Age estimation method and device and electronic equipment

Abstract

The invention provides an age estimation method and device and electronic equipment, and the method comprises the steps: inputting a to-be-estimated image containing a target object into an age estimation model, and obtaining an output result; determining the age of the target object based on the output result, wherein the age estimation model training process comprises the following steps: training an initial model based on a first sample set of a sample image carrying an age label to obtain an intermediate model; adding an age label to a second sample set only containing the sample image through the intermediate model and the first sample set; and training the intermediate model or the initial model through the first sample set and the second sample set added with the age label to obtain an age estimation model. According to the mode, the intermediate model is obtained by training the sample images carrying the age labels, the age labels are added to the images without labels based on the intermediate model, so that the number of the sample images is expanded, the age estimation model is obtained by training based on a large number of expanded sample images, the generalization ability of the model is improved, and the accuracy of model age estimation is also improved.

Description

technical field [0001] The present invention relates to the technical field of image processing, in particular to an age estimation method, device and electronic equipment. Background technique [0002] As an important face attribute, age has broad application prospects in the fields of human-computer interaction, intelligent business, security monitoring and entertainment. In related technologies, the age of the person in the face image is usually identified through a trained neural network model. The neural network model is trained on an age data set. However, since age involves personal privacy, collecting age-labeled people Face samples are very difficult and time-consuming, which makes the amount of data in the age data set limited, resulting in the model trained on the age data set is prone to overfitting, making the model generalization ability poor, thus affecting the accuracy of the model's age estimation Spend. Contents of the invention [0003] The object of t...

AI Technical Summary

Problems solved by technology

In related technologies, the age of the person in the face image is usually recognized through a trained neural network model. The neural network model is trained on an age data set. However, since age involves personal privacy, collecting age-labeled people Face samples are very difficult and time-consuming, which makes the amount of data in the age data set limited, resulting in the model trained on the age data set is easy to overfit, making the model generalization ability poor, thus affecting the accuracy of the model age estimation Spend

Images