CN104933391B - Method and apparatus for carrying out face recognition and electronic equipment - Google Patents
Method and apparatus for carrying out face recognition and electronic equipment Download PDFInfo
- Publication number
- CN104933391B CN104933391B CN201410105651.7A CN201410105651A CN104933391B CN 104933391 B CN104933391 B CN 104933391B CN 201410105651 A CN201410105651 A CN 201410105651A CN 104933391 B CN104933391 B CN 104933391B
- Authority
- CN
- China
- Prior art keywords
- information
- target
- face
- image
- database
- 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.)
- Active
Links
Landscapes
- Image Analysis (AREA)
Abstract
An embodiment of the present invention provides a kind of method and apparatus for carrying out face recognition and the electronic equipments of the device including the carry out face recognition.The method for carrying out face recognition includes：The information of the target face is detected from the target image including the target face to be identified；It obtains and the relevant image related information of the target image；The database for including multiple facial characteristics templates is established based on described image related information；The target face is identified based on the information of the target face detected and the database.In technical solution according to the ... of the embodiment of the present invention, by based on establishing multiple facial characteristics templates with target image relevant image related information and identifying the target face based on the facial characteristics template established, can more accurately and efficiently carry out face recognition.
Description
Technical field
The present invention relates to information technology field, more particularly, to a kind of method and apparatus for carrying out face recognition,
And the electronic equipment of the device including the carry out face recognition.
Background technology
In recent years, various technologies have been developed to carry out identification, including facial recognition techniques, fingerprint identification technology etc..
Face recognition uses general video camera as identification information acquisition device, obtains the face of identification object in a non-contact manner
Then image completes identification process by the way that acquired image is compared with database images.
Face recognition is the identification method based on biological characteristic, compared with other identification methods such as fingerprint recognition, is had real
When, high-precision, stability it is high, it is difficult counterfeit, cost-effective and it is non-the characteristics such as invade and harass, be easier to be easily accepted by a user.However, different
Difference between body is little, and the structure of all faces is all similar or even construction profile of face organ is all much like.It is such
Feature for using face carry out positioning be advantageous, but for be using facial area decomposing biological individual it is unfavorable, especially
In the case of the database to be compared is bigger.
Accordingly, it is desired to provide a kind of facial recognition techniques accurately carry out identification.
Invention content
An embodiment of the present invention provides it is a kind of for carry out face recognition method, carry out face recognition device and
The electronic equipment of device including the carry out face recognition makes it possible to accurately and efficiently carry out face recognition.
On the one hand, a kind of method for carrying out face recognition is provided, it may include：From including the target face to be identified
Target image in detect the information of the target face；It obtains and the relevant image related information of the target image；It is based on
Described image related information includes the database of multiple facial characteristics templates to establish；Information based on the target face detected
The target face is identified with the database.
In the method for carrying out face recognition, the target image can be the picture in video frequency program, the figure
As related information can be the title of the video frequency program.
It is described to be established including multiple faces based on described image related information in the method for carrying out face recognition
The database of feature templates may include：Performance is participated in the video frequency program to obtain based on the title of the video frequency program
The name of performer；The database of the facial characteristics template including each performer is established based on the name of acquired performer.
In the method for carrying out face recognition, the video frequency program can be the section obtained by video on-demand system
Mesh, the acquisition may include with the relevant image related information of the target image：Pass through the electronic programming of the video frequency program
Guide obtains and the relevant image related information of the target image.
In the method for carrying out face recognition, it may also include：It obtains corresponding with the target face identified attached
Information；The satellite information is synchronously shown with the target face.The mesh that the satellite information for example may include and be identified
At least one of the corresponding Real Name of mark face, alias, age, occupation, representative work, hobby.
In the method for carrying out face recognition, the target image can be the picture shot in a social activities
Face, described image related information can be the titles of the social activities.
It is described to be established including multiple faces based on described image related information in the method for carrying out face recognition
The template database of feature may include：The name for the participant for participating in the social activities is obtained based on the title of the social activities
It is single；The database of the facial characteristics template including each participant is established based on the list of acquired participant.
On the other hand, a kind of device for carrying out face recognition is provided, it may include：Detection unit, for from including
The information of the target face is detected in the target image for the target face to be identified；First acquisition unit, for acquisition and institute
State the relevant image related information of target image；Template establishes unit, for based on described image related information come establish including
The database of multiple facial characteristics templates；Recognition unit, for based on the target face detected information and the database
To identify the target face.
