AI-based talking-head videoconferencing systems reduce bandwidth by transmitting a latent representation of a speaker’s pose and expression, which is used to synthesize frames on the receiver's end. However, these systems are vulnerable to “puppeteering” attacks, where an adversary controls the identity of another person in real-time. Traditional deepfake detectors fail here, as all video content is synthetic. We propose a novel biometric defense that detects identity leakage in the transmitted latent representation. Our metric-learning approach disentangles identity cues from pose and expression, enabling detection of unauthorized swaps. Experiments across multiple talking-head models show that our method consistently outperforms prior defenses, operates in real time on consumer GPUs, and generalizes well to out-of-distribution data. By targeting the latent features shared during normal operation, our method offers a practical and robust safeguard against puppeteering.

Online free-viewpoint video (FVV) streaming is a challenging problem, which is relatively under-explored. It requires incremental on-the-fly updates to a volumetric representation, fast training and rendering to satisfy realtime constraints and a small memory footprint for efficient transmission. If acheived, it can enhance user experience by enabling novel applications, e.g., 3D video conferencing and live volumetric video broadcast, among others. In this work, we propose a novel framework for QUantized and Efficient ENcoding (QUEEN) for streaming FVV using 3D Gaussian Splatting (3D-GS). QUEEN directly learns Gaussian attribute residuals between consecutive frames at each time-step without imposing any structural constraints on them, allowing for high quality reconstruction and generalizability. To efficiently store the residuals, we further propose a quantization-sparsity framework, which contains a learned latent-decoder for effectively quantizing attribute residuals other than Gaussian positions and a learned gating module to sparsify position residuals. We propose to use the Gaussian viewspace gradient difference vector as a signal to separate the static and dynamic content of the scene. It acts as a guide for effective sparsity learning and speeds up training. On diverse FVV benchmarks, QUEEN outperforms the state-of-the-art online FVV methods on all metrics. Notably, for several highly dynamic scenes, it reduces the model size to just 0.7 MB per frame while training in under 5 sec and rendering at ~350 FPS.

Computer graphics seeks to deliver compelling images, generated within a computing budget, targeted at a specific display device, and ultimately viewed by an individual user. The foveated nature of human vision offers an opportunity to efficiently allocate computation and compression to appropriate areas of the viewer’s visual field, of particular importance with the rise of high-resolution and wide field-of-view display devices. However, while variations in acuity and contrast sensitivity across the field of view have been well-studied and modeled, a more consequential variation concerns peripheral vision’s degradation in the face of clutter, known as crowding. Understanding of peripheral crowding has greatly advanced in recent years, in terms of both phenomenology and modeling. Accurately leveraging this knowledge is critical for many applications, as peripheral vision covers a majority of pixels in the image. We advance computational models for peripheral vision aimed toward their eventual use in computer graphics. In particular, researchers have recently developed high-performing models of peripheral crowding, known as “pooling” models, which predict a wide range of phenomena but are computationally inefficient. We reformulate the problem as a dataflow computation, which enables faster processing and operating on larger images. Further, we account for the explicit encoding of “end stopped” features in the image, which was missing from previous methods. We evaluate our model in the context of perception of textures in the periphery, including a novel texture dataset and updated textural descriptors. Our improved computational framework may simplify development and testing of more sophisticated, complete models in more robust and realistic settings relevant to computer graphics.

Gaze tracking is an essential component of next generation displays for virtual reality and augmented reality applications. Traditional camera-based gaze trackers used in next generation displays are known to be lacking in one or multiple of the following metrics: power consumption, cost, computational complexity, estimation accuracy, latency, and form-factor. We propose the use of discrete photodiodes and light-emitting diodes (LEDs) as an alternative to traditional camera-based gaze tracking approaches while taking all of these metrics into consideration. We begin by developing a rendering-based simulation framework for understanding the relationship between light sources and a virtual model eyeball. Findings from this framework are used for the placement of LEDs and photodiodes. Our first prototype uses a neural network to obtain an average error rate of 2.67° at 400 Hz while demanding only 16 mW. By simplifying the implementation to using only LEDs, duplexed as light transceivers, and more minimal machine learning model, namely a light-weight supervised Gaussian process regression algorithm, we show that our second prototype is capable of an average error rate of 1.57° at 250 Hz using 800 mW.

