Ray Traversal Research Articles

Sorted Deferred Shading for Production Path Tracing

AbstractRay‐traced global illumination (GI) is becoming widespread in production rendering but incoherent secondary ray traversal limits practical rendering to scenes that fit in memory. Incoherent shading also leads to intractable performance with production‐scale textures forcing renderers to resort to caching of irradiance, radiosity, and other values to amortize expensive shading. Unfortunately, such caching strategies complicate artist workflow, are difficult to parallelize effectively, and contend for precious memory. Worse, these caches involve approximations that compromise quality. In this paper, we introduce a novel path‐tracing framework that avoids these tradeoffs. We sort large, potentially out‐of‐core ray batches to ensure coherence of ray traversal. We then defer shading of ray hits until we have sorted them, achieving perfectly coherent shading and avoiding the need for shading caches.

WE‐C‐108‐06: Accelerating Collapsed Cone Convolution/Superposition Dose Calculation On GPU Using Spatial Decomposition

Purpose: The performance of all known Collapsed Cone Convolution/Superposition (CCCS) dose calculation methods on Graphics Processing Units (GPUs) is bounded by memory throughput. One major cause of this bottleneck is the traditional uniform grid representation of the density volume for human body, which generally contains many homogeneous areas. The common ray traversal process of CCCS within a homogeneous area in a uniform grid repeatedly reads the same density information voxel by voxel, hence increasing the memory accessing workloads. We propose a new spatial decomposition based data structure, called Shell, to improve the CCCS performance on GPU. Our idea is to group voxels in homogeneous areas into rectangular regions such that the density of each such homogeneous region is acquired by one read. An efficient ray traversal method across different regions is developed to reduce the number of memory accesses. Methods: We use a spatial decomposition (e.g., octree) algorithm to adaptively partition a uniform grid density volume into a set of approximately homogeneous regions. Our Shell data structure consists of these regions, each containing its average density and pointers for its neighboring regions. The memory addresses of such pointers are associated with the voxels lying on the region boundaries, which allow any ray traversing in a region to efficiently find the next neighboring region to traverse using table lookup schemes. Results: We integrated our method with a CCCS program and tested using various clinical cases on NVIDIA GTX570 GPU. The program using Shell runs 1.32X to 1.44X faster than using a uniform grid, with less than 2% variance in the dose calculation results. Conclusion: The results show our proposed method effectively reduces the memory bottleneck and improves the CCCS performance on GPU. There is a trade off between the speedup and dose accuracy, which is determined by the spatial decomposition methods. The research of D.Z. Chen was supported in part by NSF under Grants CCF‐0916606 and CCF‐1217906.

ElVis: A System for the Accurate and Interactive Visualization of High-Order Finite Element Solutions.

This paper presents the Element Visualizer (ElVis), a new, open-source scientific visualization system for use with high-order finite element solutions to PDEs in three dimensions. This system is designed to minimize visualization errors of these types of fields by querying the underlying finite element basis functions (e.g., high-order polynomials) directly, leading to pixel-exact representations of solutions and geometry. The system interacts with simulation data through runtime plugins, which only require users to implement a handful of operations fundamental to finite element solvers. The data in turn can be visualized through the use of cut surfaces, contours, isosurfaces, and volume rendering. These visualization algorithms are implemented using NVIDIA's OptiX GPU-based ray-tracing engine, which provides accelerated ray traversal of the high-order geometry, and CUDA, which allows for effective parallel evaluation of the visualization algorithms. The direct interface between ElVis and the underlying data differentiates it from existing visualization tools. Current tools assume the underlying data is composed of linear primitives; high-order data must be interpolated with linear functions as a result. In this work, examples drawn from aerodynamic simulations-high-order discontinuous Galerkin finite element solutions of aerodynamic flows in particular-will demonstrate the superiority of ElVis' pixel-exact approach when compared with traditional linear-interpolation methods. Such methods can introduce a number of inaccuracies in the resulting visualization, making it unclear if visual artifacts are genuine to the solution data or if these artifacts are the result of interpolation errors. Linear methods additionally cannot properly visualize curved geometries (elements or boundaries) which can greatly inhibit developers' debugging efforts. As we will show, pixel-exact visualization exhibits none of these issues, removing the visualization scheme as a source of uncertainty for engineers using ElVis.

