 Hierarchical Stochastic Motion Blur Rasterization by the Intel Advanced Rendering Team We present a hierarchical traversal algorithm for stochastic rasterization of motion blur, which efficiently reduces the number of inside tests needed to resolve spatio-temporal visibility. Our method is based on novel tile against moving primitive tests that also provide temporal bounds for the overlap. The algorithm works entirely in homogeneous coordinates, supports MSAA, facilitates efficient hierarchical spatio-temporal occlusion culling, and handles typical game workloads with widely varying triangle sizes. |
| |
 An Optimizing Compiler for Automatic Shader Bounding by the Intel Advanced Rendering Team Programmable shading provides artistic control over materials and geometry, but the black box nature of shaders makes some rendering optimizations difficult to apply. In many cases, it is desirable to compute bounds of shaders in order to speed up rendering. A bounding shader can be automatically derived from the original shader by a compiler using interval analysis, but creating optimized interval arithmetic code is non-trivial. |
 Efficient Depth of Field Rasterization using a Tile Test based on Half-Space Culling by the Intel Advanced Rendering Team For depth of field rasterization, it is often desired to have an efficient tile versus triangle test, which can conservatively compute which samples on the lens that need to execute the inside-triangle test. We present a novel test for this, which is optimal in the sense that the region on the lens cannot be further reduced. Our test is based on removing half-space regions of the (u,v)-space on the lens from where the triangle to be rendered definitely cannot be seen through a tile of pixels. |
 Backface Culling for Motion Blur and Depth of Field by the Intel Advanced Rendering Team For triangles with linear vertex motion, common practice is to backface cull a triangle if it is backfacing at both the start and end of the motion. However, this is not conservative. We derive conservative tests that guarantee that a moving triangle is backfacing over an entire time interval and over the area of a lens. In addition, we present tests for the special cases of only motion blur and only depth of field. Our techniques apply to real-time and offline rendering, and to both stochastic point sampling and analytical visibility methods. |
 Efficient Bounding of Displaced Bézier Patches by the Intel Advanced Rendering Team In this paper, we present a new approach to conservative bounding of displaced Bézier patches. These surfaces are expected to be a common use case for tessellation in interactive and real-time rendering. Our algorithm combines efficient normal bounding techniques, min-max mipmap hierarchies and oriented bounding boxes. This results in substantially faster convergence for the bounding volumes of displaced surfaces, prior to tessellation and displacement shading. Our work can be used for different types of culling, ray tracing, and to sort higher order primitives in tiling architectures. |
 Automatic Pre-Tessellation Culling by the Intel Advanced Rendering Team Graphics processing units supporting tessellation of curved surfaces with displacement mapping exist today. Still, to our knowledge, culling only occurs after tessellation, i.e., after the base primitives have been tessellated into triangles. We introduce an algorithm for automatically computing tight positional and normal bounds on the fly for a base primitive. These bounds are derived from an arbitrary vertex shader program. The obtained bounds are used for backface, view frustum, and occlusion culling before tessellation. For highly tessellated scenes, we show that up to 80% of the vertex shader instructions can be avoided, which implies an "instruction speedup" of 5x. |
 Efficient Multi-View Ray Tracing Using Edge Detection and Shader Reuse We present a specialized algorithm for effcient multi-view image generation from a camera line using ray tracing, which builds on previous methods for multi-dimensional adaptive sampling and reconstruction of light fields. We introduce multiview silhouette edges to detect sharp geometrical discontinuities in the radiance function. These are used to signi cantly improve the quality of the reconstruction. In addition, we exploit shader coherence by computing analytical visibility between shading points and the camera line, and by sharing shading computations over the camera line. |
 Scalable Virtual Environments (Code Sample), Intel Labs Virtual environments have applications from gaming to disaster response training. However, today's approach typically limits each environment to running on only one server. Intel researchers have developed a new software architecture called the Distributed Scene Graph (DSG) that breaks the environment into separately executable components, which, when combined with a cloud computing model, allows applications to scale user experiences far beyond existing limits. |
