CN106203306A - The Forecasting Methodology at age, device and terminal - Google Patents
The Forecasting Methodology at age, device and terminal Download PDFInfo
- Publication number
- CN106203306A CN106203306A CN201610512766.7A CN201610512766A CN106203306A CN 106203306 A CN106203306 A CN 106203306A CN 201610512766 A CN201610512766 A CN 201610512766A CN 106203306 A CN106203306 A CN 106203306A
- Authority
- CN
- China
- Prior art keywords
- age
- prediction
- image
- age value
- value
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 63
- 238000012549 training Methods 0.000 claims abstract description 27
- 238000013527 convolutional neural network Methods 0.000 claims description 8
- 241000208340 Araliaceae Species 0.000 claims description 7
- 235000005035 Panax pseudoginseng ssp. pseudoginseng Nutrition 0.000 claims description 7
- 235000003140 Panax quinquefolius Nutrition 0.000 claims description 7
- 235000008434 ginseng Nutrition 0.000 claims description 7
- 238000012706 support-vector machine Methods 0.000 claims description 7
- 230000006870 function Effects 0.000 description 10
- 230000008569 process Effects 0.000 description 10
- 238000004891 communication Methods 0.000 description 9
- 238000004422 calculation algorithm Methods 0.000 description 7
- 238000010586 diagram Methods 0.000 description 7
- 238000005516 engineering process Methods 0.000 description 6
- 230000008859 change Effects 0.000 description 5
- 230000000694 effects Effects 0.000 description 4
- 230000003287 optical effect Effects 0.000 description 4
- 230000005236 sound signal Effects 0.000 description 4
- 230000000712 assembly Effects 0.000 description 3
- 238000000429 assembly Methods 0.000 description 3
- 230000004048 modification Effects 0.000 description 3
- 238000012986 modification Methods 0.000 description 3
- 230000001133 acceleration Effects 0.000 description 2
- 230000006978 adaptation Effects 0.000 description 2
- 238000007726 management method Methods 0.000 description 2
- 238000003909 pattern recognition Methods 0.000 description 2
- 230000002093 peripheral effect Effects 0.000 description 2
- 238000012545 processing Methods 0.000 description 2
- KLDZYURQCUYZBL-UHFFFAOYSA-N 2-[3-[(2-hydroxyphenyl)methylideneamino]propyliminomethyl]phenol Chemical compound OC1=CC=CC=C1C=NCCCN=CC1=CC=CC=C1O KLDZYURQCUYZBL-UHFFFAOYSA-N 0.000 description 1
- 230000009471 action Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 238000007796 conventional method Methods 0.000 description 1
- 230000001186 cumulative effect Effects 0.000 description 1
- 238000013500 data storage Methods 0.000 description 1
- 201000001098 delayed sleep phase syndrome Diseases 0.000 description 1
- 208000033921 delayed sleep phase type circadian rhythm sleep disease Diseases 0.000 description 1
- 239000000686 essence Substances 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 239000004973 liquid crystal related substance Substances 0.000 description 1
- 230000004807 localization Effects 0.000 description 1
- 239000012092 media component Substances 0.000 description 1
- 230000001537 neural effect Effects 0.000 description 1
- 230000002085 persistent effect Effects 0.000 description 1
- 230000004044 response Effects 0.000 description 1
- 230000003068 static effect Effects 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V40/00—Recognition of biometric, human-related or animal-related patterns in image or video data
- G06V40/10—Human or animal bodies, e.g. vehicle occupants or pedestrians; Body parts, e.g. hands
- G06V40/16—Human faces, e.g. facial parts, sketches or expressions
- G06V40/172—Classification, e.g. identification
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/214—Generating training patterns; Bootstrap methods, e.g. bagging or boosting
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- General Physics & Mathematics (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Evolutionary Biology (AREA)
- Evolutionary Computation (AREA)
- Bioinformatics & Computational Biology (AREA)
- General Engineering & Computer Science (AREA)
- Artificial Intelligence (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Health & Medical Sciences (AREA)
- General Health & Medical Sciences (AREA)
- Oral & Maxillofacial Surgery (AREA)
- Human Computer Interaction (AREA)
- Multimedia (AREA)
- Image Analysis (AREA)
Abstract
The disclosure is directed to the Forecasting Methodology at age, device and terminal, the method includes: the multiframe face-image of the user in collection view area；The each frame face-image collected is input in the forecast model of pre-training, described user is carried out the prediction at age, obtain multiple prediction age value of user；Determine current reference age value based on the plurality of prediction age value, and show this age reference value.Application disclosure embodiment, terminal by determining the reference age value of user to be shown to based on multiple prediction age value, the accuracy of age prediction can be improved, avoid differing the biggest between multiple age value of prediction, it also is able to the reference age value avoiding display too big with actual age value deviation, optimizes Consumer's Experience.
Description
Technical field
It relates to technical field of face recognition, particularly relate to the Forecasting Methodology at a kind of age, device and terminal.
Background technology
In correlation technique, generally use the method for pattern recognition that user carries out the prediction at age, and pattern recognition side
In method, common problem is: the change of face's angle of user, light, expression etc. all can cause the change that the age predicts the outcome
Change.If change is too violent, illustrate that algorithm is unstable；If predicted the outcome, the deviation with the correct age is too big, illustrates that algorithm is not
Accurately, predict the outcome the big too many situation than actual age particularly with age of Ms, bring the most bad use to user
Family is experienced.
Summary of the invention
For overcoming problem present in correlation technique, present disclose provides the Forecasting Methodology at age, device and terminal.