Emergent in the field of head mounted display design is a desire to leverage the limitations of the human visual system to reduce the computation, communication, and display workload in power and form-factor constrained systems. Fundamental to this reduced workload is the ability to match display resolution to the acuity of the human visual system, along with a resulting need to follow the gaze of the eye as it moves, a process referred to as foveation. A display that moves its content along with the eye may be called a Foveated Display, though this term is also commonly used to describe displays with non-uniform resolution that attempt to mimic human visual acuity. We therefore recommend a definition for the term Foveated Display that accepts both of these interpretations. Furthermore, we include a simplified model for human visual Acuity Distribution Functions (ADFs) at various levels of visual acuity, across wide fields of view and propose comparison of this ADF with the Resolution Distribution Function of a foveated display for evaluation of its resolution at a particular gaze direction. We also provide a taxonomy to allow the field to meaningfully compare and contrast various aspects of foveated displays in a display and optical technology-agnostic manner.

Gaze-contingent rendering and display could help meet the increasing resolution and frame rate demands of modern displays while reducing the required latency, bandwidth, and power. However, it is still unclear how degradation of the peripheral image impacts behavior, particularly for the important task of reading. We examined changes in reading speed with different levels of peripheral degradation, varying the size of the text, foveal region, and sub-sampling kernel. We found a wide spread of responses across subjects, with the average change in reading speed ranging from -123 words per minute (WPM) to +67 WPM. We did not find significant effects across types of peripheral degradation, but the change in reading speed was significantly inversely correlated with baseline reading speed (r=-0.513, n=17, p=0.0352), indicating that faster readers were more negatively impacted.

Quality, diversity, and size of training data are critical factors for learning-based gaze estimators. We create two datasets satisfying these criteria for near-eye gaze estimation under infrared illumination: a synthetic dataset using anatomically-informed eye and face models with variations in face shape, gaze direction, pupil and iris, skin tone, and external conditions (2M images at 1280x960), and a real-world dataset collected with 35 subjects (2.5M images at 640x480). Using these datasets we train neural networks performing with sub-millisecond latency. Our gaze estimation network achieves 2.06(±0.44)° of accuracy across a wide 30°×40° field of view on real subjects excluded from training and 0.5° best-case accuracy (across the same FOV) when explicitly trained for one real subject. We also train a pupil localization network which achieves higher robustness than previous methods.

Virtual reality systems often make use of spatially tracked handheld props in the form of controllers or specialized objects to add realism and interaction. Tracking these objects today relies on the use of expensive, bulky, and power-consuming trackers that must be attached to an object. We propose a passive tracking technique that works with existing low-cost, off-the-shelf optical tracking components and is capable of turning any object into a tracked virtual reality prop. Our method utilizes paper-thin retro-reflective markers that can be placed in any free-form on everyday objects. The proof-of-concept prototype acts as a simple add-on for an existing tracking system and requires only a minimal amount of compute overhead. We demonstrate that our method allows bringing physical real-world objects to virtual worlds with ease, and provides an object identification technique using patterned retro-reflective markers.

We present a near-eye augmented reality display with resolution and focal depth dynamically driven by gaze tracking. The display combines a traveling microdisplay relayed off a concave half-mirror magnifier for the high-resolution foveal region, with a wide field-of-view peripheral display using a projector-based Maxwellian-view display whose nodal point is translated to follow the viewer's pupil during eye movements using a traveling holographic optical element. The same optics relay an image of the eye to an infrared camera used for gaze tracking, which in turn drives the foveal display location and peripheral nodal point. Our display supports accommodation cues by varying the focal depth of the microdisplay in the foveal region, and by rendering simulated defocus on the "always in focus" scanning laser projector used for peripheral display. The resulting family of displays significantly improves on the field-of-view, resolution, and form-factor tradeoff present in previous augmented reality designs. We show prototypes supporting 30, 40 and 60 cpd foveal resolution at a net 85° × 78° field of view per eye.

Traditional optical manufacturing poses a great challenge to near-eye display designers due to large lead times in the order of multiple weeks, limiting the abilities of optical designers to iterate fast and explore beyond conventional designs. We present a complete near-eye display manufacturing pipeline with a day lead time using commodity hardware. Our novel manufacturing pipeline consists of several innovations including a rapid production technique to improve surface of a 3D printed component to optical quality suitable for near-eye display application, a computational design methodology using machine learning and ray tracing to create freeform static projection screen surfaces for near-eye displays that can represent arbitrary focal surfaces, and a custom projection lens design that distributes pixels non-uniformly for a foveated near-eye display hardware design candidate. We have demonstrated untethered augmented reality near-eye display prototypes to assess success of our technique, and show that a ski-goggles form factor, a large monocular field of view$(30^{o}\times 55^{o})$, and a resolution of 12 cycles per degree can be achieved.

