A Face Recognition Method Against Expression Interference Based on Generative Adversarial Network
A generative and expressive technology, applied in the field of face recognition with anti-expression interference, can solve the problems that need to be further improved, and achieve the effect of improving the recognition accuracy, improving the accuracy and preventing the matching failure.
Active Publication Date: 2021-07-30
PEKING UNIV
View PDF6 Cites 0 Cited by
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
However, the accuracy of existing face recognition technology needs to be further improved when performing face recognition.
Method used
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View moreImage
Smart Image Click on the blue labels to locate them in the text.
Smart ImageViewing Examples
Examples
Experimental program
Comparison scheme
Effect test
Embodiment Construction
[0059] The following will clearly and completely describe the technical solutions in the embodiments of the present application with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are part of the embodiments of the present application, not all of them. Based on the embodiments in this application, all other embodiments obtained by persons of ordinary skill in the art without creative efforts fall within the protection scope of this application.
[0060] The inventors of the present application found that the existing face recognition methods can achieve better recognition accuracy under strong constraints, but the accuracy will drop significantly when there are no constraints or weak constraints. For example, the face database of the existing face recognition system only stores face pictures without expressions, but in many face recognition application scenarios, such as face tracking and recognition, the object to ...
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More PUM
Login to View More Abstract
This application relates to the field of face recognition technology, and provides a face recognition method based on generative confrontation network to resist expression interference, including the following steps: S11 obtains a sample set; S12 inputs the sample pictures with expressions to the training The generator generates a synthetic picture; S13 inputs the synthetic picture and at least one sample picture into the discriminator to be trained to train and update the discriminator; S14 generates a synthetic picture again through the generator to be trained; S15 will again The generated synthetic picture and the expressionless sample picture corresponding to the synthetic picture are input into the updated discriminator to obtain a feedback value and update the generator to be trained; S16 uses the updated generator as the generator to be trained, Repeat S12 to S15 multiple times to obtain a trained generator; S17 inputs the image to be recognized into the trained generator to obtain a non-expressive image to be recognized; S18 inputs the expressionless image to be recognized to the face recognition system for human face recognition. face recognition.
Description
technical field [0001] The present application relates to the technical field of face recognition, in particular, to a face recognition method based on a generative adversarial network to resist expression interference. Background technique [0002] Face recognition technology is a biometric recognition technology based on facial features. For a static image or a dynamic video, first judge whether there is a human face in it, and if so, further determine the orientation information of the human face, then extract the feature information of the human face according to the orientation information and image information, and then compare it with the existing Compared with known faces, the corresponding identity of the face is finally identified. [0003] Since the emergence of face recognition technology in the 1960s, it has been one of the most in-depth research topics in the field of computer vision, and has achieved remarkable achievements in both academic research and comme...
Claims
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More Application Information
Patent Timeline
Login to View More Patent Type & Authority Patents(China)
IPC IPC(8): G06K9/00G06K9/62G06N3/04G06N3/08
CPCG06N3/088G06V40/172G06N3/044G06N3/045G06F18/2411G06F18/214
Inventor 王韬蒋天夫
Owner PEKING UNIV