Section: New Results

High Level Transforms for SIMD and low-level computer vision algorithms

Participants : Lionel Lacassagne, Daniel Etiemble, Alain Dominguez, Pascal Vezolle.

This paper presents a review of algorithmic transforms called High Level Transforms for IBM, Intel and ARM SIMD multi-core processors to accelerate the implementation of low level image processing algorithms. We show that these optimizations provide a significant acceleration. A first evaluation of 512-bit SIMD XeonPhi is also presented. We focus on the point that the combination of optimizations leading to the best execution time cannot be predicted, and thus, systematic benchmarking is mandatory. Once the best configuration is found for each architecture, a comparison of these performances is presented. The Harris points detection operator is selected as being representative of low level image processing and computer vision algorithms. Being composed of five convolutions, it is more complex than a simple filter and enables more opportunities to combine optimizations. The presented work can scale across a wide range of codes using 2D stencils and convolutions. Such High Level Transforms provide a speedup of x89 on a $2\times 4$ core Intel Xeon processor versus a code that is already SIMDized and OPenMPized.[26]