CN108256975A - Wearing for 3-D effect is provided for virtual fitting person take system and method based on artificial intelligence - Google Patents
Wearing for 3-D effect is provided for virtual fitting person take system and method based on artificial intelligence Download PDFInfo
- Publication number
- CN108256975A CN108256975A CN201810064913.8A CN201810064913A CN108256975A CN 108256975 A CN108256975 A CN 108256975A CN 201810064913 A CN201810064913 A CN 201810064913A CN 108256975 A CN108256975 A CN 108256975A
- Authority
- CN
- China
- Prior art keywords
- user
- clothes
- artificial intelligence
- module
- collocation
- 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
-
- 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
- G06Q30/06—Buying, selling or leasing transactions
- G06Q30/0601—Electronic shopping [e-shopping]
- G06Q30/0641—Shopping interfaces
- G06Q30/0643—Graphical representation of items or shoppers
-
- 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
- G06Q30/06—Buying, selling or leasing transactions
- G06Q30/0601—Electronic shopping [e-shopping]
- G06Q30/0631—Item recommendations
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T17/00—Three dimensional [3D] modelling, e.g. data description of 3D objects
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T19/00—Manipulating 3D models or images for computer graphics
- G06T19/006—Mixed reality
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T19/00—Manipulating 3D models or images for computer graphics
- G06T19/20—Editing of 3D images, e.g. changing shapes or colours, aligning objects or positioning parts
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2219/00—Indexing scheme for manipulating 3D models or images for computer graphics
- G06T2219/20—Indexing scheme for editing of 3D models
- G06T2219/2024—Style variation
Abstract
The invention discloses the system of taking is worn for what virtual fitting person provided 3-D effect based on artificial intelligence, including user management module, 3D modeling module, artificial intelligence module, database, expert system module, module of arranging in pairs or groups and 3D fitting modules, user management module are used for managing user information；The 3D modeling module is scanned by mobile phone to be obtained generation and meets the 3D models of type data information, and the database, which includes forming a series of standard garments for meeting various user demands wearing, takes template；The expert system module, which includes to meet Expert Rules information specialist's clothes and wear, takes template；The standard garments that the collocation module recommendation meets user preference, which are worn, takes template and expert's clothes are worn and take template for user's selection；The 3D fittings module forms virtual dynamic 3D clothes effects.Simplicity of the present invention improves, and by three-dimensional fitting true to nature, recommends Al or independently the displaying of matching of DIY has more glamour.It solves the hard nut to crack that can also be fitted on line, can accurately recommend specification.
Description
Technical field
The invention belongs to intelligent collocation fields, and in particular to one kind provides three-dimensional effect based on artificial intelligence for virtual fitting person
System and method is taken in wearing for fruit.
Background technology
When people want to go to net purchase dress ornament on line, open electric business web page and magnanimity dress ornament is displayed, many people can encounter two
A problem：First, finding, the dress ornament to be bought is too time-consuming, and efficiency is too low.Second is that not knowing which dress ornament is suitble to oneself, oneself should
Which is bought.Traditional shopping guide and customer service consulting can not meet the demand that young women upgrading is brought and improved.Then, some take
Decorations retailers are particularly electric business and start to consider that manually intelligent method helps people to solve collocation of wearing the clothes, and save shopping and select
Worry in time and reduction shopping process.Artificial intelligence application wears field to fashion and also just starts to walk.Positive place at this stage
In the stage of fumbling.
Artificial intelligence by data, information it is continuous absorb, accumulation, operation and digestion, constantly by self-teaching,
Function can become stronger day by day.It has two in the application major way that fashion dress ornament wears field：First, it is filled in by client by businessman
The questionnaire of transmission.Content covers stature situation (height, weight, measurements of the chest, waist and hips etc.) purchasing habits (dress ornament liked of user
The information such as color matching, style, collocation, brand, specification) form data, input system, the phase uploaded with proposed algorithm and client
Piece, the clothes liked with the photograph exhibition method of planar come recommended user.Second approach is with above-mentioned first in principle
It is upper roughly the same, it is a difference in that the means for collecting data are not all of filling in questionnaires acquirement by user, but electric business net purchase
Platform has been reached to a certain degree by the purchaser record of young women person user, produces data.(including your navigation patterns and
Buying behavior).Specific steps are substantially：
1) personal information questionnaire is filled in by client or purchaser record generates data.
