CN111104541A - Efficient face picture retrieval method and device - Google Patents

Efficient face picture retrieval method and device Download PDF

Info

Publication number
CN111104541A
CN111104541A CN201911411087.0A CN201911411087A CN111104541A CN 111104541 A CN111104541 A CN 111104541A CN 201911411087 A CN201911411087 A CN 201911411087A CN 111104541 A CN111104541 A CN 111104541A
Authority
CN
China
Prior art keywords
face
feature vector
picture
person
average
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Pending
Application number
CN201911411087.0A
Other languages
Chinese (zh)
Inventor
常鹏
王国威
陈志飞
魏超
朱海勇
张永光
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Xiamen Meiya Pico Information Co Ltd
Original Assignee
Xiamen Meiya Pico Information Co Ltd
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Xiamen Meiya Pico Information Co Ltd filed Critical Xiamen Meiya Pico Information Co Ltd
Priority to CN201911411087.0A priority Critical patent/CN111104541A/en
Publication of CN111104541A publication Critical patent/CN111104541A/en
Pending legal-status Critical Current

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/50—Information retrieval; Database structures therefor; File system structures therefor of still image data
    • G06F16/58—Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually
    • G06F16/583—Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually using metadata automatically derived from the content
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/50—Information retrieval; Database structures therefor; File system structures therefor of still image data
    • G06F16/58—Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually
    • G06F16/5866—Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually using information manually generated, e.g. tags, keywords, comments, manually generated location and time information
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F18/00—Pattern recognition
    • G06F18/20—Analysing
    • G06F18/22—Matching criteria, e.g. proximity measures
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V40/00—Recognition of biometric, human-related or animal-related patterns in image or video data
    • G06V40/10—Human or animal bodies, e.g. vehicle occupants or pedestrians; Body parts, e.g. hands
    • G06V40/16—Human faces, e.g. facial parts, sketches or expressions
    • G06V40/168—Feature extraction; Face representation
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V40/00—Recognition of biometric, human-related or animal-related patterns in image or video data
    • G06V40/10—Human or animal bodies, e.g. vehicle occupants or pedestrians; Body parts, e.g. hands
    • G06V40/16—Human faces, e.g. facial parts, sketches or expressions
    • G06V40/172—Classification, e.g. identification

Landscapes

  • Engineering & Computer Science (AREA)
  • Theoretical Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Data Mining & Analysis (AREA)
  • Library & Information Science (AREA)
  • Oral & Maxillofacial Surgery (AREA)
  • Health & Medical Sciences (AREA)
  • General Engineering & Computer Science (AREA)
  • General Health & Medical Sciences (AREA)
  • Databases & Information Systems (AREA)
  • Multimedia (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Human Computer Interaction (AREA)
  • Bioinformatics & Cheminformatics (AREA)
  • Artificial Intelligence (AREA)
  • Bioinformatics & Computational Biology (AREA)
  • Evolutionary Biology (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Evolutionary Computation (AREA)
  • Collating Specific Patterns (AREA)
  • Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
  • Image Analysis (AREA)

Abstract

The invention relates to a high-efficiency face picture retrieval method and device. The method comprises the following steps: creating a face comparison sample data table; after receiving a new face picture, comparing the feature vector of the newly received face picture with the average feature vector of each person in a face comparison sample data table and calculating the similarity; if the similarity between the newly received face picture and the face picture of the existing person in the data table reaches the threshold value, the newly received face picture is considered to be the same person as the existing person, the newly received face picture is marked as the unique serial number of the existing person, the average characteristic vector of the serial number is updated, if the similarity does not reach the threshold value, a new unique serial number is distributed, the newly received face picture characteristic vector is used as the average characteristic vector of the serial number and is stored in a face comparison sample data table, and then the picture and the serial number are stored in a face snapshot record table. The apparatus includes a memory and a processor that implements the method when executing instructions stored in the memory.

