Novel face detection method based on Gabor features

Abstract

Gabor-based Face representation has achieved great success in face recognition, while whether and how it can be applied to face detection is rarely studied. This paper originally investigates the Gabor feature based face detection method, and proposes a coarse-to-fine hierarchical face detector combining the high efficiency of Harr features and the excellent discriminating power of the Gabor features. Gabor features are AdaBoosted to form the final verifier after the cascade of Harr-based AdaBoost face detector. Extensive experiments are conducted on several face databases and verified the effectiveness of the proposed approach. © Springer-Verlag Berlin Heidelberg 2004.