Optimizing Gaussian filtering of volumetric data using SSE

Abstract

SUMMARYGaussian filtering is a basic operation commonly used in numerous image and volume processing algorithms. It is, therefore, desirable to perform it as efficiently as possible. Over the last decade CPUs have been successfully extended with several SIMD (Single Instruction Multiple Data) extensions, such as MMX, 3DNow!, and SSE series. In this paper we introduce a new technique for Gaussian filtering of volume data sets-the extended volume-together with its SIMD implementation using the SSE technology. We further introduce a SIMD optimized recursive IIR implementation of the Gaussian filter, and finally, we parallelize the SSE versions with the help of OpenMP (Open Multi-Processing). Experimental evaluation indicates that the SIMD implementation can significantly speed up both versions of the Gaussian filtering and that the non-recursive extended volume version is faster than the recursive IIR one for small widths of the Gaussian filter.

Optimizing 3D Convolutions for Wavelet Transforms on CPUs with SSE Units and GPUs

“…For works considering 3D convolutions with vectorization, we can cite Intel [11] but the filter is very small (3x3x3) and the data limited to 16bit data. Gaussian 3D filtering has also been ported to SSE [12], but in this case the filters are symmetric and the authors limit the use of buffers. As for 3D convolutions on GPU, [13] use a 2D+1D method but for very small filter size.…”