Face recognition method and device, storage medium and terminal equipment

Abstract

A face recognition method provided by the invention comprises the following steps: acquiring a target image comprising face information of a current user, and extracting a target feature vector in thetarget image; correcting the target feature vector according to a preset feature correction algorithm to obtain a to-be-compared feature vector; and determining a similar face image based on the to-be-compared feature vector. Through the method of the present application, a face image similar to a user can be obtained; five sense organ contours and relative position information of the five senseorgan contours in the face image are extracted; therefore, the obtained image is more similar to the face image of the user, the five sense organ contours of the user in the image and the relative positions of the five sense organ contours are corrected, the face image similar to the user is obtained based on the corrected image, the image obtained through the method is more attractive, and the confidence of the user to the user is improved. Through the method, the friend circle of the user can be further expanded, and the interest of the user in the recommended user is improved.

Description

technical field [0001] The present application relates to the technical fields of software and image processing, and particularly relates to a face recognition method, device, storage medium, and terminal equipment. Background technique [0002] With the rapid development of the mobile Internet, emerging social media platforms such as WeChat and Weibo have profoundly affected and changed people's work and lifestyle, and have shown trends such as rapid growth in the number of users and massive generation of user data, providing relevant research Effective data sources and application scenarios. [0003] Now people usually like to upload their own photos to the application, and make up their own appearance in the image through the preset makeup program in the application. However, under normal circumstances, only the user's own face is extracted and directly pasted At other fixed makeup positions, although the image obtained by this method is the same as the user's face, it i...

AI Technical Summary

Problems solved by technology

[0003] Now people usually like to upload their own photos to the application, and make up their own appearance in the image through the preset makeup program in the application. However, under normal circumstances, only the user's own face is extracted and directly pasted To other fixed positions of makeup, although the image obtained by this method is the same as the user's face, it is not beautiful, which reduces the user's confidence in their appearance

In addition, people now share their photos with friends through social media software, so their photos are often posted in the circle of friends. However, under normal circumstances, some people do not notice that they have a certain similarity with their friends in the circle of friends. Spend

And under normal circumstances, the similarity between friends can only be based on individual subjective feelings

At the same time, people's friends now often come from people in their own work circles, and it is difficult to get in touch with people outside their own circles, which limits the existing social methods of people. At the same time, the conventional method of recommending other friends is to recommend directly based on the user's personal information. However, this method also has certain disadvantages, that is, the user's interest in the recommended user cannot be stimulated only through the introduction of the user's personal information, and it also has certain restrictions on the user's social interaction.

Images