c © Copyright by Haiping Lu, 2008 Multilinear Subspace Learning for Face and Gait Recognition

Face and gait recognition problems are challenging due to largely varying appear-ances, highly complex pattern distributions, and insufficient training samples. This dis-sertation focuses on multilinear subspace learning for face and gait recognition, where low-dimensional representations are learned directly from tensorial face or gait objects. This research introduces a unifying multilinear subspace learning framework for sys-tematic treatment of the multilinear subspace learning problem. Three multilinear pro-jections are categorized according to the input-output space mapping as: vector-to-vector projection, tensor-to-tensor projection, and tensor-to-vector projection. Techniques for subspace learning from tensorial data are then proposed and analyzed. Multilinear prin-cipal component analysis (MPCA) seeks a tensor-to-tensor projection that maximizes the variation captured in the projected space, and it is further combined with linear dis-criminant analysis and boosting for better recognition performance. Uncorrelated MPCA (UMPCA) solves for a tensor-to-vector projection that maximizes the captured variation in the projected space while enforcing the zero-correlation constraint. Uncorrelated mul