Facial expression recognition based on improved residual network

Abstract

Facial expressions are an important part of human emotional signals and their recognition has become an important topic of research in the field of pattern recognition. Deep learning based methods have achieved great success in the recognition of facial expressions. However, with the evolution of convolution neural networks and the increased network depth, these methods suffer from problems such as degraded network performance and loss of feature information. To address these problems, a novel facial expression recognition algorithm based on an improved residual neural network is proposed. First, a residual neural network is designed to extract deep features while retaining the shallow ones. This can effectively prevent the degradation of network performance. Moreover, when the gradient of the Rectified Linear Units activation function used in the residual module is 0, it will inactivate the neurons and cause a loss of feature information. To address this, the Mish activation function is used instead. The slight allowance for negative values in Mish improves the gradient flow. Next, an inception module is introduced to obtain richer feature information under the same receptive field. Finally, by conducting verification experiments on the public datasets CK+ and KDEF, the authors manage to solve the problems of degraded network performance and insufficient information from extracted features, achieving recognition accuracy rates of 96.37% and 93.38% on the two datasets, respectively.

Modelling appearance variations in expressive and neutral face image for automatic facial expression recognition

“…The improved residual neural network (RNN) is introduced for AFER in [13]. The degradation in network performance is effectively prevented by the designed RNN, which derives deep features and preserves the shallow ones.…”

“…The end-to-end training of DML-Net reduces multiple metric losses and suppresses overfitting, thereby enhancing the recognition rate. The improved residual neural network (RNN) is introduced for AFER in [13]. The degradation in network performance is effectively prevented by the designed RNN, which derives deep features and preserves the shallow ones.…”

A multi‐scale feature fusion convolutional neural network for facial expression recognition

“…Guo et al proposed a double-code LBP-layer spatial attention network (DLSANet) that improved traditional recognition models by incorporating a double-code LBP (DLBP) layer and spatial attention network into the emotion network (ENet). This approach achieved superior results compared to contemporary facial emotion recognition methods on commonly used facial expression recognition datasets (Guo et al, 2023). Nan et al incorporated the MobileNetV1 model with an attention mechanism to enhance the model's ability to extract local features from facial expression samples.…”

Robust facial landmark detection by probability‐guided hourglass network

“…Facial landmark detection, also known as face alignment, is an essential topic in computer vision and is widely used in many fields, such as face recognition [1], face reconstruction [2], and facial expression recognition [3]. Unlike the traditional features characterized by feature descriptors, facial landmarks include pupils, nose tips, and eye corners, which are visible to the naked eye and have human structural properties.…”