First aspect according to disclosure embodiment, it is provided that the Forecasting Methodology at a kind of age, including:
The multiframe face-image of the user in collection view area；
The each frame face-image collected is input in the forecast model of pre-training, described user is carried out the pre-of age
Survey, obtain multiple prediction age value of described user；
Determine current reference age value based on the plurality of prediction age value, and show described with reference to age value.
Optionally, described method also includes:
Gathering the feature of multiple user's sample, described feature includes sex, age and face-image；
Feature based on the plurality of user's sample is trained, and obtains described forecast model.
Optionally, described feature based on the plurality of user's sample is trained, including:
By the method for support vector machine bonding position histogram of gradients, or the method for convolutional neural networks is to the plurality of
The feature of user's sample is trained.
Optionally, described determine current reference age value based on the plurality of age value, including:
For current nth frame face-image, by the prediction age value-nth frame face-image of described 1st frame face-image
Prediction age value be weighted average computation, obtain the age reference value corresponding to described nth frame face-image, N is more than 2
Natural number.
Optionally, the prediction age value of described prediction age value-nth frame face-image by described 1st frame face-image
It is weighted average computation, including:
Based on the weight allocation rule pre-seted, for the prediction age value-nth frame face figure of described 1st frame face-image
The prediction age value of picture is respectively allocated weight；
Based on the weight distributed, described 1st frame face-image is predicted the prediction year of age value-nth frame face-image
Age, value was weighted average computation.
Second aspect according to disclosure embodiment, it is provided that the prediction means at a kind of age, including:
First acquisition module, is configured to gather the multiframe face-image of the user in view area；
Prediction module, is configured to be input in the forecast model of pre-training, to institute each frame face-image collected
State user and carry out the prediction at age, obtain multiple prediction age value of described user；
Determine module, be configured to determine current reference age value based on the plurality of prediction age value, and show institute
State with reference to age value.
Optionally, described device also includes:
Second acquisition module, is configured to gather the feature of multiple user's sample, described feature include sex, the age and
Face-image；
Training module, is configured to feature based on the plurality of user's sample and is trained, obtain described forecast model.
Optionally, described training module includes:
Training submodule, the method being configured to support vector machine bonding position histogram of gradients, or convolutional Neural
The feature of the plurality of user's sample is trained by the method for network.
Optionally, described determine that module includes:
First calculating sub module, is configured to for current nth frame face-image, by described 1st frame face-image
The prediction age value of prediction age value-nth frame face-image is weighted average computation, obtains corresponding to described nth frame face
The age reference value of image, N is the natural number more than 2.
Optionally, described first calculating sub module includes:
Distribution sub module, is configured to based on the weight allocation rule pre-seted, for the prediction of described 1st frame face-image
The prediction age value of age value-nth frame face-image is respectively allocated weight；
3rd calculating sub module, was configured to based on the weight distributed prediction age to described 1st frame face-image
The prediction age value of value-nth frame face-image is weighted average computation.
The third aspect according to disclosure embodiment, it is provided that a kind of terminal, including: processor；Can for storing processor
Perform the memorizer of instruction；Wherein, described processor is configured to:
The multiframe face-image of the user in collection view area；
The each frame face-image collected is input in the forecast model of pre-training, described user is carried out the pre-of age
Survey；
Determine current reference age value based on the plurality of prediction age value, and show described with reference to age value.
Embodiment of the disclosure that the technical scheme of offer can include following beneficial effect:
In the disclosure, terminal by determining the reference age value of user to be shown to based on multiple prediction age value,
The accuracy of age prediction can be improved, it is to avoid between multiple age value of prediction, differ the biggest, it is also possible to avoid the ginseng of display
Examine age value too big with actual age value deviation, optimize Consumer's Experience.And can will be shown to user with reference to age value, for
The family reference camera treatment effect to face-image, in order to user selects desirable filter, optimizes Consumer's Experience.
The disclosure was affected by light, expression, angle, posture etc. are many due to the prediction age, thus uses target
User is tracked by track algorithm, thus gathers multiframe face-image, and the frame face-image of collection is the most, last reference year
Age is worth the real age closer to user, and accuracy is the highest.
In the disclosure, the age of user can be predicted by terminal by the forecast model of pre-training, and terminal can be based on
These features of age, sex and face-image of multiple user's samples train forecast model, and sample is the most, and model accuracy is more
High.
In the disclosure, terminal can use the SVM+HOG method that accuracy is higher, or sample is instructed by CNN scheduling algorithm
Practice, thus obtain forecast model.
In the disclosure, terminal can determine ginseng to be shown based on the age that multiple frame face-images are doped
Examine age value, such as, multiple prediction age value be weighted average computation, thus improve precision and the accuracy of age prediction,
Real age closer to user, it is to avoid predictive value has big difference with actual age, optimizes Consumer's Experience.
In the disclosure, terminal can be each prediction age value distribution weight based on default weight allocation rule, thus just
In reasonably prediction age value being calculated, obtain the reference age value closer to real age.
It should be appreciated that it is only exemplary and explanatory, not that above general description and details hereinafter describe
The disclosure can be limited.
Accompanying drawing explanation
Accompanying drawing herein is merged in description and constitutes the part of this specification, it is shown that meet the enforcement of the disclosure
Example, and for explaining the principle of the disclosure together with description.
Fig. 1 is the disclosure Forecasting Methodology flow chart according to a kind of age shown in an exemplary embodiment.
Fig. 2 is the disclosure Forecasting Methodology flow chart according to the another kind of age shown in an exemplary embodiment.
Fig. 3 is that the disclosure is illustrated according to the application scenarios of the Forecasting Methodology at a kind of age shown in an exemplary embodiment
Figure.
Fig. 4 is the disclosure prediction means block diagram according to a kind of age shown in an exemplary embodiment.
Fig. 5 is the disclosure prediction means block diagram according to the another kind of age shown in an exemplary embodiment.
