DocumentCode :
3315506
Title :
Simplified Static Image Face Detection Algorithm
Author :
Zaidi, Syed Ali Raza ; Zaidi, S.M.H. ; Samad, Kashan ; Iqbal, A.A.
Author_Institution :
NUST Institute of Information Technology, Rawalpindi, Pakistan, 47zaidi@niit.edu.pk
fYear :
2005
fDate :
27-28 Aug. 2005
Firstpage :
343
Lastpage :
343
Abstract :
Although there are thousands of algorithms available for face extraction from static images based on histogram intersection technique or feature based face detection but these are still complex and it is not a trivial task to implement these algorithm, as a matter of fact most programmers regard these techniques as time chewing monsters because most of newly available programming languages does not provide any specific support for achieving the task. Hence we require a simplified algorithm, which can easily operate under normal circumstances and can easily be implemented by a programmer having technical knowledge instead of getting in to very coarse-grained mathematical details. The algorithm we discuss in this paper is some how uses histogram technique but it is not the purely dependent on histogram intersection or some other technique. There are some pros and cons of every system and thus this system is also not hundred percent perfect, how ever over the time we can develop a system with capability of detecting multiple faces in a single image and naturally combining couple of more algorithms, we can also achieve 3D face detection. In this paper a prototype is implemented using Microsoft GDI+ within limitations of managed and safe code.
Keywords :
Assembly; Bandwidth; Communication system traffic control; Computer languages; Computer vision; Data mining; Delay; Diffserv networks; Face detection; Histograms; Information technology; Programming profession; Prototypes; Round robin; Scheduling algorithm; Telecommunication traffic; Traffic control;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Information and Communication Technologies, 2005. ICICT 2005. First International Conference on
Print_ISBN :
0-7803-9421-6
Type :
conf
DOI :
10.1109/ICICT.2005.1598618
Filename :
1598618
Link To Document :
بازگشت