Unlock instant, AI-driven research and patent intelligence for your innovation.

Face clustering method, device, computer equipment and readable storage medium

A clustering method and computer program technology, applied in the field of face recognition, can solve the problems of different training, achieve the effect of reducing probability, reducing data collection cost, and avoiding the transmission of wrong information

Active Publication Date: 2022-04-08
北京奇信智联科技有限公司
View PDF7 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0003] However, in a face clustering method in the prior art, using the corresponding scene data to train the model for face recognition not only needs to collect a large amount of data corresponding to the scene, but also needs to train different models for different application scenarios

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 more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • Face clustering method, device, computer equipment and readable storage medium
  • Face clustering method, device, computer equipment and readable storage medium
  • Face clustering method, device, computer equipment and readable storage medium

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0035] Embodiment 1 of the present invention provides a face clustering method. In this method, the stranger center and the identified person center are preset as the clustering centers to cluster the target face pictures. Among them, the identified person The center includes the main picture and auxiliary picture belonging to the same face. Compared with the auxiliary picture, the main picture can better reflect the characteristics of the face. Usually the main picture is a frontal photo of the face, which is more accurate, such as a ID photo, etc. The auxiliary picture can be Faces captured in various scenarios. In one scenario, the method is used to determine whether the target face picture belongs to the stranger center. In another application scenario, after it is determined that the target face picture belongs to the identified person center, the method is used to judge the result, and correct it when the judgment result is wrong, so as to avoid the transmission of wrong...

Embodiment 2

[0075] Corresponding to Embodiment 1 above, Embodiment 2 of the present invention provides a face clustering device. For details of corresponding technical features and corresponding technical effects, reference may be made to Embodiment 1 above, which will not be repeated in this embodiment. figure 2 For the face clustering device provided in Embodiment 2 of the present invention, a stranger center and an identified person center are preset, wherein the identified person center includes a main picture and an auxiliary picture belonging to the same face, such as figure 2 As shown, the device includes: a calculation module 201 , a comparison module 202 and an identification module 203 .

[0076] Wherein, the calculation module 201 is used to calculate the similarity between the target human face picture and the main picture to obtain the first similarity, and calculate the similarity between the target human face picture and the auxiliary picture to obtain the second similarit...

Embodiment 3

[0085] Embodiment 3 also provides a computer device, such as a smart phone, a tablet computer, a notebook computer, a desktop computer, a rack server, a blade server, a tower server or a cabinet server (including an independent server, Or a server cluster composed of multiple servers), etc. Such as image 3 As shown, the computer device 01 in this embodiment at least includes but is not limited to: a memory 012 and a processor 011 that can communicate with each other through a system bus, such as image 3 shown. It should be pointed out that, image 3 Only the computer device 01 is shown with components memory 012 and processor 011, but it should be understood that implementing all of the illustrated components is not a requirement and that more or fewer components may instead be implemented.

[0086] In this embodiment, the memory 012 (that is, a readable storage medium) includes a flash memory, a hard disk, a multimedia card, a card-type memory (for example, SD or DX memo...

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

No PUM Login to View More

Abstract

The invention provides a face clustering method, device, computer equipment and readable storage medium. The stranger center and the identified person center are preset. The identified person center includes the main picture and the auxiliary picture belonging to the same face. The face clustering method includes: calculating the similarity between the target face picture and the main picture, and obtaining the first A similarity, calculate the similarity between the target face picture and the auxiliary picture, and obtain the second similarity; compare the first similarity with the preset first threshold, the size relationship between the first similarity and the second similarity respectively; and if If the first similarity reaches a preset first threshold, and the magnitude relationship between the first similarity and the second similarity satisfies a first preset rule, then the category of the target face picture is identified as a stranger center. Through the present invention, there is no need to collect training data and train different models for different training scenarios, and cluster errors can be identified.

Description

technical field [0001] The present invention relates to the technical field of face recognition, in particular to a face clustering method, device, computer equipment and readable storage medium. Background technique [0002] Face clustering refers to dividing the face pictures of the same person into the same group by clustering. When implementing face clustering, the features of the face pictures are usually extracted, and the face pictures are mapped (embedding) is a high-dimensional vector, and the distribution of the mapped face pictures in the high-dimensional space is different, so the distance in the high-dimensional space can be used to measure the similarity between face pictures and realize face clustering. [0003] However, in a face clustering method in the prior art, using corresponding scene data to train a model for face recognition requires not only collecting a large amount of data corresponding to the scene, but also training different models for different...

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
no application Login to View More
Patent Type & Authority Patents(China)
IPC IPC(8): G06V40/16G06V10/762G06V10/774G06K9/62
CPCG06V40/16G06F18/23G06F18/214
Inventor 李四伟孔德亮许斯亮梁杰
Owner 北京奇信智联科技有限公司