2) data and personal photograph are obtained.
3) input system.
4) clothes liked according to proposed algorithm come recommended user.
The prior art there are the defects of it is as follows：
1. single use proposed algorithm pattern cannot really " understand " user.Only more user may like
Before clothes are pushed to user plane.
2. the buying habit of not all client is all correct.And it is produced based on the data that this custom of user is collected
Raw dress collocation is recommended be exactly correct.
3. the dress ornament shown the dress of most critical is all the effect of two dimensional surface photo type, it can neither reflect worn dress ornament
The whether fit , You No of specification have it is vivid substitute into sense fitting experience.
4. the style of recommendation hitherto is also considerably less, recommended levels are also very low.
5. its purpose of current pattern is from the angle of purchase rather than from the angle how user should wear on earth in itself
Degree, before the clothes that user may like more only are pushed to user plane, still has many consumers which type of is not known
It wears to take and is most suitable for oneself.
6. because the technical support of each system is inadequate, without forming an effective operation closed loop between each system.User
Feel also inconvenient, shopping experience is bad.
Invention content
For overcome the deficiencies in the prior art, three are provided for virtual fitting person the present invention provides one kind based on artificial intelligence
Wearing for effect of dimension takes system and method.
The technical proposal of the invention is realized in this way：Based on artificial intelligence wearing for 3-D effect is provided for virtual fitting person
System is taken, including user management module, 3D modeling module, database, artificial intelligence module, expert system module, module of arranging in pairs or groups,
3D fitting modules and purchase module, wherein：
The user management module is used for managing users registration, personal data and consumption information；
The 3D modeling module obtains the two-dimentional mug shot and skin of target user by mobile phone shooting and mobile phone scanning
Color, build, height, human body measurements of the chest, waist and hips, shoulder breadth, brachium, shape of face and hair style build data information generate the build data information
The 3D models of target user's build, will obtain mug shot and the storage of 3D models simultaneously passes 3D models to user management module
It is defeated by database, artificial intelligence module and expert system module；
The database：Including magnanimity 3D modeling, the collocation module, which is worn by collecting human body, takes image data, 3D human bodies
Sample data is taken with 3D dress ornaments data and actually wearing and is added to the artificial intelligence module, and artificial intelligence module is constantly using deep
Degree neural network is worn with reference to human body takes image data by 3D clothing matchings to 3D human bodies, deep neural network model is carried out anti-
Refreshment is practiced, and continues to optimize and forms a series of standard garments for meeting various user demands and wear and takes template；
The artificial intelligence module is initially used for the sample data arranged in pairs or groups using human body to train deep neural network, made
Neural network is into deep learning, while for whether test clothing matching reasonable；Secondly, the artificial intelligence module is based on user
Purchaser record, the specification of user is met using Collaborative Filtering Recommendation Algorithm recommendation and the clothes liked of user are worn and take mould
Plate；
The expert system module collects and merges anthroposomatology, aesthetics, ceremony and dress and weather and wears knowledge, and
The dressing contact between these knowledge is established, formation, which objectively meets aesthetic Expert Rules information and generates, meets Expert Rules letter
Breath expert's clothes, which are worn, takes template；
The collocation module includes artificial intelligence collocation unit and DIY collocation units, and target user selects phase according to demand
The collocation unit answered, if target user selects artificial intelligence collocation unit and for first purchase, the artificial intelligence collocation unit
The specification of recommended user is given birth to according to the 3D models of target user first, then recommendation meets the mesh from expert system module
Expert's clothes of mark user specification size, which are worn, takes template for user's selection；If target user selects artificial intelligence collocation unit and not
It is first purchase, then artificial intelligence collocation unit is pushed away based on the purchaser record of target user by Collaborative Filtering Recommendation Algorithm calculating
It recommends and wear the clothes collocation and and mesh of target user's phase purchaser record similarity higher than other similar users of setting first threshold
The similar clothes that user preference similarity is marked higher than setting second threshold are selected for user；If user selects DIY collocation units,
DIY collocation units allow user to select the dress ornament style of oneself preference and not preference first with selection function, then exclude automatically
User not recommend to meet user preference from the database and expert system module using preferentially simultaneously by the dress ornament style of preference
Standard garments wear and take template and expert's clothes are worn and take template and selected for user；
The 3D fittings module is used to the artificial intelligence module and artificial intelligence collocation unit recommending user's
Specification and clothes, which are worn, to be taken template and tries on 3D models, forms virtual dynamic 3D clothes effects；
The purchase module is connect with the user management module, for carrying out purchase operation for user.
