M. Mohrekesh, S. Azizi and S. Samavi, "Accelerating GPU implementation of contourlet transform," In Proceeding of Iranian Conference on Machine Vision and Image Processing (MVIP), September 2013.

Abstract

The widespread usage of the contourlet-transform (CT) and today’s real-time needs demand faster execution of CT. Solutions are available, but due to lack of portability or computational intensity, they are disadvantageous in real-time applications. In this paper we take advantage of modern GPUs for the acceleration purpose. GPU is well-suited to address data-parallel computation applications such as CT. The convolution part of CT, which is the most computational intensive step, is reshaped for parallel processing. Then the whole transform is transported into GPU to avoid multiple time consuming migrations between the host and device. Experimental results show that with existing GPUs, CT execution achieves more than 19x speedup as compared to its non-parallel CPU-based method. It takes approximately 40ms to compute the transform of a 512×512 image, which should be sufficient for real-time applications.

Keywords—contourlet transform; real-time; convolution; GPU; CUDA.

Type: 
Conference
Upload: 
Date
Year: 
2013
Month: 
SEPTEMBER