Multi-View Cosine Similarity Learning with Application to Face Verification

An instance can be easily depicted from different views in pattern recognition, and it is desirable to exploit the information of these complement each other. However, most metric learning or similarity methods are developed for single-view feature representation over past two decades, which not suitable dealing with multi-view data directly. In this paper, we propose a cosine (MVCSL) approach efficiently utilize apply face verification. The proposed MVCSL method able leverage both common private view, jointly learns view transformed subspace integrates similarities all unified framework. Specifically, employs constraints that joint positive pairs greater than negative pairs. Experiments on fine-grained verification kinship tasks demonstrate superiority our approach.