Research and Application of Facial Expression Recognition Based on Attention Mechanism

The scale of the existing convolutional neural network is getting larger and larger, resulting in too large amounts of parameters, and the structure is not light enough. Moreover, existing convolutional neural networks are difficult to recognize the subtle changes of facial expressions and cannot extract facial expression features accurately. Therefore, the performance of facial expression recognition needs to be improved. To solve the above problems, a new deep learning network model is proposed for facial expression recognition. Based on the deep residual network, the attention mechanism module (Convolutional Block Attention Module, CBAM) is added to the last layer of convolution and the first layer of convolution of the network. The spatial attention mechanism and channel attention mechanism are used to suppress the unimportant feature information and focus on the effective feature information. In the bottom layer, the influence of other factors is eliminated as much as possible, and more attention is paid to the extraction of facial expression features, which enriches the learning of facial expression features and improves the accuracy of facial expression recognition. The method proposed in this paper has been tested and verified on two public data sets FER2013 and CK+, and the results prove that the method has a high accuracy rate.