Specific embodiment
The disclosure is described in further detail with reference to the accompanying drawings and examples.It is understood that this place is retouched
The specific embodiment stated is used only for explaining that correlation is open, rather than the restriction to the disclosure.It also should be noted that in order to
Convenient for description, is illustrated only in attached drawing and disclose relevant part to related.
It should be noted that in the absence of conflict, the feature in embodiment and embodiment in the disclosure can phase
Mutually combination.The disclosure is described in detail below with reference to the accompanying drawings and in conjunction with the embodiments.
Fig. 1 is shown can be using the method or the implementation of the device at age for identification at the age for identification of the disclosure
The exemplary system architecture 100 of example.
As shown in Figure 1, system architecture 100 may include server 101,103 and network 102.Network 102 is to service
The medium of communication link is provided between device 101 and server 103.Network 102 may include various connection types, such as wired,
Wireless communication link or fiber optic cables etc..
Server 101 can be to provide the server of various services, such as the information of the video for storing user's publication
Server is used in storage.Server 103 can be to provide the server of various services, such as personage's view for being issued based on user
Frequency carries out the server of age identification, which for example can be related with personage's video of user's publication in response to receiving
Age identification request obtains personage's video from server 101 based on the request, and carries out age identification based on personage's video
Operation.
Wherein, server can be hardware, be also possible to software.When server is hardware, multiple clothes may be implemented into
The distributed server cluster of business device composition, also may be implemented into individual server.When server is software, may be implemented into
Multiple softwares or software module (such as providing Distributed Services), also may be implemented into single software or software module.?
This is not specifically limited.
It should be noted that the method at the age for identification that some embodiments of the present disclosure provide is generally by server
103 execute, and correspondingly, the device at age is generally positioned in server 103 for identification.
It should be pointed out that when personage's video acquired in the server 103 is stored in advance in server 103, or
When including in the received age identification request of institute, system architecture 100 can not include server 101.
It should be understood that the number of network and server in Fig. 1 is only schematical.According to needs are realized, can have
There are any number of network and server.
With continued reference to Fig. 2, it illustrates the processes according to one embodiment of the method at age for identification of the disclosure
200.The process 200 of the method at age for identification, comprising the following steps:
Step 201, face image set is extracted from personage's video to be processed.
In the present embodiment, the executing subject (such as server 103 shown in FIG. 1) of the method at age can be with for identification
Face image set is extracted from personage's video to be processed.
In practice, above-mentioned executing subject can for example be asked in response to receiving age identification related with above-mentioned personage's video
It asks, and extracts face image set from above-mentioned personage's video.Wherein, age identification request may include following at least one
: the video identifier of above-mentioned personage's video, above-mentioned personage's video.When the age, identification request did not include above-mentioned personage's video,
Above-mentioned executing subject can be based on the video identifier from server (such as server 101 shown in FIG. 1) that is local or being connected
Above-mentioned personage's video is obtained, then extracts face image set from above-mentioned personage's video.
It should be noted that the user that above-mentioned personage's video is belonged to can be the user of specified social application.Specifically
Ground, the user that above-mentioned personage's video is belonged to, which for example can be, to be not provided with real age or is not provided in the social application
The user at age.Wherein, display someone that the user that above-mentioned personage's video can be that it is belonged to is issued by the social application
The video of object.In addition, above-mentioned personage's video can be the video that the user that it is belonged to uses the social application to record.
In the present embodiment, if above-mentioned personage's video has corresponded to Face datection in advance as a result, above-mentioned executing subject can be with base
In the face testing result, face image set is extracted from above-mentioned personage's video.Wherein, include in above-mentioned personage's video
Face datection result corresponding to the image of human face region may include the location information of the human face region.Specifically, above-mentioned to hold
Row main body first can carry out pumping frame to above-mentioned personage's video, such as extract all images, or according to interval frame number (such as 1
Or 2 etc.) interval pumping frame is carried out to above-mentioned personage's video.It should be understood that interval frame number can be set according to actual needs,
This is not specifically limited.Then, the image extracted can be formed image collection by above-mentioned executing subject.Then, above-mentioned execution
Main body can extract face image set based on the location information in Face datection result from image collection.
