Deep learning-based multi-dimensional output face quality evaluation method and electronic equipment

Abstract

The invention discloses a deep learning-based multi-dimensional output face quality evaluation method and electronic equipment. The method comprises the following steps: preparing a training data set, a Score training set, a Class training set, a Mask training set and a Pose training set; training a network model, randomly selecting a part of pictures from each of the four training sets, merging the pictures into a batch of pictures, sending the pictures into a neural network model, obtaining output values of four branches through forward reasoning of a neural network, calculating loss values of the corresponding branches according to which data set the input pictures come from, and calculating the loss values of the corresponding branches according to the loss values; finally, adding the loss value of each branch according to different weights to obtain a total loss value for network back transmission, and updating network parameters; predicting a to-be-detected face image, inputting a face image, preprocessing the face image, sending the face image into the neural network model obtained through training for forward reasoning, outputting prediction values of four branches, and finally adding the output values of the four branches according to weights to obtain a final face quality comprehensive evaluation score. The invention further provides the corresponding electronic equipment.

Description

technical field [0001] The invention belongs to the technical field of image recognition, and more specifically relates to a multi-dimensional output face quality evaluation method and electronic equipment based on deep learning. Background technique [0002] During the face capture process of the edge device, due to the influence of environmental changes and human movement, there are low-quality face images such as blurring, occlusion, and posture changes in the captured face images, and these low-quality face images will be greatly reduced. Accuracy of face recognition system. At the same time, the storage space and transmission bandwidth of edge devices are very limited, and a large number of low-quality face pictures are not conducive to the storage and transmission of face pictures. In order to be able to select one or more high-quality face images from a large number of face images, it is necessary to use a face quality assessment method. [0003] The factors that af...

AI Technical Summary

Problems solved by technology

[0004] In view of the above defects or improvement needs of the prior art, the present invention provides a face quality assessment method based on multi-dimensional output of deep learning, which is used to solve the problem of increasing time-consuming and computing resources when multiple models run at the same time, increasing human Accuracy of face quality assessment

Images