A comparative study of multilinear principal component analysis for face recognition

Motivated by the application of the 2D principal component analysis (PCA) for face recognition, this study proposes a modified multilinear PCA method as means to provide higher accuracy with comparable processing time in contrast to the results of contemporary methods. This comparative study includes an assessment of the accuracy and processing time of the independent component analysis (ICA), the kernel PCA (KPCA) and the 2DPCA. The mathematical foundation for evaluating the computational complexity and the memory requirements for feature bases of these methods is provided.