Redirected walking techniques can enhance the immersion and visual-vestibular comfort of virtual reality (VR) navigation, but are often limited by the size, shape, and content of the physical environments. We propose a redirected walking technique that can apply to small physical environments with static or dynamic obstacles. Via a head- and eye-tracking VR headset, our method detects saccadic suppression and redirects the users during the resulting temporary blindness. Our dynamic path planning runs in real-time on a GPU, and thus can avoid static and dynamic obstacles, including walls, furniture, and other VR users sharing the same physical space. To further enhance saccadic redirection, we propose subtle gaze direction methods tailored for VR perception. We demonstrate that saccades can significantly increase the rotation gains during redirection without introducing visual distortions or simulator sickness. This allows our method to apply to large open virtual spaces and small physical environments for room-scale VR. We evaluate our system via numerical simulations and real user studies.

We introduce a new data structure and algorithms that employ it to compute real-time global illumination from static environments. Light field probes encode a scene's full light field and internal visibility. They extend current radiance and irradiance probe structures with per-texel visibility information similar to a G-buffer and variance shadow map. We apply ideas from screen-space and voxel cone tracing techniques to this data structure to efficiently sample radiance on world space rays, with correct visibility information, directly within pixel and compute shaders. From these primitives, we then design two GPU algorithms to efficiently gather real-time, viewer-dependent global illumination onto both static and dynamic objects. These algorithms make different tradeoffs between performance and accuracy. Supplemental GLSL source code is included.

Foveated rendering is a performance optimization based on the well-known degradation of peripheral visual acuity. It reduces computational costs by showing a high-quality image in the user’s central (foveal) vision and a lower quality image in the periphery. Foveated rendering is a promising optimization for Virtual Reality (VR) graphics, and generally requires accurate and low-latency eye tracking to ensure correctness even when a user makes large, fast eye movements such as saccades. However, due to the phenomenon of saccadic omission, it is possible that these requirements may be relaxed. In this article, we explore the effect of latency for foveated rendering in VR applications. We evaluated the detectability of visual artifacts for three techniques capable of generating foveated images and for three different radii of the high-quality foveal region. Our results show that larger foveal regions allow for more aggressive foveation, but this effect is more pronounced for temporally stable foveation techniques. Added eye tracking latency of 80--150ms causes a significant reduction in acceptable amount of foveation, but a similar decrease in acceptable foveation was not found for shorter eye-tracking latencies of 20--40ms, suggesting that a total system latency of 50--70ms could be tolerated.

We present a new optical design for see-through near-eye displays that is simple, compact, varifocal, and provides a wide field of view with clear peripheral vision and large eyebox. Key to this effort is a novel see-through rear-projection screen. We project an image to the see-through screen using an off-axis path, which is then relayed to the user's eyes through an on-axis partially-reflective magnifying surface. Converting the off-axis path to a compact on-axis imaging path simplifies the optical design. We establish fundamental trade-offs between the quantitative parameters of resolution, field of view, and the form-factor of our design. We demonstrate a wearable binocular near-eye display using off-the-shelf projection displays, custom-designed see-through spherical concave mirrors, and see-through screen designs using either custom holographic optical elements or polarization-selective diffusers.

Increasing resolution and dynamic range of digital color displays is challenging with designs confined by cost and power specifications. This necessitates modern displays to trade-off spatial and temporal resolution for color reproduction capability. In this work we explore the idea of joint hardware and algorithm design to balance such trade-offs. We introduce a system that uses content-adaptive and compressive factorizations to reproduce colors. Each target frame is factorized into two products of high-resolution monochromatic and low-resolution color images, which then get integrated through temporal or spatial multiplexing. As our framework minimizes the error in colorimetric space, the perceived color rendition is high, and thanks to GPU acceleration, the results are generated in real-time. We evaluate our system with a LCD prototype that uses LED backlight array and temporal multiplexing to reproduce color images. Our approach enables high effective resolution and dynamic range without increasing power consumption. We also demonstrate low-cost extensions to hyperspectral and light-field imaging, which are possible due to compressive nature of our system.

