High-Performance Graphics 2014

Permanent URI for this collection

https://diglib.eg.org/handle/10.2312/7756

Browse

Browsing High-Performance Graphics 2014 by Subject "I.3.7 [Computer Graphics]"

Now showing 1 - 5 of 5
  • Thumbnail Image
    Item
    Exploiting Local Orientation Similarity for Efficient Ray Traversal of Hair and Fur
    (The Eurographics Association, 2014) Woop, Sven; Benthin, Carsten; Wald, Ingo; Johnson, Gregory S.; Tabellion, Eric; Ingo Wald and Jonathan Ragan-Kelley
    Hair and fur typically consist of a large number of thin, curved, and densely packed strands which are difficult to ray trace efficiently. A tight fitting spatial data structure, such as a bounding volume hierarchy (BVH), is needed to quickly determine which hair a ray hits. However, the large number of hairs can yield a BVH with a large memory footprint (particularly when hairs are pre-tessellated), and curved or diagonal hairs cannot be tightly bounded within axis aligned bounding boxes. In this paper, we describe an approach to ray tracing hair and fur with improved efficiency, by combining parametrically defined hairs with a BVH that uses both axis-aligned and oriented bounding boxes. This BVH exploits similarity in the orientation of neighboring hairs to increase ray culling efficiency compared to purely axis-aligned BVHs. Our approach achieves about 2× the performance of ray tracing pre-tessellated hair models, while requiring significantly less memory.
  • Thumbnail Image
    Item
    High-Performance Rendering of Realistic Cumulus Clouds Using Pre-computed Lighting
    (The Eurographics Association, 2014) Yusov, Egor; Ingo Wald and Jonathan Ragan-Kelley
    We present a new method for rendering realistic cumulus clouds in real time. The clouds in our approach consist of randomly rotated and scaled copies of a single reference particle. During the pre-processing, we pre-compute optical depth, single and multiple scattering inside the reference particle for every camera position, orientation and light direction, and store the information in the look-up tables. At run time, information from the look-up tables is used to compute the cloud shading, avoiding any ray marching or slicing. To control the level of detail, we introduce a new technique which provides high fidelity for close clouds while using a coarse representation for distant regions. In addition to this, we present a new method for blending particles. Compared to traditional alpha-blending, this method produces more accurate visual results by accounting for volumetric intersection. The method merges collection of individual particles into a continuous medium, and also eliminates temporal artifacts. Our technique is able to produce realistic images at high frame rates.
  • Thumbnail Image
    Item
    Layered Reflective Shadow Maps for Voxel-based Indirect Illumination
    (The Eurographics Association, 2014) Sugihara, Masamichi; Rauwendaal, Randall; Salvi, Marco; Ingo Wald and Jonathan Ragan-Kelley
    We introduce a novel voxel-based algorithm that interactively simulates both diffuse and glossy single-bounce indirect illumination. Our algorithm generates high quality images similar to the reference solution while using only a fraction of the memory of previous methods. The key idea in our work is to decouple occlusion data, stored in voxels, from lighting and geometric data, encoded in a new per-light data structure called layered reflective shadow maps (LRSMs). We use voxel cone tracing for visibility determination and integrate outgoing radiance by performing lookups in a pre-filtered LRSM. Finally we demonstrate that our simple data structures are easy to implement and can be rebuilt every frame to support both dynamic lights and scenes.
  • Thumbnail Image
    Item
    Real-Time Deformation of Subdivision Surfaces from Object Collisions
    (The Eurographics Association, 2014) Schäfer, Henry; Keinert, Benjamin; Nießner, Matthias; Buchenau, Christoph; Guthe, Michael; Stamminger, Marc; Ingo Wald and Jonathan Ragan-Kelley
    We present a novel real-time approach for fine-scale surface deformations resulting from collisions. Deformations are represented by a high-resolution displacement function. When two objects collide, these offsets are updated directly on the GPU based on a dynamically generated binary voxelization of the overlap region. Consequently, we can handle collisions with arbitrary animated geometry. Our approach runs entirely on the GPU, avoiding costly CPU-GPU memory transfer and exploiting the GPU's computational power. Surfaces are rendered with the hardware tessellation unit, allowing for adaptively-rendered, high-frequency surface detail. Ultimately, our algorithm enables fine-scale surface deformations from geometry impact with very little computational overhead, running well below a millisecond even in complex scenes. As our results demonstrate, our approach is ideally suited to many real-time applications such as video games and authoring tools.
  • Thumbnail Image
    Item
    Reduced Precision for Hardware Ray Tracing in GPUs
    (The Eurographics Association, 2014) Keely, Sean; Ingo Wald and Jonathan Ragan-Kelley
    We propose a high performance, GPU integrated, hardware ray tracing system. We present and make use of a new analysis of ray traversal in axis aligned bounding volume hierarchies. This analysis enables compact traversal hardware through the use of reduced precision arithmetic. We also propose a new cache based technique for scheduling ray traversal. With the addition of our compact fixed function traversal unit and cache mechanism, we show that current GPU architectures are well suited for hardware accelerated ray tracing, requiring only small modifications to provide high performance. By making use of existing GPU resources we are able to keep all rays and scheduling traffic on chip and out of caches. We used simulations to estimate the performance of our architecture. Our system achieves an average ray rate of 3.4 billion rays per second while path tracing our test scenes.