10-06-2014, 02:49 PM
MPCA: Multilinear Principal Component Analysis of Tensor Objects
ABSTRACT
Which introduces a multi linear principal component analysis (MPCA) framework for tensor object feature extraction.
determines a multi linear projection onto a tensor subspace of lower dimensionality that captures most of the signal variation.
Even without a fully optimized design, an MPCA-based gait recognition module achieves highly competitive performance and compares favourably to the state-of-the-art gait recognizers.