A Novel Supervised Multiset Integrated Canonical Correlation Analysis for Multi-feature Fusion and Recognition

Multiset integrated canonical correlation analysis (MICCA) has been employed as a powerful tool for multiple feature extraction and it can distinctly express the integral correlation among multi-group feature. However, MICCA is the unsupervised feature extraction and it does not include the class information of the samples, resulting in the constraint of the recognition performance. In this paper, the class information is incorporated into the framework of MICCA, and the novel supervised method is presented for multi-view dimensionality reduction and recognition, called generalized multiset integrated canonical correlations (GMICC). Extensive experimental results on Extended Yale B and AT&T face images databases and COIL-20 object database show that the proposed method is more effective and discriminative than the existing methods.