Abstract
In face recognition, there exists significant challenges like scale, pose, illumination and occlusions in images acquired from real-world conditions. In this work, to cope with these challenges robust, real-time executable, person-independent, component-based, scale and pose invariant a face recognition system has been proposed. In order to align face images, Constrained Local Models (CLM) has been employed. Features have been extracted using Gabor Wavelets from face images aligned with CLM as holistic-based and component-based. After features extraction, the features have been classified by linear Support Vector Machines. Successes of classification acquired using by holistic-based and component-based methods on IMM face database has been evaluated by 5-fold cross-validation and the results have been shown comparatively.