Program
Following is the tentative program for HPG 2016:
Sunday, June 19 |
|
| 6:00-9:00 PM | HPG 2016 – Welcome drink and registration (Odessa Club & Restaurant) |
Monday, June 20 |
|
| 9:30-11:00 | Registration and breakfast (Hamilton Building – Ground Floor Concourse) |
| 11:00-11:15 | Opening remarks (Hamilton Building – Joly Theatre) |
| 11:15-12:30 | Papers: Hidden surfaces (Hamilton Building – Joly Theatre) |
| Exploring and Expanding the Continuum of OIT Algorithms Chris Wyman |
|
| SVGPU: Real Time 3D Rendering to Vector Graphics Formats Apollo Ellis, Warren Hunt, John Hart |
|
| Masked Software Occlusion Culling Jon Hasselgren, Magnus Andersson, Tomas Akenine-Möller |
|
| 12:30-2:00 | Lunch (O’Reilly Institute – Foyer) |
| 2:00-3:15 | Papers: Better BVHs (Hamilton Building – Joly Theatre) |
| Watertight Ray Traversal with Reduced Precision and Bounding Plane Reuse Karthik Vaidyanathan, Tomas Akenine-Möller, Marco Salvi |
|
| Efficient Stackless Hierarchy Traversal on GPUs with Backtracking in Constant Time Nikolaus Binder, Alexander Keller |
|
| Bandwidth-Efficient BVH Layout for Incremental Hardware Traversal Gabor Liktor, Karthik Vaidyanathan |
|
| 3:15-3:45 | Afternoon Break (O’Reilly Institute – Foyer) |
| 3:45-5:00 | Panel (Hamilton Building – Joly Theatre) |
| 7:00-11:00 | Whiskey Tasting (Irish Whiskey Museum Dublin) |
Tuesday, June 21 |
|
| 9:00-9:45 | Registration and Breakfast (Hamilton Building – Ground Floor Concourse) |
| 9:45-10:45 | Keynote: Bryan Catanzaro (Hamilton Building – Joly Theatre) Scaling Deep Learning View Abstract / Bio During the past few years, Deep Learning has made significant progress towards solving many difficult Artificial Intelligence tasks. Although the techniques behind deep learning have been studied for decades, they rely on large datasets and large computational resources, and so have only recently become practically applicable. Training deep neural networks is very computationally intensive: training one model takes tens of exaflops of work. The more models we train, the more hypotheses we can evaluate about how to solve our problems, and the more research progress we can make. Accordingly, we care a great deal about reducing training time – so High Performance Computing is central to our work. Once we have a good model, we deploy it to users, which is also a computationally intensive problem. Therefore, throughput oriented processors and associated programming models are central to the current and future success of deep learning. In this talk, I will discuss the use of GPUs for training and deploying deep learning models. I’ll talk about the directions I think deep learning is leading GPU hardware and programming models. Bio Bryan is a research scientist at Baidu, working with Adam Coates and Andrew Ng to create next generation systems for training and deploying deep learning. He earned his PhD from Berkeley, under the direction of Kurt Keutzer, where he built the Copperhead language and compiler, which allows Python programmers to use nested data parallel abstractions and gain high efficiency on contemporary parallel platforms. He earned his MS and BS from Brigham Young University, where he worked with Brent Nelson on higher radix floating-point representations for FPGAs. |
| 10:45-11:15 | Morning break (Hamilton Building – Ground Floor Concourse) |
| 11:15-12:30 | Papers: Fast GI |
| DIRT: Deferred Image-based Ray Tracing Konstantinos Vardis, Andreas-Alexandros Vasilakis, Georgios Papaioannou |
|
| Photon Splatting Using a View-Sample Cluster Hierarchy Pierre Moreau, Erik Sintorn, Viktor Kämpe, Ulf Assarsson, Michael Doggett |
|
| Deep G-Buffers for Stable Global Illumination Approximation Michael Mara, Morgan McGuire, David Luebke |
|
| 12:30-2:00 | Lunch (Hamilton Building – Ground Floor Concourse) |
| 2:00-3:40 | Papers: Ray tracing (Hamilton Building – Joly Theatre) |
| Lightcuts Interpolation Hauke Rehfeld, Carsten Dachsbacher |
|
| GVDB: Raytracing Sparse Voxel Database Structures on the GPU Rama Hoetzlein |
|
