Face Detection on Embedded Systems Abbas Bigdeli 1 , Colin Sim 2 , Morteza Biglari-Abhari 2 and Brian C. Lovell 1 1 Safeguarding Australia Program, NICTA, Brisbane, QLD 4000, Australia {abbas.bigdeli, brian.lovell}@nicta.com.au 2 Department of Electrical and Computer Engineering The University of Auckland, Auckland, New Zealand {csim036, m.abhari}@auckland.ac.nz Abstract. Over recent years automated face detection and recognition (FDR) have gained significant attention from the commercial and research sectors. This paper presents an embedded face detection solution aimed at addressing the real-time image processing requirements within a wide range of applications. As face detection is a computationally intensive task, an embedded solution would give rise to opportunities for discrete economical devices that could be applied and integrated into a vast majority of applications. This work focuses on the use of FPGAs as the embedded prototyping technology where the thread of execution is carried out on an embedded soft- core processor. Custom instructions have been utilized as a means of applying software/hardware partitioning through which the computational bottlenecks are moved to hardware. A speedup by a factor of 110 was achieved from employing custom instructions and software optimizations. 1 Introduction The identification and localization of a face or faces from either an image or video stream is a branch of computer vision known as face detection [1, 2]. Face detection has attracted considerable attention over recent years in part due to the wide range of applications in which it forms the preliminary stage. Some of the main application areas include: human- computer interaction, biometrics, content-based image retrieval systems (CBIRS), video conferencing, surveillance systems, and more recently, photography. The existing visual sensing and computing technologies are at a state where reliable, inexpensive, and accurate solutions for non-intrusive and natural means of human- computer interactions are feasible. Biometrics is an evolving application domain for face detection and is concerned with the use of physiological information to identify and verify a person’s identity. In most cases, face recognition algorithms are designed to operate on images assumed to only contain frontal faces [2]. Therefore, face detection is required to first extract faces from an image prior to the recognition step. Examples of commercial biometric systems are BioID [3] and ViiSage 1 . HumanScan is the company that developed BioID; a multimodal system incorporating voice, lip movement and face recognition to 1 www.viisage.com