Image processing method and system for updating template library based on life value of facial template images

Patent metadata

Field	Value
Publication number	US-12347164-B2
Application number	US-202217950698-A
Country	US
Kind code	B2
Filing date	Sep 22, 2022
Priority date	Mar 24, 2020
Publication date	Jul 1, 2025
Grant date	Jul 1, 2025

Abstract

This application discloses image processing methods and systems. One method includes: obtaining a first image, obtaining a template image having a life value that determines whether the template image is valid, comparing the first image with the template image, and storing the first image in a template library as a new template image in response to determining that the first image matches the template image.

First claim

What is claimed is:

1. An image processing method, comprising:
obtaining a first image;
obtaining a template image having a life value that determines whether the template image is valid, wherein the life value of the template image in a template library decreases as existence time of the template image in the template library increases, and wherein the template image comprises a first template image and a second template image, the first template image comprises a face-dominant image, and the second template image comprises a special face image;
comparing the first image with the template image; and
storing the first image in the template library as a new template image in response to determining that the first image matches the template image.

2. The method according to claim 1 , further comprising:
in response to determining that the template library comprises one or more template images, deleting a template image having a life value less than a predetermined threshold from the template library.

3. The method according to claim 1 , further comprising:
increasing the life value of the template image in response to determining that the first image matches the template image.

4. The method according to claim 1 , wherein the face-dominant image comprises an unblocked face image in a limited angle range, and the special face image comprises at least one of the following: a face image with a blocked face, a face image with an accessory, and a face image with a large deviation angle.

5. The method according to claim 4 , wherein the template image and the first image are compared by using a recognition model based on feature parameters of the template image and the first image, and wherein the recognition model is at least one of Arcface or Facenet.

6. The method according to claim 5 , wherein the recognition model determines, by using a cosine distance or a Euclidean distance between the feature parameters, whether the template image matches the first image.

7. An image processing system, comprising:
a storage apparatus, configured to store a template image having a life value that determines whether the template image is valid, wherein the life value of the template image in a template library decreases as existence time of the template image in the template library increases, and wherein the template image comprises a first template image and a second template image, the first template image comprises a face-dominant image, and the second template image comprises a special face image; and
a processing apparatus, the processing apparatus comprises:
at least one processor; and
a memory coupled to the at least one processor and storing programming instructions for execution by the at least one processor to perform operations comprising:
obtaining a first image and the template image from the storage apparatus;
comparing the first image with the template image; and
storing the first image in the template library as a new template image in response to determining that the first image matches the template image.

8. The system according to claim 7 , wherein the operations comprising:
obtaining at least one template image from a cloud server, wherein the storage apparatus is configured to store the at least one template image obtained from the cloud server.

9. The system according to claim 7 , wherein the life value of the template image increases when the first image matches the template image.

10. The system according to claim 7 , wherein the life value of the template image in the template library decreases as existence time of the template image in the template library increases.

11. The system according to claim 10 , wherein the template image comprises a first template image and a second template image, the first template image comprises a face-dominant image, and the second template image comprises a special face image.

12. The system according to claim 11 , wherein the face-dominant image comprises an unblocked face image in a limited angle range, and the special face image comprises at least one of the following: a face image with a blocked face, a face image with an accessory, and a face image with a large deviation angle.

13. The system according to claim 12 , wherein the template image and the first image are compared by using a recognition model based on feature parameters of the template image and the first image, and wherein the recognition model is at least one of Arcface or Facenet.

14. The system according to claim 13 , wherein the recognition model determines, by using a cosine distance or a Euclidean distance between the feature parameters, whether the template image matches the first image.

15. A computer program product comprising computer-executable instructions stored on a non-transitory computer-readable storage medium that, when executed by a processor, cause an apparatus to perform operations comprising:
obtaining a first image;
obtaining a template image having a life value that determines whether the template image is valid, wherein the life value of the template image in a template library decreases as existence time of the template image in the template library increases, and wherein the template image comprises a first template image and a second template image, the first template image comprises a face-dominant image, and the second template image comprises a special face image;
comparing the first image with the template image; and
storing the first image in the template library as a new template image in response to determining that the first image matches the template image.

16. The computer program product according to claim 15 , further comprising:
in response to determining that the template library comprises one or more template images, deleting a template image having a life value less than a predetermined threshold from the template library.

17. The computer program product according to claim 15 , further comprising:
increasing the life value of the template image in response to determining that the first image matches the template image.

18. The computer program product according to claim 15 , wherein the face-dominant image comprises an unblocked face image in a limited angle range, and the special face image comprises at least one of the following: a face image with a blocked face, a face image with an accessory, and a face image with a large deviation angle.

19. The computer program product according to claim 18 , wherein the template image and the first image are compared by using a recognition model based on feature parameters of the template image and the first image, and wherein the recognition model is at least one of Arcface or Facenet.

20. The computer program product according to claim 19 , wherein the recognition model determines, by using a cosine distance or a Euclidean distance between the feature parameters, whether the template image matches the first image.

Assignees

• Shenzhen Yinwang Intelligent Technology Co Ltd

Inventors

• Huang Xiaolin
• Huang Wei
• Xu Wenkang
• Zhang Junhao

Classifications

• G06N3/09

  Supervised learning · CPC title

• G06N3/0464

  Convolutional networks [CNN, ConvNet] · CPC title

• G06V40/161 Primary

  Detection; Localisation; Normalisation · CPC title

• G06V10/454

  Integrating the filters into a hierarchical structure, e.g. convolutional neural networks [CNN] · CPC title

• G06V10/82

  using neural networks · CPC title