Stable and compact face recognition via unlabeled data driven sparse representation-based classification

Sparse representation-based classification (SRC) has attracted much attention by casting the recognition problem as simple linear regression problem. SRC methods, however, still is limited to enough labeled samples per category, insufficient use of unlabeled samples, and instability representation. For tackling these problems, an data driven inverse projection pseudo-full-space model proposed with low-rank sparse constraints. The aims mine hidden semantic information intrinsic structure all available data, which suitable for few proportion imbalance between problems in frontal face recognition. mixed Gauss–Seidel Jacobian ADMM algorithm introduced solve model. convergence, representation capability stability are analyzed. Experiments on three public datasets show that LR-S-PFSRC achieves stable results, especially samples. • An projection-based small sample imbalanced data. solved a algorithm.