Face Recognition using Approximate Arithmetic

Karol Marso, Alberto Nannarelli (Supervisor)

Research output: Book/ReportReportResearch

1120 Downloads (Pure)

Abstract

Face recognition is image processing technique which aims to identify human faces and found its use in various different fields for example in security. Throughout the years this field evolved and there are many approaches and many different algorithms which aim to make the face recognition as effective as possible. The use of different approaches such as neural networks and machine learning can lead to fast and efficient solutions however, these solutions are expensive in terms of hardware resources and power consumption. A possible solution to this problem can be use of approximate arithmetic. In many image processing applications the results do not need to be completely precise and use of the approximate arithmetic can lead to reduction in terms of delay, space and power consumption. In this paper we examine possible use of approximate arithmetic in face recognition using Eigenfaces algorithm.
Original languageEnglish
PublisherDTU Compute
Number of pages25
Publication statusPublished - 2018
SeriesDTU Compute Technical Report-2018
Volume02

Cite this

Marso, K., & Nannarelli, A. (2018). Face Recognition using Approximate Arithmetic. DTU Compute. DTU Compute Technical Report-2018, Vol.. 02