CN108255998A - A kind of hair style recommends method, apparatus and equipment - Google Patents
A kind of hair style recommends method, apparatus and equipment Download PDFInfo
- Publication number
- CN108255998A CN108255998A CN201711481867.3A CN201711481867A CN108255998A CN 108255998 A CN108255998 A CN 108255998A CN 201711481867 A CN201711481867 A CN 201711481867A CN 108255998 A CN108255998 A CN 108255998A
- Authority
- CN
- China
- Prior art keywords
- information
- collocation
- hair style
- face
- shape
- 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
Classifications
-
- A—HUMAN NECESSITIES
- A45—HAND OR TRAVELLING ARTICLES
- A45D—HAIRDRESSING OR SHAVING EQUIPMENT; EQUIPMENT FOR COSMETICS OR COSMETIC TREATMENTS, e.g. FOR MANICURING OR PEDICURING
- A45D44/00—Other cosmetic or toiletry articles, e.g. for hairdressers' rooms
- A45D44/005—Other cosmetic or toiletry articles, e.g. for hairdressers' rooms for selecting or displaying personal cosmetic colours or hairstyle
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/50—Information retrieval; Database structures therefor; File system structures therefor of still image data
- G06F16/58—Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually
- G06F16/583—Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually using metadata automatically derived from the content
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/90—Details of database functions independent of the retrieved data types
- G06F16/95—Retrieval from the web
- G06F16/953—Querying, e.g. by the use of web search engines
- G06F16/9535—Search customisation based on user profiles and personalisation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q30/00—Commerce
- G06Q30/02—Marketing; Price estimation or determination; Fundraising
- G06Q30/0241—Advertisements
- G06Q30/0251—Targeted advertisements
- G06Q30/0269—Targeted advertisements based on user profile or attribute
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q30/00—Commerce
- G06Q30/02—Marketing; Price estimation or determination; Fundraising
- G06Q30/0282—Rating or review of business operators or products
Abstract
The present invention provides a kind of hair styles to recommend method, apparatus and equipment, the method includes：The shape of face identification model that the face image data input of user to be recommended is pre-established obtains the shape of face information of user to be recommended；Generation a variety of collocation informations corresponding with the shape of face information, wherein, each collocation information includes shape of face information and a kind of corresponding hair style information；A variety of collocation informations are inputted to the collocation Rating Model pre-established respectively, and is obtained according to the scoring that collocation Rating Model exports and recommends hair style information.The present invention carries out a variety of hair style collocation, and score each collocation by obtaining the shape of face information of user, according to shape of face information, recommends hair style information so as to obtain, can provide accurate hairstyle suggesting to the user.
Description
Technical field
The present invention relates to machine learning techniques fields, and in particular to a kind of hair style recommends method, apparatus and equipment.
Background technology
With the raising of social common aesthetic taste and the concern to fashion, people increasingly pursue hair style diversity and
It is personalized.And in daily life, many do not have professional aesthetics and hair style knowledge ordinary people can not recognize be most suitable for from
The hair style of own shape of face is suggested so as to be overly dependent upon hairdresser or other people subjective；But hairdresser is to need to pass through study
A large amount of aesthetics knowledge frequentlies provide really rational suggestion, and most of ordinary peoples is caused to be difficult to obtain accurate hairstyle suggesting.
Invention content
In view of the above defects of the prior art, the present invention provides a kind of hair style recommendation method, apparatus and equipment.
An aspect of of the present present invention provides a kind of hair style and recommends method, including：The face image data of user to be recommended is defeated
Enter the shape of face identification model pre-established, obtain the shape of face information of user to be recommended；Generation is corresponding more with the shape of face information
Kind collocation information, wherein, each collocation information includes shape of face information and a kind of corresponding hair style information；By a variety of collocation informations point
The collocation Rating Model pre-established is not inputted, and is obtained according to the scoring that collocation Rating Model exports and recommended hair style information.
Wherein, before the face image data by user to be recommended inputs the step of shape of face identification model pre-established
It further includes：The first training set is obtained, first training set includes multiple face image datas and corresponding shape of face information；
Face image data and corresponding shape of face information are subjected to model training as input data and output data, obtain training
The shape of face identification model of completion.
Wherein, it was further included before described the step of a variety of collocation informations are inputted the collocation Rating Model pre-established respectively：
The second training set is obtained, second training set includes a variety of collocation informations and corresponding scoring, a variety of collocation letters
Breath by a variety of shape of face information and a variety of hair style information random alignment combination producings, each collocation information include a kind of shape of face information and
A kind of hair style information, the scoring are determined by expert estimation；Using collocation information and corresponding scoring as input data
Model training is carried out with output data, obtains the collocation Rating Model that training is completed.
