Institute of Electrical & Electronic Engineers
Graphics Processing Units (GPUs) are capable of achieving remarkable performance improvements for a broad range of applications. However, they have not been widely adopted in embedded systems and mobile devices as accelerators mainly due to their relatively higher power consumption compared with embedded microprocessors. In this paper, the authors conduct a comprehensive analysis regarding the feasibility and potential of accelerating applications using GPUs in low-power domains. They use two different categories of benchmarks: the Level 3 BLAS subroutines, and the computer vision algorithms, i.e., mean shift image segmentation and Scale-Invariant Feature Transform (SIFT).