Description

Efficient face picture retrieval method and device
Technical Field
The invention relates to a face picture retrieval method and device, and belongs to the technical field of image recognition software.
Background
Human face recognition is a typical technology of biological identity recognition, and active cooperation of detected individuals is not needed, so that the human face recognition technology is widely applied to human-computer interaction, security protection, identity verification and the like in recent years.
In systems such as smart cities, security and public security, the number of collected face pictures can be as many as billions or more. Generally, a face picture to be inquired is submitted to a system, and the system inquires and returns information such as which cameras the face picture to be inquired appears at, the appearance time, the snap shot at that time and the like. If the feature vector of the face picture to be inquired is compared with all stored feature vectors of the face picture in the database in a ratio of 1:1 according to the traditional method, the comparison is carried out for billions or more, and the efficiency is very low.
In the existing scheme, each serial number records a face picture group, and each time a new face picture is received, the new face picture is compared with all face picture feature vectors of each person in a sample data table to determine whether the similarity of the existing person in the sample data table meets the requirement. Similarly, each time the system submits a face picture to be queried, the system needs to compare with all face picture feature vectors of each person in the sample data table to obtain a number meeting the conditions. In the two processes, comparison with all face image feature vectors of each person in a sample data table is involved, the comparison mode has high performance consumption and low query speed, and particularly, with the continuous increase of the number of new face images and the number of times of snapshot record query every day, the query time delay is increased under the condition of the same resource, so that the requirement of high real-time performance cannot be met.
Disclosure of Invention
The invention provides a high-efficiency face picture retrieval method and device, and aims to at least solve one of the technical problems in the prior art.
The technical scheme of the invention relates to a face picture retrieval method, which comprises the following steps:
A. creating a face comparison sample data table, limiting the same person to be assigned with a unique number in the data table, and then recording the average characteristic vector of all different face pictures of each person and the number of the face pictures as K;
B. after a new face picture is received and feature extraction is carried out, comparing the feature vector of the newly received face picture with the average feature vector of each person in a face comparison sample data table, and calculating the similarity;
C. judging whether the similarity of the face picture of the existing person in the newly received face picture and the face comparison sample data table reaches a threshold value, if so, considering that the person in the newly received face picture and the existing person are the same person, marking the newly received face picture as the unique number of the existing person, updating the average feature vector of the number and the number of the corresponding face pictures to be K +1, then storing the picture and the number into a face snapshot record table, and if not, executing the next step;
D. distributing a new unique number, taking the feature vector of the newly received face picture as the average feature vector of the number, recording the number of the face picture corresponding to the number as 1, storing the number in a face comparison sample data table, and storing the picture and the number in a face snapshot record table.
In some aspects of the invention, the method further comprises:
E. for the face picture to be inquired, firstly, comparing the feature vector of the face picture with the average feature vector of each person in the face comparison sample data table, if the similarity of the existing person reaches a certain threshold value, acquiring the serial number of the existing person, and inquiring all matched face snapshot records in the face snapshot record table by using the serial number.
In some aspects of the invention, step C comprises:
c1, when a snapshot picture is received, calculating a feature vector of the picture to obtain a first feature vector;
c2, comparing the first feature vector with the face comparison sample data table;
c3, if the first average characteristic vector and the first characteristic vector are found to exceed the similarity threshold, the personnel information ID corresponding to the first average characteristic vector is the first ID;
and C4, if the average characteristic vector and the first characteristic vector cannot be found to exceed the similarity threshold, creating a new person information in the face comparison sample data table, and assigning the ID of the person information to be a second ID.
In some aspects of the invention, step C3 includes: recording the personnel number of the piece of face snapshot data as a first ID, and storing the first ID in a face snapshot record table; and performing update calculation on the first average feature vector.
In some aspects of the invention, step C4 includes: recording the personnel number of the piece of face snapshot data as a second ID, and storing the second ID in a face snapshot record table; taking the first feature vector as a second average feature vector of a second ID of the newly created person; and associating the second average characteristic vector and the second ID into a record and storing the record into a face comparison sample data table.
In some aspects of the invention, step C further comprises:
p-dimensional feature vector configuration for newly received pictures as X1＝(x11,x12,…,x1p) And the p-dimensional average feature vector of the image of the existing personnel number with the similarity meeting the requirement is configured to be X2＝(x21,x22,…,x2p) Wherein the dimension P corresponds to the number of pictures K;
to X1And X2Calculating an average feature vector, wherein the average of each one-dimensional component is calculated as