Wherein, the scoring exported according to collocation Rating Model further includes after obtaining the step of recommending hair style information：Root
Three-dimensional facial model is established according to the face image data of user to be recommended；According to hair style recommendation information, in the three-dimensional face mould
Three-dimensional hair style model is established in type.
Wherein, the scoring exported according to collocation Rating Model further includes after obtaining the step of recommending hair style information：It obtains
Take application scenarios information；Hair style information is recommended into scoring to described according to the application scenarios information；Correspondingly, it is described according to hair
Type recommendation information is further included in the step of three-dimensional hair style model of foundation in the three-dimensional facial model, according to application scenarios information
Three-dimensional scene models are established at the rear portion of three-dimensional facial model and three-dimensional hair style model.
Wherein, the scoring exported according to collocation Rating Model further includes after obtaining the step of recommending hair style information：It obtains
Take the current hair style information of user to be recommended；Hair style information and current hair style information will be recommended to be compared, and be tied according to comparing
Fruit generates corresponding hair style and arranges strategy.
Another aspect of the present invention provides a kind of hair style recommendation apparatus, including：Shape of face identification module, for by use to be recommended
The shape of face identification model that the face image data input at family pre-establishes obtains the shape of face information of user to be recommended；Collocation generation
Module, for generating corresponding with the shape of face information a variety of collocation informations, wherein, each collocation information including shape of face information and
A kind of corresponding hair style information；Hair style recommending module, for a variety of collocation informations to be inputted to the collocation pre-established scoring respectively
Model, and obtained according to the scoring that collocation Rating Model exports and recommend hair style information.
Described device further includes：Model building module, it is three-dimensional for being established according to the face image data of user to be recommended
Facial model；According to hair style recommendation information, three-dimensional hair style model is established in the three-dimensional facial model.
Another aspect of the present invention provides a kind of hair style recommendation apparatus, including：At least one processor；And with the place
At least one processor of device communication connection is managed, wherein：The memory, which is stored with, to be referred to by the program that the processor performs
It enables, the hair style that the processor calls described program instruction to be able to carry out the above-mentioned aspect offer of the present invention recommends method, such as wrap
It includes：The shape of face identification model that the face image data input of user to be recommended is pre-established obtains the shape of face of user to be recommended
Information；Generation a variety of collocation informations corresponding with the shape of face information, wherein, each collocation information includes shape of face information and correspondence
A kind of hair style information；A variety of collocation informations are inputted to the collocation Rating Model pre-established respectively, and according to collocation scoring mould
The scoring of type output, which obtains, recommends hair style information.
Another aspect of the present invention provides a kind of non-transient computer readable storage medium storing program for executing, and the non-transient computer is readable
Storage medium stores computer instruction, and the computer instruction makes the computer perform the hair style that the above-mentioned aspect of the present invention provides
Recommendation method, such as including：The shape of face identification model that the face image data input of user to be recommended is pre-established, acquisition are treated
The shape of face information of recommended user；Generation a variety of collocation informations corresponding with the shape of face information, wherein, each collocation information includes
Shape of face information and a kind of corresponding hair style information；A variety of collocation informations are inputted to the collocation Rating Model pre-established respectively, and
The scoring exported according to collocation Rating Model, which obtains, recommends hair style information.
Hair style provided by the invention recommends method, apparatus and equipment, by obtaining the shape of face information of user, is believed according to shape of face
Breath carries out a variety of hair style collocation, and scores each collocation, recommends hair style information so as to obtain, it is accurate to provide to the user
Hairstyle suggesting.
Description of the drawings
In order to illustrate more clearly about the embodiment of the present invention or technical scheme of the prior art, to embodiment or will show below
There is attached drawing needed in technology description to be briefly described, it should be apparent that, the accompanying drawings in the following description is this hair
Some bright embodiments, for those of ordinary skill in the art, without creative efforts, can be with root
Other attached drawings are obtained according to these attached drawings.
Fig. 1 is the flow diagram that hair style provided in an embodiment of the present invention recommends method；
Fig. 2 is the flow diagram that the hair style that another embodiment of the present invention provides recommends method；
Fig. 3 is the structure diagram of hair style recommendation apparatus provided in an embodiment of the present invention；
Fig. 4 is the structure diagram of hair style recommendation apparatus provided in an embodiment of the present invention.
