Publication Details

2-D Discrete Wavelet Transform Using GPU

KUČIŠ, M.; BAŘINA, D.; KULA, M.; ZEMČÍK, P. 2-D Discrete Wavelet Transform Using GPU. In 5th Workshop on Application for Multi-Core Architectures. Paris: IEEE Computer Society, 2014. p. 1-6. ISBN: 978-1-4799-7014-8.
Czech title
2D diskrétní vlnková transformace na GPU
Type
conference paper
Language
English
Authors
Kučiš Michal, Ing.
Bařina David, Ing., Ph.D. (DCGM)
Kula Michal, Ing., Ph.D. (DCGM)
Zemčík Pavel, prof. Dr. Ing., dr. h. c. (DCGM)
Keywords

discrete wavelet transform, lifting scheme, graphics card

Abstract

This paper introduces a novel algorithm suitable for 2-D DWT computation using the lifting scheme and OpenCL on GPU.

Annotation

With the wide spread of the discrete wavelet transform, the need for its efficient implementation becomes increasingly important. This work presents an improved version of an algorithm suitable to compute the 2-D discrete wavelet transform on GPU. Depending on the GPU platform, it is suitable to split the 2-D transform computation into separated horizontal and vertical passes. Considering the horizontal passes, we have examined and chosen the best performing method among the already known ones. Furthermore, we have adapted this method for an existing algorithm computing the vertical transform pass. This step helps to reduce several synchronizations and arithmetic operations in the utilized computation scheme. For large data, the proposed vertical method achieves speed-up about 30% compared to the current state of the art methods. In contrast to previously published works, the presented approach is built on the OpenCL parallel programming framework.

Published
2014
Pages
1–6
Proceedings
5th Workshop on Application for Multi-Core Architectures
ISBN
978-1-4799-7014-8
Publisher
IEEE Computer Society
Place
Paris
DOI
10.1109/SBAC-PADW.2014.13
EID Scopus
2-s2.0-84982156104
BibTeX 
@inproceedings{BUT111507,
  author="Michal {Kučiš} and David {Bařina} and Michal {Kula} and Pavel {Zemčík}",
  title="2-D Discrete Wavelet Transform Using GPU",
  booktitle="5th Workshop on Application for Multi-Core Architectures",
  year="2014",
  pages="1--6",
  publisher="IEEE Computer Society",
  address="Paris",
  doi="10.1109/SBAC-PADW.2014.13",
  isbn="978-1-4799-7014-8",
  url="https://www.fit.vut.cz/research/publication/10721/"
}
Files
Back to top