In the device for carrying out face recognition, the detection unit can detect institute from the picture in video frequency program
The information of target face is stated, the name that the first acquisition unit can obtain the video frequency program is referred to as described image association letter
Breath.
In the device for carrying out face recognition, the template, which establishes unit, to include multiple by operating foundation as follows
The database of facial characteristics template：Drilling for performance is participated in the video frequency program to obtain based on the title of the video frequency program
The name of member；The database of the facial characteristics template including each performer is established based on the name of acquired performer.
In the device for carrying out face recognition, the video frequency program can be the section obtained by video on-demand system
Mesh, the first acquisition unit can be obtained relevant with the target image using the electronic program guides of the video frequency program
Image related information.
In the device for carrying out face recognition, it may also include：Second acquisition unit, the mesh for obtaining Yu being identified
The corresponding satellite information of mark face；Display unit, for synchronously showing the satellite information with the target face.It is described attached
It may include Real Name corresponding with the target face identified, alias, age, occupation, representative work, hobby to belong to information
At least one of.
In the device for carrying out face recognition, picture that the detection unit can be shot from a social activities
The information of the middle detection target face, the name that the first acquisition unit can obtain the social activities are referred to as described image
Related information.Correspondingly, the template establish unit can by operate as follows establish include multiple facial characteristics templates data
Library：The list for the participant for participating in the social activities is obtained based on the title of the social activities；Based on acquired participation
The list of person establishes the database of the facial characteristics template including each participant.
Another aspect, provides a kind of electronic equipment, it may include the device as described above for carrying out face recognition.
According to the ... of the embodiment of the present invention above-mentioned for carrying out the method and apparatus of face recognition and including the carry out face
In the electronic equipment of the device of portion's identification, multiple facial characteristics are established with the relevant image related information of target image by being based on
Template simultaneously identifies the target face based on the facial characteristics template established, and can more accurately and efficiently carry out face
Portion identifies.
Description of the drawings
In order to illustrate the technical solution of the embodiments of the present invention more clearly, below will be in embodiment or description of the prior art
Required attached drawing is briefly described, it should be apparent that, the accompanying drawings in the following description is only some realities of the present invention
Example is applied, it for those of ordinary skill in the art, without creative efforts, can also be according to these attached drawings
Obtain other attached drawings.
Fig. 1 is to illustrate the Organization Chart according to the ... of the embodiment of the present invention for carrying out face recognition；
Fig. 2 is the flow chart for schematically illustrating the method according to the ... of the embodiment of the present invention for carrying out face recognition；
Fig. 3 is the flow chart for schematically illustrating another method for carrying out face recognition according to the ... of the embodiment of the present invention；
Fig. 4 is the block diagram for schematically illustrating the device according to the ... of the embodiment of the present invention for carrying out face recognition.
Specific implementation mode
Following will be combined with the drawings in the embodiments of the present invention, and technical solution in the embodiment of the present invention carries out clear, complete
Site preparation describes, it is clear that described embodiments are some of the embodiments of the present invention, instead of all the embodiments, is not conflicting
In the case of, the features in the embodiments and the embodiments of the present application mutually can be combined arbitrarily.
Fig. 1 is to illustrate the Organization Chart according to the ... of the embodiment of the present invention for carrying out face recognition.
As shown in Figure 1, the face-image 20 of someone is displayed in display equipment 10, which can be all
Can also be the image-carrier of such as printing paper, poster etc such as the display screen of liquid crystal display, projection screen, video camera etc.It is any
The substance of image can be presented or device all can serve as the display equipment 10.When presenting institute in the display equipment 10
When paying close attention to the face-image of face, face recognition is carried out to the face-image using the face recognition device 30, with determination
The piece identity of the face-image.
Fig. 2 is the flow chart for schematically illustrating the method 200 according to the ... of the embodiment of the present invention for carrying out face recognition.It is described
The method 200 for carrying out face recognition is the method for carrying out face recognition based on the face-image 20 in the display equipment 10 in Fig. 1.
As shown in Fig. 2, the method 200 of the carry out face recognition may include：From the target for including the target face to be identified
The information of the target face is detected in image（S210）；It obtains and the relevant image related information of the target image
（S220）；The database for including multiple facial characteristics templates is established based on described image related information（S230）；Based on being examined
The information of the target face of survey and the database identify the target face（S240）.
