Face tracking using skin-color model and Gabor wavelets

This paper describes an approach for real-time segmentation and tracking of human faces in color video sequences. First, a skin-color model is selected and used to segment face-like regions in a color image with complex background. Second, facial features are extracted by Gabor wavelets. Third, a face model is constructed for verifying face-like regions and tracking faces. Our experimental results show that the proposed approach can be used to track faces robustly under cluttered background.