Age estimation method based on mixed features, electronic equipment and storage medium

Abstract

The invention discloses an age estimation method based on mixed features, electronic equipment and a storage medium. The method comprises the following steps: acquiring a target face image; performing face key point positioning on the target face image; cutting according to a preset cutting resolution to obtain a face local image taking each face key point as a center, and splicing each face local image in sequence to obtain a first input image; cutting the aligned target face image according to the face key point information to obtain a minimum enclosing rectangle containing all reserved face key points as a face global image, and zooming the face global image to obtain a second input image; and inputting the first input image and the second input image into different branches of an age prediction model, respectively performing feature extraction, performing superposition and merging to form a fusion feature, and performing feature decoding to output a one-dimensional Gaussian heat map. According to different receptive field requirements, a double-branch model structure is designed, so that the feature information of the model is richer, and the robustness of the model is higher.

Description

technical field [0001] The invention relates to the technical field of image processing, in particular to an age estimation method based on mixed features, an electronic device and a storage medium. Background technique [0002] In recent years, the recommendation system has accompanied people's life, and uses the collected information to build user portraits, so as to provide more accurate and customized recommendation services. Among them, the age information used is particularly important, and can also be used in behavior monitoring, anti-addiction of underage games and other fields. In many practical scenarios, it is impossible to collect the user's identity information, and it is difficult to obtain the user's age. By using the camera to collect the user's face information, estimating the age has become a feasible alternative. [0003] At present, the existing face age estimation models all use face images as input, and output age category information or regression val...

AI Technical Summary

Problems solved by technology

The age estimation method based on the regression task reduces the generalization ability of the model to the space by directly regressing the corresponding value, making the model more unstable in the convergence process; the age estimation method based on the classification task treats each age as an independent interaction. This method cannot take into account the correlation between adjacent categories, and the convergence speed is slow and the accuracy is difficult to guarantee.

Images