It should be pointed out that the facial image in face image set can be dedicated tunnel (such as 3 channels), size
For the image for specifying size (such as 224*224).
Step 202, based on the matching relationship between facial image, face image set is divided at least one face figure
As group.
In the present embodiment, above-mentioned executing subject can be based on the matching relationship between facial image, by face image set
Conjunction is divided at least one facial image group.Wherein, different facial image groups corresponds to different personages.For example, working as face figure
When image set conjunction is divided into multiple facial image groups, it is meant that above-mentioned personage's video shows multiple personages, multiple face figure
Each of picture group face image group corresponds to a personage in multiple personage, and multiple facial image group is corresponding
Personage is different.
It should be noted that above-mentioned executing subject can for example use the face alignment algorithm based on deep learning, to people
Facial image in face image set is compared two-by-two, to judge whether any two facial images belong to same people.Then,
Above-mentioned executing subject can be based on comparison result, and the facial image for belonging to same people is divided into same person's face image group.
Optionally, above-mentioned personage's video in advance in corresponding Face datection result, it is corresponding include human face region image
Face datection result can also include human face region face characteristic information.Wherein, face characteristic information can be used for characterizing
Human face region corresponding to it.Above-mentioned executing subject can determine that each facial image in face image set is corresponding
Face image set is divided at least by the similarity of face characteristic information between any two then based on identified similarity
One facial image group.Here, above-mentioned executing subject can be using various similarity calculating methods (such as Euclidean distance, cosine
Similarity etc.) calculate face characteristic information between similarity.
As an example it is supposed that face image set includes facial image A, B, C, D, E.Wherein, in facial image A, B, C
Similarity between the corresponding face characteristic information of any two facial images is greater than similarity threshold, facial image D, E
Similarity between corresponding face characteristic information is greater than similarity threshold.Above-mentioned executing subject can by facial image A,
B, C is included into same person's face image group, and facial image D, E are included into same person's face image group.
Step 203, for each of at least one facial image group face image group, which is inputted pre-
First trained Three dimensional convolution neural network, obtains age recognition result corresponding with personage corresponding to the facial image group.
In the present embodiment, for each of at least one above-mentioned facial image group face image group, above-mentioned execution master
The facial image group can be inputted Three dimensional convolution neural network trained in advance by body, obtain Three dimensional convolution neural network output
Age recognition result, and the age recognition result is determined as the age corresponding with personage corresponding to the facial image group
Recognition result.Wherein, the Three dimensional convolution neural network is for carrying out age identification.In addition, the Three dimensional convolution neural network can be with
It runs in above-mentioned executing subject.
It should be noted that above-mentioned Three dimensional convolution neural network can be model training end (such as above-mentioned executing subject or
The server that above-mentioned executing subject is connected) it is obtained by following steps training:
Firstly, obtaining training sample set.Wherein, training sample may include sample facial image group and with sample face
The corresponding age value of image group.Each sample facial image in sample facial image group is the face figure of same sample of users
Picture, age value corresponding to the sample facial image group are the real age of the sample of users.For in training sample set
Each training sample, the sample facial image group in the training sample can be through the target person to corresponding sample of users
What video was pre-processed.The target person video for example can be sample of users by carrying out video record to he or she
It obtains.In addition, sample facial image can be dedicated tunnel (such as 3 channels), size be specified size (such as 224*
224) image.
Then, it using sample facial image group included by the training sample in training sample set as input, will correspond to
The age value of the sample facial image group of input obtains above-mentioned Three dimensional convolution neural network as output, training.Specifically, above-mentioned
Model training end can will be corresponded to using sample facial image group included by the training sample in training sample set as input
The age value of the sample facial image group of input is trained initial Three dimensional convolution neural network as output.When model is instructed
After white silk, can will be trained after initial Three dimensional convolution neural network be determined as the Three dimensional convolution for being used to carry out age identification
Neural network.It should be pointed out that the three-dimensional volume that initial Three dimensional convolution neural network can be indiscipline or training is not completed
Product neural network.