Figure BDA0002349969840000021
Wherein j is more than or equal to 1 and less than or equal to p;
updating the average feature vector corresponding to the number as
Figure BDA0002349969840000022
In some aspects of the invention, step E comprises: when a user submits a query picture, calculating a feature vector of the picture to obtain a first query feature vector; comparing the first feature vector with the face comparison sample data table by using the query: if the first average characteristic vector is found to exceed a threshold value, inquiring a face snapshot record table by using the corresponding first ID of the person, and returning an inquiry result as a face track; if no average feature vector is found that exceeds the threshold, a result is returned that the track is empty.
The invention also relates to a computer device comprising a memory and a processor, wherein the processor executes a computer program stored in the memory to implement the method.
The invention also relates to a computer-readable storage medium, on which computer program instructions are stored, which, when executed by a processor, implement the above-mentioned method.
The invention provides an improved and efficient human face picture retrieval scheme, wherein the number of features of each person in a comparison sample data table is not required to be more, and the capacity of the comparison sample data table is saved. Meanwhile, when a track of one person is searched, the track is compared with a comparison sample data table only, and the comparison with all the snap-shot picture characteristics is not needed, so that the query performance is expected to be improved by multiple times. Therefore, the efficiency of the face picture query can be further improved, the query response speed of a user is improved, the requirement on hardware computing resources is reduced, and the hardware investment is reduced.
Taking 100W population in a city as an example, assuming that 1000W face snapshot pictures are taken every day and 3 hundred million data are taken in one month, the traditional scheme needs to compare 3 hundred million features by inquiring data in one month, and the scheme only needs to compare with a comparison sample data table of 100W population, so that the comparison efficiency is improved by hundreds of times.
Drawings
Fig. 1 shows a general flow chart of the method according to the invention.
Fig. 2 is a flowchart illustrating a process of newly receiving a face picture according to a first embodiment of the present invention.
Fig. 3 is a flowchart illustrating a process of a face picture to be queried according to a second embodiment of the invention.
Detailed Description
The conception, the specific structure and the technical effects of the present invention will be clearly and completely described in conjunction with the embodiments and the accompanying drawings to fully understand the objects, the schemes and the effects of the present invention.
It will be understood that, although the terms first, second, third, etc. may be used herein to describe various elements, these elements should not be limited by these terms. These terms are only used to distinguish one element of the same type from another. For example, a first element could be termed a second element, and, similarly, a second element could be termed a first element, without departing from the scope of the present disclosure. The use of any and all examples, or exemplary language ("e.g.," such as "or the like") provided herein, is intended merely to better illuminate embodiments of the invention and does not pose a limitation on the scope of the invention unless otherwise claimed.
Referring to fig. 1, a face image retrieval method according to the present invention generally includes the following steps:
A. creating a face comparison sample data table, limiting the same person to be assigned with a unique number in the data table, and then recording the average characteristic vector of all different face pictures of each person and the number of the face pictures as K;
B. after a new face picture is received and feature extraction is carried out, comparing the feature vector of the newly received face picture with the average feature vector of each person in a face comparison sample data table, and calculating the similarity;
C. judging whether the similarity of the face picture of the existing person in the newly received face picture and the face comparison sample data table reaches a threshold value, if so, considering that the person in the newly received face picture and the existing person are the same person, marking the newly received face picture as the unique number of the existing person, updating the average feature vector of the number and the number of the corresponding face pictures to be K +1, then storing the picture and the number into a face snapshot record table, and if not, executing the next step;
D. distributing a new unique number, taking the feature vector of the newly received face picture as the average feature vector of the number, recording the number of the face picture corresponding to the number as 1, storing the number in a face comparison sample data table, and storing the picture and the number in a face snapshot record table.
E. For the face picture to be inquired, firstly, comparing the feature vector of the face picture with the average feature vector of each person in the face comparison sample data table, if the similarity of the existing person reaches a certain threshold value, acquiring the serial number of the existing person, and inquiring all matched face snapshot records in the face snapshot record table by using the serial number.
In an embodiment, the average feature vector in the method according to the invention is calculated as follows.
Suppose the feature vector of the newly received picture is X1＝(x11,x12,…,x1p) (P dimension characteristic vector) and inquiring the existing personnel number with the similarity meeting the requirement, wherein the average characteristic vector is X2＝(x21,x22,…,x2p) (P dimension feature vector), the average feature vector is the average feature vector of K feature vectors (the number of corresponding face pictures is K).
To X1And X2Calculating an average feature vector, wherein the average value of each dimension component is:
