Jason Yang

A real-time distributed light field camera

We present the design and implementation of a real-time, distributed light field camera. Our system allows multiple viewers to navigate virtual cameras in a dynamically changing light field that is captured in real-time. Our light field camera consists of 64 commodity video cameras that are connected to off-the-shelf computers. We employ a distributed rendering algorithm that allows us to overcome the data bandwidth problems inherent in dynamic light fields. Our algorithm works by selectively transmitting only those portions of the video streams that contribute to the desired virtual views. This technique not only reduces the total bandwidth, but it also allows us to scale the number of cameras in our system without increasing network bandwidth. We demonstrate our system with a number of examples.

Real-time concurrent linked list construction on the GPU

We introduce a method to dynamically construct highly concurrent linked lists on modern graphics processors. Once constructed, these data structures can be used to implement a host of algorithms useful in creating complex rendering effects in real time. We present a straightforward way to create these linked lists using generic atomic operations available in APIs such as OpenGL 4.0 and DirectX 11. We also describe several possible applications of our algorithm. One example uses per‐pixel linked lists for order‐independent transparency; as a consequence, we are able to directly implement fully programmable blending, which frees developers from the restrictions imposed by current graphics APIs. The second uses linked lists to implement real‐time indirect shadows.

Symmetric key cryptography on modern graphics hardware

GPUs offer a tremendous amount of computational bandwidth that was until now largely unusable for cryptographic computations due to a lack of integer arithmetic and user-friendly programming APIs that provided direct access to the GPUs computing resources. The latest generation of GPUs, which introduces integer/binary arithmetic, has been leveraged to create several implementations of the AES and DES symmetric key algorithms. Both conventional and bitsliced implementations are described that achieve data rates on the order of 3-30 Gbps from a single AMD HD 2900 XT graphics card, yielding speedups of 6-60x over equivalent implementations on high-performance CPUs.

A unified header compression framework for low-bandwidth links

Compressing protocol headers has traditionally been an attractive way of conserving bandwidth over low-speed links, including those in wireless systems. However, despite the growth in recent years in the number of end-to-end protocols beyond TCP/IP, header compression deployment for those protocols has not kept pace. This is in large part due to complexities in implementation, which often requires a detailed knowledge of kernel internals, and a lack of a common way of pursuing the general problem across a variety of end-to-end protocols. To address this, rather than defining several new protocol-specific standards, we present a unified framework for header compression. This framework includes a simple, platform-independent header description language that protocol implementors can use to describe high-level header properties, and a platform-specific code generation tool that produces kernel source code automatically from this header specification. Together, the high-level description language and code generator free protocol designers from having to understand any details of the target platform, enabling them to implement header compression with relatively little effort. We analyze the performance of compression produced using this framework for TCP/IP in the Linux 2.0 kernel and demonstrate that unified, automatically-generated header compression without significant performance penalty is viable.

Filtering approaches for real-time anti-aliasing

For more than a decade, supersample anti-aliasing (SSAA) and multisample anti-aliasing (MSAA) have been the gold-standard anti-aliasing solutions in games. However, these techniques are not well suited for deferred shading or fixed environments like the current generation of consoles. In recent years, industry and academia have been exploring alternative approaches, where anti-aliasing is performed as a post-processing step. The original, CPU-based morphological anti-aliasing (MLAA) method gave birth to an explosion of real-time anti-aliasing techniques that rival MSAA. Most of these techniques share concepts and ideas, so the main goal of this course is to establish a conceptual link between them, identifying novelties and differences. The presenters explain how sub-pixel data can be used to improve quality and performance tradeoffs at post-processing steps, which is a cutting-edge research area today. The course includes an overview of both research and industry filter-based anti-aliasing techniques in games for all modern platforms (AMD and NVIDIA GPUs, PlayStation 3, and Xbox 360), low-level insight to ease adoption of these techniques and give attendees a complete concept-to-implementation roadmap, and deep quality, performance, and ease-of-integration comparisons of each technique.

