基于多方向特征融合的动态人脸微表情识别方法

Dynamic Facial Micro-expression Recognition Method Based on Multi-directional Feature Fusion

由于当前已有方法未考虑多方向融合动态人脸特征,导致人脸微表情识别效果较差,识别精度较低,识别耗时较长。为此,提出基于多方向特征融合的动态人脸微表情识别方法。估计噪声动态人脸图像,分块处理噪声动态人脸图像,并求解动态人脸图像块系数,对动态人脸图像完成重构。增强去噪后动态人脸图像,通过AAM模型提取动态人脸图像中距离特征信息,对全部特征进行多方向融合,实现动态人脸微表情识别。实验结果表明,所提方法的人脸微表情识别效果较好,能够精准完成动态人脸微表情识别,减小识别耗时。

Because the existing methods do not consider the multi-directional fusion of dynamic facial features,the effect of facial micro expression recognition is poor,the recognition accuracy is low,and the recognition time is large. Therefore,a dynamic facial micro expression recognition method based on multi-directional feature fusion is proposed. The paper estimates the noise dynamic face image,processes the noise dynamic face image in blocks,solves the block coefficients of the dynamic face image,and reconstructs the dynamic face image. The denoised dynamic face image is enhanced,the distance feature information in the dynamic face image is extracted through AAM model,and all features in multiple directions are fused to realize dynamic face micro expression recognition. The experimental results show that the proposed method has a good effect on facial micro expression recognition,can accurately complete dynamic facial micro expression recognition and reduce the recognition time.