Implementation of Kd-Trees on the GPU to Achieve Real Time Graphics Processing
University of Minnesota
UMM CSci Senior Seminar Conference Morris, 2011
@article{martin2011implementation,
title={Implementation of Kd-Trees on the GPU to Achieve Real Time Graphics Processing},
author={Martin, W.W.},
year={2011}
}
This paper examines the parallelization of ray tracing algorithms with the goal of running the whole process on the graphics processing unit (GPU) rather than the central processing unit (CPU). The motivation behind this endeavour is to utilize the massively parallel nature of the GPU. This parallelism allows the construction of 3-dimensional images to take place in real time. To achieve this we focus on how to create and process multi-dimensional tree structures (kd-trees) to model image data. Kd-trees organize multi-dimensional data in a searchable data structure, lending itself to the efficient creation of lighting effects.
January 8, 2012 by hgpu