In S210, the information of the target face is detected from the target image including the target face to be identified.Institute
The image shown in the display equipment 10 that target image is Fig. 1 is stated, can be the image shown on a liquid crystal display, also may be used
To be the image presented on film poster.There are the face of personage of interest in the target image, that is, the mesh to be identified
Mark face.For example, the target image can be the picture in video frequency program, the picture shot in a social activities can also be
Face.The target image for example can be the still image or dynamic image acquired by pick-up lens, and it includes described that can be
The multiple images of target face, multiple image can have different positions, different expressions etc..
The information of the target face is the letter that can characterize the feature of the target face to be distinguished with other faces
Breath.As an example, the information of the target face can be the organ for characterizing face（For example, eyes, nose, mouth, chin
Deng）Shape and the distance between each organ characteristic, such as including face characteristic point between Euclidean distance,
Curvature and angle etc..Information of any other characteristic as the target face can also be used, as long as it being capable of table
Levy the feature of face.
In the S210, the information of the target face can be detected based on the target image.It specifically, can be from institute
Position and the size for determining face in target image are stated, that is, carries out Face location；The image for the target face being originally taken due to
It is limited by various conditions and random disturbances, tends not to directly use, it is necessary to it be pre-processed, such as to face
Image carries out light compensation, greyscale transformation, histogram equalization, normalization, filtering and sharpening etc.；Then the spy of face is extracted
Levy the information as the target face.In practice, it for the information of different types of target face, can take different
Technology is detected, and the mode for specifically detecting the information of the target face does not constitute limitation to the embodiment of the present invention.
In S220, obtain and the relevant image related information of the target image.The image related information is in addition to institute
State except the content in target image with the relevant information of target image.The feelings of picture in target image is video frequency program
In condition, described image related information can be the title of the video frequency program；It is in a social activities in the target image
In the case of the picture of shooting, described image related information can be the title of the social activities.For example, in variety show
The picture of the video frequency program of " I is singer " includes the face-image of a singer（That is target face）In situation, described image is closed
Connection information can be the title " I is singer " of the variety show.
If the face module of the information of the target face obtained in S210 and all personnel compared to carry out face
Identification will consume a large amount of time and resource since there are many face template for comparison, and similar due to face structure
Property, it may not be possible to the accurately face in recognition target image.Described image related information is obtained in S220, then it can be true
Surely the relevant information of the target image is obtained, to be compared in the personnel involved in described image related information to know
Other target face, this greatly reduces the range of comparison, and improves the accuracy of identification.
In the video broadcast system of such as cable television, video frequency program is generally included in the video programming signal received
Electronic program guides, therefore can easily obtain described image related information using the electronic program guides.For example,
In the case of the video frequency program is the program obtained by video on-demand system, in the S220, the video can be passed through
The electronic program guides of program（EPG, electronic program guide）To obtain and the relevant figure of the target image
As related information.That is, in the case of with electronic program guides, the S220 can be automatically realized.
In the case of the target image is the picture shot in a social activities, the on-the-spot meeting of usual social activities
The advertisement information about the social activities is puted up, can be determined from the advertisement information about the social activities in target image
The title of the social activities, as image related information.The social activities can be recorded according to target image and manually
Title, Text region can also be carried out to word in the target image and determine the title of the social activities.Therefore, may be used
The suitable means are taken to obtain described image related information with the concrete condition of combining target image, the image related information
Specific acquisition modes do not constitute the limitation to the embodiment of the present invention.
In S230, the database for including multiple facial characteristics templates is established based on described image related information.The number
It is used to be compared with the information of the target face according to each facial characteristics template in library, to carry out face recognition.
It is described based on described image association letter in the case of described image related information is the title of the video frequency program
Breath may include to establish the database including multiple facial characteristics templates：It is obtained described based on the title of the video frequency program
The name of the performer of performance is participated in video frequency program；The face for including each performer is established based on the name of acquired performer
The database of feature templates.By taking the video frequency program is " I is singer " as an example, all of participation " I is singer " can be established
All staff on board in the facial characteristics template of singer, " I is singer "（Including host, musician etc.）Facial characteristics mould
Plate etc..The database may include all facial characteristics templates for being likely to occur personnel in the target image, greatly contract
Subtracted will with the quantity of the facial characteristics template of the information comparison of the target face obtained in S210, and due to target image
In target face participated in " I is singer ", so the personnel of the target face should be in the database in, to
It can accurately be identified.