The clothes that the artificial intelligence collocation unit is bought according to target user, establish clothes --- user's matrix normalizing
Change the table of falling Check, the similarity of the purchaser record with target user of other users is calculated using vectorial COS distance.
The value of the first threshold is not less than 70%.
The clothes that the artificial intelligence collocation unit was bought in the past according to target user, establish clothes --- clothes matrix
Normalize the table of falling Check, calculated using vectorial COS distance the purchaser record to target user of other clothes clothes it is similar
Degree.
The value of the second threshold is not less than 70%.
Selection function in the DIY collocation unit is realized using navigation and human-computer dialogue.
The dress ornament that the clothes are worn in taking includes：Clothes, footwear, cap, scarf, tippet, silk scarf, necktie, belt, waistband, hand
Set, scarf, handbag, hosiery, headwear, decorative glasses, decorative jewelry and hair style.
Virtual fitting person based on artificial intelligence wears the clothes matching method, and this method is applied is based on artificial intelligence above-mentioned
Virtual fitting person offer 3-D effect is worn in the system of taking, and this method includes the following steps：
S1. user's registration step：User is registered by the user management module；
S2.3D modeling procedures：After register account number, user is logged in by account number cipher, then utilizes the 3D modeling
Module by mobile phone shooting and mobile phone scanning obtain user two-dimentional mug shot and the colour of skin, build, height, human body measurements of the chest, waist and hips,
Build data information generation is met the 3D moulds of type data information by shoulder breadth, brachium, shape of face and hair style build data information
Type, will obtain mug shot and 3D models are transferred to database, artificial intelligence by the storage of 3D models simultaneously to user management module
It can module and expert system module；
S3. wear the clothes collocation step：If target user selects artificial intelligence collocation unit and for first purchase, the artificial intelligence
The unit that can arrange in pairs or groups gives birth to the specification of recommended user according to the 3D models of target user first, then recommends from expert system module
Meet target user's specification expert's clothes wear take template for user select；
If target user selects artificial intelligence collocation unit and is not first purchase, artificial intelligence collocation unit is based on mesh
The purchaser record of mark user is calculated to recommend to be higher than with target user's phase purchaser record similarity by Collaborative Filtering Recommendation Algorithm and be set
Determine wearing the clothes for other similar users of first threshold to arrange in pairs or groups and with target user's preference similarity higher than setting second threshold
Similar clothes are selected for user；
If user selects DIY collocation units, DIY collocation units allow user to select oneself preference first with selection function
With the dress ornament style of not preference, then exclude automatically user not preference dress ornament style simultaneously using preferentially from the database and
The standard garments for meeting user preference is recommended to wear in expert system module and take template and expert's clothes are worn and take template for user's choosing
It selects；
S4. step is selected：User wears the template point for taking stencil-chosen satisfaction according to the clothes recommended in collocation step of wearing the clothes
It hits and tries on.
S5.3D fitting steps：User clicks try on after, 3D fitting module be used for the specification for selecting user with
And clothes are worn and take template and try on 3D models, form virtual dynamic 3D clothes effects；
S6. purchase or exit step：For user according to after 3D fits step, user selects satisfaction according to clothes effect
Clothes wear the dress ornament taken in template and are bought and paid or exited by the user management module.
Beneficial effects of the present invention：Existing system is usually only with single proposed algorithm.Based on user and clothes
Collaborative Filtering Recommendation Algorithm, the data that it is used are that the passing purchaser record of user or questionnaire are tabled look-up.And this system is in addition to pushing away
It recommends outside algorithm, also uses Al (artificial intelligence) recommendations, with a large amount of human bodies collocation sample data training deep neural network, make clothes
The recommendation collocation of dress is more reasonable.This system has also merged expert system.The collocation of recommendation is made to have more specialist.This is in the field
Also belong to and using for the first time.
