Browsing by Author "Fellner, Dieter W."

Now showing 1 - 4 of 4
  • Thumbnail Image
    Item
    A GPU Ray Tracing Implementation for Triangular Grid Primitives
    (The Eurographics Association, 2023) Buelow, Max von; Kuijper, Arjan; Fellner, Dieter W.; Abey Campbell; Claudia Krogmeier; Gareth Young
    Triangular grid primitives are a new technique for more efficient handling of memory intensive meshes, also called micro meshes in recent proprietary hardware implementations. This makes it a technique with high potential in the area of virtual environments where hardware capabilities are typically limited. In this poster, we focus on software ray tracing on GPUs and present a novel, easy-to-implement approach that uses a two-level bounding volume hierarchy (BVH) to accelerate these grids. The primary goal of our work is to make the technology more accessible by focusing on standard GPU devices without hardware ray tracing units. With our approach, we are able to encode geometry and BVH with approximately 7.5 bytes per triangle, reducing standard representations by a factor of 3.73 while reducing BVH construction time. Our data structure achieves a peak performance impact of 16% for a three-level subdivision.
  • Thumbnail Image
    Item
    GPU-Parallel Constant-Time Limit Evaluation of Catmull-Clark Solids
    (The Eurographics Association, 2021) Besler, Sebastian; Altenhofen, Christian; Stork, André; Fellner, Dieter W.; Andres, Bjoern and Campen, Marcel and Sedlmair, Michael
    Subdivision solids, such as Catmull-Clark (CC) solids, are versatile volumetric representation schemes that can be employed for geometric modeling, physically based simulation, and multi-material additive manufacturing. With volumetric limit evaluation still being the performance bottleneck for these applications, we present a massively parallel approach to Altenhofen et al.'s constant-time limit evaluation method for CC solids. Our algorithm exploits the computational power of modern GPUs, while maintaining the mathematical concepts of Altenhofen et al.'s method. Distributing the computations for a single cell across multiple streaming multiprocessors (SMs) increases the utilization of the GPU's resources compared to straightforward parallelization. Specialized compute kernels for different topological configurations optimize shared memory usage and memory access. Our hybrid approach dynamically chooses the best kernel based on the topology and the evaluation parameters, resulting in speedups of between 5.75x and 61.58x compared to a CPU-parallel implementation of Altenhofen et al.'s method.
  • No Thumbnail Available
    Item
    Real-time Indexing of Point Cloud Data During LiDAR Capture
    (The Eurographics Association, 2022) Bormann, Pascal; Dorra, Tobias; Stahl, Bastian; Fellner, Dieter W.; Peter Vangorp; Martin J. Turner
    We introduce a software system that is capable of indexing point cloud data in real-time as it is being captured by a LiDAR (Light Detection and Ranging) sensor. Our system extends the popular MNO (modifiable nested octree) structure so that it can be built progressively without knowing the bounding box of the point cloud. Using a task-based parallel algorithm incoming points are continuously processed and distributed to the octree nodes using grid-based sampling. Different task priority functions enable prioritization of either high point throughput or low latency. We provide a reference implementation of this system and evaluate it using both a synthetic and a real-world test scenario. The synthetic test demonstrates good scalability up to 16 threads, with maximum point throughputs of up to 1.8 million points per second. These numbers are verified on a sensor system using a Velodyne VLP-16 LiDAR sensor, where our system is able to index all data produced by the scanner in real-time.
  • Thumbnail Image
    Item
    Reconstructing Bounding Volume Hierarchies from Memory Traces of Ray Tracers
    (The Eurographics Association, 2022) Buelow, Max von; Stensbeck, Tobias; Knauthe, Volker; Guthe, Stefan; Fellner, Dieter W.; Yang, Yin; Parakkat, Amal D.; Deng, Bailin; Noh, Seung-Tak
    The ongoing race to improve computer graphics leads to more complex GPU hardware and ray tracing techniques whose internal functionality is sometimes hidden to the user. Bounding volume hierarchies and their construction are an important performance aspect of such ray tracing implementations. We propose a novel approach that utilizes binary instrumentation to collect memory traces and then uses them to extract the bounding volume hierarchy (BVH) by analyzing access patters. Our reconstruction allows combining memory traces captured from multiple ray tracing views independently, increasing the reconstruction result. It reaches accuracies of 30% to 45% when comparing against the ground-truth BVH used for ray tracing a single view on a simple scene with one object. With multiple views it is even possible to reconstruct the whole BVH, while we already achieve 98% with just seven views. Because our approach is largely independent of the data structures used internally, these accurate reconstructions serve as a first step into estimation of unknown construction techniques of ray tracing implementations.