Efficient implementation of the 3D‐DDA ray traversal algorithm on GPU and its application in radiation dose calculation

The three-dimensional digital differential analyzer (3D-DDA) algorithm is a widely used ray traversal method, which is also at the core of many convolution∕superposition (C∕S) dose calculation approaches. However, porting existing C∕S dose calculation methods onto graphics processing unit (GPU) has brought challenges to retaining the efficiency of this algorithm. In particular, straightforward implementation of the original 3D-DDA algorithm inflicts a lot of branch divergence which conflicts with the GPU programming model and leads to suboptimal performance. In this paper, an efficient GPU implementation of the 3D-DDA algorithm is proposed, which effectively reduces such branch divergence and improves performance of the C∕S dose calculation programs running on GPU. The main idea of the proposed method is to convert a number of conditional statements in the original 3D-DDA algorithm into a set of simple operations (e.g., arithmetic, comparison, and logic) which are better supported by the GPU architecture. To verify and demonstrate the performance improvement, this ray traversal method was integrated into a GPU-based collapsed cone convolution∕superposition (CCCS) dose calculation program. The proposed method has been tested using a water phantom and various clinical cases on an NVIDIA GTX570 GPU. The CCCS dose calculation program based on the efficient 3D-DDA ray traversal implementation runs 1.42 ∼ 2.67× faster than the one based on the original 3D-DDA implementation, without losing any accuracy. The results show that the proposed method can effectively reduce branch divergence in the original 3D-DDA ray traversal algorithm and improve the performance of the CCCS program running on GPU. Considering the wide utilization of the 3D-DDA algorithm, various applications can benefit from this implementation method.

Visibility driven BVH build up algorithm for ray tracing

The minimization of traversal cost using surface area heuristic is extensively used to build high quality spatial subdivisions and bounding volume hierarchies for ray tracing. Despite the fair performance of trees built with the cost model, it is known that the underlying assumptions for surface area heuristics are not realistic. In this paper we show how the cost function of the surface area heuristic can be improved on using the assumed visibility of geometric primitives such as triangles. This way the build algorithm utilizes the exact or assumed visibility to construct more efficient BVHs by traversing smaller portion of the hierarchy. We show that by these inexpensive modifications to the cost function we can speed up the ray traversal by approximately 102% on average for path tracing of highly occluded scenes compared to standard surface area heuristics. Moreover, it is also possible to lower the construction time and memory usage by subdividing only those parts of the animated scene through which rays are expected to be traversed.

An Efficient CPU-GPU Implementation of the Multiple Absorption Coefficient Zonal Method (MACZM)

The multiple absorption coefficient zonal method (MACZM) is an efficient radiative heat transfer modeling method in nonisothermal inhomogeneous media. The method is of high interest for dynamic applications because of its ability to asses semitransparent radiative heat transfer in very short computation time. In this work, an efficient algorithm for MACZM is implemented. A connectivity control study is presented for taking into account the connectivity considerations required by the method. An identified ray traversal algorithm corresponding to part of the MACZM implementation is then selected among three different approaches presented in the article, based on well-known ray traver sal algorithms, the 6-tripod line algorithm and the 6-parametric line algorithm. On the other hand, the MACZM is highly parallel and is implemented in CUDA, a parallel computing architecture that enables easy use of a powerful graphics processing unit (GPU). An efficient implementation is discussed consisting of an optimal solution for exploiting the method parallelism (threading) and the use of the memory resources available on the GPU. Speed-ups going from 300 to 600 times are achieved, using a NVIDIA Tesla C 1060 GPU and an Intel Xeon CPU E5507 at 2.27 GHz. Radiative heat transfer is then simulated in a steel reheating furnace using the optimized GPU implementation. The computation time is further reduced by using a multigrid approach.