With continued reference to the signal that Fig. 3, Fig. 3 are according to the application scenarios of the method at the age for identification of the present embodiment
Figure.In the application scenarios of Fig. 3, personage's video C that user A is issued by social application B can store on server 301.
Wherein, personage's video C can be user A and carry out the resulting video of video record to its people using social application B.In addition, with
Family A not set age in social application B.When server 302 receives the age identification of the video identifier including personage's video C
When request, server 302 can obtain personage's video C from server 301 according to the video identifier.Then, server 302 can
To extract face image set from personage's video C.Later, server 302 can be closed based on the matching between facial image
System, is divided into a facial image group for face image set.Wherein, facial image group corresponds to user A.Then, server 302
The facial image group can be inputted to Three dimensional convolution neural network 303 trained in advance, obtain age identification corresponding with user A
As a result.Wherein, Three dimensional convolution neural network 303 is for carrying out age identification.
The method provided by the above embodiment of the disclosure, by extracting face image set from personage's video to be processed
It closes, then based on the matching relationship between facial image, face image set is divided at least one facial image group,
In, different facial image groups corresponds to different personages, then for each of at least one facial image group face figure
It is right with the facial image group institute to obtain by facial image group input Three dimensional convolution neural network trained in advance as group
The corresponding age recognition result of the personage answered, wherein Three dimensional convolution neural network is for carrying out age identification.The disclosure it is upper
The scheme for stating embodiment offer may be implemented to carry out age identification to one or more personages shown by personage's video.In addition,
By the utilization to Three dimensional convolution neural network, age recognition efficiency can be improved, and obtain the year with high accuracy
Age recognition result.
With further reference to Fig. 4, it illustrates the processes 400 of another embodiment of the method at age for identification.The use
In the process 400 of the method at identification age, comprising the following steps:
Step 401, pumping frame is carried out to personage's video to be processed, and the image extracted is formed into image collection.
In the present embodiment, the executing subject (such as server 103 shown in FIG. 1) of the method at age can be with for identification
Pumping frame is carried out to personage's video to be processed, and the image extracted is formed into image collection.Here, above-mentioned executing subject is for example
All images can be extracted from personage's video, or personage's video is carried out according to interval frame number (such as 1 or 2 etc.)
Take out frame in interval.It should be understood that interval frame number can be set according to actual needs, it is not specifically limited herein.
In practice, above-mentioned executing subject can for example be asked in response to receiving age identification related with above-mentioned personage's video
It asks, and pumping frame is carried out to above-mentioned personage's video, and the image extracted is formed into image collection.Wherein, age identification request
It may include at least one of following: the video identifier of above-mentioned personage's video, above-mentioned personage's video.When the age, identification request is not wrapped
When including above-mentioned personage's video, above-mentioned executing subject can be based on the video identifier from server that is local or being connected (such as Fig. 1
Shown in server 101) obtain above-mentioned personage's video, pumping frame, and the image that will be extracted then are carried out to above-mentioned personage's video
Form image collection.
It should be noted that the user that above-mentioned personage's video is belonged to can be the user of specified social application.Specifically
Ground, the user that above-mentioned personage's video is belonged to, which for example can be, to be not provided with real age or is not provided in the social application
The user at age.Wherein, display someone that the user that above-mentioned personage's video can be that it is belonged to is issued by the social application
The video of object.In addition, above-mentioned personage's video can be the video that the user that it is belonged to uses the social application to record.
Step 402, Face datection is carried out to each image in image collection respectively, obtains Face datection result, wherein
Face datection result corresponding to image including human face region includes the location information and face characteristic information of human face region.
In the present embodiment, above-mentioned executing subject can carry out Face datection to each image in image collection respectively,
Obtain Face datection result, wherein Face datection result corresponding to the image including human face region includes the position of human face region
Confidence breath and face characteristic information.Face characteristic information can be used for characterizing the human face region corresponding to it.
