU-based Volume
  Rendering by Krueger and Westermann
• present background on volume rendering
• earlier paper High-Quality Pre-Integrated Volume
  Rendering Using Hardware-Accelerated Pixel
  Shading by Engel et al.

10
Advanced Volume Rendering

• High-Quality Two-Level Volume Rendering of
  Segmented Data Sets on Consumer Graphics
  Hardware by Hadwiger et al.
• should also present the original paper on
  two-level volume rendering by Hauser et al.
• and Hardware-accelerated high-quality filtering
  on PC hardware by Hadwiger et al.

11
Unstructured Grids

• A Fast High Accuracy Volume Renderer for
  Unstructured Data by Moreland and Angel
• present background on unstructured grid rendering
  and summarize the highlights of these papers
• Hardware-based view-independent cell projection
  by Weiler et al.
• High-quality unstructured volume rendering on
  the PC platform by Guthe et al.
• Hardware-Assisted Visibility Sorting for
  Unstructured Volume Rendering by Callahan et
  al.

12
Signal and Image Processing
13
Transforms

• "The FFT on a GPU by Moreland and Angel
• GPU-based Frequency Domain Volume Rendering by
  Viola et al.
• also overview the original paper of Frequency
  domain volume rendering by Totsuka and Levoy
• The Discrete Wavelet Transform on a GPU by Wang
  et al.

14
Images, Textures, and Sound

• Fast and Accurate Color Image Processing Using
  3D Graphics Cards by Colantoni et al.
• GPU Image Inpainting via Texture Synthesis by
  Chong
• Tile-Based Texture Mapping on Graphics Hardware
  by Wei
• Computation of room acoustics using programmable
  video
  hardware by Jedrzejewski

15
Image-Based Flow Visualization

• 3D IBFV Hardware-Accelerated 3D Flow
  Visualization by Telea and van Wijk
• the first paper Image Based Flow Visualization
  by van Wijk
• GPU-Based 3D Texture Advection for the
  Visualization of Unsteady Flow Fields by
  Weiskopf and Ertl

16
Segmentation and Computer Vision
17
Segmentation with Level Sets

• Interactive Deformation and Visualization of
  Level Set Surfaces using Graphics Hardware by
  Lefohn 
• earlier segmentation paper
• Fast Volume Segmentation With Simultaneous
  Visualization Using Programmable Graphics
  Hardware by Sherbondy
• medical application paper GIST An Interactive,
  GPU-Based Level-Set Segmentation
  Tool for 3D Medical Images by Lefohn

18
System and API Issues

• A Streaming Narrow Band Algorithm Interactive
  Computation and Visualization of Level Sets by
  Lefohn
• touches on CPU-GPU load balancing and memory
  management
• Other papers
• Brook for GPUs Stream Computing on Graphics
  Hardware by Buch et al.
• Shader Algebra by McCool et al.
• Scout A Hardware-Accelerated System for
  Quantitatively Driven Visualization and Analysis
  by McCormick et al.

19
Image Analysis

• Generalized Distance Transforms and Skeletons in
  Graphics Hardware by Strzodka and Telea
• also summarize the Siggraph paper by Hoff
• A graphics hardware implementation of the
  Generalized Hough Transform for fast object
  recognition, scale, and 3D pose detection by
  Strzodka et al.

20
Computer Vision

• Computer Vision Signal Processing on Graphics
  Processing Units by Fung and Mann
• Mediated Reality Using Computer Graphics Hardware
  for Computer Vision by Fung et al.
• Using Multiple Graphics Cards as a General
  Purpose Parallel Computer Applications to
  Computer Vision by Fung and Mann
• The OpenVIDIA project Parallel GPU Computer
  Vision

21
Computer Vision (contd)
22
General Purpose Computing
23
General Scientific Computing

• SIMD Optimization of Linear Expressions for
  Programmable Graphics Hardware by Bajaj et al.
• General Mathematics in Graphics Hardware by
  Trendall and Steward
• Using Modern Graphics Architectures for
  General-Purpose Computing A Framework and
  Analysis by Thompson et al.
• GPU Cluster for High Performance Computing by
  Fan et al.
• Parallel Computing with Multiple GPUs on a
  Single Machine to Achieve Performance Gains by
  Gulde at al.

24
Visual Simulations

• Visual Simulation of Ice Crystal Growth by Kim
  and Lin
• Simulation of cloth (see nVidia demo)
• Benchmarking and Implementation
  ofProbability-Based Simulations on Programmable
  Graphics Cards by Tomov et al.

25
Medical Applications

• Accelerating Popular Tomographic Reconstruction
  Algorithms On Commodity PC Graphics Hardware by
  Xu and Mueller
• Ultra-Fast 3D Filtered Backprojection on
  Commodity Graphics Hardware by Xu and Mueller

26
Databases

• Fast Database Operations using Graphics
  Processors Govindaraju et al.

27
or suggest your own topic