Fig. 6 is the disclosure prediction means block diagram according to the another kind of age shown in an exemplary embodiment.
Fig. 7 is the disclosure prediction means block diagram according to the another kind of age shown in an exemplary embodiment.
Fig. 8 is the disclosure prediction means block diagram according to the another kind of age shown in an exemplary embodiment.
Fig. 9 is a disclosure structural representation according to a kind of prediction means for the age shown in an exemplary embodiment
Figure.
Detailed description of the invention
Here will illustrate exemplary embodiment in detail, its example represents in the accompanying drawings.Explained below relates to
During accompanying drawing, unless otherwise indicated, the same numbers in different accompanying drawings represents same or analogous key element.Following exemplary embodiment
Described in embodiment do not represent all embodiments consistent with the disclosure.On the contrary, they are only with the most appended
The example of the apparatus and method that some aspects that described in detail in claims, the disclosure are consistent.
The term used in the disclosure is only merely for describing the purpose of specific embodiment, and is not intended to be limiting the disclosure.
" a kind of ", " described " and " being somebody's turn to do " of singulative used in disclosure and the accompanying claims book is also intended to include majority
Form, unless context clearly shows that other implications.It is also understood that term "and/or" used herein refers to and wraps
Any or all containing one or more projects of listing being associated may combination.
Although should be appreciated that in the disclosure possible employing term first, second, third, etc. to describe various information, but this
A little information should not necessarily be limited by these terms.These terms are only used for same type of information is distinguished from each other out.Such as, without departing from
In the case of disclosure scope, the first information can also be referred to as the second information, and similarly, the second information can also be referred to as
One information.Depend on linguistic context, word as used in this " if " can be construed to " ... time " or " when ...
Time " or " in response to determining ".
As it is shown in figure 1, Fig. 1 is the Forecasting Methodology flow chart according to a kind of age shown in an exemplary embodiment, the party
Method may be used for, in terminal, comprising the following steps:
Step 101, the multiframe face-image of the user gathered in view area.
Terminal in the disclosure can be any intelligent terminal with function of surfing the Net, for example, it is possible to specially mobile phone, flat
Plate computer, PDA (Personal Digital Assistant, personal digital assistant) etc..Wherein, terminal can pass through wireless office
Territory net couple in router, and by the server on router access public network.
In the camera of terminal, age forecast function can be carried, in the case of opening this age forecast function, can base
The face-image of the user in view area carries out the prediction at age to user.Wherein, this age forecast function can be manual
Open, it is also possible to be configured in advance, automatically turn on when opening camera function every time.
Step 102, each frame face-image collected is input in the forecast model of pre-training, user is carried out the age
Prediction, obtain multiple prediction age value of user.
In disclosure embodiment, for each frame face-image gathered, terminal all carries out the prediction at age to it, thus
Obtain multiple prediction age value.Each frame face-image can all be stored by terminal, it is also possible to does not stores, the feelings not stored
Under condition, after obtaining based on each frame face-image predicting age value, corresponding face-image is deleted.Terminal is permissible
Multiple prediction age value are stored, in order to follow-up based on multiple prediction age value calculating reference age value.
Step 103, determine current reference age value based on multiple prediction age value, and show this reference age value.
In disclosure embodiment, terminal references predicts the multiple prediction age value obtained, and determines user to be shown to
Reference age value.
In above-described embodiment, terminal by determining the reference year of user to be shown to based on multiple prediction age value
Age is worth, it is possible to increase the accuracy of age prediction, it is to avoid differ the biggest between multiple prediction age value of prediction, it is also possible to avoid
The reference age value of display and actual age value deviation are too big, optimize Consumer's Experience.
As in figure 2 it is shown, Fig. 2 is the Forecasting Methodology flow chart according to the another kind of age shown in an exemplary embodiment, should
Method may be used for, in terminal, comprising the following steps:
Step 201, the user identified in view area.
In disclosure embodiment, the user in photographic head view area can be identified by terminal, and by face
Image carries out contrast and determines whether it is same user.
Step 202, in the case of user is same people, based on target tracking algorism gather user multiframe face figure
Picture.
In disclosure step, terminal can use target tracking algorism, such as meansift algorithm, Camshift algorithm etc.
The face-image of user is tracked, and, as long as the face of user does not leaves the viewfinder range of photographic head, just it is always maintained at
Tracking mode, thus can continuous collecting multiframe face-image for same user.The purpose using target tracking algorism is right
Same user gathers multiframe face-image, in order to every two field picture all carries out age prediction, then to the some prediction years obtained
Age, value carried out comprehensive value, obtained final reference age value, due to the prediction age by light, expression, angle, posture etc. in many ways
The impact in face, thus the frame face-image gathered is the most, last reference age value is closer to the real age of user, accuracy
The highest.
Step 203, for each frame face-image, forecast model based on pre-training carries out the prediction at age.
In disclosure embodiment, terminal training in advance forecast model.First obtaining multiple user's sample, user's sample is mark
Note has age and the face-image of sex, terminal gathers the feature of these user's samples, feature i.e.: face-image and the property of correspondence
Not and the age, being then based on these features is trained, it is possible to obtain forecast model.Wherein, terminal can pass through such as SVM
(Support Vector Machine, support vector machine)+HOG (Histogram of Oriented Gradient, direction ladder
Degree rectangular histogram) mode, or CNN (Convolutional Neural Network, convolutional neural networks) carries out the degree of depth
Practising, train and obtain forecast model, the forecast model that different methods obtains is incomplete same, and terminal can select its one side
Formula is trained, and obtains a forecast model, and the forecast model that distinct methods obtains may be used to the age to user and carries out
Prediction.
Step 204, multiple prediction age value are calculated, obtain with reference to age value.