Specific embodiment
Purpose, technical scheme and advantage to make the embodiment of the present invention are clearer, below in conjunction with the embodiment of the present invention
In attached drawing, the technical solution in the embodiment of the present invention is explicitly described, it is clear that described embodiment be the present invention
Part of the embodiment, instead of all the embodiments.Based on the embodiments of the present invention, those of ordinary skill in the art are not having
All other embodiments obtained under the premise of creative work are made, shall fall within the protection scope of the present invention.
Fig. 1 is the flow diagram that hair style provided in an embodiment of the present invention recommends method, as shown in Figure 1, including：Step
101, the face image data of user to be recommended is inputted to the shape of face identification model pre-established, obtains the shape of face of user to be recommended
Information；Step 102, a variety of collocation informations corresponding with the shape of face information are generated, wherein, each collocation information is believed including shape of face
Breath and a kind of corresponding hair style information；Step 103, a variety of collocation informations are inputted to the collocation Rating Model pre-established respectively,
And it is obtained according to the scoring that collocation Rating Model exports and recommends hair style information.
Wherein, shape of face refers to the profile of face；The first half of face is made of maxilla, cheekbone, temporal bone, frontal bone and parietal bone
Arc-shaped structure, lower half depend on mandibular form；There are many sorting technique of type, such as waveforms method is by the shape of face of the mankind
It is divided into ten kinds of shapes of face：Round shape of face, oval shape of face, oval shape of face, the shape of face of falling oval, rectangular shape of face, rectangle shape of face,
Trapezoidal shape of face, inverted trapezoidal shape of face, diamond shape shape of face and pentagon shape of face.Wherein, hair style refers to the length of hair, color and shape,
Naked eyes institute it is observed that, be generally divided into 5 types, i.e., straight hair, wave hair, curly hair, ulotrichy curly hair and minor spiral shape are sent out.This
Arbitrary shape of face sorting technique and hair style sorting technique may be used in inventive embodiments, it is not limited.
In a step 101, producing equipment by images such as camera or video cameras first and get needs progress hair style to push away
The face image data of the user recommended；The image data includes at least the face for needing recommended user；The acquisition of shape of face information can
To be extraction face image, then dimensional parameters are calculated, so as to obtain shape of face information；Specifically can mould be identified by shape of face
Type, shape of face identification model are obtained after neural network model is trained；The specific calculation that VGG Face Descriptor may be used
Face is identified in method model.Shape of face information can include multiple data of description face, such as the data such as face is long, face is wide.
In a step 102, according to the shape of face information obtained in step 101, a variety of collocation informations are generated；Each collocation information
All it is the collocation carried out by a kind of hair style information and the shape of face information of user to be recommended.
In step 103, it according to the collocation information obtained in step 102, scores each collocation information；Scoring tool
Body can be realized by collocation Rating Model；Collocation Rating Model is obtained based on neural metwork training, and the input value of model is one
Kind collocation information, output valve are scoring；After above-mentioned collocation information is inputted respectively, commenting for each collocation information can be got
Point；It can be using the hair style information to score corresponding to higher one or more collocation informations as recommendation hair style information.
Hair style provided in an embodiment of the present invention recommends method, by obtaining the shape of face information of user, according to shape of face information into
The a variety of hair style collocation of row, and score each collocation, recommend hair style information so as to obtain, accurate hair can be provided to the user
Type suggestion.
On the basis of any of the above-described embodiment, the face image data by user to be recommended inputs what is pre-established
It was further included before the step of shape of face identification model：The first training set is obtained, first training set includes multiple face image datas
And corresponding shape of face information；Using face image data and corresponding shape of face information as input data and output data
Model training is carried out, obtains the shape of face identification model that training is completed.
Specifically, it before shape of face identification is carried out using image identification model, needs first to create model, recycles the first training
The set pair analysis model is trained；In first training set, input value of multiple face image datas for training, and each face image data
Corresponding shape of face information is output valve or label；Finally obtain the shape of face identification model that training is completed.
It is described that a variety of collocation informations are inputted to the collocation scoring pre-established respectively on the basis of any of the above-described embodiment
It was further included before the step of model：Obtain the second training set, second training set includes a variety of collocation informations and corresponding
Scoring, a variety of collocation informations are by a variety of shape of face information and a variety of hair style information random alignment combination producings, each collocation letter
Breath includes a kind of shape of face information and a kind of hair style information, and the scoring is determined by expert estimation；By collocation information and corresponding
Scoring carries out model training respectively as input data and output data, obtains the collocation Rating Model that training is completed.