Alternatively, described to be based on the figure in the case of described image related information is the title of the social activities
As related information may include to establish the template database including multiple facial characteristics：It is obtained based on the title of the social activities
Take the list for the participant for participating in the social activities；It is established including each participant's based on the list of acquired participant
The database of facial characteristics template.The social activities is such as can be training, a meeting, a product news conference.It is logical
Often, the information that the organizer of social activities can be with worker therein, invitation personnel（For example, name, position etc.）, being based on should
Information can be obtained the list of the participant, and establish the database for the facial characteristics template for including each participant.
Each facial characteristics template in the database can be obtained from other databases, can also be based on each
Other images of personnel representated by facial characteristics template are established.It therefore, can based on described image related information in S230
With the facial characteristics template to be compared of the smaller range of determination, the target face is promptly identified so as to more acurrate.
In S240, the target face is identified based on the information of the target face detected and the database.Example
It such as, can be by the information of the target face detected in S210 and the facial characteristics mould in the database established in S230
Plate scans for matching, with the similarity of the information of the determination target face and each facial characteristics template, when similarity is super
When crossing a threshold value, then the target face detected is identified as to the personnel representated by the facial characteristics template.Alternatively, Ke Yicong
Each similarity determines the facial characteristics template with maximum similarity, and the target face detected is identified as with maximum
Personnel representated by the facial characteristics template of similarity.Therefore, the S240 be exactly by the information of target face to be identified with
Obtained facial characteristics template is compared, and is judged the identity information of target face according to similarity degree.
Based on the target face identified in S240, other applications can be further executed.For example, by safeguard protection
Area can pass through recognition of face recognize the person of seeking entry into identity；Monitoring system is installed to prevent terrorist from stepping on airport
Machine；Using recognition of face ancillary credit card network payment, to prevent the owner of non-credit card from using credit card etc..In addition,
After S240, satellite information corresponding with the target face identified can be obtained；Institute is synchronously shown with the target face
State satellite information.The satellite information may include Real Name corresponding with the target face identified, alias, the age, occupation,
At least one of representative work, hobby.In this way, can be provided a user about the mesh identified after identification face
The satellite information of face is marked, to promote user to be best understood from the relevant information of target face of interest, this leads to currently
It crosses especially advantageous in the activity of display viewing video frequency program.
According to the ... of the embodiment of the present invention above-mentioned in carrying out the method for face recognition, by being based on and target image phase
The image related information of pass establishes multiple facial characteristics templates and identifies the target based on the facial characteristics template established
Face can more accurately and efficiently carry out face recognition.
Fig. 3 is the flow chart for schematically illustrating another method for carrying out face recognition according to the ... of the embodiment of the present invention.
An exemplary application of the method for carrying out face recognition has been shown in particular in Fig. 3.S210, S220, S240 points of step in Fig. 3
Not with it is corresponding in Fig. 2, step S231 and S232 in Fig. 3 correspond to the step S230 in Fig. 2.In addition, compared with Fig. 2,
Step S250 and S260 are also added in Fig. 3.
As shown in figure 3, detecting the information of the face of the singer A from the video frequency program picture of the face including singer A
（S210）；Obtain the title with the relevant video frequency program of video frequency program picture（S220）；Title based on the video frequency program come
Obtain the name (S231) for the performer that performance is participated in the video frequency program；Packet is established based on the name of acquired performer
Include the database (S232) of the facial characteristics template of each performer；Information based on the target face detected and the database
To identify the target face (S240)；Obtain Real Name corresponding with the target face identified, age, representative work
(S250)；The Real Name, age, representative work are synchronously shown with the target face（S260）.
According to fig. 3 diagram as can be seen that using the embodiment of the present invention carries out face recognition method can accurately with
Face recognition is efficiently carried out, and can more effectively apply the result of the face recognition.
Fig. 4 is the block diagram for schematically illustrating the device 400 according to the ... of the embodiment of the present invention for carrying out face recognition.It is described into
The operation of the device 400 of row face recognition can be executed by the face recognition device 30 in Fig. 1.
As shown in figure 4, the device 400 for carrying out face recognition may include：Detection unit 410, for from including to know
The information of the target face is detected in the target image of other target face；First acquisition unit 420, for obtain with it is described
The relevant image related information of target image；Template establishes unit 430, for based on described image related information come establish including
The database of multiple facial characteristics templates；Recognition unit 440, for based on the target face detected information and the data
Library identifies the target face.
The target image to be detected of the detection unit 410, can corresponding to the image shown in the display equipment 10 of Fig. 1
Can also be the image presented on film poster to be the image shown on a liquid crystal display.In the target image
There are the face of personage of interest, that is, the target face to be identified.For example, the detection unit 410 can be from video frequency program
Picture in detect the information of the target face, detect the target in the picture that can also be shot from a social activities
The information of face.