Owing to the purpose of disclosure embodiment is the ginseng that comprehensive multiple prediction age value determines user to be shown to
Examine age value, thus for the former frame face-images gathered, be not usually required to display and predict the outcome, but as calculating ginseng
Examine the basis of age value.Generally, numerical value of N can be set, such as 5, from the beginning of the 5th frame face-image, based on the 1st frame the-the 5 pattern
The prediction age value of portion's image determines the reference age value of the 5th frame face-image, the most current reference age value.
In a possible implementation, terminal prediction age value-nth frame face-image to the 1st frame face-image
Prediction age value be averaged calculating, ask for its meansigma methods, as with reference to age value.The 5 frame face-images gathered are divided
5 the prediction age value not being predicted obtaining are averaged calculating, and the result obtained is as current reference age value.
Such as, shown reference age value can calculate based on following formula (1):
Wherein gi is the prediction age value that the i-th predicted Model Identification of frame face-image goes out, and showage is for showing at screen
On reference age value.
In the implementation that another is possible, prediction age value-nth frame face that terminal is corresponding to the 1st frame face-image
The prediction age value that portion's image is corresponding is weighted average computation, using result of calculation as with reference to age value.Wherein weighted value can
To determine based on the weight assignment rule pre-seted, such as, acquisition time is the closer to the prediction year of the face-image of current time
Age, the weighted value of value correspondence was the biggest, thus, owing to having considered multiple prediction age value, it is possible to increase the accuracy of prediction,
Avoid predicting the outcome and too great deviations occurs, also avoid, with actual age value, too great deviations occurs.
In a possible implementation, it is also possible to after nth frame, to each two field picture, all combine the most each
The age value of the prediction of two field picture determines with reference to age value.Such as arranging N is 10, then for the 11st two field picture, based on
The age value that 1-the 11st two field picture predicts respectively obtains with reference to age value, and shows this reference age value；For the 12nd pattern
Portion's image, the age value predicted respectively based on 1-the 12nd two field picture obtains with reference to age value, and shows this reference age value.
Such as, shown reference age value can calculate based on following formula (2):
Wherein wi is weight, wi cumulative and be 1.
Step 205, display are with reference to age value.
In disclosure step, user will be shown to reference to age value, owing to the function at this prediction age is generally opening
Use during skin Caring function, it is thus possible to providing the user preferable reference, user can see current skin Caring effect intuitively
And the age of correspondence.
Using said method, when the quantity of the face-image acquired is more, the final reference age value calculated is the most gradually
Gradually tend to be steady, and close to the real age of user, thus improve stability and the accuracy of age display.
As it is shown on figure 3, Fig. 3 is the disclosure application according to the Forecasting Methodology at a kind of age shown in an exemplary embodiment
Scene schematic diagram.In the scene shown in Fig. 3, including: as the smart mobile phone of terminal.
When user opens the skin Caring camera function of smart mobile phone, photographic head starts to gather the user's in view area
Face-image, and to each frame face-image gathered, it is input in the forecast model of pre-training carry out the prediction at age, thus
Obtain multiple age predictive value.It is 6 owing to being provided with threshold value, thus for the 6th frame face-image, is calculated by forecast model
After its prediction age value, by the prediction age corresponding for prediction age value the-the 6 frame face-image corresponding for the 1st frame face-image
Value is averaged calculating, obtains with reference to age value, and this reference age value is shown on screen by terminal, for reference.And
The each frame face-image gathered afterwards, all carries out above-mentioned calculating, such as to the 10th frame face-image, by the 1st frame face-image
The prediction age value that corresponding prediction age value the-the 10 frame face-image is corresponding is averaged calculating, using the value that obtains as ginseng
Examine age value to show.
In application scenarios shown in Fig. 3, it is achieved the detailed process of the prediction at age may refer to aforementioned to retouching in Fig. 1-2
State, do not repeat them here.
Corresponding with the Forecasting Methodology embodiment at aforementioned age, the disclosure additionally provides the prediction means at age and is answered
The embodiment of terminal.
As shown in Figure 4, Fig. 4 is the disclosure prediction means block diagram according to a kind of age shown in an exemplary embodiment,
This device can be applied in the terminal, and for the method performing embodiment illustrated in fig. 1, this device may include that the first collection
Module 410, prediction module 420 and determine module 430.
Wherein, the first acquisition module 410, it is configured to gather the multiframe face-image of the user in view area；
Prediction module 420, is configured to be input in the forecast model of pre-training each frame face-image collected, right
Described user carries out the prediction at age, obtains multiple prediction age value of user；
Determine module 430, be configured to determine current reference age value based on multiple prediction age value, and show this ginseng
Examine age value.
In above-described embodiment, terminal by determining the reference year of user to be shown to based on multiple prediction age value
Age is worth, it is possible to increase the accuracy of age prediction, it is to avoid differ the biggest between multiple age value of prediction, it is also possible to avoid display
Reference age value and actual age value deviation too big, optimize Consumer's Experience.And owing to the prediction age is by light, expression, angle
Many impacts such as degree, posture, thus use target tracking algorism that user is tracked, thus gather multiframe face figure
Picture, the frame face-image of collection is the most, and last reference age value is closer to the real age of user, and accuracy is the highest.Another
Aspect, terminal can will be shown to user, the camera for reference treatment effect to face-image with reference to age value, in order to uses
Family selects desirable filter, optimizes Consumer's Experience.
As it is shown in figure 5, Fig. 5 is the disclosure prediction means frame according to the another kind of age shown in an exemplary embodiment
Figure, this embodiment is on the basis of aforementioned embodiment illustrated in fig. 4, and this device can also include: the second acquisition module 440 and training
Module 450.
