Pose Invariant Face Recognition Using DT-CWT Partitioning and KPCA
In this paper the suitability of Dual Tree Complex Wavelet Transform for pose invariant Face Recognition is studied and a feature extraction frame work is proposed. This proposed framework will aid in design of Face Recognition system to address the challenging issue like Pose Variation. In contrast to the discrete wavelet Transform (DWT) the design of Dual Tree Complex Wavelet Transform is rugged to shift Invariance and poses good directional properties. These features of DT-CWT motivated to study their suitability for Face Feature Extraction, as the features of face are oriented in different directions. In this proposed frame work the Image is decomposed using DT-CWT and the features are extracted from low frequency band using Kernel Principal Component analysis (KPCA). To show the performance, the proposed method is tested on ORL Database. Satisfactory results are obtained using proposed method compared to existing state of art.