Figure BDA0002349969840000041
then the average eigenvector for X1 and X2 is
Figure BDA0002349969840000042
Wherein, K/10 and 1/10 are weights (weight values can be adjusted as required) corresponding to the two feature vectors, X2 is an average feature vector of K face pictures, and X1 is a feature vector corresponding to a newly received picture, which is only a feature vector of 1 face picture, and the weights are different.
Referring to fig. 2, in an embodiment of the method according to the present invention, a specific example and steps of a processing procedure when a face snapshot picture is received are as follows:
1. when a snapshot picture is received, calculating a feature vector of the picture to obtain a feature vector 1;
2. and comparing the feature vector 1 with a face comparison sample data table:
a) if the average characteristic vector 1 and the characteristic vector 1 are found to exceed the similarity threshold, the personnel information ID corresponding to the average characteristic vector 1 is ID1
i. Recording the personnel number of the piece of face snapshot data as ID1, and storing the personnel number in a face snapshot record table;
performing updating calculation on the average characteristic vector 1, wherein the updating algorithm is as the formula;
b) if the average feature vector and the feature vector 1 cannot be found to exceed the similarity threshold, creating new personnel information in the face comparison sample data table, and assigning the ID of the personnel information to be ID2
i. Recording the personnel number of the piece of face snapshot data as ID2, and storing the personnel number in a face snapshot record table;
using feature vector 1 as the average feature vector of the newly created person ID2, i.e., average feature vector 2;
and iii, storing the { average feature vector 2, ID2} as a record into a face comparison sample data table.
Referring to fig. 3, in an embodiment of the method according to the present invention, a specific example and steps of a processing procedure for querying a face snapshot track are as follows:
1. when a user submits a query picture, calculating a feature vector of the picture to obtain a query feature vector 1;
2. comparing the query feature vector 1 with a face comparison sample data table:
a) if the average feature vector 1 is found to exceed the threshold value, querying a face snapshot record table by using the corresponding personnel ID1, and returning a query result as a face track;
b) if no average feature vector is found that exceeds the threshold, a result is returned that the track is empty.
It should be recognized that the method steps in embodiments of the present invention may be embodied or carried out by computer hardware, a combination of hardware and software, or by computer instructions stored in a non-transitory computer readable memory. The method may use standard programming techniques. Each program may be implemented in a high level procedural or object oriented programming language to communicate with a computer system. However, the program(s) can be implemented in assembly or machine language, if desired. In any case, the language may be a compiled or interpreted language. Furthermore, the program can be run on a programmed application specific integrated circuit for this purpose.
Further, the operations of processes described herein can be performed in any suitable order unless otherwise indicated herein or otherwise clearly contradicted by context. The processes described herein (or variations and/or combinations thereof) may be performed under the control of one or more computer systems configured with executable instructions, and may be implemented as code (e.g., executable instructions, one or more computer programs, or one or more applications) collectively executed on one or more processors, by hardware, or combinations thereof. The computer program includes a plurality of instructions executable by one or more processors.
Further, the method may be implemented in any type of computing platform operatively connected to a suitable interface, including but not limited to a personal computer, mini computer, mainframe, workstation, networked or distributed computing environment, separate or integrated computer platform, or in communication with a charged particle tool or other imaging device, and the like. Aspects of the invention may be embodied in machine-readable code stored on a non-transitory storage medium or device, whether removable or integrated into a computing platform, such as a hard disk, optically read and/or write storage medium, RAM, ROM, or the like, such that it may be read by a programmable computer, which when read by the storage medium or device, is operative to configure and operate the computer to perform the procedures described herein. Further, the machine-readable code, or portions thereof, may be transmitted over a wired or wireless network. The invention described herein includes these and other different types of non-transitory computer-readable storage media when such media include instructions or programs that implement the steps described above in conjunction with a microprocessor or other data processor. The invention may also include the computer itself when programmed according to the methods and techniques described herein.
A computer program can be applied to input data to perform the functions described herein to transform the input data to generate output data that is stored to non-volatile memory. The output information may also be applied to one or more output devices, such as a display. In a preferred embodiment of the invention, the transformed data represents physical and tangible objects, including particular visual depictions of physical and tangible objects produced on a display.
The above description is only a preferred embodiment of the present invention, and the present invention is not limited to the above embodiment, and any modifications, equivalent substitutions, improvements, etc. within the spirit and principle of the present invention should be included in the protection scope of the present invention as long as the technical effects of the present invention are achieved by the same means. The invention is capable of other modifications and variations in its technical solution and/or its implementation, within the scope of protection of the invention.

