CONF heusch:AFGR:2006/IDIAP Local Binary Patterns as an Image Preprocessing for Face Authentication Heusch, Guillaume Rodriguez, Yann Marcel, Sébastien EXTERNAL http://publications.idiap.ch/attachments/papers/2006/heusch-AFGR-2006.pdf PUBLIC http://publications.idiap.ch/index.php/publications/showcite/heusch:rr05-76 Related documents IEEE Int. Conf. on Automatic Face and Gesture Recognition (AFGR) 2006 9-14 IDIAP-RR 05-76 One of the major problem in face authentication systems is to deal with variations in illumination. In a \mbox{realistic} scenario, it is very likely that the lighting conditions of the probe image does not correspond to those of the gallery image, hence there is a need to handle such variations. In this work, we present a new preprocessing algorithm based on Local Binary Patterns (LBP): a texture representation is derived from the input face image before being forwarded to the classifier. The efficiency of the proposed approach is empirically demonstrated using both an appearance-based (LDA) and a feature-based (HMM) face authentication systems on two databases: BANCA and XM2VTS (with its darkened set). Conducted experiments show a significant improvement in terms of verification error rates and compare to results obtained with state-of-the-art preprocessing techniques. REPORT heusch:rr05-76/IDIAP Local Binary Patterns as an Image Preprocessing for Face Authentication Heusch, Guillaume Rodriguez, Yann Marcel, Sébastien EXTERNAL http://publications.idiap.ch/attachments/reports/2005/heusch-idiap-rr-05-76.pdf PUBLIC Idiap-RR-76-2005 2005 IDIAP One of the major problem in face authentication systems is to deal with variations in illumination. In a \mbox{realistic} scenario, it is very likely that the lighting conditions of the probe image does not correspond to those of the gallery image, hence there is a need to handle such variations. In this work, we present a new preprocessing algorithm based on Local Binary Patterns (LBP): a texture representation is derived from the input face image before being forwarded to the classifier. The efficiency of the proposed approach is empirically demonstrated using both an appearance-based (LDA) and a feature-based (HMM) face authentication systems on two databases: BANCA and XM2VTS (with its darkened set). Conducted experiments show a significant improvement in terms of verification error rates and compare to results obtained with state-of-the-art preprocessing techniques.