Here, face detection model can have been run in above-mentioned executing subject, above-mentioned executing subject can be by image collection
In every image input the face detection model, obtain corresponding Face datection result.Wherein, Face datection model for example may be used
To be using model-naive Bayesian (Naive Bayesian Model, NBM), support vector machines (Support Vector
Machine, SVM), XGBoost (eXtreme Gradient Boosting) or convolutional neural networks (Convolutional
Neural Networks, CNN) etc. models be trained.The image that Face datection model can be used for detecting input is
No includes that human face region is also used to further position the human face region when detecting that the image includes human face region
And face characteristic extracts.
Step 403, based on the location information in Face datection result, face image set is extracted from image collection.
In the present embodiment, above-mentioned executing subject can be based on the location information in Face datection result, from image collection
In extract face image set.Wherein, the facial image in face image set can be dedicated tunnel (such as 3 channels)
, the image that size is specified size (such as 224*224).
Step 404, the corresponding face characteristic information of each facial image in face image set is clustered,
And it is based on cluster result, face image set is divided at least one facial image group.
In the present embodiment, above-mentioned executing subject can for example use preset clustering algorithm, such as K mean cluster algorithm
(k-means clustering algorithm) etc., to the corresponding face of each facial image in face image set
Characteristic information is clustered, and is based on cluster result, and face image set is divided at least one facial image group.Here,
Above-mentioned executing subject can will include facial image corresponding to face characteristic information in same class cluster be divided into it is same
In facial image group.
Step 405, for each of at least one facial image group face image group, which is inputted pre-
First trained Three dimensional convolution neural network, obtains age recognition result corresponding with personage corresponding to the facial image group,
Wherein, Three dimensional convolution neural network is used to carry out every facial image in the facial image group of input in feature extraction, and base
Age identification is carried out in the characteristic information extracted, characteristic information includes the corresponding age information of facial image and following at least one
: 3 d pose information, quality information.
In the present embodiment, for each of at least one above-mentioned facial image group face image group, above-mentioned execution master
The facial image group can be inputted Three dimensional convolution neural network trained in advance by body, obtain Three dimensional convolution neural network output
Age recognition result, and the age recognition result is determined as the age corresponding with personage corresponding to the facial image group
Recognition result.Wherein, which is used to carry out every facial image in the facial image group of input special
Sign is extracted, and carries out age identification based on the characteristic information extracted.This feature information includes the corresponding age letter of facial image
Breath and at least one of following: 3 d pose information, quality information.
3 d pose information may include pitch angle angle value, yaw angle angle value and rolling angle value.Pitch angle angle value can be with
It is the angle value of pitch angle (pitch).Yaw angle angle value can be the angle value of yaw angle (yaw).Rolling angle value can be
The angle value of roll angle (roll).Quality information may include fog-level value.Fog-level value for example can be in [0,
100] numerical value in.Fog-level value is bigger, and the facial image that can be characterized corresponding to it is fuzzyyer.Fog-level value is lower,
The facial image that can be characterized corresponding to it is more clear.
In practice, every facial image in facial image group for inputting above-mentioned Three dimensional convolution neural network is above-mentioned
Three dimensional convolution neural network can be corresponding at least one of following based on the facial image: 3 d pose information, quality information, really
Fixed weighted value corresponding with the facial image.Then, above-mentioned Three dimensional convolution neural network can be based in the facial image group
The corresponding weighted value of multiple facial images and age information carry out age identification.
Every facial image in facial image group for inputting above-mentioned Three dimensional convolution neural network, if above-mentioned three-dimensional volume
The characteristic information that product neural network is extracted for the facial image includes the corresponding 3 d pose information of the facial image, three-dimensional
Posture information includes that pitch angle angle value, yaw angle angle value and rolling angle value, above-mentioned Three dimensional convolution neural network can calculate
Summation between pitch angle angle value, yaw angle angle value corresponding to the facial image and the absolute value for the angle value that rolls, and by the
Ratio between one preset value (such as 1) and the summation is determined as weighted value corresponding to the facial image.If this feature information
Including the corresponding quality information of the facial image, quality information includes fog-level value, and above-mentioned Three dimensional convolution neural network can be with
Ratio between first preset value and the fog-level value is determined as weighted value corresponding to the facial image.If this feature is believed
Breath includes 3 d pose information and quality information, and 3 d pose information includes pitch angle value, yaw angle angle value and rolling angle
Value, quality information includes fog-level value, and above-mentioned Three dimensional convolution neural network can calculate bows corresponding to the facial image
Summation between elevation angle angle value, yaw angle angle value and the absolute value and corresponding fog-level value of the angle value that rolls, and will
Ratio between first preset value and identified summation is determined as weighted value corresponding to the facial image.