Second acquisition module 440, is configured to gather the feature of multiple user's sample, and described feature includes sex, age
And face-image；
Training module 450, is configured to feature based on the plurality of user's sample and is trained, obtain described prediction mould
Type.
In above-described embodiment, terminal can these features of age, sex and face-image based on multiple user's samples be come
Training forecast model, sample is the most, and model accuracy is the highest.
As shown in Figure 6, Fig. 6 is the disclosure prediction means frame according to the another kind of age shown in an exemplary embodiment
Figure, this embodiment is on the basis of aforementioned embodiment illustrated in fig. 5, and training module 450 may include that training submodule 451.
Training submodule 451, the method being configured to support vector machine bonding position histogram of gradients, or convolution god
Through the method for network, the feature of the plurality of user's sample is trained.
In above-described embodiment, terminal can use the SVM+HOG that accuracy is higher, or sample is instructed by CNN scheduling algorithm
Practice, thus obtain forecast model.
As it is shown in fig. 7, Fig. 7 is the disclosure prediction means frame according to the another kind of age shown in an exemplary embodiment
Figure, this embodiment, on the basis of aforementioned embodiment illustrated in fig. 4, determines that module 430 may include that the first calculating sub module 431.
First calculating sub module 431, is configured to for current nth frame face-image, by described 1st frame face-image
The prediction age value of prediction age value-nth frame face-image be weighted average computation, obtain corresponding to described nth frame face
The age reference value of portion's image, N is the natural number more than 2.
In above-described embodiment, terminal can determine based on the age that multiple frame face-images are doped and finally to show
Reference age value, such as multiple prediction age value are weighted average computation, thus improve precision and the standard of age prediction
Exactness, closer to the real age of user, it is to avoid predictive value has big difference with actual age, optimizes Consumer's Experience.
As shown in Figure 8, Fig. 8 is the disclosure prediction means frame according to the another kind of age shown in an exemplary embodiment
Figure, this embodiment is on the basis of aforementioned embodiment illustrated in fig. 7, and the first calculating sub module 431 may include that
Distribution sub module 433 and the 3rd calculating sub module 434.
Distribution sub module 433, is configured to based on the weight allocation rule pre-seted, for described 1st frame face-image
The prediction age value of prediction age value-nth frame face-image is respectively allocated weight；
3rd calculating sub module 434, was configured to based on the weight distributed prediction year to described 1st frame face-image
Age, the prediction age value of value-nth frame face-image was weighted average computation.
Wherein, weight allocation rule includes:
The acquisition time distance current time of face-image is the nearest, and weight is the biggest.
In above-described embodiment, terminal can be each prediction age value distribution weight based on default weight allocation rule,
Consequently facilitating reasonably prediction age value is calculated, obtain the reference age value closer to real age.
The prediction means embodiment at the age shown in above-mentioned Fig. 4 to Fig. 8 can be applied in the terminal.
In said apparatus, the function of unit and the process that realizes of effect specifically refer to corresponding step in said method
Realize process, do not repeat them here.
For device embodiment, owing to it corresponds essentially to embodiment of the method, so relevant part sees method in fact
The part executing example illustrates.Device embodiment described above is only schematically, wherein said as separating component
The unit illustrated can be or may not be physically separate, and the parts shown as unit can be or can also
It not physical location, i.e. may be located at a place, or can also be distributed on multiple NE.Can be according to reality
Need to select some or all of module therein to realize the purpose of disclosure scheme.Those of ordinary skill in the art are not paying
In the case of going out creative work, i.e. it is appreciated that and implements.
Corresponding with Fig. 4, the disclosure also provides for a kind of terminal, and described terminal includes processor；For storing processor
The memorizer of executable instruction；Wherein, described processor is configured to:
The multiframe face-image of the user in collection view area；
The each frame face-image collected is input in the forecast model of pre-training, described user is carried out the pre-of age
Survey；
Determine current reference age value based on the plurality of prediction age value, and show described with reference to age value.
As it is shown in figure 9, Fig. 9 is that the disclosure is according to a kind of prediction means for the age shown in an exemplary embodiment
One structural representation (end side) of 900.Such as, device 900 can be the mobile phone with routing function, computer, number
Word broadcast terminal, messaging devices, game console, tablet device, armarium, body-building equipment, personal digital assistant etc..
With reference to Fig. 9, device 900 can include following one or more assembly: processes assembly 902, memorizer 904, power supply
Assembly 906, multimedia groupware 908, audio-frequency assembly 910, the interface 912 of input/output (I/O), sensor cluster 914, and
Communications component 916.
Process assembly 902 and generally control the integrated operation of device 900, such as with display, call, data communication, phase
The operation that machine operation and record operation are associated.Process assembly 902 and can include that one or more processor 920 performs to refer to
Order, to complete all or part of step of above-mentioned method.Additionally, process assembly 902 can include one or more module, just
Mutual in process between assembly 902 and other assemblies.Such as, process assembly 902 and can include multi-media module, many to facilitate
Media component 908 and process between assembly 902 mutual.
Memorizer 904 is configured to store various types of data to support the operation at device 900.Showing of these data
Example includes any application program for operation on device 900 or the instruction of method, contact data, telephone book data, disappears
Breath, picture, video etc..Memorizer 904 can be by any kind of volatibility or non-volatile memory device or their group
Close and realize, such as static RAM (SRAM), Electrically Erasable Read Only Memory (EEPROM), erasable compile
Journey read only memory (EPROM), programmable read only memory (PROM), read only memory (ROM), magnetic memory, flash
Device, disk or CD.
The various assemblies that power supply module 906 is device 900 provide electric power.Power supply module 906 can include power management system
System, one or more power supplys, and other generate, manage and distribute, with for device 900, the assembly that electric power is associated.
