Mapping Vision Algorithms on SIMD Architecture Smart Cameras

SIMD (single-instruction multiple-data) processors have demonstrated high performance for vector-based image processing, thereby facilitating real-time vision applications. However, to fully exploit the advantages of the SIMD architecture, implementation of a given vision algorithm needs to undergo a mapping from a general purpose CPU programming style to a pixel parallel style. This paper describes how part of a given gesture analysis algorithm is mapped on a smart camera with the SIMD processor to achieve real-time operation. The pixel parallel nature of the SIMD processing restricts diversified treatment of pixels. Therefore, in this paper we show how to modify the algorithm and discuss improvements in the architecture in order to achieve the intended functionality. Mapping of background removal, segmentation, and labeling functions is described. We also discuss robustness issues since the mapping to smart cameras aims for practical, real-time applications.