Real-time reflection mapping with parallax

We present a novel algorithm to efficiently render accurate reflections on programmable graphics hardware. Our algorithm overcomes problems that commonly occur in environment mapping such as the lack of motion parallax and inaccuracies when objects are close to the reflectors. In place of a 2D environment map, which only represents points infinitely far away from the reflector, we use six 4D light field slabs to represent the surrounding scene. Each reflected ray is rendered by indexing into these precaptured environment light fields. We are able to render accurate reflections with motion parallax at interactive frame rates independent of the reflector geometry and the scene complexity. Furthermore, we can move the reflectors within a constrained region of space and guarantee that the environment light field provides the necessary rays. We benefit from the programmability of existing graphics hardware to efficiently compute the reflected rays and transform them into the appropriate light field index. We also take advantage of the large texture memories and memory bandwidth available in todays graphics card to store and query hardware-compressed light fields.

Forward+: Bringing Deferred Lighting to the Next Level

This paper presents Forward+, a method of rendering many lights by culling and storing only lights that contribute to the pixel. Forward+ is an extension to traditional forward rendering. Light culling, implemented using the compute capability of the GPU, is added to the pipeline to create lists of lights; that list is passed to the final rendering shader, which can access all information about the lights. Although Forward+ increases workload to the final shader, it theoretically requires less memory traffic compared to compute-based deferred lighting. Furthermore, it removes the major drawback of deferred techniques, which is a restriction of materials and lighting models. Experiments are performed to compare the performance of Forward+ and deferred lighting.

A framework for rendering complex scattering effects on hair

The appearance of hair plays a critical role in synthesizing realistic looking human characters. However, due to the high complexity in hair geometry and the scattering nature of hair fibers, rendering hair with photorealistic quality and at interactive speeds remains as an open problem in computer graphics. Previous approaches attempt to simplify the scattering model to only tackle a specific aspect of the scattering effects. In this paper, we present a new approach to simultaneously render complex scattering effects including volumetric shadows, transparency, and antialiasing under a unified framework. Our solution uses a shadow-ray path to produce volumetric self-shadows and an additional view-ray path to produce transparency. To compute and accumulate the contribution of individual hair fibers along each (shadow or view) path, we develop a new GPU-based k-buffer technique that can efficiently locate the K nearest scattering locations and combine them in the correct order. Compared with existing multi-layer based approaches[Kim and Neumann 2001; Yuksel and Keyser 2008; Sintorn and Assarsson 2009], we show that our k-buffer solution can more accurately reproduce the shadowing and transparency effects. Further, we present an anti-aliasing scheme that directly builds upon the k-buffer. We implement all three effects (volumetric shadows, transparency, and anti-aliasing) under a unified rendering pipeline. Experiments on complex hair models demonstrate that our new solution produces near photorealistic hair rendering at very interactive speed.

A directionally adaptive edge anti-aliasing filter

The latest generation of graphics hardware provides direct access to multisample anti-aliasing (MSAA) rendering data. By taking advantage of these existing pixel subsample values, an intelligent reconstruction filter can be computed using programmable GPU shader units. This paper describes an adaptive anti-aliasing (AA) filter for real-time rendering on the GPU. Improved quality is achieved by using information from neighboring pixel samples to compute both an approximation of the gradient of primitive edges and the final pixel color.

Antibody-modified conduits for highly selective cytokine elimination from blood

Cytokines play an important role in dysregulated immune responses to infection, pancreatitis, ischemia/reperfusion injury, burns, hemorrhage, cardiopulmonary bypass, trauma, and many other diseases. Moreover, the imbalance between inflammatory and antiinflammatory cytokines can have deleterious effects. Here, we demonstrated highly selective blood-filtering devices - antibody-modified conduits (AMCs) - that selectively eliminate multiple specific deleterious cytokines in vitro. AMCs functionalized with antibodies against human vascular endothelial growth factor A or tumor necrosis factor α (TNF-α) selectively eliminated the target cytokines from human blood in vitro and maintained them in reduced states even in the face of ongoing infusion at supraphysiologic rates. We characterized the variables that determine AMC performance, using anti-human TNF-α AMCs to eliminate recombinant human TNF-α. Finally, we demonstrated selective cytokine elimination in vivo by filtering interleukin 1 β from rats with lipopolysaccharide-induced hypercytokinemia.