Facial expression recognition method, convolutional neural network model training method, devices and electronic apparatus

Abstract

The embodiments of the present invention provide a facial expression recognition method, a convolutional neural network model training method, a facial expression recognition device, a convolutional neural network model training device and an electronic apparatus. The facial expression recognition method includes the following steps that: facial expression features are extracted from a face imageto be detected by means of the convolutional layer portion of a convolutional neural network model and acquired face key points in the face image to be detected, so that a facial expression feature image is obtained; ROI (regions of interest) corresponding to the face key points in the facial expression feature image are determined; pooling processing is performed on the determined ROIs through adopting the pooling layer of the convolutional neural network model, so that a pooled ROI feature image can be obtained; and the facial expression recognition result of the face image is obtained at least according to the ROI feature map. With the facial expression recognition method provided by the embodiments of the present invention adopted, subtle facial expression changes can be effectively captured, and at the same time, differences caused by different facial gestures can be better processed; and the detailed information of the changes of the plurality of regions of a face are fully utilized, so that subtle facial expression changes and faces in different postures can be recognized more accurately.

Description

technical field [0001] The embodiments of the present invention relate to the technical field of artificial intelligence, and in particular to an expression recognition method, device and electronic equipment, and a convolutional neural network model training method, device and electronic equipment. Background technique [0002] Facial expression recognition technology refers to assigning an expression category to a given face image, including: anger, disgust, happiness, sadness, fear, surprise, etc. At present, facial expression recognition technology has gradually shown broad application prospects in the fields of human-computer interaction, clinical diagnosis, distance education, investigation and interrogation, and is a popular research direction of computer vision and artificial intelligence. [0003] An existing facial expression recognition technology is a recognition technology based on a traditional machine learning framework. Expression recognition using this trad...

AI Technical Summary

Problems solved by technology

However: first, face feature extraction requires manual design and extraction, which requires expertise in a specific field; second, compared with deep feature (feature map), classic geometric features such as Gabor filter, SIFT, etc. Weak; third, traditional machine learning methods are difficult to use more and more training data, the training time is long, and the training process is scattered and complicated

[0004] As a result, the cost of existing expression recognition is high, and the accuracy of expression recognition is low

Images