Holograms display a 3D image in high resolution and allow viewers to focus freely as if looking through a virtual window, yet computer generated holography (CGH) hasn't delivered the same visual quality under plane wave illumination and due to heavy computational cost. Light field displays have been popular due to their capability to provide continuous focus cues. However, light field displays must trade off between spatial and angular resolution, and do not model diffraction. We present a light field-based CGH rendering pipeline allowing for reproduction of high-definition 3D scenes with continuous depth and support of intra-pupil view-dependent occlusion. Our rendering accurately accounts for diffraction and supports various types of reference illuminations for hologram. We avoid under- and over-sampling and geometric clipping effects seen in previous work. We also demonstrate an implementation of light field rendering plus the Fresnel diffraction integral based CGH calculation which is orders of magnitude faster than the state of the art [Zhang et al. 2015], achieving interactive volumetric 3D graphics. To verify our computational results, we build a see-through, near-eye, color CGH display prototype which enables co-modulation of both amplitude and phase. We show that our rendering accurately models the spherical illumination introduced by the eye piece and produces the desired 3D imagery at the designated depth. We also analyze aliasing, theoretical resolution limits, depth of field, and other design trade-offs for near-eye CGH.

A variety of applications such as virtual reality and immersive cinema require high image quality, low rendering latency, and consistent depth cues. 4D light field displays support focus accommodation, but are more costly to render than 2D images, resulting in higher latency. The human visual system can resolve higher spatial frequencies in the fovea than in the periphery. This property has been harnessed by recent 2D foveated rendering methods to reduce computation cost while maintaining perceptual quality. Inspired by this, we present foveated 4D light fields by investigating their effects on 3D depth perception. Based on our psychophysical experiments and theoretical analysis on visual and display bandwidths, we formulate a content-adaptive importance model in the 4D ray space. We verify our method by building a prototype light field display that can render only 16% -- 30% rays without compromising perceptual quality.

Accommodative depth cues, a wide field of view, and ever-higher resolutions all present major hardware design challenges for near-eye displays. Optimizing a design to overcome one of these challenges typically leads to a trade-off in the others. We tackle this problem by introducing an all-in-one solution - a new wide field of view, gaze-tracked near-eye display for augmented reality applications. The key component of our solution is the use of a single see-through, varifocal deformable membrane mirror for each eye reflecting a display. They are controlled by airtight cavities and change the effective focal power to present a virtual image at a target depth plane which is determined by the gaze tracker. The benefits of using the membranes include wide field of view (100° diagonal) and fast depth switching (from 20 cm to infinity within 300 ms). Our subjective experiment verifies the prototype and demonstrates its potential benefits for near-eye see-through displays.

Foveated rendering synthesizes images with progressively less detail outside the eye fixation region, potentially unlocking significant speedups for wide field-of-view displays, such as head mounted displays, where target framerate and resolution is increasing faster than the performance of traditional real-time renderers. To study and improve potential gains, we designed a foveated rendering user study to evaluate the perceptual abilities of human peripheral vision when viewing today's displays. We determined that filtering peripheral regions reduces contrast, inducing a sense of tunnel vision. When applying a postprocess contrast enhancement, subjects tolerated up to 2× larger blur radius before detecting differences from a non-foveated ground truth. After verifying these insights on both desktop and head mounted displays augmented with high-speed gaze-tracking, we designed a perceptual target image to strive for when engineering a production foveated renderer. Given our perceptual target, we designed a practical foveated rendering system that reduces number of shades by up to 70% and allows coarsened shading up to 30° closer to the fovea than Guenter et al. [2012] without introducing perceivable aliasing or blur. We filter both pre- and post-shading to address aliasing from undersampling in the periphery, introduce a novel multiresolution- and saccade-aware temporal antialising algorithm, and use contrast enhancement to help recover peripheral details that are resolvable by our eye but degraded by filtering. We validate our system by performing another user study. Frequency analysis shows our system closely matches our perceptual target. Measurements of temporal stability show we obtain quality similar to temporally filtered non-foveated renderings.

