Real Time Object Recognition Using SIFT Algorithm on FPGA

Guven S., ALKAR A. Z.

22nd IEEE Signal Processing and Communications Applications Conference (SIU), Trabzon, Turkey, 23 - 25 April 2014, pp.1503-1506 identifier identifier

  • Publication Type: Conference Paper / Full Text
  • Volume:
  • Doi Number: 10.1109/siu.2014.6830526
  • City: Trabzon
  • Country: Turkey
  • Page Numbers: pp.1503-1506
  • Hacettepe University Affiliated: Yes

Abstract

SIFT (Scale Invariant Feature Transform) is an algorithm commonly used in object recognition applications. The main advantage of SIFT algorithm is its scale and orientation invariance. Processes in SIFT algorithm require high processing power. But these processes can be parallelized which allow high speed increase in algorithm performance. In this report, we discuss the work performed on parallelizing and speeding up of SIFT algorithm using FPGA.