Institute of Electrical & Electronic Engineers
FPGAs are commonly used as execution platforms for signal and image processing applications because they provide a good tradeoff between the programmability of CPUs and DSPs and the performance of ASICs. While the computational power of Field Programmable Gate Arrays (FPGA) makes them attractive as code accelerators, the lack of high-level language programming tools is a major obstacle to their wider use. Graphics Processing Units (GPUs), on the other hand, have benefitted from advanced and widely used high-level programming tools. This paper evaluates the performance, throughput and energy of both FPGAs and GPUs on image processing codes using high-level language programming tools for both.