Two‐Level Grids for Ray Tracing on GPUs

AbstractWe investigate the use of two‐level nested grids as acceleration structure for ray tracing of dynamic scenes. We propose a massively parallel, sort‐based construction algorithm and show that the two‐level grid is one of the structures that is fastest to construct on modern graphics processors. The structure handles non‐uniform primitive distributions more robustly than the uniform grid and its traversal performance is comparable to those of other high quality acceleration structures used for dynamic scenes. We propose a cost model to determine the grid resolution and improve SIMD utilization during ray‐triangle intersection by employing a hybrid packetization strategy. The build times and ray traversal acceleration provide overall rendering performance superior to previous approaches for real time rendering of animated scenes on GPUs.

Micropolygon ray tracing with defocus and motion blur

We present a micropolygon ray tracing algorithm that is capable of efficiently rendering high quality defocus and motion blur effects. A key component of our algorithm is a BVH (bounding volume hierarchy) based on 4D hyper-trapezoids that project into 3D OBBs (oriented bounding boxes) in spatial dimensions. This acceleration structure is able to provide tight bounding volumes for scene geometries, and is thus efficient in pruning intersection tests during ray traversal. More importantly, it can exploit the natural coherence on the time dimension in motion blurred scenes. The structure can be quickly constructed by utilizing the micropolygon grids generated during micropolygon tessellation. Ray tracing of defocused and motion blurred scenes is efficiently performed by traversing the structure. Both the BVH construction and ray traversal are easily implemented on GPUs and integrated into a GPU-based micropolygon renderer. In our experiments, our ray tracer performs up to an order of magnitude faster than the state-of-art rasterizers while consistently delivering an image quality equivalent to a maximum-quality rasterizer. We also demonstrate that the ray tracing algorithm can be extended to handle a variety of effects, such as secondary ray effects and transparency.

Micropolygon ray tracing with defocus and motion blur

We present a micropolygon ray tracing algorithm that is capable of efficiently rendering high quality defocus and motion blur effects. A key component of our algorithm is a BVH (bounding volume hierarchy) based on 4D hyper-trapezoids that project into 3D OBBs (oriented bounding boxes) in spatial dimensions. This acceleration structure is able to provide tight bounding volumes for scene geometries, and is thus efficient in pruning intersection tests during ray traversal. More importantly, it can exploit the natural coherence on the time dimension in motion blurred scenes. The structure can be quickly constructed by utilizing the micropolygon grids generated during micropolygon tessellation. Ray tracing of defocused and motion blurred scenes is efficiently performed by traversing the structure. Both the BVH construction and ray traversal are easily implemented on GPUs and integrated into a GPU-based micropolygon renderer. In our experiments, our ray tracer performs up to an order of magnitude faster than the state-of-art rasterizers while consistently delivering an image quality equivalent to a maximum-quality rasterizer. We also demonstrate that the ray tracing algorithm can be extended to handle a variety of effects, such as secondary ray effects and transparency.

Effficient stackless ray traversal for bounding sphere hierarchies with CUDA

Abstract One of the challenges of GPU-based ray tracing is the hierarchical space indexation structure traversal. Stackless approaches to this problem offer benefits both in terms of maximum representable scene size and performances [5] , [9] . We improve on the current stackless traversal methods by increasing memory access locality and accelerating both the primary and secondary ray shooting through specific optimizations achieved at the cost of additional precomputations. Those supplementary steps can be performed on the host and have their results transferred asynchronously on the GPU while the rendering kernel is running, hence being completely hidden.

Ray Tracing of Dynamic Scenes by Managing Empty Regions in Adaptive Boxes

This paper presents a new ray-tracing technique to efficiently speed up ray traversal in empty regions. It first produces grid cells in a new method,and then aggregates empty cells adaptively into fewer empty boxes,so that much acceleration can be obtained by reducing computation on ray going through empty regions. The new acceleration structure can be constructed in O(n) time,and its storage complexity is also O(n). By such a structure,rays can be traced in O(log n) time,similar to ray tracing via a kd-tree . When this structure is integrated with existing acceleration structures,large dynamic scenes can be efficiently rendered. For instance,a dynamic scene with multiple buddha models in up to 6G triangles can be interactively rendered in high photorealism on a common PC,where rays are traced individually and second rays are also computed.

