Title :
An improved algorithm for dynamic face detection
Author :
Goras, B.T. ; Vieriu, R.L.
Author_Institution :
Gheorghe Asachi Tech. Univ., Iaşi, Romania
Abstract :
The purpose of this paper is to present an improved version of Viola & Jones [1] face detection algorithm, characterized by the fact that the weak classifier generation uses two threshold levels, depending on the positive examples distributions only. The algorithm was implemented in Matlab, and is able to dynamically detect faces in 240×320 webcam frames within 100ms.
Keywords :
face recognition; learning (artificial intelligence); AdaBoost; Matlab; dynamic face detection; picture size 240 pixel; picture size 320 pixel; threshold levels; weak classifier generation; webcam frames; Face; Face detection; Feature extraction; Heuristic algorithms; Memory management; Optimization; Testing; Adaboost; Face detection; weak classifier;
Conference_Titel :
Electronics and Telecommunications (ISETC), 2012 10th International Symposium on
Conference_Location :
Timisoara
Print_ISBN :
978-1-4673-1177-9
DOI :
10.1109/ISETC.2012.6408062