A fast and accurate face detector for indexation of face images

New fast normalized neural networks for pattern detection

“…Furthermore, another idea to increase the speed of these fast neural networks through image decomposition is presented. Moreover, the problem of sub-image (local) normalization in the Fourier space which presented in [4] is solved. The number of computation steps required for weight normalization is proved to be less than that needed for image normalization.…”

“…Its reliability and performance have a major influence in a whole pattern recognition system. Nowadays, neural networks have shown very good results for detecting a certain pattern in a given image [2,4,6,[8][9][10]12]. Among other techniques [3,5,7], neural networks are efficient pattern detectors [2,4,6,9].…”

Design of High Speed Neural Networks for Fast Pattern Detection by using Cross Correlation and Matrix Decomposition

“…Nowadays, neural networks have shown very good results for detecting a certain pattern in a given image (Rowley et al 1998;Feraud et al 2000;Anifantis et al 1999;Lang et al 1988;El-Bakry 2001). Among other techniques (Schneiderman & Kanade 1998;Zhu et al 2000;Srisuk & Kurutach 2002;Bao et al 2006), neural networks are efficient pattern detectors (Rowley et al 1998;Feraud et al 2000;ElBakry 2002,a;El-bakry 2002,b;Essannouni and Ibn Elhaj 2006;Roth et al 2006;Ramasubramanian & Kannan 2006). But the problem with neural networks is that the computational complexity is very high because the networks have to process many small local windows in the images (Zhu et al 2000;Srisuk & Kurutach 2002;Yang et al 2002).…”

A symmetric transformation for LDA-based face verification

“…Feature extraction 3.1.1. Face modeling In a real application, the face bounding box will be provided by an accurate face detector [13,6] but here the bounding box is computed using manually located eyes coordinates, assuming a perfect face detection. In this paper, the face bounding box is determined using face/head anthropometry measures [5] according to a face model ( The face bounding box w/h crops the face from the glabella to the subnasale and do not includes the ears in order to minimize the influence of the hair-cut and of the lip movement.…”

“…The 12 sessions were separated into 3 different scenarios (Fig. 4): controlled (for sessions 1-4), degraded (for sessions [5][6][7][8], and adverse (for sessions 9-12).…”