Accessibility analysis for automatic inspection in CMMs by using bounding volume hierarchies

Accessibility analysis represents one of the most critical tasks in inspection planning. This analysis determines those probe orientations that can touch an inspection point without collision. This paper presents a methodology based on part discretization and the application of space partitioning techniques (kd-tree) in order to reduce the number of intersection tests between probe and part. A STL model has been used for discretizing the inspection part in a set of triangles, which permits the application of the developed system to any type of part, regardless of its shape and its complexity. Likewise, a recursive ray traversal algorithm has been used in order to speed up the traversal of the kd-tree hierarchical structure and to calculate exclusively the intersection of each probe orientation with those part triangles that can potentially interfere with it. In a further step of the analysis, the real geometry of the probe has been considered. Hence, a simplified model has been developed for each probe component (column, head, touch probe, stylus and tip) using different basic geometrical shapes. Finally, collision-free probe orientations are clustered for minimizing the orientation changes during the inspection process. Furthermore, the applied algorithm allows for determining different valid combinations of clusters. The developed system was applied to two example parts in order to prove that this methodology is adequate for the solution of real cases.

Laser scan planning based on visibility analysis and space partitioning techniques

This paper presents a methodology to develop an automatic process planning system applied for scanning parts with free-form surfaces by using a laser stripe system mounted on a coordinate measuring machine (CMM). The part has been modelled using a STL format that permits the automatic recognition of any part surface. The valid orientations of the scanning device are obtained in order to guarantee the visibility of the zone to be scanned and also to be compatible with the constraints imposed by the process. With the aim to speed up the calculation of valid orientations, we apply different methods like space partitioning techniques base on kd-tree as well as back-face culling algorithms. Once the space occupied by the part is partitioned in regions, recursive ray traversal algorithms are used in order to exclusively check for intersection the part triangles of the STL model that can potentially be traversed by each laser beam direction. In order to reduce the scanning time related to laser orientation changes, part triangles must be classified into a set of clusters based on their common visibility orientations. Finally, the scanning paths for each cluster are generated as well as the joining paths between them by taking into consideration depth of field and laser beam width.

Stackless KD‐Tree Traversal for High Performance GPU Ray Tracing

Abstract Significant advances have been achieved for realtime ray tracing recently, but realtime performance for complex scenes still requires large computational resources not yet available from the CPUs in standard PCs. Incidentally, most of these PCs also contain modern GPUs that do offer much larger raw compute power. However, limitations in the programming and memory model have so far kept the performance of GPU ray tracers well below that of their CPU counterparts. In this paper we present a novel packet ray traversal implementation that completely eliminates the need for maintaining a stack during kd‐tree traversal and that reduces the number of traversal steps per ray. While CPUs benefit moderately from the stackless approach, it improves GPU performance significantly. We achieve a peak performance of over 16 million rays per second for reasonably complex scenes, including complex shading and secondary rays. Several examples show that with this new technique GPUs can actually outperform equivalent CPU based ray tracers.

Accelerated regular grid traversals using extended anisotropic chessboard distance fields on a parallel stream processor

Modern graphics processing units (GPUs) are an implementation of parallel stream processors. In recent years, there have been a few studies on mapping ray tracing to the GPU. Since graphics processors are not designed to process complex data structures, it is crucial to explore data structures and algorithms for efficient stream processing. In particular ray traversal is one of the major bottlenecks in ray tracing and direct volume rendering methods. In this work we focus on the efficient regular grid based ray traversals on GPU. A new empty space skipping traversal method is introduced. Our method extends the anisotropic chessboard distance structure and employs a GPU friendly traversal algorithm with minimal dynamic branching. Additionally, several previous techniques have been redesigned and adapted to the stream processing model. We experimentally show that our traversal method is considerably faster and better suited to the parallel stream processing than the other grid based techniques.