In addition, above-mentioned Three dimensional convolution neural network can use following formula, based on multiple people in the facial image group
The corresponding weighted value of face image (such as all or part of facial image) and age information, carry out age identification:
Wherein, C represents age recognition result;N is the quantity of multiple facial images;I is the nature in [1, n]
Number;W represents weighted value, WiRepresent the corresponding weighted value of i-th facial image in multiple facial images, WnRepresent this multiple
The corresponding weighted value of n-th facial image in facial image;V represents the age value in age information, ViRepresent multiple people
The age value in the corresponding age information of i-th facial image in face image.
It should be noted that above-mentioned Three dimensional convolution neural network can be obtained by following steps training: obtaining training sample
This set, wherein training sample includes sample facial image group and age value corresponding with sample facial image group, sample face
Each sample facial image in image group is the facial image of same sample of users;By the training sample in training sample set
Included sample facial image group is as input, using the age value of the sample facial image group of corresponding input as output, instruction
Get above-mentioned Three dimensional convolution neural network.
Figure 4, it is seen that compared with the corresponding embodiment of Fig. 2, the method at the age for identification in the present embodiment
Process 400 highlight the step of being extended to face image set extracting method;To the division methods of facial image group into
The step of row extension;And the step of function of Three dimensional convolution neural network is defined.The side of the present embodiment description as a result,
The diversity of information processing may be implemented in case, and can be further improved the accuracy of age recognition result.
With further reference to Fig. 5, as the realization to method shown in above-mentioned each figure, present disclose provides a kind of years for identification
One embodiment of the device in age, the Installation practice is corresponding with embodiment of the method shown in Fig. 2, which can specifically answer
For in various electronic equipments.
As shown in figure 5, the device 500 at the age for identification of the present embodiment may include: that extraction unit 501 is configured to
Face image set is extracted from personage's video to be processed;Division unit 502 is configured to based between facial image
Face image set is divided at least one facial image group by matching relationship, wherein different facial image groups is corresponding different
Personage;Recognition unit 503 is configured to for each of at least one above-mentioned facial image group face image group, by the people
Face image group input Three dimensional convolution neural network trained in advance, obtains corresponding with personage corresponding to the facial image group
Age recognition result, wherein Three dimensional convolution neural network is for carrying out age identification.
In the present embodiment, for identification in the device 500 at age: extraction unit 501, division unit 502 and identification are single
The specific processing of member 503 and its brought technical effect can be respectively with reference to step 201, the steps in embodiment shown in Fig. 2
202 and step 203 related description, details are not described herein.
In some optional implementations of the present embodiment, extraction unit 501 can be further configured to: to personage
Video carries out pumping frame, and the image extracted is formed image collection;Face is carried out respectively to each image in image collection
Detection, obtains Face datection result, wherein Face datection result corresponding to the image including human face region includes human face region
Location information;Based on the location information in Face datection result, face image set is extracted from image collection.
In some optional implementations of the present embodiment, above-mentioned Three dimensional convolution neural network is used for the face to input
Every facial image in image group carries out feature extraction, and carries out age identification based on the characteristic information extracted, wherein special
Reference breath includes the corresponding age information of facial image and at least one of following: 3 d pose information, quality information.
In some optional implementations of the present embodiment, Face datection knot corresponding to the image including human face region
Fruit further includes face characteristic information;And division unit 502 can be further configured to: to each in face image set
The corresponding face characteristic information of facial image is clustered, and be based on cluster result, by face image set be divided into
A few facial image group.