2) simplicity improves.
Present system need not bother user to fill in a large amount of personal information and dress ornament purchase hobby.Only need to operate navigation and
Good in interactive function.With regard to purchase efficiency can be increased substantially.
3) in addition to Al (artificial intelligence) recommends collocation, also with the autonomous matching functions of DIY.User to think oneself to match carries
Platform is supplied.
4) virtual fitting of stereoscopic three-dimensional effect is a big characteristic of this system.By three-dimensional fitting true to nature, push away Al
It recommends or the displaying of matching of autonomous DIY has more glamour.Solves the hard nut to crack that can also be fitted on line
5) specification can accurately be recommended.
Description of the drawings
Fig. 1 is the structure diagram for wearing the system of taking for providing 3-D effect for virtual fitting person based on artificial intelligence.
Fig. 2 is that the virtual fitting person based on artificial intelligence wears the clothes the step flow chart of matching method.
Specific embodiment
The specific embodiment of the present invention is described further below：
As shown in Figure 1, the system of taking is worn for what virtual fitting person provided 3-D effect based on artificial intelligence, including user management
Module, 3D modeling module, database, artificial intelligence module, expert system module, module of arranging in pairs or groups, 3D fitting modules and purchase mould
Block, wherein：
The user management module is used for managing users registration, personal data and consumption information；
The 3D modeling module obtains the two-dimentional mug shot and skin of target user by mobile phone shooting and mobile phone scanning
Color, build, height, human body measurements of the chest, waist and hips, shoulder breadth, brachium, shape of face and hair style build data information, and the build data information is given birth to
Into the 3D models of target user's build, two-dimentional mug shot and the storage of 3D models will be obtained to user management module simultaneously by 3D
Model is transferred to database, artificial intelligence module and expert system module；The 3D models of formation include specification, model, match
The information such as color, classification, there is 3D human body model datas, and system can accurately recommend specification.Net purchase dress ornament on line
The dress ornament that consumer no longer worries to buy misfits.
The database includes the 3D modeling of magnanimity, and the collocation module is worn by collecting human body and takes image data, 3D people
It body and 3D dress ornaments data and actually wears and takes sample data and be added to the artificial intelligence module, pass through sample number of arranging in pairs or groups to human body
According to training deep neural network, repetition training is carried out to deep neural network model, continue to optimize and forms a series of meet respectively
The standard garments of kind user demand, which are worn, takes template；In order to reduce between recommended clothes and actual user's trial assembly in database
Difference, with largely 3 D human body garment data sample training deep neural network in practice.With wear take sample data constantly train,
Continuous tuning neural network model, enables deep neural network to extract clothes and wears the feature taken, finally make based on deep learning
The clothes that artificial intelligence is recommended are the very satisfied clothes of user.It collects a large amount of human body and wears and take image data collection, for training
Deep neural network so that the human body of deep neural network output, which is worn to take, meets public mainstream aesthetic conceptions.
Wherein, deep neural network training needs a kind of deep learning frame such as Tensor Flow (deep learnings of mainstream
Tool) and deep learning algorithm such as DNN etc. (deep neural network algorithm) is constantly abstract and feature extraction garment coordination picture,
And then it can judge whether recommended clothes meet mainstream aesthetic conceptions.
The artificial intelligence module is initially used for the sample data arranged in pairs or groups using human body to train deep neural network, made
Neural network is into deep learning, while for whether test clothing matching reasonable；Secondly, the artificial intelligence module is based on user
Purchaser record, the specification of user is met using Collaborative Filtering Recommendation Algorithm recommendation and the clothes liked of user are worn and take mould
Plate；
The expert system module collects and merges anthroposomatology, aesthetics, ceremony and dress and weather and wears knowledge, and
The dressing contact between these knowledge is established, formation, which objectively meets aesthetic Expert Rules information and generates, meets Expert Rules letter
Breath expert's clothes are worn and take template, it cans be compared to be the experience for having gathered plurality of professional clothing matching teacher, wisdom and knowledge in
All over the body.It can recommend clothing matching with more professional aesthetic conceptions to user faster.
