A scene-level context-aware emotion recognition deep network method

Abstract

The invention discloses a scene-level context-aware emotion recognition deep network method, by reading the training sample set X in body tag value and original emotion tag value to get the body part image set X B ; to X in and X B After normalization processing, they are sent to the upper and lower convolutional neural networks to extract emotional features T F and contextual sentiment features T C , the T F and T C Send them to the upper and lower adaptive layers respectively to get the fusion weight λ F and λ C , the T F , T C , λ F and λ C Fusion to get emotional fusion feature T A , T A Through the linear mapping of the fully connected layer, the initial prediction values ​​of arousal and valence are obtained, and the loss between the two initial prediction values ​​and the original emotional label value is measured, and the network model is obtained after the training is completed and the test sample set is processed and sent to the network. model, get the test sample set X tn Predict label values. The method of the invention takes into account the degree of influence of features of different attributes on the emotions of characters when merging features, and improves the prediction performance of the model on the basis of enriching the research work of emotion recognition based on images.

Description

technical field [0001] The invention belongs to the technical field of pattern recognition, and in particular relates to a scene-level context-aware emotional recognition deep network method. Background technique [0002] Emotion is a necessary form of expression for people to express their feelings. In daily life, understanding and identifying a person's emotions from the actual scene they are in can help to perceive their mental state and predict behavior, and interact effectively. As early as the 1990s, the concept of affective computing was proposed by the MIT Media Laboratory, and then scientists began to work on converting complex human emotions into numerical information that can be recognized by computers, so as to better realize human-computer interaction. Computers tend to be more intelligent, which has become one of the key issues to be solved urgently in the era of artificial intelligence. [0003] Traditionally, emotion recognition tasks for static images are ...

AI Technical Summary

Problems solved by technology

[0006] The purpose of the present invention is to provide a scene-level context-aware emotion recognition deep network method, which solves the problem of the narrow range of emotion analysis based on static images in the prior art, only for face images, and the method of directly splicing different attribute features Limitations of Emotion Recognition

Images