Specifically, it before scoring to collocation information using Rating Model of arranging in pairs or groups, needs to create model, and utilize second
Training the set pair analysis model is trained；Second training set specifically includes a variety of collocation informations and corresponding scoring；Wherein, collocation information
Being composed of a variety of shape of face information and a variety of hair style information random alignments (can also directly collect really taking in multiple reality
With image)；Scoring is then carried out that (professional person directly can score to collocation information or according to collocation information by professional person
Corresponding photo scores)；Collocation information is the input data of model training, and the output data to score as model training passes through
Collocation Rating Model can be obtained after training.
On the basis of any of the above-described embodiment, the scoring exported according to collocation Rating Model, which obtains, recommends hair style letter
It is further included after the step of breath：Three-dimensional facial model is established according to the face image data of user to be recommended；According to hair style recommendation
Breath establishes three-dimensional hair style model in the three-dimensional facial model.
Specifically, the practical effect of recommendation hair style information can be more intuitively shown to user by establishing threedimensional model
Fruit.
On the basis of any of the above-described embodiment, the scoring exported according to collocation Rating Model, which obtains, recommends hair style letter
It is further included after the step of breath：Obtain application scenarios information；According to the application scenarios information to the recommendation hair style information into commenting
Point；Correspondingly, it is described according to hair style recommendation information, it is also wrapped in the step of three-dimensional hair style model of foundation in the three-dimensional facial model
It includes, three-dimensional scene models is established at the rear portion of three-dimensional facial model and three-dimensional hair style model according to application scenarios information.Specifically,
In the occasion that can be combined with when hair style recommendation user to be recommended and will attend, i.e. application scenarios information, to recommending
Hair style information is further scored；The scoring, which can reflect, recommends hair style if appropriate for application scenarios.
Wherein, application scenarios information can include the field of relatively formal occasion (such as attending a meeting) and opposite leisure
It closes (such as attending wedding dinner or blind date), and can corresponding Score index be set with the characteristics of connected applications scene information.
Also, corresponding three-dimensional scene models are added in the background of threedimensional model, there is entire model more true
Effect；In the case where there are multiple recommendation hair styles, user is helped to make a choice.
On the basis of any of the above-described embodiment, the scoring exported according to collocation Rating Model, which obtains, recommends hair style letter
It is further included after the step of breath：Obtain the current hair style information of user to be recommended；To recommend hair style information and current hair style information into
Row compares, and generates corresponding hair style according to comparison result and arrange strategy.
Specifically, hair style arranges strategy and refers to, how by the current hair style of user, be organized into recommendation hair style, i.e., it is specific how
Arrange hair.Such as：Bang is cut, a little bob is cut by face, covers cheek side, the crown is flat, and both sides are fluffy.
In order to be illustrated to the method that the embodiments of the present invention provide, Fig. 2 is what another embodiment of the present invention provided
Hair style recommends the flow diagram of method, as shown in Fig. 2, including：
Step 0, model is created, and is trained using training set；In model training, appropriate fine tune are carried out
It is changed to desired classification (i.e. the classification of shape of face) later)；Shape of face identification model is obtained after training；
Step 1, in identification process, camera extraction portrait or photo are as input, using VGG
Face is identified in the algorithm model of FaceDescriptor；
Step 2, obtaining a variety of hair style samples, to carry out analysis classification similar with the method for step 1, and nerve net may be used
Network carries out hair style classification to a large amount of hair styles；
Step 3, the classification information obtained in step 1 and step 2 is combined training, val_data includes professional person
Good case in the excellent case and reality that are combined into from various classification, by the training for recommended models of giving a mark, output is several
Outstanding adaptation result；
Step 4, after obtaining model output, weight is carried out to the different collocation hair styles of different shapes of face according to professional person and is beaten
Point, founding mathematical models score to the output result of step 3, select the high several adaptation modes of score and export；
Wherein, marking mode is illustrated：
Long face：Face length/face is wide>1.5, forehead width=cheekbone width=chin width
Purpose:Shorten shape of face
Method:Bang is cut, a little bob is cut by face, covers cheek side, the crown is flat, and both sides are fluffy
Round face：Face length/face is wide close to 1, forehead width=cheekbone width
Purpose:Is elongated to be suitble to:4,6 points；Side point, avoids：In point
Square face：Face length/face is wide<1.5, forehead width=cheekbone is wide
Method:The long bang of comb on one side is cut, forehead is made to narrow, the crown is fluffy, is suitble to side point
Step 5, output result is re-converted into image data, then 3d virtually demonstration and offer design recommendation is provided,
It is intuitive to see result；
Step 6, meanwhile, the model be also in continuous renewal, by change or increase newly data carry out incremental training can
To export result according to the variation of trend more new system in time.