 New Intel Center Driving the Future of Visual Computing, Intel Labs Intel is launching a new research center to advance the field of visual computing. The Intel Science and Technology Center for Visual Computing (ISTC-VC) will develop innovations in lifelike computer graphics, natural user interfaces, and realistic virtual humans that will make people's technology experiences more immersive in the future. The goal is to drive visual computing applications that look, act and feel real, and to make the technology broadly accessible to consumers. |
 Egocentric Affordance Fields in Pedestrian Steering, Mubbasir Kapadia, UCLA Research in the area of pedestrian simulation has seen a dramatic rise in recent years. With the potential for this work being realized in a wide variety of areas, ranging from urban planning and training simulations to games, life-like steering motions for each individual have become critical for a truly immersive and realistic experience. In this paper we propose a general framework for local path-planning and steering that can be easily extended to perform high-level behaviors. Our framework is based on the concept of affordances: the possible ways an agent can interact with its environment. |
 Adaptive Order Independent Transparency, Intel Advanced Rendering Team: We introduce a new technique for real-time order independent transparency called Adaptive Transparency (AT) that closely approximates the ground-truth results obtained from A-buffer compositing but is 5x to 40x faster. The key contribution of Adaptive Transparency is the introduction of an adaptively compressed visibility representation that can be efficiently constructed and queried during rendering. It can be applied to a widerange of transparent geometry (e.g., foliage, windows, hair, and smoke). AT closely matches the ground-truth A-buffer solution and is both higher quality and faster than other approximate OIT techniques. (presented at GDC 2011) |
 Adaptive Volumetric Shadow Maps, Intel Advanced Rendering Team: Adaptive volumetric shadow maps (AVSM) is an introduction to a real-time shadow algorithm that supports high-quality shadowing from dynamic volumetric media such as hair and smoke. AVSM includes a streaming simplification algorithm that generates an accurate volumetric light attenuation function using a small fixed memory footprint. (presented at Eurographics 2010 and SIGGRAPH 2010) |
 Deferred Rendering for Current and Future Rendering Pipelines, Intel Advanced Rendering Team: Deferred Rendering demonstrates a number of deferred rendering techniques including conventional deferred shading, deferred lighting and tile-based deferred shading. Tile-based techniques in particular use efficient user-space scheduling to apply per-sample shading to only the edge pixels that require it. (presented at SIGGRAPH 2010) |
 Sample Distribution Shadow Maps, Intel Advanced Rendering Team: Sample Distribution Shadow Maps (SDSMs) showcase an extension to Z-partitioning (cascaded shadow maps). SDSMs optimize the placement and size of a fixed number of Z-partitions by analyzing the shadow sample distribution required by the current frame and build on the advantages of current state of the art techniques, including predictable performance and constant memory usage, while removing tedious and ultimately suboptimal parameter tuning. (presented at SIGGRAPH 2010) |
 Footstep Navigation, Shawn Singh, UCLA Navigation is an essential ability of autonomous virtual characters. The standard approach is to model each character as a point-sized particle with a single collision radius, outputting a desired force or velocity vector to an animation system. In this project, we propose a more detailed, biomechanically inspired model of how a character can move, outputting footsteps to an animation system. Footstep-based navigation offers far more control and less ambiguity compared to the traditional vector-based navigation, and steps can be easily validated for correctness. The coarse-grain nature of footsteps makes it possible to plan sequences of steps, including foot orientation and timing, for hundreds of characters in real-time. |
 Experimental Cloud-based Ray Tracing Using Intel® MIC Architecture for Highly Parallel Visual Processing, Daniel Pohl, Intel Labs Ray tracing is a rendering method that has so far mostly been used in the professional graphics space like in the automotive or movie industry. It enables more accurate and more realistic images. Through offloading the performance-intensive work into a “cloud” consisting of Intel’s MIC cards those benefits can also be applied to a regular gaming machine like a small notebook. The paper looks at several benefits of using ray tracing in games and also has a discussion on the latencies that occur in a cloud-based gaming setup. |