The information for the target face that the detection unit 410 is detected be can characterize the feature of the target face with
The information that other faces distinguish.As an example, the information of the target face can be the organ for characterizing face（Example
Such as, eyes, nose, mouth, chin etc.）Shape and the distance between each organ characteristic, such as including face
Euclidean distance, curvature and angle between characteristic point etc..The detection unit 410 can also detect any other characteristic
As the information of the target face, as long as it can characterize the feature of face.
The detection unit 410 can determine position and the size of face from the target image, then to determining
Face position and size pre-processed, such as light compensation, greyscale transformation, histogram equalization are carried out to the image of face
Change, normalize, filter and sharpen etc.；And letter of the feature of face as the target face is extracted from treated information
Breath.In practice, for the information of different types of target face, the detection unit 410 can take different technologies
It is detected, the mode for specifically detecting the information of the target face does not constitute limitation to the embodiment of the present invention.
The first acquisition unit 420 obtains and the relevant image related information of the target image.Image association letter
Breath be other than the content in the target image with the relevant information of target image.The detection unit 410 from regarding
In the case of the information for detecting the target face in picture in frequency program, the first acquisition unit 420 can obtain institute
The name for stating video frequency program is referred to as described image related information；It is shot from a social activities in the detection unit 410
In the case of the information for detecting the target face in picture, the first acquisition unit 420 can obtain the social activities
Name be referred to as described image related information.
If the information of target face and the face module of all personnel that detection unit 410 is detected compared to carry out
Face recognition will consume a large amount of time and resource, and due to face structure since there are many face template for comparison
Similitude, it may not be possible to the accurately face in recognition target image.The first acquisition unit 420 obtains described image pass
Join information, then can determine the relevant information for obtaining the target image, in the people involved in described image related information
It is compared in member to identify target face, this greatly reduces the range of comparison, and improves the accuracy of identification.
In the video broadcast system of such as cable television, video frequency program is generally included in the video programming signal received
Electronic program guides, therefore can easily obtain described image related information using the electronic program guides.For example,
In the case of the video frequency program is the program obtained by video on-demand system, the first acquisition unit 420 can utilize
The electronic program guides of the video frequency program come obtain with the relevant image related information of the target image, i.e., automatically obtain
With the relevant image related information of the target image.
In the case of the target image is the picture shot in a social activities, the on-the-spot meeting of usual social activities
Put up the advertisement information about the social activities, the first acquisition unit 420 can from target image about the society
The title that the social activities is determined in movable advertisement information, as image related information.The first acquisition unit 420 can
To receive the title of the manually recorded social activities of user, Text region can also be carried out to word in the target image
To determine the title of the social activities.Therefore, the first acquisition unit 420 can be come with the concrete condition of combining target image
Suitable means are taken to obtain described image related information, specific acquisition modes do not constitute the limitation to the embodiment of the present invention.
The template establishes unit 430 based on described image related information to establish the number for including multiple facial characteristics templates
According to library.Each facial characteristics template of the database be used to be compared with the information of the target face, to carry out face
Identification.The template establishes unit 430 can obtain each facial characteristics template to be compared from other databases, also may be used
To establish the facial characteristics template based on other images of each personnel.
The case where name of the video frequency program is referred to as described image related information is obtained in the first acquisition unit 420
In, the template establishes unit 430 can be by operating the database that foundation includes multiple facial characteristics templates as follows：Based on described
The title of video frequency program obtains the name for the performer that performance is participated in the video frequency program；Surname based on acquired performer
Name establishes the database of the facial characteristics template including each performer.By taking the video frequency program is " I is singer " as an example, institute
Stating template and establishing unit 430 can establish in the facial characteristics template of all singers of participation " I is singer ", " I is singer "
All staff on board（Including host, musician etc.）Facial characteristics template etc., and form database.The database can wrap
All facial characteristics templates for being likely to occur personnel in the target image are included, the face to be compared greatly is reduced
The quantity of feature templates, and since the target face in target image has participated in " I is singer ", so the target face
Personnel should be in the database in, so as to accurately be identified.
Alternatively, the name for social activities being obtained in the first acquisition unit 420 is referred to as described image related information
In situation, the template establishes unit 430 can be by operating the database that foundation includes multiple facial characteristics templates as follows：It is based on
The title of the social activities come obtain participate in the social activities participant list；List based on acquired participant
To establish the database of the facial characteristics template including each participant.The social activities for example can be training, one
A meeting, product news conference etc..In general, the information that the organizer of social activities can have worker therein, invite personnel（Example
Such as, name, position etc.）, the list of the participant is can be obtained based on the information, and foundation includes the face of each participant
The database of feature templates.
