Uniform grids are a spatial subdivision acceleration structure well suited for ray tracing. They are known for their fast build times and ease of use, but suffer from slow traversals in the presence of empty space. To address this issue, we present macro 64-regions, a new GPU based approach for finding and storing empty volumes in an underlying uniform grid. This allows for fast traversals through regions that do not contain scene geometry. Further, unlike previous solutions to this problem, we do not store a hierarchical structure and therefore the traversal steps are simplified. Because macro 64-regions are dependent on an underlying grid, we also introduce an improvement in the grid construction process. Our method does not rely on sorti...
We investigate the use of two-level nested grids as acceleration structure for ray tracing of dynami...
Domain decomposition based on spatial locality is a classical data-parallel problem whose solution m...
We propose a new massively parallel algorithm for constructing high-quality bounding volume hierarch...
Uniform grids are a spatial subdivision acceleration structure well suited for ray tracing. They are...
We present a fast, parallel GPU algorithm for construction of uni-form grids for ray tracing, which ...
In this thesis we investigate the use of uniform grids as acceleration structures for ray tracing on...
Commonly used hierarchical data structures such as bounding volume hierarchies and kd-trees have rat...
The focus of research in acceleration structures for ray tracing recently shifted from render time t...
We investigate the use of two-level nested grids as acceleration structure for ray tracing of dynami...
We present a novel approach for GPU-based high quality volume rendering of large out-of-core volume ...
The focus of research in acceleration structures for ray tracing recently shifted from render time t...
Ray tracing is the only image synthesis technique which allows the rendering of specular effects. In...
Commonly used hierarchical data structures such as bounding volume hierarchies and kd-trees have rat...
Ray tracing on GPUs is an area of ongoing research. GPUs are well suited for this parallel rendering...
Regular grid spatial subdivision is frequently used for fast ray tracing of scanned models. Scanned ...
We investigate the use of two-level nested grids as acceleration structure for ray tracing of dynami...
Domain decomposition based on spatial locality is a classical data-parallel problem whose solution m...
We propose a new massively parallel algorithm for constructing high-quality bounding volume hierarch...
Uniform grids are a spatial subdivision acceleration structure well suited for ray tracing. They are...
We present a fast, parallel GPU algorithm for construction of uni-form grids for ray tracing, which ...
In this thesis we investigate the use of uniform grids as acceleration structures for ray tracing on...
Commonly used hierarchical data structures such as bounding volume hierarchies and kd-trees have rat...
The focus of research in acceleration structures for ray tracing recently shifted from render time t...
We investigate the use of two-level nested grids as acceleration structure for ray tracing of dynami...
We present a novel approach for GPU-based high quality volume rendering of large out-of-core volume ...
The focus of research in acceleration structures for ray tracing recently shifted from render time t...
Ray tracing is the only image synthesis technique which allows the rendering of specular effects. In...
Commonly used hierarchical data structures such as bounding volume hierarchies and kd-trees have rat...
Ray tracing on GPUs is an area of ongoing research. GPUs are well suited for this parallel rendering...
Regular grid spatial subdivision is frequently used for fast ray tracing of scanned models. Scanned ...
We investigate the use of two-level nested grids as acceleration structure for ray tracing of dynami...
Domain decomposition based on spatial locality is a classical data-parallel problem whose solution m...
We propose a new massively parallel algorithm for constructing high-quality bounding volume hierarch...