We present a novel design for an optical see-through augmented reality display that offers a wide field of view and supports a compact form factor approaching ordinary eyeglasses. Instead of conventional optics, our design uses only two simple hardware components: an LCD panel and an array of point light sources (implemented as an edge-lit, etched acrylic sheet) placed directly in front of the eye, out of focus. We code the point light sources through the LCD to form miniature see-through projectors. A virtual aperture encoded on the LCD allows the projectors to be tiled, creating an arbitrarily wide field of view. Software rearranges the target augmented image into tiled sub-images sent to the display, which appear as the correct image when observed out of the viewer's accommodation range. We evaluate the design space of tiled point light projectors with an emphasis on increasing spatial resolution through the use of eye tracking. We demonstrate feasibility through software simulations and a real-time prototype display that offers a 110° diagonal field of view in the form factor of large glasses and discuss remaining challenges to constructing a practical display.

We demonstrate that layered spatial light modulators (SLMs), subject to fixed lateral displacements and refreshed at staggered intervals, can synthesize images with greater spatiotemporal resolution than that afforded by any single SLM used in their construction. Dubbed cascaded displays , such architectures enable superresolution flat panel displays (e.g., using thin stacks of liquid crystal displays (LCDs)) and digital projectors (e.g., relaying the image of one SLM onto another). We introduce a comprehensive optimization framework, leveraging non-negative matrix and tensor factorization, that decomposes target images and videos into multi-layered, time-multiplexed attenuation patterns---offering a flexible trade-off between apparent image brightness, spatial resolution, and refresh rate. Through this analysis, we develop a real-time dual-layer factorization method that quadruples spatial resolution and doubles refresh rate. Compared to prior superresolution displays, cascaded displays place fewer restrictions on the hardware, offering thin designs without moving parts or the necessity of temporal multiplexing. Furthermore, cascaded displays are the first use of multi-layer displays to increase apparent temporal resolution. We validate these concepts using two custom-built prototypes: a dual-layer LCD and a dual-modulation liquid crystal on silicon (LCoS) projector, with the former emphasizing head-mounted display (HMD) applications.

We describe the design space for real-time photon density estimation, the key step of rendering global illumination (GI) via photon mapping. We then detail and analyze efficient GPU implementations of four best-of-breed algorithms. All produce reasonable results on NVIDIA GeForce 670 at 1920 × 1080 for complex scenes with multiple-bounce diffuse effects, caustics, and glossy reflection in real-time. Across the designs we conclude that tiled, deferred photon gathering in a compute shader gives the best combination of performance and quality.

We propose near-eye light field displays that enable thin, lightweight head-mounted displays (HMDs) capable of presenting nearly correct convergence, accommodation, binocular disparity, and retinal defocus depth cues. Sharp images are depicted by out-of-focus elements by synthesizing light fields corresponding to virtual objects within a viewer's natural accommodation range. We formally assess the capabilities of microlens arrays to achieve practical near-eye light field displays. Building on concepts shared with existing integral imaging displays and light field cameras, we optimize performance in the context of near-eye viewing. We establish fundamental trade-offs between the quantitative parameters of resolution, field of view, and depth of field, as well as the ergonomic parameters of form factor and ranges of allowed eye movement. As with light field cameras, our design supports continuous accommodation of the eye throughout a finite depth of field; as a result, binocular configurations provide a means to address the accommodation-convergence conflict occurring with existing stereoscopic displays. We construct a complete prototype display system, comprising: a custom-fabricated HMD using modified off-the-shelf parts and real-time, GPU-accelerated light field renderers (including a general ray tracing method and a "backward compatible" rasterization method supporting existing stereoscopic content). Through simulations and experiments, we motivate near-eye light field displays as thin, lightweight alternatives to conventional near-eye displays.

Subpixel Reconstruction Antialiasing (SRAA) combines singlepixel (1x) shading with subpixel visibility to create antialiased images without increasing the shading cost. SRAA targets deferred-shading renderers, which cannot use multisample antialiasing.

Stochastic renderers produce unbiased but noisy images of scenes that include the advanced camera effects of motion and defocus blur and possibly other effects such as transparency. We present a simple algorithm that selectively adds bias in the form of image space blur to pixels that are unlikely to have high frequency content in the final image. For each pixel, we sweep once through a fixed neighborhood of samples in front to back order, using a simple accumulation scheme. We achieve good quality images with only 16 samples per pixel, making the algorithm potentially practical for interactive stochastic rendering in the near future.