The screen of one output interface of offer that multimedia groupware 908 is included between described device 900 and user.One
In a little embodiments, screen can include liquid crystal display (LCD) and touch panel (TP).If screen includes touch panel, screen
Curtain may be implemented as touch screen, to receive the input signal from user.Touch panel includes one or more touch sensing
Device is with the gesture on sensing touch, slip and touch panel.Described touch sensor can not only sense touch or sliding action
Border, but also detect the persistent period relevant to described touch or slide and pressure.In certain embodiments, many matchmakers
Body assembly 908 includes a front-facing camera and/or post-positioned pick-up head.When device 900 is in operator scheme, such as screening-mode or
During video mode, front-facing camera and/or post-positioned pick-up head can receive the multi-medium data of outside.Each front-facing camera and
Post-positioned pick-up head can be a fixing optical lens system or have focal length and optical zoom ability.
Audio-frequency assembly 910 is configured to output and/or input audio signal.Such as, audio-frequency assembly 910 includes a Mike
Wind (MIC), when device 900 is in operator scheme, during such as call model, logging mode and speech recognition mode, mike is joined
It is set to receive external audio signal.The audio signal received can be further stored at memorizer 904 or via communication set
Part 916 sends.In certain embodiments, audio-frequency assembly 910 also includes a speaker, is used for exporting audio signal.
I/O interface 912 provides interface for processing between assembly 902 and peripheral interface module, above-mentioned peripheral interface module can
To be keyboard, put striking wheel, button etc..These buttons may include but be not limited to: home button, volume button, start button and lock
Set button.
Sensor cluster 914 includes one or more sensor, for providing the state of various aspects to comment for device 900
Estimate.Such as, what sensor cluster 914 can detect device 900 opens/closed mode, the relative localization of assembly, such as described
Assembly is display and the keypad of device 900, and sensor cluster 914 can also detect device 900 or 900 1 assemblies of device
Position change, the presence or absence that user contacts with device 900, device 900 orientation or acceleration/deceleration and device 900
Variations in temperature.Sensor cluster 914 can include proximity transducer, is configured to when not having any physical contact detect
The existence of neighbouring object.Sensor cluster 914 can also include optical sensor, such as CMOS or ccd image sensor, is used for becoming
Use as in application.In certain embodiments, this sensor cluster 914 can also include acceleration transducer, gyro sensors
Device, Magnetic Sensor, pressure transducer, microwave remote sensor or temperature sensor.
Communications component 916 is configured to facilitate the communication of wired or wireless mode between device 900 and other equipment.Device
900 can access wireless network based on communication standard, such as WiFi, 2G or 3G, or combinations thereof.An exemplary enforcement
In example, communications component 916 receives the broadcast singal from external broadcasting management system or broadcast related information via broadcast channel.
In one exemplary embodiment, described communications component 916 also includes near-field communication (NFC) module, to promote junction service.Example
As, can be based on RF identification (RFID) technology in NFC module, Infrared Data Association (IrDA) technology, ultra broadband (UWB) technology,
Bluetooth (BT) technology and other technologies realize.
In the exemplary embodiment, device 900 can be by one or more application specific integrated circuits (ASIC), numeral letter
Number processor (DSP), digital signal processing appts (DSPD), PLD (PLD), field programmable gate array
(FPGA), controller, microcontroller, microprocessor or other electronic components realize, be used for performing said method.
In the exemplary embodiment, a kind of non-transitory computer-readable recording medium including instruction, example are additionally provided
As included the memorizer 904 of instruction, above-mentioned instruction can have been performed said method by the processor 920 of device 900.Such as,
Described non-transitory computer-readable recording medium can be ROM, random access memory (RAM), CD-ROM, tape, floppy disk
With optical data storage devices etc..
The disclosure additionally provides a kind of non-transitory computer-readable recording medium, when the instruction in described storage medium by
When the processor of mobile terminal performs so that mobile terminal is able to carry out the Forecasting Methodology at a kind of age, and described method includes:
The multiframe face-image of the user in collection view area；
The each frame face-image collected is input in the forecast model of pre-training, described user is carried out the pre-of age
Survey, obtain multiple prediction age value of described user；
Determine current reference age value based on the plurality of prediction age value, and show described with reference to age value.
Those skilled in the art, after considering description and putting into practice invention disclosed herein, will readily occur to its of the disclosure
Its embodiment.The disclosure is intended to any modification, purposes or the adaptations of the disclosure, these modification, purposes or
Person's adaptations is followed the general principle of the disclosure and includes the undocumented common knowledge in the art of the disclosure
Or conventional techniques means.Description and embodiments is considered only as exemplary, and the true scope of the disclosure and spirit are by following
Claim is pointed out.
The foregoing is only the preferred embodiment of the disclosure, not in order to limit the disclosure, all essences in the disclosure
Within god and principle, any modification, equivalent substitution and improvement etc. done, should be included within the scope of disclosure protection.
Claims (11)
1. the Forecasting Methodology at an age, it is characterised in that including:
The multiframe face-image of the user in collection view area；
The each frame face-image collected is input in the forecast model of pre-training, described user is carried out the prediction at age,
Obtain multiple prediction age value of described user；
Determine current reference age value based on the plurality of prediction age value, and show described with reference to age value.
Method the most according to claim 1, it is characterised in that described method also includes:
Gathering the feature of multiple user's sample, described feature includes sex, age and face-image；
Feature based on the plurality of user's sample is trained, and obtains described forecast model.
Method the most according to claim 2, it is characterised in that described feature based on the plurality of user's sample is instructed
Practice, including:
By the method for support vector machine bonding position histogram of gradients, or the method for convolutional neural networks is to the plurality of user
The feature of sample is trained.
