program + presentations

Most of the presentations have their slides posted below. With one or two exceptions we expect to havee all papers, posters and special presentations posted within a week of the conference. Check back soon if those you desire are not yet online.

Thursday, 24 June 2010

15:00-18:00: Onsite Registration

Friday, 25 June 2010

08:00: Onsite Registration

10:30-11:00: Opening Remarks

11:00-11:50: Micropolygons I

  • Hardware Implementation of Micropolygon Rasterization with Motion and Defocus Blur
    John Brunhaver, Kayvon Fatahalian, Pat Hanrahan
    (Slides)
  • Space-Time Hierarchical Occlusion Culling for Micropolygon Rendering with Motion Blur
    Solomon Boulos, Edward Luong, Kayvon Fatahalian, Henry Moreton, Pat Hanrahan
    (Slides)

11:50-14:00: Lunch

14:00-14:50: Micropolygons II

  • A Lazy Object-Space Shading Architecture With Decoupled Sampling
    Christopher A. Burns, Kayvon Fatahalian, William R. Mark
    (Slides)
  • Task Management for Irregular-Parallel Workloads on the GPU
    Stanley Tzeng, Anjul Patney, John Owens
    (Slides)

14:50-15:15: Posters fast forward

15:15-15:45: Break

15:45-17:00: Rendering with Volumes

  • Real Time Volumetric Shadows using Polygonal Light Volumes
    Markus Billeter, Erik Sintorn, Ulf Assarsson
    (Slides | Movie)
  • Ambient Occlusion Volumes
    Morgan McGuire
  • (Slides)
  • Large Data Visualization on Distributed Memory Multi-GPU Clusters
    Thomas Fogal, Hank Childs, Siddharth Shankar, Jens Krueger, Dan Bergeron, Philip J. Hatcher
  • (Slides)

Saturday, 26 June 2010

08:30-10:00: Keynote I, Turner Whitted (Microsoft Research)

  • Title: Disaggregated Graphics: Rich Clients for Clouds
  • Abstract: We sometimes forget that the famous "wheel of reincarnation" translates as it rotates, transporting us to unfamiliar technological territory even if we recognize historical similarities. So it is with the emergence of cloud computing with its concentrated computation and wide bandwidth interconnection. It is not, however, a return to the mainframe computer centers of the 1960s or the client/server model of the 1980s. Instead we are offered more computation, more pixels, more modes of interaction, .... more of everything. We are given so much more that the change of experience is qualitative, not merely quantitative.
    Microsoft Research's vX project is an experiment devised to explore the client side of this new computing environment. Radically rich visual computing calls for radically new architectures, programming models, and approaches to interaction. We are re-examining these venues simultaneously rather than independently. At the highest level, the vX model insists on interaction being local even if it is shared among a heterogeneous collection of devices. At a lower level, the vX programming model emphasizes local memory access within each of many processing cores. This philosophy extends to the lowest level of the graphics engine with memory intensive passive representations being replaced with processor intensive functional representations.
    As we progress with this project we find our alignment with technological trends for processing and interconnection takes us far from conventional graphics practice. This should be no surprise. It is time for change.
  • Turner Whitted's Biography: As a researcher and former manager at Microsoft Research, Turner Whitted has explored topics in hardware devices, HCI, and computer graphics. He was a member of the computer science faculty at the University of North Carolina at Chapel Hill from 1983 until 2001 as well as a cofounder and director of Numerical Design Limited. Prior to that he was a member of the technical staff in Bell Labs' computer systems research laboratory where he introduced the notion of using recursive ray tracing to implement global illumination. He earned BSE and MS degrees from Duke University and a PhD from North Carolina State University, all in electrical engineering. In the past he has served on the editorial boards of IEEE Computer Graphics and Applications and ACM Transactions on Graphics, and was papers chair for SIGGRAPH 97. He is an ACM Fellow and a member of the National Academy of Engineering.

10:00-10:30: Break

10:30-11:45: Ray Tracing I

  • Edge Avoiding A-Trous Wavelet Transform for fast Global Illumination Filtering
    Holger Dammertz, Daniel Sewtz, Johannes Hanika, Hendrik P. Lensch
    (Slides)
  • Parallel SAH k-D Tree Construction
    Byn Choi, Rakesh Komuravelli, Victor Lu, Hyojin Sung, Robert L. Bocchino, Sarita V. Adve, John C. Hart
    (Slides)
  • HLBVH: Hierarchical LBVH Construction for Real-Time Ray Tracing of Dynamic Geometry
    Jacopo Pantaleoni, David Luebke
    (Slides)

11:45-13:45: Lunch

13:45-15:00: Ray Tracing II

  • AnySL: Efficient and Portable Shading for Ray Tracing
    Ralf Karrenberg, Dmitri Rubinstein, Philipp Slusallek, Seabastian Hack
    (Slides)
  • Restart Trail for Stackless BVH Traversal
    Samuli Laine
    (Slides)
  • Architecture Considerations for Tracing Incoherent Rays
    Timo Aila, Tero Karras
    (Slides)