The recognition unit 440 identifies the target face based on the information of the target face detected and the database
Portion.For example, the recognition unit 440 can be by the information of the target face detected in detection unit 410 and by template
The facial characteristics template established in the database of the foundation of unit 430 scans for matching, with the information of the determination target face
The target face detected is then identified as this when similarity is more than a threshold value with the similarity of each facial characteristics template
Personnel representated by facial characteristics template.Alternatively, the recognition unit 440 can be determined from each similarity has maximum phase
Like the facial characteristics template of degree, the target face detected is identified as having representated by the facial characteristics template of maximum similarity
Personnel.Therefore, the recognition unit 440 carries out the information of target face to be identified and obtained facial characteristics template
Compare, and the identity information of target face is judged according to similarity degree.
Based on the target face that the recognition unit 440 is identified, other applications can be further executed.For example,
The identity for the person of seeking entry into can be recognized by recognition of face by the area of safeguard protection；Monitoring system is installed to prevent on airport
Terrorist's boarding；Using recognition of face ancillary credit card network payment, to prevent the owner of non-credit card from using credit card
Deng.For example, optionally, can also include second acquisition unit 450 in the device 400 for carrying out face recognition, be used for
Obtain satellite information corresponding with the target face identified；Display unit 460, for synchronously being shown with the target face
The satellite information.The satellite information may include Real Name corresponding with the target face identified, alias, age, duty
At least one of industry, representative work, hobby.In this way, can be provided a user after identification face about being identified
The satellite information of target face, to promote user to be best understood from the relevant information of target face of interest, this is current
It is watched by display especially advantageous in the activity of video frequency program.The display unit 460 be not necessarily included in it is described for into
The device 400 of row face recognition, and it is attached to the device 400 for carrying out face recognition, such as Fig. 1 can be utilized
In display equipment 10 show.
According to the ... of the embodiment of the present invention above-mentioned in carrying out the device of face recognition, by being based on and target image phase
The image related information of pass establishes multiple facial characteristics templates and identifies the target based on the facial characteristics template established
Face can more accurately and efficiently carry out face recognition.
In the case of disclosing the device according to the ... of the embodiment of the present invention for carrying out face recognition above in conjunction with Fig. 4, own
Electronic equipment including the device for carrying out face recognition is also in the open scope of the embodiment of the present invention.
Those of ordinary skill in the art may realize that lists described in conjunction with the examples disclosed in the embodiments of the present disclosure
Member and algorithm steps can be realized with the combination of electronic hardware or computer software and electronic hardware.These functions are actually
It is implemented in hardware or software, depends on the specific application and design constraint of technical solution.Professional technician
Each specific application can be used different methods to achieve the described function, but this realization is it is not considered that exceed
The scope of the present invention.
It is apparent to those skilled in the art that for convenience and simplicity of description, the device of foregoing description
It with the specific work process of unit, can refer to corresponding processes in the foregoing method embodiment, details are not described herein.
In several embodiments provided herein, it should be understood that disclosed device and method can pass through it
Its mode is realized.For example, the apparatus embodiments described above are merely exemplary, for example, the division of the unit, only
Only a kind of division of logic function, formula that in actual implementation, there may be another division manner, such as multiple units or component can be tied
Another equipment is closed or is desirably integrated into, or some features can be ignored or not executed.
The unit illustrated as separating component may or may not be physically separated, aobvious as unit
The component shown may or may not be physical unit.It can select according to the actual needs therein some or all of
Unit achieves the purpose of the solution of this embodiment.
It, can be with if the function is realized in the form of SFU software functional unit and when sold or used as an independent product
It is stored in a computer read/write memory medium.Based on this understanding, technical scheme of the present invention is substantially in other words
The part of the part that contributes to existing technology or the technical solution can be expressed in the form of software products, the meter
Calculation machine software product is stored in a storage medium, including some instructions are used so that a computer equipment（Can be
People's computer, server or network equipment etc.）It performs all or part of the steps of the method described in the various embodiments of the present invention.
And storage medium above-mentioned includes：USB flash disk, mobile hard disk, read-only memory, random access memory, magnetic disc or CD etc. are each
Kind can store the medium of program code.
