Publication Details
Block-based Approach to 2-D Wavelet Transform on GPUs
Bařina David, Ing., Ph.D. (DCGM FIT BUT)
Zemčík Pavel, prof. Dr. Ing. (DCGM FIT BUT)
discrete wavelet transforms, image processing, lifting scheme, graphics card, memory barrier
This paper introduces two novel approaches suitable for 2-D discrete wavelet transform (DWT) computation using the lifting scheme and modern GP-GPUs.
We are introducing a new approach to computation of 2-D discrete wavelet transform on modern GPUs. The proposed approach involves block-based processing enabling one seamless transform even for high resolution input data. Inside the blocks, two distinct methods can be used - either separable or non-separable 2-D lifting scheme. Furthermore, we compare the proposed approach under different conditions to the best of the existing methods, whereas our approach consistently outperforms the others. Our methods are implemented in the OpenCL framework and tested on a wide range GPUs from NVIDIA as well as AMD.
@INPROCEEDINGS{FITPUB11054,
author = "Michal Kula and David Ba\v{r}ina and Pavel Zem\v{c}\'{i}k",
title = "Block-based Approach to 2-D Wavelet Transform on GPUs",
pages = "643--653",
booktitle = "Information Technology: New Generations",
series = "Advances in Intelligent Systems and Computing",
volume = 448,
year = 2016,
location = "Las Vegas, Nevada, US",
publisher = "Springer International Publishing",
ISBN = "978-3-319-32467-8",
doi = "10.1007/978-3-319-32467-8\_56",
language = "english",
url = "https://www.fit.vut.cz/research/publication/11054"
}