Accelerating space variant Gaussian filtering on graphics processing unit

Abstract

In this paper we examine the performance advantages of using a CPU to execute the space variant Gaussian filtering. Our results show that the straightforward convolution GPU implementation obtains up to 8 times better performance than the best recursive algorithm (the Deriche's filter) executed on a CPU, for useful maximum a values. GPUs have turned out a useful option to obtain high execution performance, specially due to the emergence of high level languages for graphics hardware.