Claims (9)

1. A face picture retrieval method is characterized by comprising the following steps:
A. creating a face comparison sample data table, limiting the same person to be assigned with a unique number in the data table, and then recording the average characteristic vector of all different face pictures of each person and the number of the face pictures as K;
B. after a new face picture is received and feature extraction is carried out, comparing the feature vector of the newly received face picture with the average feature vector of each person in a face comparison sample data table, and calculating the similarity;
C. judging whether the similarity of the face picture of the existing person in the newly received face picture and the face comparison sample data table reaches a threshold value, if so, considering that the person in the newly received face picture and the existing person are the same person, marking the newly received face picture as the unique number of the existing person, updating the average feature vector of the number and the number of the corresponding face pictures to be K +1, then storing the picture and the number into a face snapshot record table, and if not, executing the next step;
D. distributing a new unique number, taking the feature vector of the newly received face picture as the average feature vector of the number, recording the number of the face picture corresponding to the number as 1, storing the number in a face comparison sample data table, and storing the picture and the number in a face snapshot record table.
2. The method of claim 1, wherein the method further comprises:
E. for the face picture to be inquired, firstly, comparing the feature vector of the face picture with the average feature vector of each person in the face comparison sample data table, if the similarity of the existing person reaches a certain threshold value, acquiring the serial number of the existing person, and inquiring all matched face snapshot records in the face snapshot record table by using the serial number.
3. The method according to claim 1 or 2, wherein said step C comprises:
c1, when a snapshot picture is received, calculating a feature vector of the picture to obtain a first feature vector;
c2, comparing the first feature vector with the face comparison sample data table;
c3, if the first average characteristic vector and the first characteristic vector are found to exceed the similarity threshold, the personnel information ID corresponding to the first average characteristic vector is the first ID;
and C4, if the average characteristic vector and the first characteristic vector cannot be found to exceed the similarity threshold, creating a new person information in the face comparison sample data table, and assigning the ID of the person information to be a second ID.
4. The method of claim 3, wherein said step C3 includes:
recording the personnel number of the piece of face snapshot data as a first ID, and storing the first ID in a face snapshot record table;
and performing update calculation on the first average feature vector.
5. The method of claim 3, wherein said step C4 includes:
recording the personnel number of the piece of face snapshot data as a second ID, and storing the second ID in a face snapshot record table;
taking the first feature vector as a second average feature vector of a second ID of the newly created person;
and associating the second average characteristic vector and the second ID into a record and storing the record into a face comparison sample data table.
6. The method according to claim 1 or 2, wherein said step C further comprises:
p-dimensional feature vector configuration for newly received pictures as X1＝(x11,x12,…,x1p) And the p-dimensional average feature vector of the image of the existing personnel number with the similarity meeting the requirement is configured to be X2＝(x21,x22,…,x2p) Wherein the dimension P corresponds to the number of pictures K;
to X1And X2Calculating an average feature vector, wherein the average of each one-dimensional component is calculated as
Figure FDA0002349969830000021
Wherein j is more than or equal to 1 and less than or equal to p;
updating the average feature vector corresponding to the number as
Figure FDA0002349969830000022
7. The method of claim 2, wherein the step E comprises:
when a user submits a query picture, calculating a feature vector of the picture to obtain a first query feature vector;
comparing the first feature vector with the face comparison sample data table by using the query:
if the first average characteristic vector is found to exceed a threshold value, inquiring a face snapshot record table by using the corresponding first ID of the person, and returning an inquiry result as a face track;
if no average feature vector is found that exceeds the threshold, a result is returned that the track is empty.
8. A computer arrangement comprising a memory and a processor, wherein the processor implements the method of any one of claims 1 to 7 when executing a computer program stored in the memory.
9. A computer readable storage medium having stored thereon program instructions which, when executed by a processor, implement the method of any one of claims 1 to 7.
CN201911411087.0A 2019-12-31 2019-12-31 Efficient face picture retrieval method and device Pending CN111104541A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201911411087.0A CN111104541A (en) 2019-12-31 2019-12-31 Efficient face picture retrieval method and device

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201911411087.0A CN111104541A (en) 2019-12-31 2019-12-31 Efficient face picture retrieval method and device

Publications (1)

Publication Number Publication Date
CN111104541A true CN111104541A (en) 2020-05-05

