SIGGRAPH 2010 By Focus
In SIGGRAPH 2010 Courses, attendees learn from the experts in the field and gain inside knowledge that is critical to career advancement. Courses are short (1.5 hours) or half-day (3.25 hours) structured sessions that often include elements of interactive demonstration, performance, or other imaginative approaches to teaching.
The spectrum of Courses ranges from an introduction to the foundations of computer graphics and interactive techniques for those new to the field to advanced instruction on the most current techniques and topics. Courses include core curricula taught by invited instructors as well as Courses selected from juried proposals.
James L. Mohler
SIGGRAPH 2010 Director for Education
Spectral Mesh Processing
Spectral mesh processing was proposed at the beginning of the 1990s to port the “signal processing toolbox” to 3D mesh models. Now, with recent advances in computing power and numerical software, this vision can be fully implemented. In this course, attendees learn how to transfer the underlying concepts to the mesh-model setting, how to implement the “spectral mesh processing” toolbox, and how to use it for real applications, including filtering, shape matching, remeshing, segmentation, and parameterization.
What is so Spectral?
Do Your Own Spectral Mesh Processing at Home?
Applications -- What Can We Do With It ?
Segmentation, Shape Retrieval, Non-Rigid Matching, Symmetry Detection
Applications -- What Can We Do With It ?
Zhang and Levy
Simon Fraser University
Processing for Visual Artists and Designers
Processing is a free language designed for artists, designers, musicians, educators, students, and anyone else who wants to create expressive, meaningful images, animations, and interactive graphics. Unlike most programming languages, it has been designed from the ground up for creating modern visual works appropriate for print, the web, and installations. With Processing, even people who are new to computers can create rich, exciting graphics that can be used in applications as diverse as interactive demonstrations, personally expressive art, interactive educational environments, gaming, procedural art, text manipulation, and animation.
Using Processing to create graphics and animations is intellectually and emotionally rewarding. Even beginners can create imagery that's wildly complex or serenely simple and elegant. It can move as though alive, or change in only the most subtle ways. Computers give us a whole new world of tools and media for communicating and expressing ourselves. When your intellectual programming mind and your intuitive image-making mind are purring together in harmony, the process is pure joy and the results are often beautiful. Best of all, Processing is free!
This course introduces the basic ideas of modern programming, but it does not focus on abstract theory. Every idea is illustrated, and Processing's value in making pictures and animations is made clear. The course provides all the tools, from context to motivation, required to continue the learning adventure and become a Processing master.
Welcome and Overview
Why Processing Is Useful and Cool
A Closet Full of Shoes: Variables
Rinse, Lather, Repeat: Loops and Routines
Smooth Moves: Using Curves
Say It Like You Mean It: Using Type
Building A Project, Start To Finish
Resources: Where To Go From Here
Coyote Wind Studios
Physically Based Shading Models in Film and Game Production
Physically grounded shading models have been known for many years, but they have only recently started to replace the "ad-hoc" models in common use for both film and game production. Compared to "ad-hoc" models, which require laborious tweaking to produce high-quality images, physically-based, energy-conserving shading models easily create materials that hold up under a variety of lighting environments. These advantages apply to both photorealistic and stylized scenes, and to game development as well as production of CG animation and computer VFX. Surprisingly, physically based models are not more difficult to implement or evaluate than the traditional "ad-hoc" ones.
This course begins with a short explanation of the physics of light-matter interaction and how it is expressed in simple shading models. Then several speakers discuss specific examples of how shading models have been used in film and game production. In each case, the advantages of the new models are demonstrated, and drawbacks or issues arising from their usage are discussed. The course also includes descriptions of specific production techniques related to physically based shading.
Background: The Physics of Shading
Practical Implementation of Physically Based Shading Models at tri-Ace
Crafting Physically Motivated Shading Models for Game Development
Terminators and Iron Men: Image-Based Lighting and Physical Shading at ILM
Faster Photorealism in Wonderland: Physically Based Shading and Lighting at Sony Pictures Imageworks
Gotanda, Hoffman, Martinez
Sony Pictures Imageworks
Industrial Light & Magic
Perceptually Motivated Graphics, Visualization, and 3D Displays
This course provides an overview of how knowledge of the human visual system (HVS) and perception are applied to several aspects of computer graphics, virtual environments, visualization, and 3D display technologies. It explains the role HVS and human perception play in optimization of rendering algorithms, display algorithms, virtual environments design, fidelity, and engineering. Example applications include real-time rendering, high-quality rendering, material editing using images, and training and knowledge transfer in virtual environments. The course also surveys recent research results presented at the ACM SIGGRAPH Symposium on Applied Perception in Graphics and Visualization and other conferences.