Fig. 3 is the structure diagram of hair style recommendation apparatus provided in an embodiment of the present invention, as shown in figure 3, including：Shape of face is known
Other module 301 for the shape of face identification model for pre-establishing the input of the face image data of user to be recommended, obtains to be recommended
The shape of face information of user；Collocation generation module 302, for generating a variety of collocation informations corresponding with the shape of face information, wherein,
Each collocation information includes shape of face information and a kind of corresponding hair style information；Hair style recommending module 303, for a variety of collocation to be believed
Breath inputs the collocation Rating Model pre-established respectively, and is obtained according to the scoring that collocation Rating Model exports and recommend hair style letter
Breath.
Wherein, being got first by shape of face identification module 301 needs the face image number for the user for carrying out hair style recommendation
According to；The image data includes at least the face for needing recommended user；The acquisition of shape of face information can be extraction face image, then right
Dimensional parameters are calculated, so as to obtain shape of face information；Shape of face identification module 301 can specifically pass through shape of face identification model, face
Type identification model is obtained after neural network model is trained；The specific algorithm mould that VGG Face Descriptor may be used
Face is identified in type.Shape of face information can include multiple data of description face, such as the data such as face is long, face is wide.
Wherein, collocation generation module 302 generates a variety of collocation according to the shape of face information obtained in shape of face identification module 301
Information；Each collocation information is all the collocation carried out by a kind of hair style information and the shape of face information of user to be recommended.
Wherein, hair style recommending module 303 believes each collocation according to the collocation information obtained in collocation generation module 302
Breath scores；Scoring can specifically be realized by collocation Rating Model；Collocation Rating Model is obtained based on neural metwork training,
The input value of model is a kind of collocation information, and output valve is scoring；After above-mentioned collocation information is inputted respectively, hair style recommending module
303 can get the scoring of each collocation information；Hair style recommending module 303 can be by higher one or more collocation of scoring
Hair style information corresponding to information is as recommendation hair style information.
Hair style recommendation apparatus provided in an embodiment of the present invention, by obtaining the shape of face information of user, according to shape of face information into
The a variety of hair style collocation of row, and score each collocation, recommend hair style information so as to obtain, accurate hair can be provided to the user
Type suggestion.
On the basis of any of the above-described embodiment, described device further includes：First training module, for obtaining the first training
Collection, first training set include multiple face image datas and corresponding shape of face information；By face image data and right
The shape of face information answered carries out model training respectively as input data and output data, obtains the shape of face identification mould that training is completed
Type.
On the basis of any of the above-described embodiment, described device further includes：Second training module, for obtaining the second training
Collection, second training set include a variety of collocation informations and corresponding scoring, and a variety of collocation informations are by a variety of shapes of face
Information and a variety of hair style information random alignment combination producings, each collocation information includes a kind of shape of face information and a kind of hair style is believed
Breath, the scoring are determined by expert estimation；Using collocation information and corresponding scoring as input data and output data
Model training is carried out, obtains the collocation Rating Model that training is completed.
On the basis of any of the above-described embodiment, described device further includes：Model building module, for according to use to be recommended
The face image data at family establishes three-dimensional facial model；According to hair style recommendation information, three are established in the three-dimensional facial model
Tie up hair style model.
On the basis of any of the above-described embodiment, described device further includes：Scene grading module, for obtaining application scenarios
Information；Hair style information is recommended into scoring to described according to the application scenarios information；The model building module is additionally operable to basis
Application scenarios information establishes three-dimensional scene models at the rear portion of three-dimensional facial model and three-dimensional hair style model.
On the basis of any of the above-described embodiment, described device further includes arrangement policy module, for obtaining use to be recommended
The current hair style information at family；Hair style information and current hair style information will be recommended to be compared, and generated and corresponded to according to comparison result
Hair style arrange strategy.