Method the most according to claim 1, it is characterised in that described determine current based on the plurality of prediction age value
Reference age value, including:
For current nth frame face-image, pre-by the prediction age value-nth frame face-image of described 1st frame face-image
Surveying age value and be weighted average computation, obtain the age reference value corresponding to described nth frame face-image, N is oneself more than 2
So number.
Method the most according to claim 4, it is characterised in that the described prediction age value by described 1st frame face-image-
The prediction age value of nth frame face-image is weighted average computation, including:
Based on the weight allocation rule pre-seted, for the prediction age value-nth frame face-image of described 1st frame face-image
Prediction age value is respectively allocated weight；
Based on the weight distributed, described 1st frame face-image is predicted the prediction age value of age value-nth frame face-image
It is weighted average computation.
6. the prediction means at an age, it is characterised in that including:
First acquisition module, is configured to gather the multiframe face-image of the user in view area；
Prediction module, is configured to be input in the forecast model of pre-training, to described use each frame face-image collected
Family carries out the prediction at age, obtains multiple prediction age value of described user；
Determine module, be configured to determine current reference age value based on the plurality of prediction age value, and show described ginseng
Examine age value.
Device the most according to claim 6, it is characterised in that described device also includes:
Second acquisition module, is configured to gather the feature of multiple user's sample, and described feature includes sex, age and face
Image；
Training module, is configured to feature based on the plurality of user's sample and is trained, obtain described forecast model.
Device the most according to claim 7, it is characterised in that described training module includes:
Training submodule, the method being configured to support vector machine bonding position histogram of gradients, or convolutional neural networks
Method the feature of the plurality of user's sample is trained.
Device the most according to claim 6, it is characterised in that described determine that module includes:
First calculating sub module, is configured to for current nth frame face-image, by the prediction of described 1st frame face-image
The prediction age value of age value-nth frame face-image is weighted average computation, obtains corresponding to described nth frame face-image
Age reference value, N is the natural number more than 2.
Device the most according to claim 9, it is characterised in that described first calculating sub module includes:
Distribution sub module, is configured to based on the weight allocation rule pre-seted, for the prediction age of described 1st frame face-image
The prediction age value of value-nth frame face-image is respectively allocated weight；
3rd calculating sub module, is configured to based on the weight the distributed prediction age value-the to described 1st frame face-image
The prediction age value of N frame face-image is weighted average computation.
11. 1 kinds of terminals, it is characterised in that including: processor；For storing the memorizer of processor executable；Wherein,
Described processor is configured to:
The multiframe face-image of the user in collection view area；
The each frame face-image collected is input in the forecast model of pre-training, described user is carried out the prediction at age；
Determine current reference age value based on the plurality of prediction age value, and show described with reference to age value.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610512766.7A CN106203306A (en) | 2016-06-30 | 2016-06-30 | The Forecasting Methodology at age, device and terminal |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610512766.7A CN106203306A (en) | 2016-06-30 | 2016-06-30 | The Forecasting Methodology at age, device and terminal |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN106203306A true CN106203306A (en) | 2016-12-07 |
Family
ID=57464060
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201610512766.7A Pending CN106203306A (en) | 2016-06-30 | 2016-06-30 | The Forecasting Methodology at age, device and terminal |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN106203306A (en) |
Cited By (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108021863A (en) * | 2017-11-01 | 2018-05-11 | 平安科技（深圳）有限公司 | Electronic device, the character classification by age method based on image and storage medium |
| CN108108668A (en) * | 2017-12-01 | 2018-06-01 | 北京小米移动软件有限公司 | Age Forecasting Methodology and device based on image |
| CN108197592A (en) * | 2018-01-22 | 2018-06-22 | 百度在线网络技术（北京）有限公司 | Information acquisition method and device |
| CN108345604A (en) * | 2017-01-22 | 2018-07-31 | 阿里巴巴集团控股有限公司 | Data processing method, recommends method and relevant device at searching method |
| CN109376932A (en) * | 2018-10-30 | 2019-02-22 | 平安医疗健康管理股份有限公司 | Age prediction method, device, server and storage medium based on prediction model |
| CN109829415A (en) * | 2019-01-25 | 2019-05-31 | 平安科技（深圳）有限公司 | Gender identification method, device, medium and equipment based on depth residual error network |
| CN110473171A (en) * | 2019-07-18 | 2019-11-19 | 上海联影智能医疗科技有限公司 | Brain age detection method, computer equipment and storage medium |
| CN110709856A (en) * | 2017-05-31 | 2020-01-17 | 宝洁公司 | System and method for determining apparent skin age |
| CN115862597A (en) * | 2022-06-17 | 2023-03-28 | 南京地平线集成电路有限公司 | Method and device for determining character type, electronic equipment and storage medium |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102567719A (en) * | 2011-12-26 | 2012-07-11 | 东南大学 | Human age automatic estimation method based on posterior probability neural network |
| CN102663413A (en) * | 2012-03-09 | 2012-09-12 | 中盾信安科技(江苏)有限公司 | Multi-gesture and cross-age oriented face image authentication method |