In some optional implementations of the present embodiment, above-mentioned Three dimensional convolution neural network can pass through following steps
Training obtain: obtain training sample set, wherein training sample include sample facial image group and with sample facial image group pair
The age value answered, each sample facial image in sample facial image group are the facial image of same sample of users;It will train
Sample facial image group included by training sample in sample set is as input, by the sample facial image group of corresponding input
Age value as output, training obtain above-mentioned Three dimensional convolution neural network.
The device provided by the above embodiment of the disclosure, by extracting face image set from personage's video to be processed
It closes, then based on the matching relationship between facial image, face image set is divided at least one facial image group,
In, different facial image groups corresponds to different personages, then for each of at least one facial image group face figure
It is right with the facial image group institute to obtain by facial image group input Three dimensional convolution neural network trained in advance as group
The corresponding age recognition result of the personage answered, wherein Three dimensional convolution neural network is for carrying out age identification.The disclosure it is upper
The scheme for stating embodiment offer may be implemented to carry out age identification to one or more personages shown by personage's video.In addition,
By the utilization to Three dimensional convolution neural network, age recognition efficiency can be improved, and obtain the year with high accuracy
Age recognition result.
Below with reference to Fig. 6, it illustrates the electronic equipment that is suitable for being used to realize embodiment of the disclosure, (example is as shown in figure 1
Server 103) 600 structural schematic diagram.Terminal device in embodiment of the disclosure can include but is not limited to such as move
Phone, laptop, digit broadcasting receiver, PDA (personal digital assistant), PAD (tablet computer), PMP (portable more matchmakers
Body player), the mobile terminal of car-mounted terminal (such as vehicle mounted guidance terminal) etc. and number TV, desktop computer etc.
Deng fixed terminal.Electronic equipment shown in Fig. 6 is only an example, should not function and use to embodiment of the disclosure
Range band carrys out any restrictions.
As shown in fig. 6, electronic equipment 600 may include processing unit (such as central processing unit, graphics processor etc.)
601, random access can be loaded into according to the program being stored in read-only memory (ROM) 602 or from storage device 608
Program in memory (RAM) 603 and execute various movements appropriate and processing.In RAM 603, it is also stored with electronic equipment
Various programs and data needed for 600 operations.Processing unit 601, ROM 602 and RAM603 are connected with each other by bus 604.
Input/output (I/O) interface 605 is also connected to bus 604.
In general, following device can connect to I/O interface 605: including such as touch screen, touch tablet, keyboard, mouse, taking the photograph
As the input unit 606 of head, microphone, accelerometer, gyroscope etc.;Including such as liquid crystal display (LCD), loudspeaker, vibration
The output device 607 of dynamic device etc.;Storage device 608 including such as hard disk etc.;And communication device 609.Communication device 609 can
To allow electronic equipment 600 wirelessly or non-wirelessly to be communicated with other equipment to exchange data.Although Fig. 6 is shown with various
The electronic equipment 600 of device, it should be understood that being not required for implementing or having all devices shown.It can be alternatively
Implement or have more or fewer devices.Each box shown in Fig. 6 can represent a device, also can according to need
Represent multiple devices.
Particularly, in accordance with an embodiment of the present disclosure, it may be implemented as computer above with reference to the process of flow chart description
Software program.For example, embodiment of the disclosure includes a kind of computer program product comprising be carried on computer-readable medium
On computer program, which includes the program code for method shown in execution flow chart.In such reality
It applies in example, which can be downloaded and installed from network by communication device 609, or from storage device 608
It is mounted, or is mounted from ROM 602.When the computer program is executed by processing unit 601, the implementation of the disclosure is executed
The above-mentioned function of being limited in the method for example.
