Face clustering method and device

A clustering method and clustering technology, which are applied in the fields of still image data clustering/classification, still image data retrieval, special data processing applications, etc., can solve the problem of classifying faces into multiple files, and achieve accurate face filing , Improve the quality of the face, the effect of high efficiency

Active Publication Date: 2020-04-07
ZHEJIANG DAHUA TECH CO LTD
View PDF7 Cites 2 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0003] However, in the process of real-time capture, the angle of the face, light, etc. have a large impact on the face feature image, and it is easy to classify the face of the same user into multiple files by simply performing feature matching and archiving.

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 and device
  • Face clustering method and device
  • Face clustering method and device

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0045] In order to better understand the technical solutions provided by the embodiments of the present application, a detailed description will be given below in conjunction with the accompanying drawings and specific implementation manners.

[0046] Face clustering can be simply understood as grouping together face images belonging to a person or a group of people. At present, it is generally used to compare the features of frame-by-frame images to determine which image frames belong to which person. However, in the surveillance scene, the face angles of the same user at different times are very different. When a user’s posture at two different moments is quite different, it is easy to classify the two images of the user as two users. In this way, the accuracy of face clustering will be low.

[0047] In view of this, the embodiment of this application provides a face clustering method, please refer to figure 1 , indicating an application scenario. Combine below figure 1 An ...

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 and device, which are used for improving the face clustering accuracy. The method comprises the following steps: capturing from a video frame to obtaina first captured image, and obtaining a unique tracking identifier of the first captured image according to a face tracking algorithm; if a first reference face image set composed of all face images with the same unique tracking identifier as the first snapshot image exists in a face clustering database, comparing the face quality score of the first snapshot image with the face quality score of the first reference face image, wherein the first reference face image is a reference face image with the maximum face quality score in a first reference face image set; and if the face quality score ofthe first snapshot image is smaller than or equal to the face quality score of the first reference face image, associatively storing the first snapshot image and the first reference face image set. According to the method, face clustering is carried out by combining a face tracking algorithm, so that the face clustering accuracy is improved.

Description

technical field [0001] The present application relates to the technical field of artificial intelligence, in particular to a face clustering method and device. Background technique [0002] For example, in the field of surveillance and security, face clustering can be used for file building. Face clustering can be simply understood as filing the faces of similar users in a file, which can be used for subsequent investigations. At present, the face clustering process is generally to perform face feature matching on the faces in each frame of the video, and find out the faces belonging to the same user. [0003] However, in the process of real-time capture, the angle of the face, light, etc. have a large impact on the face feature image, and simple feature matching and archiving may easily cause the same user's face to be classified into multiple files. Contents of the invention [0004] Embodiments of the present application provide a face clustering method and device for...

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 Applications(China)
IPC IPC(8): G06F16/55G06F16/783G06K9/00G06K9/62
CPCG06F16/55G06F16/784G06V40/168G06F18/22
Inventor 沈瑜阮学武
Owner ZHEJIANG DAHUA TECH CO LTD
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Try Eureka
PatSnap group products

Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic.

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap