Depression recognition based on text and facial expression

Depression is a kind of mood disorder disease characterized by significant and lasting depression, which seriously affects people's physical and mental health. In recent years, the number of people suffering from depression has gradually increased. In order to improve the recognition rate of depression and reduce the workload of doctors, this paper proposes to apply the deep learning algorithm BiLSTM (Bi-directional Long Short-Term Memory) and Attention to recognize depression. Among them, BiLSTM is used to extract contextual temporal information of text features and facial features. Attention is used to learn the correlation between vision and text modalities. This paper undertakes extensive experiments to demonstrate the network's effect. The experimental results show that this method has certain practical application value for depression recognition.