The collocation module includes artificial intelligence collocation unit and DIY collocation units, and target user selects phase according to demand
The collocation unit answered, if target user selects artificial intelligence collocation unit and for first purchase, the artificial intelligence collocation unit
The specification of recommended user is given birth to according to the 3D models of target user first, then recommendation meets the mesh from expert system module
Expert's clothes of mark user specification size, which are worn, takes template for user's selection；If target user selects artificial intelligence collocation unit and not
It is first purchase, then artificial intelligence collocation unit is pushed away based on the purchaser record of target user by Collaborative Filtering Recommendation Algorithm calculating
It recommends and wear the clothes collocation and and mesh of target user's phase purchaser record similarity higher than other similar users of setting first threshold
The similar clothes that user preference similarity is marked higher than setting second threshold are selected for user；If user selects DIY collocation units,
DIY collocation units allow user to select the dress ornament style of oneself preference and not preference first with selection function, then exclude automatically
User not recommend to meet user preference from the database and expert system module using preferentially simultaneously by the dress ornament style of preference
Standard garments wear and take template and expert's clothes are worn and take template and selected for user；
The 3D fittings module is used to the artificial intelligence module and artificial intelligence collocation unit recommending user's
Specification and clothes, which are worn, to be taken template and tries on 3D models, forms virtual dynamic 3D clothes effects；
Specifically, the clothes that the artificial intelligence collocation unit is bought according to target, establish clothes --- user's matrix
Normalize the table of falling Check (a kind of technical term of statistical table), using vectorial COS distance calculating other users and target user
Purchaser record similarity.The clothes that the artificial intelligence collocation unit is bought according to target, establish clothes --- clothes
Matrix normalization tune Check tables calculate the phase of the clothes of the purchaser record with target user of other clothes using vectorial COS distance
Like degree.The first threshold and the value of second threshold are not less than 70%.The first and second threshold values takes described in the present embodiment
Value is respectively 80% and 85%.
Selection function in the DIY collocation unit realizes that navigation can allow user fast using navigation and human-computer dialogue
Speed enters directly into the dress ornament region for oneself wanting to buy, and artificial intelligence system is avoided to recommend this excessive oneself unnecessary clothes
Decorations save the time buying for user；Another interactive function is that user can input the thing liked or do not liked, and is allowed
The hobby of system definitely user.This two functions can make shopping have more specific aim, can improve purchase efficiency.
3D fitting module is used for the specification for selecting user and clothes are worn and take template and try 3D models on,
Form virtual dynamic 3D clothes effects；The collocation recommended by artificial intelligence or the collocation independently selected by DIY can be with
Carry out the virtual fitting of 3-D effect true to nature.(because human body and dress ornament all have been subjected to 3D modeling) which solves puzzlement lines
Upper dress ornament net purchase for many years buy back a large amount of goods not to the worries of plate dress ornament because cannot try on due to.
The purchase module is connect with the user management module, for carrying out purchase operation for user.
The dress ornament that heretofore described clothes are worn in taking includes：Clothes, footwear, cap, scarf, tippet, silk scarf, necktie, belt,
Waistband, gloves, scarf, handbag, hosiery, headwear, decorative glasses, decorative jewelry and hair style.