 OpenSimulator Virtual World Server Case Study (part 1), Robert Adams, Intel Labs This is the first of several articles exploring some of the design, performance, and execution features of virtual world servers. An open source virtual world simulator is used as an example, and design requirements are deduced from its operation. This first article describes OpenSimulator's design, and future articles will explore specific facets such as workloads and power savings. |
 OpenSimulator Virtual World Server Case Study (part 2), Robert Adams, Intel Labs The previous article described the architecture of one open source virtual world server. This article shows how the architecture of that virtual world server affects the operation of various workloads. |
 OpenSimulator Virtual World Server Case Study (part 3), Robert Adams, Intel Labs Virtual world simulators advance the world state even when no one is connected to the simulation. This means that, unlike web servers, virtual world servers are always running and using power. In this article, we explore one potential method of power savings: lowering physics execution frequency to introduce more idle time. Reducing physics simulation frequency may introduce errors in the physics simulation, so we look at the tradeoffs between physics simulation correctness and platform power usage |
 Toward real-time path tracing in games, Jacco Bikker, NHTV Computer graphics for movies and high-quality animations are often rendered using a process called path tracing [1]. Using this rendering technique, single frames take minutes or even hours to render. However, path tracing manages to capture almost all physical phenomena, including indirect lighting and caustics, in a single, clean algorithm. We show that by applying algorithms borrowed from recent advances in real-time ray tracing, path tracing in real-time on consumer hardware is near. |
 Interactive Geometric Sound Propagation and Rendering, Micah Taylor, University of North Carolina We describe a novel algorithm and system for sound propagation and rendering in virtual environments and media applications. Our approach uses geometric propagation techniques for fast computation of propagation paths from a source to a listener and takes into account specular reflections, diffuse reflections, and edge diffraction. In order to perform fast path computation, we use a unified ray-based representation to efficiently trace discrete rays as well as volumetric ray-frusta. Furthermore, our propagation algorithm scales well with the number of cores, and uses interactive audio rendering technique to generate spatialized audio signals. The overall approach can render sound in dynamic scenes, allowing source, listener, and obstacle motion, and we show its performance on game-like and architectural environments. To the best of our knowledge, this is the first interactive sound rendering system that can perform plausible sound propagation and rendering in dynamic virtual environments.
|
 Accelerating polygonization of skeletal implicit surfaces on many-core architectures, Pourya Shirazian, University of Victoria Interactive design systems that use polygonization methods for rendering implicit surfaces have difficulty in responding interactively to user changes. To address this, we propose a novel approach for tessellating Blobtree skeletal implicit models using a divide and conquer strategy controlled by bounding volumes. Our algorithm leverages the parallel processing power available on Many-Core architectures. |
 Towards real-time C-arm reconstruction using multi-core CPUs, Hannes Hoffmann, Friedrich-Alexander University Three-dimensional (3-D) computed tomography (CT) imaging allows for more precise diagnosis of diseases, improved treatment planning and navigation, and insightful follow-up studies. C-arm CT systems, like the one shown in Figure 1 (left), acquire many 2-D planar projections at different angles around the patient. Today, the algorithm for reconstruction of the 3-D data, which is used in clinics most commonly, is filtered backprojection. Simply speaking, it smears all projection values back into the volume. Reconstruction is compute intense and also requires high memory bandwidth due to the huge data sets involved. Using C-arm systems, 3-D data is also available during interventions, enabling new medical applications. However, the OR environment requires real-time processing – including 2-D pre-processing, 3-D reconstruction and 3-D post-processing – of up to 60 fps. In our research we investigate the suitability of GPUs as well as multi-core servers for on-the-fly reconstruction of C-arm CT data. |
|
Ramon