| Local Shading Coherence Extraction for SIMD-Efficient Path Tracing on CPUs Attila Áfra, Carsten Benthin, Ingo Wald, Jacob Munkberg |
|
| Adaptive Sampling for On-The-Fly Ray Casting of Particle-based Fluids Hendrik Hochstetter, Jens Orthmann, Andreas Kolb |
|
| 3:40-4:10 | Afternoon break (Hamilton Building – Ground Floor Concourse) |
| 4:10-5:40 | Hot3D (Hamilton Building – Joly Theatre) |
| 7:00-11:00 | HPG Conference Dinner (Trinity College Dining Hall) |
Wednesday, June 22 |
|
| 9:00-9:30 | Registration and Breakfast (Hamilton Building – Ground Floor Concourse) |
| 9:30-10:20 | Papers: Textures and Shading (Hamilton Building – Joly Theatre) |
| Infinite Resolution Textures Alexander Reshetov, David Luebke |
|
| Filtering Distributions of Normals for Shading Antialiasing Anton S. Kaplanyan, Stephen Hill, Anjul Patney, Aaron Lefohn |
|
| 10:20-11:00 | Morning break (Hamilton Building – Ground Floor Concourse) |
| 11:00-11:50 | Papers: VR and GPU Compute (Hamilton Building – Joly Theatre) |
| Comparison of Projection Methods for Rendering Virtual Reality Robert Toth, Tomas Akenine-Möller, Jim Nilsson |
|
| A Fast, Massively Parallel Solver for Large, Irregular Pairwise Markov Random Fields Daniel Thuerck, Michael Waechter, Sven Widmer, Max von Buelow, Patrick Seemann, Marc Pfetsch, Michael Goesele |
|
| 11:50-12:15 | Break (Hamilton Building – MacNeill Theatre) |
| 12:15-12:45 | Wrap up (Hamilton Building – MacNeill Theatre) |
| 1:00-2:00 | Lunch / HPG town hall (O’Reilly Institute – Foyer) |
| 2:00-3:00 | Keynote: Markus Gross (joint w/EGSR, Hamilton Building – MacNeill Theatre) View Abstract / Bio Abstract: The Technology to Create the Magic Disney Research was launched in 2008 as a network of research laboratories that collaborate closely with academic institutions such as the Swiss Federal Institute of Technology in Zurich and Carnegie Mellon University. Its mission is to push the frontiers of technology in areas relevant to Disney’s creative entertainment businesses. Disney Research develops innovations for Parks, Film, Animation, Television, Games, and Consumer Products. Research areas include video and animation technologies, postproduction and special effects, digital fabrication, robotics, and much more. This talk gives an overview of Disney Research spiced with some examples of our latest and greatest inventions. The focus is on the collaboration between ETH Zurich and the Walt Disney Company displaying the synergies arising from this program. This talk will highlight a company perspective as well as a view from the academic angle. Bio Markus Gross is a Professor of Computer Science at the Swiss Federal Institute of Technology Zürich (ETH), head of the Computer Graphics Laboratory, and the Director of Disney Research, Zürich. He joined the ETH Computer Science faculty in 1994. His research interests include physically based modeling, computer animation, immersive displays, and video technology. Before joining Disney, Gross was director of the Institute of Computational Sciences at ETH. He received a master of science in electrical and computer engineering and a PhD in computer graphics and image analysis, both from Saarland University in Germany in 1986 and 1989. Gross serves on the boards of numerous international research institutes, societies, and governmental organizations. He received the Technical Achievement Award from EUROGRAPHICS in 2010, the Swiss ICT Champions Award in 2011 and the IEEE Visualization Technical Achievement Award in 2015. He is a fellow of the ACM and of the EUROGRAPHICS Association and a member of the German Academy of Sciences Leopoldina as well as the Berlin-Brandenburg Academy of Sciences and Humanities. In 2013 he received a Technical Achievement Award from the Academy of Motion Picture Arts and Sciences, the Konrad Zuse Medal of GI and the Karl Heinz Beckurts price. He cofounded Cyfex AG, Novodex AG, LiberoVision AG, Dybuster AG and Gimalon AG. |
| 7:00-11:00 PM | Joint Social Event – HPG, MAM, and EGSR (The chq Building) |
Questions? Click here for help in contacting us.