Fig. 4 is the structure diagram of hair style recommendation apparatus provided in an embodiment of the present invention, as shown in figure 4, the equipment includes：
At least one processor 401；And at least one processor 402 with the processor 401 communication connection, wherein：It is described to deposit
Reservoir 402 is stored with the program instruction that can be performed by the processor 401, and the processor 401 calls described program instruction energy
The hair style recommendation method that enough execution the various embodiments described above are provided, such as including：The face image data of user to be recommended is defeated
Enter the shape of face identification model pre-established, obtain the shape of face information of user to be recommended；Generation is corresponding more with the shape of face information
Kind collocation information, wherein, each collocation information includes shape of face information and a kind of corresponding hair style information；By a variety of collocation informations point
The collocation Rating Model pre-established is not inputted, and is obtained according to the scoring that collocation Rating Model exports and recommended hair style information.
The embodiment of the present invention also provides a kind of non-transient computer readable storage medium storing program for executing, the non-transient computer readable storage
Medium storing computer instructs, which makes computer perform the hair style recommendation method that corresponding embodiment is provided, example
Such as include：The shape of face identification model that the face image data input of user to be recommended is pre-established, obtains user's to be recommended
Shape of face information；Corresponding with the shape of face information a variety of collocation informations of generation, wherein, each collocation information including shape of face information and
A kind of corresponding hair style information；A variety of collocation informations are inputted to the collocation Rating Model pre-established respectively, and are commented according to collocation
The scoring of sub-model output, which obtains, recommends hair style information.
The embodiments such as hair style recommendation apparatus described above are only schematical, wherein illustrate as separating component
Unit may or may not be physically separate, and the component shown as unit may or may not be object
Manage unit, you can be located at a place or can also be distributed in multiple network element.It can select according to the actual needs
Some or all of module therein is selected to realize the purpose of this embodiment scheme.Those of ordinary skill in the art are not paying wound
In the case of the labour for the property made, you can to understand and implement.
Through the above description of the embodiments, those skilled in the art can be understood that each embodiment can
It is realized by the mode of software plus required general hardware platform, naturally it is also possible to pass through hardware.Based on such understanding, on
Technical solution is stated substantially in other words to embody the part that the prior art contributes in the form of software product, it should
Computer software product can store in a computer-readable storage medium, such as ROM/RAM, magnetic disc, CD, including several fingers
It enables and (can be personal computer, server or the network equipment etc.) so that computer equipment is used to perform each implementation
Certain Part Methods of example or embodiment.
Hair style provided in an embodiment of the present invention recommends method, apparatus and equipment, the field of deep learning has been widened, for not
For the people for understanding hair style and aesthetics, suitable opinion and scheme can be provided, the cost of inquiry professional person is reduced and encounters and know
Know the risk being likely encountered during practitioner not really up to the mark, and for the people of relevant industries, new person's culture can be largely reduced
Time cost increases trial and error means, and reciprocity communication also can be preferably carried out with customer.
Finally it should be noted that：The above embodiments are merely illustrative of the technical solutions of the present invention, rather than its limitations；Although
The present invention is described in detail with reference to the foregoing embodiments, it will be understood by those of ordinary skill in the art that：It still may be used
To modify to the technical solution recorded in foregoing embodiments or carry out equivalent replacement to which part technical characteristic；
And these modification or replace, various embodiments of the present invention technical solution that it does not separate the essence of the corresponding technical solution spirit and
Range.
Claims (10)
1. a kind of hair style recommends method, which is characterized in that including：
The shape of face identification model that the face image data input of user to be recommended is pre-established obtains the shape of face of user to be recommended
Information；
Generation a variety of collocation informations corresponding with the shape of face information, wherein, each collocation information includes shape of face information and correspondence
A kind of hair style information；
A variety of collocation informations are inputted to the collocation Rating Model pre-established, and the scoring exported according to collocation Rating Model respectively
It obtains and recommends hair style information.
2. according to the method described in claim 1, it is characterized in that, the face image data by user to be recommended inputs in advance
It was further included before the step of shape of face identification model first established：
The first training set is obtained, first training set includes multiple face image datas and corresponding shape of face information；
Face image data and corresponding shape of face information are subjected to model training as input data and output data, obtained
The shape of face identification model that training is completed.
3. according to the method described in claim 1, it is characterized in that, described input what is pre-established respectively by a variety of collocation informations
It was further included before the step of collocation Rating Model：
Obtain the second training set, second training set includes a variety of collocation informations and corresponding scoring, described a variety of to take
With information by a variety of shape of face information and a variety of hair style information random alignment combination producings, each collocation information is believed including a kind of shape of face
Breath and a kind of hair style information, the scoring are determined by expert estimation；
Collocation information and corresponding scoring are subjected to model training as input data and output data, training is obtained and completes
Collocation Rating Model.