As shown in Fig. 2, the virtual fitting person based on artificial intelligence wears the clothes matching method this method apply it is above-mentioned based on
Artificial intelligence provides the wearing in the system of taking of 3-D effect for virtual fitting person, and this method includes the following steps：
S1. user's registration step：User is registered by the user management module；
S2.3D modeling procedures：After register account number, user is logged in by account number cipher, then utilizes the 3D modeling
Module by mobile phone shooting and mobile phone scanning obtain user two-dimentional mug shot and the colour of skin, build, height, human body measurements of the chest, waist and hips,
Build data information generation is met the 3D moulds of type data information by shoulder breadth, brachium, shape of face and hair style build data information
Type, will obtain mug shot and 3D models are transferred to database, artificial intelligence by the storage of 3D models simultaneously to user management module
It can module and expert system module；
S3. wear the clothes collocation step：If target user selects artificial intelligence collocation unit and for first purchase, the artificial intelligence
The unit that can arrange in pairs or groups gives birth to the specification of recommended user according to the 3D models of target user first, then recommends from expert system module
Meet target user's specification expert's clothes wear take template for user select；
If target user selects artificial intelligence collocation unit and is not first purchase, artificial intelligence collocation unit is based on mesh
The purchaser record of mark user is calculated to recommend to be higher than with target user's phase purchaser record similarity by Collaborative Filtering Recommendation Algorithm and be set
Determine wearing the clothes for other similar users of first threshold to arrange in pairs or groups and with target user's preference similarity higher than setting second threshold
Similar clothes are selected for user；
If user selects DIY collocation units, DIY collocation units allow user to select oneself preference first with selection function
With the dress ornament style of not preference, then exclude automatically user not preference dress ornament style simultaneously using preferentially from the database and
The standard garments for meeting user preference is recommended to wear in expert system module and take template and expert's clothes are worn and take template for user's choosing
It selects；
S4. step is selected：User wears the template point for taking stencil-chosen satisfaction according to the clothes recommended in collocation step of wearing the clothes
It hits and tries on.
S5.3D fitting steps：User clicks try on after, 3D fitting module be used for the specification for selecting user with
And clothes are worn and take template and try on 3D models, form virtual dynamic 3D clothes effects；
S6. purchase or exit step：For user according to after 3D fits step, user selects satisfaction according to clothes effect
Clothes wear the dress ornament taken in template and are bought and paid or exited by the user management module.
The present invention allows the effect of virtual fitting three-dimensional to show that artificial intelligence pushes away by human body and dress ornament 3D modeling
Wearing for recommending is taken.Its effect be existing market planar photograph substitution method institute it is incomparable.It can solve to take on line
The two big puzzlements of decorations net consumer.When the fitting of the sense of reality cannot be obtained and often buy wrong clothing.Second is that dress ornament specification on line
Confusion, consumer often buy the dress ornament of off size body.
Claimed range：
To promote personal image, manually intelligent method provides wearing for dress ornament for the virtual fitting person under 3D modeling state
The guide service for matching aspect.
Dress ornament match including：Clothes, footwear, boots, cap, scarf, silk scarf, necktie, handbag, headwear and hair style and 3D people
The collocation of body.
Existing system is usually only with single proposed algorithm.Collaborative Filtering Recommendation Algorithm based on user and clothes,
The data that it is used are that the passing purchaser record of user or questionnaire are tabled look-up.And this system also uses other than proposed algorithm
Al (artificial intelligence) recommends, and with a large amount of human bodies collocation sample data training deep neural network, the recommendation of clothes is made to arrange in pairs or groups and is more closed
Reason.This system has also merged expert system.The collocation of recommendation is made to have more specialist.This also belongs in the field uses for the first time.
2) simplicity improves.
Present system need not bother user to fill in a large amount of personal information and dress ornament purchase hobby.Only need to operate navigation and
Good in interactive function.With regard to purchase efficiency can be increased substantially.
3) in addition to Al (artificial intelligence) recommends collocation, also with the autonomous matching functions of DIY.User to think oneself to match carries
Platform is supplied.
4) virtual fitting of stereoscopic three-dimensional effect is a big characteristic of this system.By three-dimensional fitting true to nature, push away Al
It recommends or the displaying of matching of autonomous DIY has more glamour.Solves the hard nut to crack that can also be fitted on line
5) specification can accurately be recommended.
According to the disclosure and teachings of the above specification, those skilled in the art in the invention can also be to above-mentioned embodiment party
Formula is changed and is changed.Therefore, the invention is not limited in specific embodiment disclosed and described above, to the one of invention
A little modifications and changes should also be as falling into the scope of the claims of the present invention.In addition, it although is used in this specification
Some specific terms, but these terms are merely for convenience of description, do not limit the present invention in any way.