The above description is merely a specific embodiment, but scope of protection of the present invention is not limited thereto, any
Those familiar with the art in the technical scope disclosed by the present invention, can easily think of the change or the replacement, and should all contain
Lid is within protection scope of the present invention.Therefore, the protection scope of the present invention shall be subject to the protection scope of the claims.
Claims (17)
1. a kind of method for carrying out face recognition, including：
The information of the target face is detected from the target image including the target face to be identified；
Acquisition and the relevant image related information of the target image, the wherein image related information is in addition to the target image
In content except with the relevant information of target image；
The database for including multiple facial characteristics templates is established based on described image related information；
The target face is identified based on the information of the target face detected and the database.
2. the method according to claim 1, wherein the target image is the picture in video frequency program, described image association letter
Breath is the title of the video frequency program.
3. method according to claim 2, wherein described to be established including multiple facial characteristics based on described image related information
The database of template includes：
The name for the performer that performance is participated in the video frequency program is obtained based on the title of the video frequency program；
The database of the facial characteristics template including each performer is established based on the name of acquired performer.
4. method according to claim 2, wherein the video frequency program is the program obtained by video on-demand system, described
It obtains with the relevant image related information of the target image and includes：
It is obtained and the relevant image related information of the target image by the electronic program guides of the video frequency program.
5. the method according to claim 1 further includes：
Obtain satellite information corresponding with the target face identified；
The satellite information is synchronously shown with the target face.
6. method according to claim 5, wherein the satellite information includes true surname corresponding with the target face identified
At least one of name, alias, age, occupation, representative work, hobby.
7. the method according to claim 1, the target image is the picture shot in a social activities, described image association
Information is the title of the social activities.
8. according to the method described in claim 7, described established including multiple facial characteristics based on described image related information
Template database include：
The list for the participant for participating in the social activities is obtained based on the title of the social activities；
The database of the facial characteristics template including each participant is established based on the list of acquired participant.
9. a kind of device for carrying out face recognition, including：
Detection unit, the information for detecting the target face from the target image including the target face to be identified；
First acquisition unit, for obtaining and the relevant image related information of the target image, wherein the image related information
Be other than the content in the target image with the relevant information of target image；
Template establishes unit, for establishing the database for including multiple facial characteristics templates based on described image related information；
Recognition unit, for identifying the target face based on the information of the target face detected and the database.
10. device according to claim 9, wherein
The detection unit detects the information of the target face from the picture in video frequency program,
The name that the first acquisition unit obtains the video frequency program is referred to as described image related information.
11. device according to claim 10, wherein it includes multiple faces that the template, which establishes unit by operating foundation as follows,
The database of feature templates：
The name for the performer that performance is participated in the video frequency program is obtained based on the title of the video frequency program；
The database of the facial characteristics template including each performer is established based on the name of acquired performer.
12. device according to claim 10, wherein the video frequency program is the program obtained by video on-demand system, institute
First acquisition unit is stated to close with the relevant image of the target image to obtain using the electronic program guides of the video frequency program
Join information.
13. device according to claim 9, further includes：
Second acquisition unit, the corresponding satellite information of target face for obtaining with being identified；
Display unit, for synchronously showing the satellite information with the target face.
14. device according to claim 13, wherein the satellite information includes corresponding with the target face identified true
At least one of name, alias, age, occupation, representative work, hobby.
15. device according to claim 9, wherein
The information of the target face is detected in the picture that the detection unit is shot from a social activities,
The name that the first acquisition unit obtains the social activities is referred to as described image related information.
16. device according to claim 15, it includes multiple faces that the template, which establishes unit by operating foundation as follows,
The database of feature templates：
The list for the participant for participating in the social activities is obtained based on the title of the social activities；
The database of the facial characteristics template including each participant is established based on the list of acquired participant.
