Expression recognition method and device and electronic device

Abstract

The invention relates to the technical field of artificial intelligence, and provides an expression recognition method and device and an electronic device. The method comprises the steps of obtaininga face image; performing feature extraction on the face image to obtain first type information and second type information; obtaining global feature information according to the first type of information, and determining first expression information according to the global feature information; and obtaining a target characteristic value according to the second type of information, and determiningsecond expression information according to the target characteristic value. According to the method, multi-task learning can be realized, and the data volume required by model training is reduced. Theemotion recognition result and the local expression recognition result can be obtained at the same time, the expression recognition efficiency and real-time performance are improved, and the user experience is further improved.

Description

technical field [0001] The present invention relates to the technical field of artificial intelligence, in particular to an expression recognition method, an expression recognition device and electronic equipment. Background technique [0002] With the rapid development of computer technology and artificial intelligence technology and related disciplines, the degree of automation of the whole society is constantly improving, and people's demand for human-computer interaction similar to the way people communicate is becoming stronger and stronger. Expression recognition is the basis of emotion understanding, the prerequisite for computers to understand people's emotions, and an effective way for people to explore and understand intelligence. [0003] There are a large number of deep learning-based methods in existing facial expression recognition, but deep learning requires a large number of labeled samples to achieve better performance, and sample search and labeling require...

AI Technical Summary

Problems solved by technology

[0003] There are a large number of deep learning-based methods in existing facial expression recognition, but deep learning requires a large number of labeled samples to achieve better performance, and sample search and labeling require a lot of cost investment; in addition, existing deep learning methods are time-consuming The degree is very high, the network structure is complex, and it is difficult to run in real time on ordinary front-end equipment and mobile terminals

Images