View Dependent Layer Sampling: An Approach to Hardware Implementation of Volume Ray Casting

Ray casting is a widely recognized method for high quality volume rendering. It traverses and samples the volume data ray by ray in image space. Traditionally, the algorithm is implemented in CPU on PC platform, resulting in slow speed and poor interactivity. This paper introduces a new technique named View Dependent Layer Sampling (VDLS), which supports a hardware implementation of ray casting by Graphics Processing Unit (GPU). VDLS organizes the ray sampling points into a set of layers which can be efficiently represented by two view dependent geometric buffers as two dynamic textures. Based on the structure of VDLS, the six steps involved in the ray casting algorithm including ray generation, ray traversal, interpolation, classification, shading and composition can be fully accomplished in GPU, taking advantage of its programmability and flexibility. In addition, two speedup techniques exploiting object space and image space coherence are proposed for fast culling of the lapsed rays. Several advanced features regarding illumination and composition are further discussed, with which VDLS is capable of reconfiguring the well-known geometric hardware engine for volume ray casting. The novel approach of GPU supported ray casting can render up to 150 million interpolated, post shaded and composed ray samples per second for perspective view. Experimental results suggest that the proposed framework can be regarded as an alternative for on-the-fly visualization and exploitation of discrete scalar data in medical visualization, physical phenomena simulation and material testing applications.

Efficient coverage mask generation for antialiasing

One challenge for computer graphics and scientific visualization is to develop techniques to more effectively display and analyze the vast amounts of information we are collating, a race we are currently losing. To meet this goal, a key initiative suggests moving away from small-scale, low-resolution displays to immersive, high-fidelity systems. These environments permit higher levels of interactivity and exploration, but require more power and bandwidth than current graphics systems can deliver. The move to larger and/or higher resolution displays will place a greater emphasis on the digitization algorithms we employ for even the simplest geometrical primitives. In this tutorial, we present a technique to describe and digitize the line as a set of runs, runs of runs, runs of runs of runs, and so on, in fact, as any level of runs within the full hierarchy of runs in the digital line. The digitization algorithms we present apply to a broad range of resolutions and applications, including geometric scan conversion, ray traversal, linear mapping and interpolation.

Priority scheduling for networked virtual environments

One challenge for computer graphics and scientific visualization is to develop techniques to more effectively display and analyze the vast amounts of information we are collating, a race we are currently losing. To meet this goal, a key initiative suggests moving away from small-scale, low-resolution displays to immersive, high-fidelity systems. These environments permit higher levels of interactivity and exploration, but require more power and bandwidth than current graphics systems can deliver. The move to larger and/or higher resolution displays will place a greater emphasis on the digitization algorithms we employ for even the simplest geometrical primitives. In this tutorial, we present a technique to describe and digitize the line as a set of runs, runs of runs, runs of runs of runs, and so on, in fact, as any level of runs within the full hierarchy of runs in the digital line. The digitization algorithms we present apply to a broad range of resolutions and applications, including geometric scan conversion, ray traversal, linear mapping and interpolation.

A Multiscale Approach to Integrated Volume Segmentation and Rendering

A number of techniques have been proposed for rendering volumetric scalar data sets. Techniques have also been proposed for analyzing the three dimensional information contents of the underlying domain, but traditionally the data analysis part is left as a post‐processing step which only involves the rendered two dimensional images. In this paper, we describe a visualization method for scalar volume data which integrates explicit knowledge of the underlying domain into the rendering process. The key of this approach lies in a hierarchical description of the discrete signal, which is decomposed into a sequence of multiscale representations. We describe a technique for the analysis of structures within the data. This allows for the segmentation and classification of the relevant features and can be used to improve their visual sensation. We also address the problem of accelerating the final rendering pass by integrating the extracted object space information into the ray traversal process.

A Multiscale Approach to Integrated Volume Segmentation and Rendering

A number of techniques have been proposed for rendering volumetric scalar data sets. Techniques have also been proposed for analyzing the three dimensional information contents of the underlying domain, but traditionally the data analysis part is left as a post‐processing step which only involves the rendered two dimensional images. In this paper, we describe a visualization method for scalar volume data which integrates explicit knowledge of the underlying domain into the rendering process. The key of this approach lies in a hierarchical description of the discrete signal, which is decomposed into a sequence of multiscale representations. We describe a technique for the analysis of structures within the data. This allows for the segmentation and classification of the relevant features and can be used to improve their visual sensation. We also address the problem of accelerating the final rendering pass by integrating the extracted object space information into the ray traversal process.