17. a kind of electronic equipment includes the device as described in any one of claim 9-16.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410105651.7A CN104933391B (en) | 2014-03-20 | 2014-03-20 | Method and apparatus for carrying out face recognition and electronic equipment |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410105651.7A CN104933391B (en) | 2014-03-20 | 2014-03-20 | Method and apparatus for carrying out face recognition and electronic equipment |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN104933391A CN104933391A (en) | 2015-09-23 |
| CN104933391B true CN104933391B (en) | 2018-08-10 |
Family
ID=54120552
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201410105651.7A Active CN104933391B (en) | 2014-03-20 | 2014-03-20 | Method and apparatus for carrying out face recognition and electronic equipment |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN104933391B (en) |
Families Citing this family (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107370981A (en) * | 2016-05-13 | 2017-11-21 | 中兴通讯股份有限公司 | The information cuing method and device of personnel participating in the meeting in a kind of video conference |
| CN106604051A (en) * | 2016-12-20 | 2017-04-26 | 广州华多网络科技有限公司 | Live channel recommending method and device |
| CN109033959B (en) * | 2018-06-20 | 2022-03-29 | 北京微播视界科技有限公司 | Method and device for adding special effect to face of object |
| CN109033958A (en) * | 2018-06-20 | 2018-12-18 | 北京微播视界科技有限公司 | Increase the method and device of special efficacy for the face of special object |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2004127285A (en) * | 2002-09-13 | 2004-04-22 | Sony Corp | Image recognition apparatus, image recognition processing method and image recognition program |
| CN101990667A (en) * | 2008-04-02 | 2011-03-23 | 谷歌公司 | Method and apparatus to incorporate automatic face recognition in digital image collections |
| CN102542249A (en) * | 2010-11-01 | 2012-07-04 | 微软公司 | Face recognition in video content |
| CN103150551A (en) * | 2013-02-06 | 2013-06-12 | 上海云联广告有限公司 | Method for automatically carrying out facial recognition |
| CN103258190A (en) * | 2013-05-13 | 2013-08-21 | 苏州福丰科技有限公司 | Face recognition method used for mobile terminal |
-
2014
- 2014-03-20 CN CN201410105651.7A patent/CN104933391B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2004127285A (en) * | 2002-09-13 | 2004-04-22 | Sony Corp | Image recognition apparatus, image recognition processing method and image recognition program |
| CN101990667A (en) * | 2008-04-02 | 2011-03-23 | 谷歌公司 | Method and apparatus to incorporate automatic face recognition in digital image collections |
| CN102542249A (en) * | 2010-11-01 | 2012-07-04 | 微软公司 | Face recognition in video content |
| CN103150551A (en) * | 2013-02-06 | 2013-06-12 | 上海云联广告有限公司 | Method for automatically carrying out facial recognition |
| CN103258190A (en) * | 2013-05-13 | 2013-08-21 | 苏州福丰科技有限公司 | Face recognition method used for mobile terminal |
Also Published As
| Publication number | Publication date |
|---|---|
| CN104933391A (en) | 2015-09-23 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Andrews et al. | Telling faces together: Learning new faces through exposure to multiple instances | |
| Chingovska et al. | Face recognition systems under spoofing attacks | |
| Mohammadi et al. | Deeply vulnerable: a study of the robustness of face recognition to presentation attacks | |
| Wang et al. | Facial recognition system using LBPH face recognizer for anti-theft and surveillance application based on drone technology | |
| Lovato et al. | Faved! biometrics: Tell me which image you like and I'll tell you who you are | |
| US8805123B2 (en) | System and method for video recognition based on visual image matching | |
| US20150033362A1 (en) | Notification and Privacy Management of Online Photos and Videos | |
| CN107644213A (en) | Video person extraction method and device | |
| CN104933391B (en) | Method and apparatus for carrying out face recognition and electronic equipment | |
| JP6969663B2 (en) | Devices and methods for identifying the user's imaging device | |
| CN102567502A (en) | Supplementing biometric identification with device identification | |
| US10423817B2 (en) | Latent fingerprint ridge flow map improvement | |
| CN108959884B (en) | Human authentication verification device and method | |
| CN108805005A (en) | Auth method and device, electronic equipment, computer program and storage medium | |
| US20190147226A1 (en) | Method, system and apparatus for matching a person in a first image to a person in a second image | |
| CN105893813A (en) | Biometric Information Registration Apparatus And Biometric Information Registration Method | |
| Mou et al. | Alone versus in-a-group: A multi-modal framework for automatic affect recognition | |
| CN110245573A (en) | A kind of register method, apparatus and terminal device based on recognition of face | |
| Fox et al. | Videoforensicshq: Detecting high-quality manipulated face videos | |
| CN108734003A (en) | Auth method, device, equipment, storage medium and program | |
| CN111241873A (en) | Image reproduction detection method, training method of model thereof, payment method and payment device | |
| Wu et al. | Hierarchical dynamic depth projected difference images–based action recognition in videos with convolutional neural networks | |
| CN103984931A (en) | Information processing method and first electronic equipment | |
| López-Gil et al. | Do deepfakes adequately display emotions? a study on deepfake facial emotion expression | |
| CN208722199U (en) | A kind of intelligence delivery system |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |