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

A search method and face image technology, applied in computer parts, computing, still image data retrieval, etc., can solve problems such as application scenarios that are not suitable for a large amount of face image feature data, and a single face search method.

Pending Publication Date: 2021-03-26
ZTE CORP
View PDF0 Cites 3 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0003] A face search method, device, device, and computer-readable storage medium provided by the embodiments of the present invention solve the problem in the related art that the face search method is single and is not suitable for application scenarios with a large amount of face image feature data

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 search method, device and equipment, and computer readable storage medium
  • Face search method, device and equipment, and computer readable storage medium
  • Face search method, device and equipment, and computer readable storage medium

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0031] The face search method provided in this embodiment can flexibly select a method of directly matching the feature information to be searched with the feature information in the corresponding feature database according to the data volume of the feature information in the image feature database to be searched currently, And / or firstly match the feature information to be searched with the index corresponding to the feature information in the corresponding feature database, and then perform fine matching between the feature information to be searched and the feature information corresponding to the initial matching index, which enriches the While the face search method is adopted, for the case of a large amount of data in the image feature database to be searched, since the feature information to be searched can be initially matched with the feature index with a small amount of data, and then the fine matching can be performed. While ensuring the search accuracy, it greatly r...

Embodiment 2

[0068] This embodiment provides a face search device, which can be installed in various face search devices, please refer to image 3 shown, including:

[0069] The message processing module 301 is configured to obtain feature information to be searched for a face image to be searched and an image feature database to be searched.

[0070] The first retrieval processing module 302 is configured to match the feature information to be searched with the feature information in the first graphic feature database to be searched when there is a first type of image feature database to be searched in the image feature database to be searched .

[0071] The second retrieval processing module 303 is configured to combine the feature information to be searched with the feature information in the second type graphic feature database to be searched when there is a second type of image feature database to be searched in the image feature database to be searched Matching the corresponding feat...

Embodiment 3

[0085] For ease of understanding, this embodiment will be described below by taking the maintenance of the image feature database and the face search process as examples.

[0086] In this example, each module shown in Embodiment 2 can be started and run according to the corresponding configuration. For example, the first number threshold can be configured according to the current server performance, such as 1,000,000; when the number of feature information currently stored in a certain type of image feature database in the system is less than or equal to 1,000,000, it is the first type of image feature database. When the number of feature information currently stored in the first-type image feature database is greater than 1,000,000, it is the second-type image feature database.

[0087] Based on the above configuration, for the maintenance process of the image feature database, please refer to Figure 5 shown, including:

[0088] The basic library maintenance module perform...

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 embodiment of the invention provides a face search method, device and equipment, and a computer readable storage medium. A mode of directly matching feature information to be searched with featureinformation in a corresponding feature database can be flexibly selected according to the data volume condition of the feature information in a current image feature database to be searched, and / or the feature information to be searched is initially matched with the index corresponding to the feature information in the corresponding feature database, and then the feature information to be searched is finely matched with the feature information corresponding to the index which is successfully matched initially, so that the face search mode is enriched; and meanwhile, as the feature informationto be searched can be initially matched with the feature index with smaller data volume and then finely matched, the number of the feature information needing to be matched one by one can be greatlyreduced, the method can be well suitable for application scenes with large data volume, and the precision of face recognition is ensured.

Description

technical field [0001] The present invention relates to the field of artificial intelligence, in particular to a face search method, device, equipment and computer-readable storage medium. Background technique [0002] In the field of artificial intelligence, the application of face recognition technology is a common application scenario. With the continuous evolution of technology and the accumulation of facial image feature data, the data level of the basic facial feature database is also increasing. And it has reached tens of millions of levels, so in a face recognition search system, it may not only need to deal with small-scale face retrieval, but also need to deal with large-scale face retrieval. The face search method in the related art is to use the post-training algorithm to extract image features from the face images in the face library, and form a picture feature library to save them, and then perform image feature extraction to the pictures that need to be retrie...

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): G06F16/583G06F16/51
CPCG06F16/583G06F16/51G06V40/172G06V40/169
Inventor 潘道华
Owner ZTE CORP
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