15:00-15:40: Poster & live demo session

  • Posters
    • PixelStrom: FPGA-Based Image Combining for Parallel Graphics Systems
      Jens Maiero, Andre Hinkenjann, Marko Winzker and Matthias Bues
    • (Poster)
    • OpenBRDF - An Open Toolkit for Reflection and Transmittance Distribution Functions
      Sebastian Alberternst, Manfred Brill, and Markus Laininger
      (Poster)
    • Toward Stream Filtered Ray Tracing on a Digital Signal Processor
      Kevin Bensema, Jesse Porch, Jared Heinly, Shawn Recker, and Christiaan Gribble
      (Poster)
    • Hybrid AABB-Frustum Culling to Accelerate Ray-Primitive Intersection Tests
      Jae-Ho Nah, Woo-Chan Park, Yoon-Sig Kang, and Tack-don Han
    • Brigade: Towards Real-time Path Tracing in Games
      Jacco Bikker
      (Poster)
    • Bridging Ray and Raster Processing on GPUs
      Kenny Mitchell, Christian Oberholzer, Jan Kautz and Peter-Pike Sloan
      (Poster)
    • Collection and Organization of Facts Surrounding the History of Graphics Hardware
      Josh Steinhurst
      (Poster)
    • Binned SAH Kd-Tree Construction on a GPU
      Piotr Danilewski, Stefan Popov, and Philipp Slusallek
    • (Poster)
    • Dynamic Thread Creation for Improving Ray-Tracing
      Mike Steffen and Joseph Zambreno
      (Poster)
    • Real-Time Ray-Tracing for Physically Simulated Rigid-Body Scenes
      Colin Fowler, Michael Manzke, and Steven Collins
    • (Poster)
    • Collision Detection Hardware Optimised for Ray-Tracers
      Muiris Woulfe, Michael Doyle, and Michael Manzke
      (Poster)
  • Live Demos
    • A Work-Efficient GPU Algorithm for Level Set Segmentation
      Mike Roberts, Jeff Packer, Mario Costa Sousa, and Joseph Ross Mitchell
      (Poster)
    • Real Time Volumetric Shadows using Polygonal Light Volumes
      Markus Billeter, Erik Sintorn, and Ulf Assarsson
    • AnySL: Efficient and Portable Shading for Ray Tracing
      Ralf Karrenberg, Dmitri Rubinstein, Philipp Slusallek, Sebastian Hack
    • (Poster)

15:40-17:00: Hot3D

  • ATI Radeon HD5000 Series : An Inside View (Mark Fowler, AMD)
    An overview and discussion of the internal architecture of the Radeon HD 5000 series of products, the first GPUs to deliver full support of dx11 and the first to deliver support in all market segments. This discussion will get into the internals of the GPU shader core and its implementation of the dx11 compute shader and OpenCL.
    (Slides)
  • Mali 400 MP: A Scalable GPU for Mobile and Embedded Devices (Tom Olson, ARM)
    Mali 400 MP is the world's first OpenGL ES 2.0 conformant multi-core GPU. Its scalable design allows ARM silicon partners to address markets ranging from mobile phones and navigation systems to HDTVs and smartbooks, and preserves ARM's traditional lead in power and bandwidth efficiency.
    (Slides)
  • Fast Tessellated Rendering on Fermi GF100 (Tim Purcell, NVIDIA)
    This presentation will describe the motivation for high-speed tessellated rendering, give details on Fermi GF100's architectural innovations (in particular, how the Fermi pipeline differs from the logical pipeline of our prior architectures), will present comparative performance data, and will conclude with a demonstration of GF100 performing tessellated rendering on synthetic scenes and scenes from current DX11 applications.
    (Slides)

17:00: HPG social event

Sunday, 27 June 2010

08:30-10:00: Keynote II, Cevat Yerli and Anton Kaplanyan (Crytek)

  • Title: Crytek's Future Game Graphics
  • Abstract: We want to share our ten-year expertise of making a generalized and balanced real-time rendering pipeline on consoles. Different algorithms for image synthesis will be discussed as well as different architectures for different workloads. The problems of the current rendering pipeline and the current generation of consoles will be discussed. Also we will talk about the new possible applications for real-time graphics such as movies industry and server-side rendering.
  • Cevat Yerli's Biography: President & CEO of Crytek. Cevat's first games and development experiences go back to the 1980s with the Commodore 64 and the Schneider CPC 6128, where he worked on simulation games. His passion has always been creating and playing games. While studying economics, he began working towards his dream of founding a game development company. The dream became reality in 1999 when he founded Crytek with his two brothers. Cevat gives creative direction for all Crytek products.
  • Anton Kaplanyan's Biography: Anton Kaplanyan is a Lead Researcher at Crytek. During the development of CryEngine 3 he was responsible for multiple researches on graphics and porting of CryEngine 2 to the current generation of consoles. Currently he is busy working on the next iteration of the engine to keep pushing both DX11 and next-gen console technology. Additionally he is working on his PhD within Stuttgart University. Prior to joining Crytek he received his M.S. in Computer Science at Moscow University of Electronic Engineering, Russia in early 2007.
  • (Slides)

10:00-10:30: Break

10:30-11:45: GPU Algorithms

  • A Work-Efficient GPU Algorithm for Level Set Segmentation
    Mike Roberts, Jeff Packer, Mario Costa Sousa, Joseph Ross Mitchell
    (Slides)
  • Random Numbers and Noise via the Tiny Encryption Algorithm
    Fahad Zafar, Marc Olano, Aaron Curtis
    (Slides)
  • Texture Compression of Light Maps using Smooth Profile Functions
    Jim Rasmusson, Jacob Strom, Per Wennersten, Michael Doggett and Tomas Akenine-Möller
    (Slides)

11:45-13:45: Lunch

13:45-15:00: Surfaces & Rasterization

  • Efficient Bounding of Displaced Bezier Patches
    Jacob Munkberg, Jon Hasselgren, Robert Toth, Tomas Akenine-Möller
    (Slides)
  • Analytical Motion Blur Rasterization with Compression
    Carl Johan Gribel, Michael Doggett, Tomas Akenine-Moller
    (Slides)
  • Real-time stochastic rasterization on conventional GPU architectures
    Morgan McGuire, Eric Enderton, Peter Shirley, David Luebke
    (Slides)

15:00-15:30: Break

15:30-16:40: Panel

16:40-17:00: Best paper presentation and Closing Remarks

17:15: Joint HPG/EGSR wine & cheese reception