The need to assure the safety of places, people and information is particularly compelling nowadays. A reliable and widely accepted solution is the use of biometric systems and of face recognition in particular, thanks to its high degree of acceptance and ease of use. Most techniques presented in literature work on the whole face and often omit to consider the occlusion problem. This work proposes two classifiers, suitably adjusted to be Component Based and robust to occlusions. The experiments that we carried out show how these approaches perform well even in presence of partial occlusions.
Component Based Face Recognition Against Occlusions: Gabor Wavelets vs. GFD / ABATE A., F; Nappi, M; Riccio, D; DE MARSICO, Maria. - ELETTRONICO. - (2006), pp. 92-97. (Intervento presentato al convegno International Conference on Distributed Multimedia Systems tenutosi a Grand Canyon - USA nel August 30, September 1, 2006).
Component Based Face Recognition Against Occlusions: Gabor Wavelets vs. GFD
DE MARSICO, Maria
2006
Abstract
The need to assure the safety of places, people and information is particularly compelling nowadays. A reliable and widely accepted solution is the use of biometric systems and of face recognition in particular, thanks to its high degree of acceptance and ease of use. Most techniques presented in literature work on the whole face and often omit to consider the occlusion problem. This work proposes two classifiers, suitably adjusted to be Component Based and robust to occlusions. The experiments that we carried out show how these approaches perform well even in presence of partial occlusions.I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.