Perceptually Motivated 3D Displays and Depth Perception Banks
Perceptually Motivated Visualization
Perceptually Motivated Rendering
Perceptually Motivated Simulation & Virtual Environments
Leading Edge Research and APGV 2010
Mania and Banks
A Look to the Future
McNamara and Mania
Texas A&M University
Technical University of Crete
North Carolina State University
University of California, Berkeley
Image Statistics: From Data Collection to Applications in Graphics
Natural images exhibit statistical regularities that differentiate them from random collections of pixels, and the human visual system appears to have evolved to exploit such statistical regularities. Because computer graphics is about producing imagery for observation by humans, it is important to understand which statistical regularities occur in nature, so they can be emulated by image-synthesis methods. This course introduces all aspects of natural image statistics, ranging from data collection to analysis and their applications in computer graphics, computational photography, and image processing.
Data Collection and Calibration
First Order Statistics
Second Order Statistics
Higher Order Statistics
University of Bristol
University of Bristol
Brandenburgische Technische Universität
Build Your Own 3D Display
Film studios are now routinely producing live-action and animated 3D content for theatrical release. This advance is primarily enabled by widespread adoption of digital projection, which allows accurate view synchronization, but the underlying 3D display technologies have changed little in the last few decades. Theatrical systems rely on stereoscopic display: projecting unique images for the right and left eyes and separating the images with various filters in viewing glasses. But now several LCD manufacturers are introducing auto-multiscopic displays, which allow view-dependent imagery to be perceived without special glasses. 3D display is poised for another resurgence.
This course provides attendees with the mathematics, software, and practical details they need to build their own low-cost stereoscopic displays. Each new concept is illustrated using a practical 3D display implemented with off-the-shelf parts. First, the course explains glasses-bound stereoscopic displays and provides detailed plans for attendees to construct their own LCD shutter glasses. Then the course explains unencumbered auto-multiscopic displays, including step-by-step directions to construct lenticular and parallax-barrier designs using modified LCDs. All the necessary software, including algorithms for rendering and calibration, is provided for each example, so attendees can quickly construct 3D displays for their own educational, amusement, and research purposes.
The course concludes by describing various methods for capturing, rendering, and viewing various multi-view imagery sources: stereoscopic OpenGL support, methods for ray-tracing multi-view imagery with POV-Ray, and techniques for capturing live-action light fields.
Introduction: History and Physiology
Representation and Display
Glasses-Bound Stereoscopic Displays
Unencumbered Automultiscopic Displays
Source Material: Rendering and Capture
Questions & Answers
Hirsch and Lanman
MIT Media Lab
Stylized Rendering in Games
As they matured, the visual arts (painting, sculpture, photography, and architecture) all developed new visual-abstraction mechanisms to go beyond "realism" . Recent advances in visual effects have put film and games into this transitional state. In a sense, we're like artists at the end of the Renaissance: we've nearly mastered photorealism, but are only at the beginning of our discoveries about expression and perception.
Some film effects are subtle, like the color shifts and post-processing in Mirror's Edge. Others, such as the graphic-novel look of Prince of Persia, dominate the entire rendering style. In games, real-time and interactive constraints require more efficient and robust solutions than are employed elsewhere in computer graphics. And to be successful, a stylized renderer must integrate with appropriately stylized models, animation, and audio to form a coherent virtual world and ultimately enhance game play.
In this course, leading game developers candidly discuss the challenges of creating and implementing a stylized artistic vision for a game. Each speaker covers a specific game aspect, such as the art pipeline, rendering algorithms, art direction, and modeling.
Monday Night Combat
Chandana "Eka" Ekanayake
The Illustrative Rendering of Prince of Persia
Personalized Cool Characters in Brink
Style and Gameplay in the Mirror's Edge
Cartoon 3D for Battlefield Heroes
Making Concept Art Real for Borderlands
Thibault and Martel
Panel Discussion and Questions
Biomedical Applications: What You Need to Know
This course takes an in-depth look at the types of models and analysis tools researchers are developing for biomedical applications. These applications are motivated by the increasing ubiquity of 3D imaging devices, which enables frequent, high-quality images of structures that researchers use, for example, to track the effects of disease. Although simply visualizing the data is useful, quantitative measurements require development of higher-level models for measuring quantities such as area, density, and shape change.
This course covers the state of the art in this area, including: how and what kinds of models are created from images, what kinds of measurements can be done both from the images themselves and from intermediate models, and common issues that arise when dealing with imaging data.