4. according to the method described in claim 1, it is characterized in that, the scoring acquisition exported according to collocation Rating Model pushes away
It is further included after the step of recommending hair style information：
Three-dimensional facial model is established according to the face image data of user to be recommended；
According to hair style recommendation information, three-dimensional hair style model is established in the three-dimensional facial model.
5. according to the method described in claim 4, it is characterized in that, the scoring acquisition exported according to collocation Rating Model pushes away
It is further included after the step of recommending hair style information：
Obtain application scenarios information；
Hair style information is recommended into scoring to described according to the application scenarios information；
Correspondingly, it is described according to hair style recommendation information, it is gone back in the step of three-dimensional hair style model of foundation in the three-dimensional facial model
Including,
Three-dimensional scene models are established at the rear portion of three-dimensional facial model and three-dimensional hair style model according to application scenarios information.
6. according to the method described in claim 1, it is characterized in that, the scoring acquisition exported according to collocation Rating Model pushes away
It is further included after the step of recommending hair style information：
Obtain the current hair style information of user to be recommended；
Hair style information and current hair style information will be recommended to be compared, and corresponding hair style is generated according to comparison result and arranges plan
Slightly.
7. a kind of hair style recommendation apparatus, which is characterized in that including：
Shape of face identification module for the shape of face identification model for pre-establishing the input of the face image data of user to be recommended, obtains
Take the shape of face information of user to be recommended；
Collocation generation module, for generating a variety of collocation informations corresponding with the shape of face information, wherein, each collocation information packet
Include shape of face information and a kind of corresponding hair style information；
Hair style recommending module, for a variety of collocation informations to be inputted to the collocation Rating Model pre-established respectively, and according to collocation
The scoring of Rating Model output, which obtains, recommends hair style information.
8. device according to claim 7, which is characterized in that further include：
Model building module, for establishing three-dimensional facial model according to the face image data of user to be recommended；It is pushed away according to hair style
Information is recommended, three-dimensional hair style model is established in the three-dimensional facial model.
9. a kind of hair style recommendation apparatus, which is characterized in that including：
At least one processor；
And at least one processor being connect with the processor communication, wherein：The memory is stored with can be by the place
The program instruction that device performs is managed, the processor calls described program instruction to be able to carry out as described in claim 1 to 6 is any
Method.
10. a kind of non-transient computer readable storage medium storing program for executing, which is characterized in that the non-transient computer readable storage medium storing program for executing is deposited
Computer instruction is stored up, the computer instruction makes the computer perform the method as described in claim 1 to 6 is any.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201711481867.3A CN108255998A (en) | 2017-12-29 | 2017-12-29 | A kind of hair style recommends method, apparatus and equipment |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201711481867.3A CN108255998A (en) | 2017-12-29 | 2017-12-29 | A kind of hair style recommends method, apparatus and equipment |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN108255998A true CN108255998A (en) | 2018-07-06 |
Family
ID=62725476
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201711481867.3A Pending CN108255998A (en) | 2017-12-29 | 2017-12-29 | A kind of hair style recommends method, apparatus and equipment |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108255998A (en) |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108984772A (en) * | 2018-07-23 | 2018-12-11 | 京东方科技集团股份有限公司 | A kind of information recommending apparatus and method |
| CN109300013A (en) * | 2018-10-25 | 2019-02-01 | 百度在线网络技术（北京）有限公司 | Method and apparatus for generating information |
| CN109493160A (en) * | 2018-09-29 | 2019-03-19 | 口碑(上海)信息技术有限公司 | A kind of virtual examination forwarding method, apparatus and system |
| CN110349202A (en) * | 2019-07-05 | 2019-10-18 | 合肥聚美网络科技有限公司 | A kind of hair styling method based on shape of face analysis |