It is situated between it should be noted that computer-readable medium described in embodiment of the disclosure can be computer-readable signal
Matter or computer readable storage medium either the two any combination.Computer readable storage medium for example can be with
System, device or the device of --- but being not limited to --- electricity, magnetic, optical, electromagnetic, infrared ray or semiconductor, or it is any more than
Combination.The more specific example of computer readable storage medium can include but is not limited to: have one or more conducting wires
Electrical connection, portable computer diskette, hard disk, random access storage device (RAM), read-only memory (ROM), erasable type are programmable
Read-only memory (EPROM or flash memory), optical fiber, portable compact disc read-only memory (CD-ROM), light storage device, magnetic are deposited
Memory device or above-mentioned any appropriate combination.In embodiment of the disclosure, computer readable storage medium, which can be, appoints
What include or the tangible medium of storage program that the program can be commanded execution system, device or device use or and its
It is used in combination.And in embodiment of the disclosure, computer-readable signal media may include in a base band or as carrier wave
The data-signal that a part is propagated, wherein carrying computer-readable program code.The data-signal of this propagation can be adopted
With diversified forms, including but not limited to electromagnetic signal, optical signal or above-mentioned any appropriate combination.Computer-readable signal is situated between
Matter can also be any computer-readable medium other than computer readable storage medium, which can be with
It sends, propagate or transmits for by the use of instruction execution system, device or device or program in connection.Meter
The program code for including on calculation machine readable medium can transmit with any suitable medium, including but not limited to: electric wire, optical cable,
RF (radio frequency) etc. or above-mentioned any appropriate combination.
Above-mentioned computer-readable medium can be included in above-mentioned electronic equipment;It is also possible to individualism, and not
It is fitted into the electronic equipment.Above-mentioned computer-readable medium carries one or more program, when said one or more
When a program is executed by the electronic equipment, so that the electronic equipment: extracting face image set from personage's video to be processed
It closes;Based on the matching relationship between facial image, face image set is divided at least one facial image group, wherein no
Same facial image group corresponds to different personages;For each of at least one facial image group face image group, by this
Facial image group input Three dimensional convolution neural network trained in advance, obtains corresponding with personage corresponding to the facial image group
Age recognition result, wherein Three dimensional convolution neural network is for carrying out age identification.
The behaviour for executing embodiment of the disclosure can be write with one or more programming languages or combinations thereof
The computer program code of work, described program design language include object oriented program language-such as Java,
Smalltalk, C++ further include conventional procedural programming language-such as " C " language or similar program design language
Speech.Program code can be executed fully on the user computer, partly be executed on the user computer, as an independence
Software package execute, part on the user computer part execute on the remote computer or completely in remote computer or
It is executed on server.In situations involving remote computers, remote computer can pass through the network of any kind --- packet
It includes local area network (LAN) or wide area network (WAN)-is connected to subscriber computer, or, it may be connected to outer computer (such as benefit
It is connected with ISP by internet).
Flow chart and block diagram in attached drawing are illustrated according to the system of the various embodiments of the disclosure, method and computer journey
The architecture, function and operation in the cards of sequence product.In this regard, each box in flowchart or block diagram can generation
A part of one module, program segment or code of table, a part of the module, program segment or code include one or more use
The executable instruction of the logic function as defined in realizing.It should also be noted that in some implementations as replacements, being marked in box
The function of note can also occur in a different order than that indicated in the drawings.For example, two boxes succeedingly indicated are actually
It can be basically executed in parallel, they can also be executed in the opposite order sometimes, and this depends on the function involved.Also it to infuse
Meaning, the combination of each box in block diagram and or flow chart and the box in block diagram and or flow chart can be with holding
The dedicated hardware based system of functions or operations as defined in row is realized, or can use specialized hardware and computer instruction
Combination realize.
Being described in unit involved in the embodiment of the present disclosure can be realized by way of software, can also be by hard
The mode of part is realized.Wherein, the title of unit does not constitute the restriction to the unit itself under certain conditions, for example, mentioning
Unit is taken to be also described as " extracting the unit of face image set from personage's video to be processed ".
Above description is only the preferred embodiment of the disclosure and the explanation to institute's application technology principle.Those skilled in the art
Member is it should be appreciated that the open scope involved in the disclosure, however it is not limited to technology made of the specific combination of above-mentioned technical characteristic
Scheme, while should also cover in the case where not departing from design disclosed above, it is carried out by above-mentioned technical characteristic or its equivalent feature
Any combination and the other technical solutions formed.Such as features described above has similar function with (but being not limited to) disclosed in the disclosure
Can technical characteristic replaced mutually and the technical solution that is formed.