| Title | Lightweight Multi-Scale Network with Attention for Facial Expression Recognition |
| Publication Type | Conference Paper |
| Year of Publication | 2021 |
| Authors | Hu, Zhibin, Yan, Chunman |
| Conference Name | 2021 4th International Conference on Advanced Electronic Materials, Computers and Software Engineering (AEMCSE) |
| Keywords | attention mechanism, convolution, convolutional neural network, convolutional neural networks, face recognition, facial expression recognition, facial recognition, feature extraction, Human Behavior, Kernel, Medical services, Metrics, Multi-scale network, pubcrawl, resilience, Resiliency, Sociology |
| Abstract | Aiming at the problems of the traditional convolutional neural network (CNN), such as too many parameters, single scale feature and inefficiency by some useless features, a lightweight multi-scale network with attention is proposed for facial expression recognition. The network uses the lightweight convolutional neural network model Xception and combines with the convolutional block attention module (CBAM) to learn key facial features; In addition, depthwise separable convolution module with convolution kernel of 3 x 3, 5 x 5 and 7 x 7 are used to extract features of facial expression image, and the features are fused to expand the receptive field and obtain more rich facial feature information. Experiments on facial expression datasets Fer2013 and KDEF show that the expression recognition accuracy is improved by 2.14% and 2.18% than the original Xception model, and the results further verify the effectiveness of our methods. |
| DOI | 10.1109/AEMCSE51986.2021.00143 |
| Citation Key | hu_lightweight_2021 |
Lightweight Multi-Scale Network with Attention for Facial Expression Recognition