| CN110598097A (en) * | 2019-08-30 | 2019-12-20 | 中国科学院自动化研究所南京人工智能芯片创新研究院 | Hair style recommendation system, method, equipment and storage medium based on CNN |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104391971A (en) * | 2014-12-05 | 2015-03-04 | 常州飞寻视讯信息科技有限公司 | Intelligent automatic dress collocation recommending method |
| CN104794275A (en) * | 2015-04-16 | 2015-07-22 | 北京联合大学 | Face and hair style matching model for mobile terminal |
| GB2523032A (en) * | 2011-02-17 | 2015-08-12 | Metail Ltd | Computer implemented methods and systems for generating virtual body models for garment fit visualisation |
| CN105045968A (en) * | 2015-06-30 | 2015-11-11 | 青岛理工大学 | Hairstyle design method and system |
| CN105956150A (en) * | 2016-05-12 | 2016-09-21 | 张家港索奥通信科技有限公司 | Method and apparatus for generating hair style and makeup matching suggestions of a user |
-
2017
- 2017-12-29 CN CN201711481867.3A patent/CN108255998A/en active Pending
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| GB2523032A (en) * | 2011-02-17 | 2015-08-12 | Metail Ltd | Computer implemented methods and systems for generating virtual body models for garment fit visualisation |
| CN104391971A (en) * | 2014-12-05 | 2015-03-04 | 常州飞寻视讯信息科技有限公司 | Intelligent automatic dress collocation recommending method |
| CN104794275A (en) * | 2015-04-16 | 2015-07-22 | 北京联合大学 | Face and hair style matching model for mobile terminal |
| CN105045968A (en) * | 2015-06-30 | 2015-11-11 | 青岛理工大学 | Hairstyle design method and system |
| CN105956150A (en) * | 2016-05-12 | 2016-09-21 | 张家港索奥通信科技有限公司 | Method and apparatus for generating hair style and makeup matching suggestions of a user |
Cited By (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108984772A (en) * | 2018-07-23 | 2018-12-11 | 京东方科技集团股份有限公司 | A kind of information recommending apparatus and method |
| CN109493160A (en) * | 2018-09-29 | 2019-03-19 | 口碑(上海)信息技术有限公司 | A kind of virtual examination forwarding method, apparatus and system |
| CN109300013A (en) * | 2018-10-25 | 2019-02-01 | 百度在线网络技术（北京）有限公司 | Method and apparatus for generating information |
| CN110349202A (en) * | 2019-07-05 | 2019-10-18 | 合肥聚美网络科技有限公司 | A kind of hair styling method based on shape of face analysis |
| CN110349202B (en) * | 2019-07-05 | 2021-07-20 | 合肥聚美网络科技有限公司 | Hairstyle design method based on face analysis |
| CN110598097A (en) * | 2019-08-30 | 2019-12-20 | 中国科学院自动化研究所南京人工智能芯片创新研究院 | Hair style recommendation system, method, equipment and storage medium based on CNN |
| CN110598097B (en) * | 2019-08-30 | 2022-07-29 | 中科南京人工智能创新研究院 | Hair style recommendation system, method, equipment and storage medium based on CNN |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108255998A (en) | A kind of hair style recommends method, apparatus and equipment | |
| CN106295476B (en) | Face key point localization method and device | |
| Brennan | Caricature generator | |
| CN106919738A (en) | A kind of hair style matching process | |
| EP1839525A1 (en) | A mannequin head | |
| CN106682958A (en) | Method and device for trying on makeup virtually | |
| CN108694266A (en) | intelligent decoration design method and system based on machine learning | |
| CN108664569A (en) | A kind of hair style recommends method, system, terminal and medium | |
| CN108197574A (en) | The recognition methods of personage's style, terminal and computer readable storage medium | |
| CN109310196A (en) | Makeup auxiliary device and cosmetic auxiliary method | |
| CN110755847B (en) | Virtual operation object generation method and device, storage medium and electronic device | |
| CN104331564B (en) | Adorn guidance method and terminal device based on terminal device | |
| CN110175505A (en) | Determination method, apparatus, storage medium and the electronic device of micro- expression type | |
| CN110136054A (en) | Image processing method and device | |
| CN107944093A (en) | A kind of lipstick color matching system of selection, electronic equipment and storage medium | |
| CN109409319A (en) | A kind of pet image beautification method, device and its storage medium | |
| CN107784630A (en) | The method, apparatus and terminal of attribute upset are carried out to facial image | |
| CN116994718B (en) | VR technology-based mental disorder auxiliary treatment method | |
| CN106339658A (en) | Data processing method and device | |
| TWI462030B (en) | An intelligent system and algorithms of facial analysis and hairstyle fitting | |
| CN1979556A (en) | Hair-style virtual design method and apparatus generated by computer software | |
| CN105187736B (en) | A kind of method, system and mobile terminal that static face picture is converted into video | |
| CN109753873A (en) | Image processing method and relevant apparatus | |
| CN108228950A (en) | A kind of information processing method and device | |
| JP6212226B1 (en) | Wig manufacturing method, wig sales method and program |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20180706 |
|
| RJ01 | Rejection of invention patent application after publication |