Washington University in St. Louis
Virginia Polytechnic Institute and State University
National Center for Microscopy and Imaging Research, University of California, San Diego
Recent Advances in Real-Time Collision and Proximity Computations for Games and Simulations
It is quite challenging for beginners to keep up with all the published papers and evolving rendering systems for collision detection and other proximity queries. This course presents an overview of existing techniques, practical solutions, and how the field will change in the coming years. It explains recent developments designed to achieve interactive performance for large-scale rigid, articulated, deforming, and fracturing models in various applications. It also covers two popular physics libraries, Bullet and PhysX, and explains how they implement various proximity queries and can be used for various simulations.
Introduction to Collision and Proximity Queries
Proximity Queries for Rigid and Articulated Characters
Collision Detection for Deformable and Fracturing Models
GPU-Based Proximity Computations
Optimizing Proximity Queries for CPU, SPU and GPU
PhysX and Proximity Queries
Korea Advanced Institute of Science and Technology
University of North Carolina at Chapel Hill
Sony Computer Entertainment US R&D
Young J. Kim
Ewha Womans University
Importance Sampling for Production Rendering
Importance sampling provides a practical, production-proven method for integrating diffuse and glossy surface reflections with arbitrary image-based environment or area lighting constructs. Functions are evaluated at random points across a domain to produce an estimate of an integral. When using a large number of sample points, the method produces a very accurate result of the integral and provides a strong basis for simulating complex problems such as light transport.
Frequently, using the necessary number of samples to reach the exact result is too computationally expensive, so fewer samples are evaluated at the cost of visual noise, or variance, within the image. Importance sampling offers a means to reduce the variance by skewing the samples toward regions of the illumination integral that provide the most energy. For instance, the direction of specular reflection or a bright light source within an environment more likely represent the final value of the integral than a random sample.
The variance can be reduced more efficiently by combining multiple components of the illumination integral, such as the lighting and material function, to determine where to sample, which is the principle of multiple importance sampling (MIS). An alternative to the noise in importance sampling, filtered importance sampling (FIS), can provide fast integration, where the lighting environment look-ups are pre-filtered to give a smoother result with a significantly smaller number of samples.
Importance sampling, MIS, and FIS have various practical implications. This quarter-day course provides the background required for using Monte Carlo-based techniques for direct lighting and explains how visual-effects companies use these shading methods in their production pipelines.
Filtered Importance Sampling (FIS)
FIS for Area Lights in "A Christmas Carol"
Importance Sampling Framework at MPC
Multiple Importance Sampling (MIS)
Moving Picture Company
Color Enhancement and Rendering in Film and Game Production
Production of convincing and compelling scene representations on limited display media is a challenge common to painting, photography, film production, and computer graphics. The core of the problem is finding a transformation from the colors in the original scene to those in the final image. For almost 200 years, this transformation has been primarily determined by the chemical and optical properties of film, which have been carefully engineered for pleasing results (the "film look"). Digital color enhancement has vastly extended the variety of possible looks, but the "film look" remains the default baseline.
Despite its importance in film and game production, the transformation from scene-referred to display-referred colors (also called "rendering"; not to be confused with the more common computer graphics meaning of the term) is little-understood by many practitioners. This course covers the relevant theory, practical production methods, techniques and considerations relating to color enhancement, and rendering in both film and game production.
From Scene to Screen
Color Spaces and Operations
Color at Pixar: Ingredients for Creativity
The Craft of Color Grading
Filmic Tonemapping for Real-time Rendering
Film Simulation for Videogames
Color Enhancement for Videogames
Pixar Animation Studios
Lilliputian Pictures LLC
Sony Pictures Imageworks
An Introduction to 3D Spatial Interaction With Videogame Motion Controllers
Three-dimensional interfaces use motion sensing, physical inputs, and spatial-interaction techniques to effectively control highly dynamic virtual content. With the advent of the Nintendo Wii, the Sony EyeToy, and a host of soon-to-be-released peripherals such as the Playstation Motion Controller, Microsoft’s Natal, and Sixense’s TrueMotion, game developers, researchers, and hobbyists are challenged to create compelling interface techniques and game-play mechanics that make use of this technology. Researchers in the fields of virtual and augmented reality as well as 3D user interfaces have been working on 3D interaction for nearly two decades. The techniques, interaction styles, and metaphors developed in these communities are directly applicable to games that make use of motion-controller hardware.