Stochastic transparency provides a unified approach to order-independent transparency, antialiasing, and deep shadow maps. It augments screen-door transparency using a random sub-pixel stipple pattern, where each fragment of transparent geometry covers a random subset of pixel samples of size proportional to alpha. This results in correct alpha-blended colors on average, in a single render pass with fixed memory size and no sorting, but introduces noise. We reduce this noise by an alpha correction pass, and by an accumulation pass that uses a stochastic shadow map from the camera. At the pixel level, the algorithm does not branch and contains no read-modify-write loops, other than traditional z-buffer blend operations. This makes it an excellent match for modern massively parallel GPU hardware. Stochastic transparency is very simple to implement and supports all types of transparent geometry, able without coding for special cases to mix hair, smoke, foliage, windows, and transparent cloth in a single scene.

Stochastic transparency provides a unified approach to order-independent transparency, anti-aliasing, and deep shadow maps. It augments screen-door transparency using a random sub-pixel stipple pattern, where each fragment of transparent geometry covers a random subset of pixel samples of size proportional to alpha. This results in correct alpha-blended colors on average, in a single render pass with fixed memory size and no sorting, but introduces noise. We reduce this noise by an alpha correction pass, and by an accumulation pass that uses a stochastic shadow map from the camera. At the pixel level, the algorithm does not branch and contains no read-modify-write loops other than traditional z-buffer blend operations. This makes it an excellent match for modern massively parallel GPU hardware. Stochastic transparency is very simple to implement and supports all types of transparent geometry, able without coding for special cases to mix hair, smoke, foliage, windows, and transparent cloth in a single scene.

Abstract We propose to enhance the capabilities of the human visual system by performing optical image processing directly on an observed scene. Unlike previous work which additively superimposes imagery on a scene, or completely replaces scene imagery with a manipulated version, we perform all manipulation through the use of a light modulation display to spatially filter incoming light. We demonstrate a number of perceptually motivated algorithms including contrast enhancement and reduction, object highlighting for preattentive emphasis, colour saturation, de‐saturation and de‐metamerization, as well as visual enhancement for the colour blind. A camera observing the scene guides the algorithms for on‐the‐fly processing, enabling dynamic application scenarios such as monocular scopes, eyeglasses and windshields.

The NVIDIA® OptiX™ ray tracing engine is a programmable system designed for NVIDIA GPUs and other highly parallel architectures. The OptiX engine builds on the key observation that most ray tracing algorithms can be implemented using a small set of programmable operations. Consequently, the core of OptiX is a domain-specific just-in-time compiler that generates custom ray tracing kernels by combining user-supplied programs for ray generation, material shading, object intersection, and scene traversal. This enables the implementation of a highly diverse set of ray tracing-based algorithms and applications, including interactive rendering, offline rendering, collision detection systems, artificial intelligence queries, and scientific simulations such as sound propagation. OptiX achieves high performance through a compact object model and application of several ray tracing-specific compiler optimizations. For ease of use it exposes a single-ray programming model with full support for recursion and a dynamic dispatch mechanism similar to virtual function calls.

Abstract We present two novel parallel algorithms for rapidly constructing bounding volume hierarchies on manycore GPUs. The first uses a linear ordering derived from spatial Morton codes to build hierarchies extremely quickly and with high parallel scalability. The second is a top‐down approach that uses the surface area heuristic (SAH) to build hierarchies optimized for fast ray tracing. Both algorithms are combined into a hybrid algorithm that removes existing bottlenecks in the algorithm for GPU construction performance and scalability leading to significantly decreased build time. The resulting hierarchies are close in to optimized SAH hierarchies, but the construction process is substantially faster, leading to a significant net benefit when both construction and traversal cost are accounted for. Our preliminary results show that current GPU architectures can compete with CPU implementations of hierarchy construction running on multicore systems. In practice, we can construct hierarchies of models with up to several million triangles and use them for fast ray tracing or other applications.

The graphics processing unit (GPU) has become an integral part of today's mainstream computing systems. Over the past six years, there has been a marked increase in the performance and capabilities of GPUs. The modern GPU is not only a powerful graphics engine but also a highly parallel programmable processor featuring peak arithmetic and memory bandwidth that substantially outpaces its CPU counterpart. The GPU's rapid increase in both programmability and capability has spawned a research community that has successfully mapped a broad range of computationally demanding, complex problems to the GPU. This effort in general-purpose computing on the GPU, also known as GPU computing, has positioned the GPU as a compelling alternative to traditional microprocessors in high-performance computer systems of the future. We describe the background, hardware, and programming model for GPU computing, summarize the state of the art in tools and techniques, and present four GPU computing successes in game physics and computational biophysics that deliver order-of-magnitude performance gains over optimized CPU applications.

