Face recognition model training method and device, equipment and medium

By dividing the face training set and increasing the proportion of difficult samples, a more attack-resistant face recognition model is trained, which solves the problem of difficult face occlusion attacks in the existing technology, and achieves higher recognition accuracy and model accuracy.

CN119942257APending Publication Date: 2025-05-06BEIJING MOMENTA TECH CO LTD
0 Cites 1 Cited by

Patent Information

Application Number
CN202311464976.X
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2023-11-06
Publication Date
2025-05-06

AI Technical Summary

Technical Problem

The prior art is difficult to effectively resist face occlusion attacks, especially in typical difficult scenarios containing a large number of intermediate samples.

Method used

By using the pre-trained face recognition model to reason the face training set, divide it into the face simple set and the face difficulty set, and increase the proportion of the difficulty set to form a new face training set, and then train the target face recognition model.

Benefits of technology

It improves the accuracy of the model for difficult face recognition, enhances its resistance to face occlusion attacks, and thus improves the overall accuracy of the model.

✦ Generated by Eureka AI based on patent content.
Patent Text Reader

Abstract

The invention relates to the technical field of face recognition, and provides a face recognition model training method and device based on face shielding, equipment and a medium, and the method comprises the steps: carrying out the reasoning of a face training set through a pre-trained face recognition model, and obtaining a reasoning result; dividing the face training set into a face simple set and a face difficult set according to the reasoning result; increasing the proportion of the face difficulty set in the face training set to obtain a new face training set; and training the pre-trained face recognition model by using the new face training set to obtain a target face recognition model. Through the technical scheme, the recognition accuracy of the model on the difficult face is improved, and the model precision is improved, so that the discrimination of attack means such as face shielding attack is realized.
Need to check novelty before this filing date? Find Prior Art