This course demystifies the workings of current videogame motion controllers and provides a thorough overview of the techniques, strategies, and algorithms used in creating 3D interfaces for tasks such as 2D and 3D navigation, object selection and manipulation, gesture-based application control, and character control. It summarizes the strengths and limitations of various motion-control sensing technologies in today’s and soon-to-be-released peripherals, including accelerometers, gyroscopes, and 2D and 3D depth cameras. It also presents techniques for compensating for their deficiencies, including gesture recognition and non-isomorphic control-to-display mappings. Course materials include detailed notes and demonstration videos.
Welcome, Introduction & Roadmap
Common Tasks in 3D User Interfaces
3D Interfaces With 2D and 3D Cameras
Working With the Nintendo Wiimote
3D Spatial Interaction with the PlayStation Move
3D Gesture Recognition Techniques
University of Central Florida
Sony Computer Entertainment America
Volumetric Methods in Visual Effects
Computer-generated volumetric elements such as clouds, fire, and whitewater are becoming commonplace in movie production. The goal of this course is to familiarize attendees with the technology behind these effects. Experienced presenters who have authored proprietary and commercial volumetrics tools summarize the basics of the technology and explain the rationales behind drastically different development choices.
The course begins with a quick introduction to generating and rendering volumes, then presents a production-usable volumetrics toolkit, focusing on the feature set and why those features are desirable, and concludes with a survey of the specific tools developed at Double Negative, DreamWorks, Sony Pictures ImageWorks, Rhythm & Hues, and Side Effects Software. The production-system presentations focus on devleopment history, how the tools are used by artists, and the strengths and weaknesses of the software. Emphasis is on strategies for tackling efficient data structures, shading architecture, multi-threading and parallelization, holdouts, and motion blurring.
Basics of Volume Modeling & Rendering
Rhythm & Hues
Side Effects Software
Penney and Kontkanen
Clifford and Graham
Sony Pictures Imageworks
Nafees Bin Zafar
Sony Pictures Imageworks
Rhythm & Hues Studios
Side Effects Software Inc.
Fundamentals of Visual Analytics
For centuries, we have been improving our ability to collect and organize data, and this process is accelerating at an unprecedented rate. Unfortunately, our ability to analyze and generate information and knowledge from the data has not kept pace with their expanding volume. Over the past 30 years, the fields of visualization and information visualization have developed to help solve this problem. These active areas of research have led to development of helpful tools for decision makers, business operators, scientists, and engineers. However, gaining insight, supporting analysis, and making decisions based on these massive, disparate, uncertain, and rapidly evolving datasets requires more than just data visualization. So a new discipline has emerged. Visual analytics is the science of analytical reasoning facilitated by interactive visual interfaces. This course provides an introduction to the fundamentals of visual analytics, describes its core components, and summarizes the field's grand challenges.
Introduction, History, and Needs of Visual Analytics
Ebert and Thomas
Cognitive Science Basis of Visual Analysis and Decision Making
Decision Making in Visual Analytics
Visual Analytics Research Problems in Scientific Visualization
Data Representations, Transformations, and Statistics for Visual Reasoning
Database and Data Analysis Issues in Visual Analytics
Pacific Northwest National Laboratory
Barbara Gans Tversky
Advances in Real-Time Rendering in 3D Graphics and Games I
Advances in real-time graphics research and the ever-increasing power of mainstream GPUs and consoles continue to generate an explosion of innovative algorithms for fast, interactive rendering of complex and engaging virtual worlds. Every year, the latest video games display a vast new variety of sophisticated algorithms for ground-breaking 3D rendering that pushes the visual boundaries and interactive experience of rich environments.
This course is designed to encourage cross-pollination of knowledge for future games and other interactive applications. As the next installment in the now-established series of SIGGRAPH Courses on real-time rendering, it focuses on the best of graphics practices and research from the game-development community, and provides practical and production-proven algorithms. Course instructors include designers and producers from the makers of several award-winning games: Bungie, Naughty Dog, Crytek, DICE, AMD, Rockstar, and others. Topics include many advanced production secrets in addition to practical advice for implementing advanced techniques. Attendees will acquire several highly optimized algorithms in various areas of real-time rendering.
Introduction: Current Trends and Future Challenges in Real-Time Rendering for Games
Rendering Techniques in Toy Story 3
Ownby, Hall and Hall
A Real-Time Radiosity Architecture for Video Games
Einarsson and Martin
Real-Time Order Independent Transparency and Indirect Illumination Using Direct3D 11
Yang and McKee
CryENGINE 3: Reaching the Speed of Light
John Paul Ownby
Disney Interactive Avalanche Studio
Advanced Micro Devices, Inc.
Advanced Micro Devices, Inc.