| CN102693418A (en) * | 2012-05-17 | 2012-09-26 | 上海中原电子技术工程有限公司 | Multi-pose face identification method and system |
| CN104036236A (en) * | 2014-05-27 | 2014-09-10 | 厦门瑞为信息技术有限公司 | Human face gender recognition method based on multi-parameter index weighting |
| CN105279499A (en) * | 2015-10-30 | 2016-01-27 | 小米科技有限责任公司 | Age recognition method and device |
| CN105335709A (en) * | 2015-10-21 | 2016-02-17 | 奇酷互联网络科技（深圳）有限公司 | Face identification display method, face identification display device and terminal |
| CN105678253A (en) * | 2016-01-04 | 2016-06-15 | 东南大学 | Semi-supervised age estimation device based on faces and semi-supervised age estimation method based on faces |
-
2016
- 2016-06-30 CN CN201610512766.7A patent/CN106203306A/en active Pending
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102567719A (en) * | 2011-12-26 | 2012-07-11 | 东南大学 | Human age automatic estimation method based on posterior probability neural network |
| CN102663413A (en) * | 2012-03-09 | 2012-09-12 | 中盾信安科技(江苏)有限公司 | Multi-gesture and cross-age oriented face image authentication method |
| CN102693418A (en) * | 2012-05-17 | 2012-09-26 | 上海中原电子技术工程有限公司 | Multi-pose face identification method and system |
| CN104036236A (en) * | 2014-05-27 | 2014-09-10 | 厦门瑞为信息技术有限公司 | Human face gender recognition method based on multi-parameter index weighting |
| CN105335709A (en) * | 2015-10-21 | 2016-02-17 | 奇酷互联网络科技（深圳）有限公司 | Face identification display method, face identification display device and terminal |
| CN105279499A (en) * | 2015-10-30 | 2016-01-27 | 小米科技有限责任公司 | Age recognition method and device |
| CN105678253A (en) * | 2016-01-04 | 2016-06-15 | 东南大学 | Semi-supervised age estimation device based on faces and semi-supervised age estimation method based on faces |
Non-Patent Citations (1)
| Title |
|---|
| ALEX NKENGNE 等: "Age Prediction using a Supervised Facial Model", 《2011 IEEE INTERNATIONAL SYMPOSIUM ON BIOMEDICAL IMAGING:FROM NANO TO MACRO》 * |
Cited By (14)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108345604B (en) * | 2017-01-22 | 2022-01-25 | 阿里巴巴集团控股有限公司 | Data processing method, searching method, recommending method and related equipment |
| CN108345604A (en) * | 2017-01-22 | 2018-07-31 | 阿里巴巴集团控股有限公司 | Data processing method, recommends method and relevant device at searching method |
| CN110709856A (en) * | 2017-05-31 | 2020-01-17 | 宝洁公司 | System and method for determining apparent skin age |
| CN110709856B (en) * | 2017-05-31 | 2023-11-28 | 宝洁公司 | System and method for determining apparent skin age |
| CN108021863A (en) * | 2017-11-01 | 2018-05-11 | 平安科技（深圳）有限公司 | Electronic device, the character classification by age method based on image and storage medium |
| CN108108668B (en) * | 2017-12-01 | 2021-04-27 | 北京小米移动软件有限公司 | Age prediction method and device based on image |
| CN108108668A (en) * | 2017-12-01 | 2018-06-01 | 北京小米移动软件有限公司 | Age Forecasting Methodology and device based on image |
| CN108197592A (en) * | 2018-01-22 | 2018-06-22 | 百度在线网络技术（北京）有限公司 | Information acquisition method and device |
| CN108197592B (en) * | 2018-01-22 | 2022-05-27 | 百度在线网络技术（北京）有限公司 | Information acquisition method and device |
| CN109376932A (en) * | 2018-10-30 | 2019-02-22 | 平安医疗健康管理股份有限公司 | Age prediction method, device, server and storage medium based on prediction model |
| CN109829415A (en) * | 2019-01-25 | 2019-05-31 | 平安科技（深圳）有限公司 | Gender identification method, device, medium and equipment based on depth residual error network |
| WO2020151300A1 (en) * | 2019-01-25 | 2020-07-30 | 平安科技（深圳）有限公司 | Deep residual network-based gender recognition method and apparatus, medium, and device |
| CN110473171A (en) * | 2019-07-18 | 2019-11-19 | 上海联影智能医疗科技有限公司 | Brain age detection method, computer equipment and storage medium |
| CN115862597A (en) * | 2022-06-17 | 2023-03-28 | 南京地平线集成电路有限公司 | Method and device for determining character type, electronic equipment and storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN106203306A (en) | The Forecasting Methodology at age, device and terminal | |
| CN110516745B (en) | Training method and device of image recognition model and electronic equipment | |
| CN105512685B (en) | Object identification method and device | |
| CN110381443A (en) | Near-field communication card Activiation method and device | |
| CN105182776A (en) | Intelligent furniture controlling method and apparatus | |
| CN106355573A (en) | Target object positioning method and device in pictures | |
| CN107582028A (en) | Sleep monitor method and device | |
| CN107832741A (en) | The method, apparatus and computer-readable recording medium of facial modeling | |
| CN106845377A (en) | Face key independent positioning method and device | |
| CN106295511A (en) | Face tracking method and device | |
| CN108010060A (en) | Object detection method and device | |
| CN107527024A (en) | Face face value appraisal procedure and device | |
| CN106157602A (en) | The method and apparatus of calling vehicle | |
| CN109819288A (en) | Determination method, apparatus, electronic equipment and the storage medium of advertisement dispensing video | |
| CN106778531A (en) | Face detection method and device | |
| CN106228158A (en) | The method and apparatus of picture detection | |
| CN109543537A (en) | Weight identification model increment training method and device, electronic equipment and storage medium | |
| CN104361486A (en) | Alarm clock reminding method and device | |
| CN109117874A (en) | Operation behavior prediction technique and device | |
| CN104063865A (en) | Classification model creation method, image segmentation method and related device | |
| CN109635920A (en) | Neural network optimization and device, electronic equipment and storage medium | |
| CN113807150B (en) | Data processing, posture prediction method, device and storage medium | |
| CN103886284A (en) | Character attribute information identification method and device and electronic device | |
| CN106339695A (en) | Face similarity detection method, device and terminal | |
| CN107194464A (en) | The training method and device of convolutional neural networks model |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20161207 |