Family

ID=70424156

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201911411087.0A Pending CN111104541A (en) 2019-12-31 2019-12-31 Efficient face picture retrieval method and device

Country Status (1)

Country Link
CN (1) CN111104541A (en)

Cited By (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN111598012A (en) * 2020-05-19 2020-08-28 恒睿(重庆)人工智能技术研究院有限公司 Picture clustering management method, system, device and medium
CN112241684A (en) * 2020-09-16 2021-01-19 四川天翼网络服务有限公司 Face retrieval distributed computing method and system
CN112487222A (en) * 2020-11-30 2021-03-12 江苏正赫通信息科技有限公司 Method for quickly searching and effectively storing similar human faces

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20170132457A1 (en) * 2014-06-27 2017-05-11 Beijing Qihoo Technology Company Limited Human face similarity recognition method and system
CN108170732A (en) * 2017-12-14 2018-06-15 厦门市美亚柏科信息股份有限公司 Face picture search method and computer readable storage medium
CN108229330A (en) * 2017-12-07 2018-06-29 深圳市商汤科技有限公司 Face fusion recognition methods and device, electronic equipment and storage medium
CN109858354A (en) * 2018-12-27 2019-06-07 厦门市美亚柏科信息股份有限公司 A kind of face identity library, the foundation of track table and face track querying method and system

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20170132457A1 (en) * 2014-06-27 2017-05-11 Beijing Qihoo Technology Company Limited Human face similarity recognition method and system
CN108229330A (en) * 2017-12-07 2018-06-29 深圳市商汤科技有限公司 Face fusion recognition methods and device, electronic equipment and storage medium
CN108170732A (en) * 2017-12-14 2018-06-15 厦门市美亚柏科信息股份有限公司 Face picture search method and computer readable storage medium
CN109858354A (en) * 2018-12-27 2019-06-07 厦门市美亚柏科信息股份有限公司 A kind of face identity library, the foundation of track table and face track querying method and system

Cited By (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN111598012A (en) * 2020-05-19 2020-08-28 恒睿(重庆)人工智能技术研究院有限公司 Picture clustering management method, system, device and medium
CN111598012B (en) * 2020-05-19 2021-11-12 恒睿(重庆)人工智能技术研究院有限公司 Picture clustering management method, system, device and medium
CN112241684A (en) * 2020-09-16 2021-01-19 四川天翼网络服务有限公司 Face retrieval distributed computing method and system
CN112487222A (en) * 2020-11-30 2021-03-12 江苏正赫通信息科技有限公司 Method for quickly searching and effectively storing similar human faces

Similar Documents

Publication Publication Date Title
CN108229314B (en) Target person searching method and device and electronic equipment
CN111104541A (en) Efficient face picture retrieval method and device
CN108830138B (en) Livestock identification method, device and storage medium
CN108170732A (en) Face picture search method and computer readable storage medium
CN109784270B (en) Processing method for improving face picture recognition integrity
CN109948735B (en) Multi-label classification method, system, device and storage medium
CN109063041B (en) Method and device for embedding relational network graph
CN106649503A (en) Query method and system based on sql
CN110738650B (en) Infectious disease infection identification method, terminal device and storage medium
CN115862088A (en) Identity recognition method and device
US20190079796A1 (en) Computational resource management device, computational resource management method, and computer-readable recording medium
US20180307720A1 (en) System and method for learning-based group tagging
CN110009662B (en) Face tracking method and device, electronic equipment and computer readable storage medium
US20150339591A1 (en) Collegial Activity Learning Between Heterogeneous Sensors
JP2021503123A (en) Video summary generation methods and devices, electronic devices and computer storage media
CN105844647A (en) Kernel-related target tracking method based on color attributes
JP2013105373A (en) Device, method, and program for data acquisition
KR101780534B1 (en) Method and system for extracting image feature based on map-reduce for searching image
JP5890340B2 (en) Image classification device and image classification program
JP2012155394A (en) Document classification learning control device, document classification device and computer program
CN112766390A (en) Method, device and equipment for determining training sample
CN110738175A (en) Face image processing method and device, computer equipment and storage medium
CN108958917A (en) The method for scheduling task and system of cloud computing
US11709798B2 (en) Hash suppression
CN115378806A (en) Flow distribution method and device, computer equipment and storage medium

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
RJ01 Rejection of invention patent application after publication

Application publication date: 20200505

RJ01 Rejection of invention patent application after publication