Method for realizing portrait clustering based on multi-data fusion analysis

A multi-data and data technology, applied in the computer field, can solve the problems of no face-related human body capture data associative aggregation, reduced clustering accuracy, and no consideration of other data fusion analysis, so as to reduce GPU computing resource consumption and reduce archive aggregation. The effect of reducing class errors and improving the efficiency of retrieval and analysis

Pending Publication Date: 2022-01-11
南京启数智能系统有限公司
View PDF0 Cites 1 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

The disadvantages of this scheme are: first, it relies too much on the existing static face database, and when static data support cannot be provided, the clustering effect is extremely general; second, it does not consider the fusion analysis of other data
The disadvantage of this scheme is: with the rapid growth of the access data volume and the archive clustering cardinality, the archive clustering error will be amplified, the clustering accuracy will decrease and the divergence will increase
[0005] There are also common problems in the above two technical solutions: first, the introduction of more auxiliary analysis data is not considered to improve the clustering accuracy; In the data environment, the accuracy of archive clustering is poor

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
  • Method for realizing portrait clustering based on multi-data fusion analysis
  • Method for realizing portrait clustering based on multi-data fusion analysis
  • Method for realizing portrait clustering based on multi-data fusion analysis

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0039] In order to better illustrate the present invention, further description will now be made in conjunction with the embodiments and accompanying drawings.

[0040] The method of realizing portrait clustering based on multi-data fusion analysis, the steps are as follows:

[0041] S01: By accessing the data module, access the required video stream data, face and body snapshot data;

[0042] S02: Call the face video structuring and picture structuring computing engines respectively for the video stream data and the human body snapshot image data to perform face and human body structuring, so as to output and obtain corresponding face and human body structured data;

[0043] S03: Connect the basic information data of all face capture devices;

[0044] S04: Call the face clustering analysis algorithm, through the fusion analysis of location, time, structured feature value and structural attribute, aggregate and associate the same faces, output face clustering data, and comple...

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 discloses a method for realizing portrait clustering based on multi-data fusion analysis, which comprises the following steps of: accessing required video stream data and face and human body snapshot picture data through an access data module; respectively calling a face video structuralization and picture structuralization calculation engine for the video stream data and the human body snapshot picture data to carry out face and human body structuralization; calling a face clustering analysis algorithm, and performing aggregation association on the same faces through fusion analysis of places, time, structured feature values and structured attributes to complete one-time portrait clustering analysis; creating and endowing a video identity ID through a video identity ID management module; and calling a human body correlation analysis algorithm to complete correlation analysis of the suspected human body data. According to the method, on the basis of face feature value data comparative analysis, time-space relation analysis is carried out by fusing the equipment basic information and the structured attribute data, so that the data clustering precision is greatly improved, and the aggregation divergence is reduced.

Description

technical field [0001] The invention relates to the field of computer technology, in particular to a method for realizing portrait clustering based on multi-data fusion analysis. Background technique [0002] Face clustering technical solutions in the prior art mainly have the following two implementations: [0003] 1. Based on a built-in static face database feature database, real-time feature value comparison is performed on the connected face snapshot data, and the captured pictures that meet the set threshold will be directly linked to the relevant personnel. The shortcomings of this solution are: first, it relies too much on the existing static face database, and when static data support cannot be provided, the clustering effect is extremely general; second, it does not consider the fusion analysis of other data. [0004] 2. A clustering analysis algorithm for face snapshot data, but only for similarity comparison of face structural feature values, and clustering and m...

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): G06V10/80G06V40/16
CPCG06F18/253
Inventor 孙靖宇高希赵伟伟
Owner 南京启数智能系统有限公司
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