The graphics processor (GPU) on today's commodity video cards has evolved into an extremely powerful and flexible processor. Modern graphics architectures provide tremendous memory bandwidth and computational horsepower, with dozens of fully programmable shading units that support vector operations and IEEE floating point precision. High-level languages have emerged for graphics hardware, making this computational power accessible. GPGPU stands for "General-Purpose Computation on GPUs". GPGPU researchers have achieved over an order of magnitude speedup over modern CPUs on some non-graphics problems.This course provides detailed coverage of general-purpose computation on graphics hardware. We emphasize core computational building blocks, ranging from linear algebra to database queries, and review the tools, perils, and strategies in GPU programming. We present analysis of GPU performance characteristics, and use this analysis to provide insight into how to build efficient GPGPU algorithms. Finally we present a set of case studies on general-purpose applications of graphics hardware.

We present a technique for interactive rendering of glossy objects in complex and dynamic lighting environments that captures interreflections and all-frequency shadows. Our system is based on precomputed radiance transfer and separable BRDF approximation. We factor glossy BRDFs using a separable decomposition and keep only a few low-order approximation terms, each consisting of a purely view-dependent and a purely light-dependent component. In the precomputation step, for every vertex, we sample its visibility and compute a direct illumination transport vector corresponding to each BRDF term. We use modern graphics hardware to accelerate this step and further compress the data using a nonlinear wavelet approximation. The direct illumination pass is followed by one or more interreflection passes, each of which gathers compressed transport vectors from the previous pass to produce global illumination transport vectors. To render at run time, we dynamically sample the lighting to produce a light vector, also represented in a wavelet basis. We compute the inner product of the light vector with the precomputed transport vectors, and the results are further combined with the BRDF view-dependent components to produce vertex colors. We describe acceleration of the rendering algorithm using programmable graphics hardware and discuss the limitations and trade-offs imposed by the hardware.

We have previously presented Qsilver, a flexible simulation system for graphics architectures. In this paper we describe our extensions to this system, which we use - instrumented with a power model and HotSpot - to analyze the application of standard CPU static and runtime thermal management techniques on the GPU. We describe experiments implementing clock gating, fetch gating, dynamic voltage scaling, multiple clock domains and permuted floor-planning on the GPU using our simulation environment, and demonstrate that these techniques are beneficial in the GPU domain. Further, we show that the inherent parallelism of GPU workloads enables significant thermal gains on chips designed employing static floorplan repartitioning

We propose an adaptive form of frameless rendering with the potential to dramatically increase rendering speed over conventional interactive rendering approaches. Without the rigid sampling patterns of framed renderers, sampling and reconstruction can adapt with very fine granularity to spatio-temporal color change. A sampler uses closed-loop feedback to guide sampling toward edges or motion in the image. Temporally deep buffers store all the samples created over a short time interval for use in reconstruction and as sampler feedback. GPU-based reconstruction responds both to sampling density and space-time color gradients. Where the displayed scene is static, spatial color change dominates and older samples are given significant weight in reconstruction, resulting in sharper and eventually antialiased images. Where the scene is dynamic, more recent samples are emphasized, resulting in less sharp but more up-to-date images. We also use sample reprojection to improve reconstruction and guide sampling toward occlusion edges, undersampled regions, and specular highlights. In simulation our frameless renderer requires an order of magnitude fewer samples than traditional rendering of similar visual quality (as measured by RMS error), while introducing overhead amounting to 15% of computation time.

The problem of localization of wireless sensor nodes has long been regarded as very difficult to solve, when considering the realities of real world environments. In this paper, we formally describe, design, implement and evaluate a novel localization system, called Spotlight. Our system uses the spatio-temporal properties of well controlled events in the network (e.g., light), to obtain the locations of sensor nodes. We demonstrate that a high accuracy in localization can be achieved without the aid of expensive hardware on the sensor nodes, as required by other localization systems. We evaluate the performance of our system in deployments of Mica2 and XSM motes. Through performance evaluations of a real system deployed outdoors, we obtain a 20cm localization error. A sensor network, with any number of nodes, deployed in a 2500m2 area, can be localized in under 10 minutes, using a device that costs less than $1000. To the best of our knowledge, this is the first report of a sub-meter localization error, obtained in an outdoor environment, without equipping the wireless sensor nodes with specialized ranging hardware.

We present a technique, based on precomputed light transport, for interactive rendering of translucent objects under all-frequency environment maps. We consider the complete BSSRDF model proposed by Jensen et al. [2001]. which includes both single and diffuse multiple scattering components. The challenge is how to efficiently precompute all-frequency light transport functions due to subsurface scattering. We apply the two-pass hierarchical technique by Jensen et al. [2002] in the space of non-linearly approximated transport vectors, which allows us to efficiently evaluate transport vectors due to diffuse multiple scattering. We then include an approximated single scattering term in the precomputation, which previous interactive systems have ignored. For an isotropic phase function, this approximation produces a diffuse transport vector per vertex, and is combined with the multiple scattering component. For a general phase function, we introduce a technique from BRDF rendering to factor the phase function using a separable decomposition to allow for view-dependent rendering. We show that our rendering results qualitatively match the appearance of translucent objects, achieving a high level of realism at interactive rates.

We present a new algorithm for best-effort simplification of polygonal meshes based on principles of visual perception. Building on previous work, we use a simple model of low-level human vision to estimate the perceptibility of local simplification operations in a view-dependent Multi-Triangulation structure. Our algorithm improves on prior perceptual simplification approaches by accounting for textured models and dynamic lighting effects. We also model more accurately the scale of visual changes resulting from simplification, using parametric texture deviation to bound the size (represented as spatial frequency) of features destroyed, created, or altered by simplifying the mesh. The resulting algorithm displays many desirable properties: it is view-dependent, sensitive to silhouettes, sensitive to underlying texture content, and sensitive to illumination (for example, preserving detail near highlight and shadow boundaries, while aggressively simplifying washed-out regions). Using a unified perceptual model to evaluate these effects automatically accounts for their relative importance and balances between them, overcoming the need for ad hoc or hand-tuned heuristics.

Interruptible rendering is a novel approach to the fidelity-versus-performance tradeoff ubiquitous in real-time rendering. Interruptible rendering unifies spatial error caused by rendering coarse approximations for speed and temporal error caused by the delay imposed by rendering into a single image-space error measure. The heart of this approach is a progressive rendering framework that renders a coarse image into the back buffer and continuously refines it while monitoring temporal error. When temporal error exceeds the spatial error caused by coarse rendering, further refinement is pointless, and the image is displayed. We discuss how to adapt different rendering algorithms for interruptible use and present implementations based on polygonal rendering and ray casting. Interruptible rendering provides a low-latency, self-tuning approach to interactive rendering. To evaluate our results we introduce a "gold standard" approach that measures dynamic visual error against a hypothetical perfect rendering and show that interruptible rendering is more accurate than standard fidelity-versus-performance schemes. This improved accuracy enables better interactive rendering, both for complex models and complex rendering modalities such as ray casting.

No abstract available.

Hierarchical dynamic simplification (HDS) is a new approach to the problem of simplifying arbitrary polygonal environments. HDS operates dynamically, retessellating the scene continuously as the user's viewing position shifts, and adaptively processing the entire database without first decomposing the environment into individual objects. The resulting system allows real-time display of very complex polygonal CAD models consisting of thousands of parts and hundreds of thousands of polygons. HDS supports various preprocessing algorithms and various run-time criteria, providing a general framework for dynamic view-dependent simplification. Briefly, HDS works by clustering vertices together in a hierarchical fashion. The simplification process continuously queries this hierarchy to generate a scene containing only those polygons that are important from the current viewpoint. When the volume of space associated with a vertex cluster occupies less than a user-specified amount of the screen, all vertices within that cluster are collapsed together and degenerate polygons filtered out. HDS maintains an active list of visible polygons for rendering. Since frame-to-frame movements typically involve small changes in viewpoint, and therefore modify the active list by only a few polygons, the method takes advantage of temporal coherence for greater speed.

We describe an approach for determining potentially visible sets in dynamic architectural models. Our scheme divides the models into cells and portals, computing a conservative estimate of which cells are visible at render time. The technique is simple to implement and can be easily integrated into existing systems, providing increased interactive performance on large architectural models.