Automatic Face Recognition Using Local Invariant Features – A Scale Space Approach

We propose automatic face recognition system based on multi-scale Harris-Laplace detector, robust to the illumination, orientation, scale and translation variations. A face region is segmented from given image by face segmentation algorithm. A segmented face region is represented by a set of interest points, detected by Harris-Laplace detector, to achieve scale invariance. Each interest point is described by a local feature vector, based on the Gabor wavelet, extracted at various scales and orientations to achieve rotation invariance. The face recognition is performed by voting algorithm. Comparison of results with existing algorithms validates the usefulness of local invariant features for face recognition at less computational cost and less storage requirements.