Multi-kernel-learning-based multi-mode emotion identification method

Abstract

The invention discloses a multi-kernel-learning-based multi-mode emotion identification method. According to the method, extraction of emotion features like an expression feature, a voice feature and a physiological feature is carried out on sample data of each mode in a multi-mode emotion database; several different kernel matrixes are constructed for each mode respectively; the kernel matrix groups corresponding to different modes are fused to obtain a fused multi-mode emotion feature; and a multi-kernel support vector machine is used as a classifier to carry out training and identification. Therefore, basic emotions like angering, disgusting, fearing, delighting, upsetting, and surprising and the like can be identified effectively.

Description

technical field [0001] The invention relates to the fields of signal processing and pattern recognition, in particular to a multi-modal emotion recognition method based on multi-core learning. Background technique [0002] Emotion recognition has been a hot topic in the field of pattern recognition, which benefits people's social communication and activities. In the process of people's life and communication, there are various ways to express inner emotions, but facial expression and voice are the most direct, easiest and most expressive ways and are easily obtained by researchers. At present, the single-modal emotion recognition technology for expression or voice is relatively mature, but the recognition effect is often not reliable and accurate due to the singleness of information. Therefore, it is worth further research to use the correlation and complementarity of multimodal features of different properties to realize a more reliable and stable emotion recognition syste...

AI Technical Summary

Problems solved by technology

Faced with the emotional characteristics of multiple modalities, using a single kernel function cannot solve the problem of uneven distribution of feature space and feature redundancy. Moreover, different kernel functions have different properties, and the effects obtained when using different kernel functions are often very different. Therefore, multi-core learning has emerged. In previous multi-core learning applications, kernel functions with different characteristics were often given and then trained to obtain the weights corresponding to the kernel functions to achieve feature fusion. It is still necessary to consider the selection of different features. Types and parameters of kernel functions

Images