Vector Quantization: A Many-Core Approach
Informatics Engineering Department, School of Technology and Management, Polytechnic Institute of Leiria, Morro do Lena, Alto do Vieiro, 2411-901 Leiria, Portugal
Polytechnic Institute of Leiria, 2013
@article{silva2013vector,
title={Vector Quantization: A Many-Core Approach},
author={Silva, Rita and Marques, Telmo and Desirat, Jorge and Domingues, Patricio},
year={2013}
}
Many-Core computing is an actual growing concept that allows the true parallelization of computational tasks. In the particular case of this paper, the vector quantization algorithm was adapted to the many-core concept with the objective of compressing images encoded in the PGM format. For that, a given sequential implementation of the algorithm was optimized and reengineered to support NVIDIA’s CUDA framework and OpenCL open framework, resulting in two similar implementations of a much faster version of the vector quantization algorithm.
May 21, 2014 by hgpu
