Emotion recognition method and device based on tensor decomposition, medium and electronic equipment

Abstract

The invention provides an emotion recognition method and device based on tensor decomposition, a computer readable storage medium and electronic equipment, and the method comprises the steps: converting physiological signal data of multiple modes into tensor data, and decomposing tensors into a plurality of low-rank matrixes corresponding to the multiple modes; constructing a plurality of graph matrixes respectively corresponding to a plurality of modes according to the tensor data, estimating missing data of the physiological signal data by utilizing the tensor data and known quantities in the graph matrixes, and calculating the missing data of the physiological signal data according to the missing data and the known quantities in the graph matrixes. Therefore, complete physiological signal data is supplemented, emotion categories are recognized according to the complete physiological signal data, emotion recognition of the complete physiological signal data is achieved, and the accuracy is guaranteed through the mutual relevance between the physiological signal data.

Description

technical field [0001] The invention relates to the field of application programs, in particular to an emotion recognition method, device, computer-readable storage medium and electronic equipment based on tensor decomposition. Background technique [0002] Emotion recognition plays an important role in both human-computer interaction and interpersonal communication in daily life. Although people have studied emotion recognition for many years, it is still a challenging problem because human emotions are not expressed in a single way, but in multiple ways, such as speech, gestures, etc. , facial expressions, and physiological signals. Since physiological signals are controlled by the sympathetic nervous system, they are generally independent of human will, and unlike other voluntary or involuntary signals, physiological signals are not easily suppressed or hidden. Compared with visual and auditory signals, physiological signals can provide more reliable information for emo...

AI Technical Summary

Problems solved by technology

This increases the level of difficulty we face with emotion recognition

[0003] Existing emotion recognition methods mainly address this challenge by devising effective fusion strategies on the premise that all forms of signals are always available, which is often unrealistic in practice.

In addition, certain factors limit data collection, and it is very difficult to collect all the data required for analysis, which also leads to the fact that existing fusion strategies for emotion recognition often cannot be used.

Images