Dual-Cross Patterns with RPCA of Key Frame for Facial Micro-expression Recognition

摘要

Fast and discriminative feature extraction has always been a critical issue for spontaneous micro-expression recognition applications. In this paper, a micro-expression analysis framework based on new facial representation is proposed. Firstly, to remove redundant information in the micro-expression video sequences, the key frame is adaptively selected on the criteria of structural similarity index (SSIM) between different face images. Then, robust principal component analysis (RPCA) obtains the sparse information of the Key frame, which not only retains the expression attributes of the micro-expression sequence, but also eliminates useless information. Furthermore, we use Dual-cross patterns (DCP) to extract features of sparse key frame. Repeated comparison experiments were performed on the SMIC database to evaluate the performance of the method. Experimental results demonstrate that our proposed method achieves promising performance for micro-expression recognition.