OLBVH: octree linear bounding volume hierarchy for volumetric meshes
- PDF / 1,490,705 Bytes
- 14 Pages / 595.276 x 790.866 pts Page_size
- 8 Downloads / 206 Views
ORIGINAL ARTICLE
OLBVH: octree linear bounding volume hierarchy for volumetric meshes Daniel Ströter1
· Johannes S. Mueller-Roemer2
· André Stork2 · Dieter W. Fellner2,3
© The Author(s) 2020
Abstract We present a novel bounding volume hierarchy for GPU-accelerated direct volume rendering (DVR) as well as volumetric mesh slicing and inside-outside intersection testing. Our novel octree-based data structure is laid out linearly in memory using space filling Morton curves. As our new data structure results in tightly fitting bounding volumes, boundary markers can be associated with nodes in the hierarchy. These markers can be used to speed up all three use cases that we examine. In addition, our data structure is memory-efficient, reducing memory consumption by up to 75%. Tree depth and memory consumption can be controlled using a parameterized heuristic during construction. This allows for significantly shorter construction times compared to the state of the art. For GPU-accelerated DVR, we achieve performance gain of 8.4×–13×. For 3D printing, we present an efficient conservative slicing method that results in a 3×–25× speedup when using our data structure. Furthermore, we improve volumetric mesh intersection testing speed by 5×–52×. Keywords Bounding volume hierarchy · GPGPU · Volumetric meshes · Direct volume rendering · Intersection detection · Slicing
1 Introduction Bounding volume hierarchies (BVHs) and spatial data structures in general are indispensable tools in computer graphics. They are used to accelerate a multitude of algorithms, including collision detection, frustum culling, and ray tracing. Due to the low cost and high performance of massively parallel manycore graphics processing units (GPUs), BVHs that can be efficiently constructed and traversed on the GPU are of particular interest in GPU-accelerated physically based animation and ray tracing. However, current research focuses on GPU BVHs for triangular surface meshes or point clouds (see Sect. 2). BVHs for volumetric meshes, e.g., tetrahedral meshes used in finite element simulations, must have different characteristics to achieve high performance. For example, volumetric meshes
B
Daniel Ströter [email protected]
1
Technische Universität Darmstadt, 64277 Darmstadt, Germany
2
Fraunhofer IGD and Technische Universität Darmstadt, Darmstadt, Germany
3
Technische Universität Graz, Graz, Austria
generally fill a space more densely than surface meshes and therefore lead to deeper hierarchies. Volumetric meshes are widely used in physically based animation, computational physics, and scientific visualization of simulation results (see Fig. 1). Furthermore, volumetric meshes are advantageous for the description of 3D printed models, as they enable the description of material gradients (see, e.g., Altenhofen et al. [2]). In this paper, we examine how construction and traversal performance as well as memory use can be improved by adapting previous GPU-optimized BVHs for volumetric meshes. We introduce a novel octree-based l
Data Loading...
