Forged face identification method, device, equipment and storage medium
The facial unit consistency characteristics of the image to be identified are obtained through the face identification model, which solves the recognition disadvantages of the prior art when processing unknown fake methods to synthesize images, and achieves better versatility and robustness.
CN116824710BActive Publication Date: 2025-04-29INST OF AUTOMATION CHINESE ACAD OF SCI +1
Patent Information
- Application Number
- CN202310587326.8
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2023-05-23
- Publication Date
- 2025-04-29
- Estimated Expiration
- 2043-05-23
AI Technical Summary
Technical Problem
The prior art shows disadvantages when processing images synthesized by unknown fraud methods, and it is difficult to generalize the recognition effect.
Method used
The facial unit consistency characteristics of the image to be identified are obtained through the face identification model, and the identification results of the image are determined based on these characteristics. The model includes a backbone network, an encoded network and a fully connected network, which acquires facial unit consistency characteristics through multiple cascading encoders.
Benefits of technology
It improves the identification effect of the synthesis of images of unknown fraud methods and improves the universality and robustness of the model.
✦ Generated by Eureka AI based on patent content.
Smart Images
Figure CN116824710B_ABST
Abstract
The present invention provides a method, device, equipment and storage medium for identifying forged faces, wherein an image to be identified is input into a face identification model; an identification result corresponding to the image to be identified output by the face identification model is obtained; wherein the face identification model is used to obtain facial unit consistency features of the image to be identified, and determine the identification result corresponding to the image to be identified based on the facial unit consistency features of the image to be identified; the facial unit consistency features of the image to be identified are used to characterize the correlation between relevant areas of each facial unit of the image to be identified; the face identification model is obtained by training based on sample images and identification labels corresponding to the sample images, thereby improving the identification effect of images synthesized by unknown counterfeiting methods.
Need to check novelty before this filing date? Find Prior Art
Citation Information
Patent Citations
Face forgery detection method and device based on visual semantic information
CN113627233A
Face forgery detection method and device, terminal and storage medium
CN115050066A