Claims (8)
1. the system of taking is worn for what virtual fitting person provided 3-D effect based on artificial intelligence, which is characterized in that including user management
Module, 3D modeling module, database, artificial intelligence module, expert system module, module of arranging in pairs or groups, 3D fitting modules and purchase mould
Block, wherein：
The user management module is used for managing users registration, personal data and consumption information；
The 3D modeling module obtains the two-dimentional mug shot of target user and the colour of skin, body by mobile phone shooting and mobile phone scanning
The build data information is generated target by type, height, human body measurements of the chest, waist and hips, shoulder breadth, brachium, shape of face and hair style build data information
3D models are transmitted the mug shot of acquisition and the storage of 3D models to user management module by the 3D models of user's build simultaneously
To database, artificial intelligence module and expert system module；
The database：The collocation module, which is worn by collecting human body, takes image data, 3D human bodies and 3D dress ornaments data and reality
Border, which is worn, takes sample data and is added to the artificial intelligence module, and artificial intelligence module is constantly using deep neural network with reference to human body
It wears and takes image data by 3D clothing matchings to 3D human bodies, repetition training is carried out to deep neural network model, is continued to optimize simultaneously
It forms a series of standard garments for meeting various user demands and wears and take template；
The artificial intelligence module is initially used for the sample data arranged in pairs or groups using human body to train deep neural network, makes nerve
Network is into deep learning, while for whether test clothing matching reasonable；Secondly, purchase of the artificial intelligence module based on user
Record is bought, the specification of user is met using Collaborative Filtering Recommendation Algorithm recommendation and the clothes liked of user are worn and take template；
The expert system module collects and merges anthroposomatology, aesthetics, ceremony and dress and weather and wears knowledge, and establish
Dressing contact between these knowledge, formation objectively meet aesthetic Expert Rules information and generate that meet Expert Rules information special
Family's clothes, which are worn, takes template；
The collocation module includes artificial intelligence collocation unit and DIY collocation units, and target user selects accordingly according to demand
It arranges in pairs or groups unit, if target user select artificial intelligence collocation unit and for first purchase, the artificial intelligence arranges in pairs or groups unit first
The specification of recommended user is given birth to according to the 3D models of target user, then recommendation meets the target use from expert system module
Expert's clothes of family specification, which are worn, takes template for user's selection；If target user selects artificial intelligence collocation unit and is not first
Secondary purchase, then artificial intelligence collocation unit based on the purchaser record of target user by Collaborative Filtering Recommendation Algorithm calculate recommend with
Target user's phase purchaser record similarity is higher than wearing the clothes for other similar users of setting first threshold and arranges in pairs or groups and used with target
Family preference similarity is selected higher than the similar clothes of setting second threshold for user；If user selects DIY collocation units, DIY
Collocation unit allows user to select the dress ornament style of oneself preference and not preference first with selection function, then excludes user automatically
The dress ornament style of preference is not simultaneously using the mark for preferentially recommending to meet user preference from the database and expert system module
Quasi- clothes, which are worn, takes template and expert's clothes are worn and take template and selected for user；
The 3D fittings module is used to recommend the artificial intelligence module and artificial intelligence collocation unit the specification of user
Size and clothes, which are worn, to be taken template and tries on 3D models, forms virtual dynamic 3D clothes effects；
The purchase module is connect with the user management module, for carrying out purchase operation for user.
2. provide the system of taking of wearing of 3-D effect, feature for virtual fitting person based on artificial intelligence as described in claim 1
It is, the clothes that the artificial intelligence collocation unit is bought according to target user establish clothes --- user's matrix normalization
The table of falling Check calculates the similarity of the purchaser record with target user of other users using vectorial COS distance.
3. provide the system of taking of wearing of 3-D effect, feature for virtual fitting person based on artificial intelligence as claimed in claim 2
It is, the value of the first threshold is not less than 70%.
4. provide the system of taking of wearing of 3-D effect, feature for virtual fitting person based on artificial intelligence as described in claim 1
It is, the clothes that the artificial intelligence collocation unit was bought in the past according to target user establish clothes --- clothes matrix is returned
One changes the table of falling Check, and the similarity of the clothes of the purchaser record with target user of other clothes is calculated using vectorial COS distance.
5. provide the system of taking of wearing of 3-D effect, feature for virtual fitting person based on artificial intelligence as claimed in claim 4
It is, the value of the second threshold is not less than 70%.
6. provide the system of taking of wearing of 3-D effect, feature for virtual fitting person based on artificial intelligence as claimed in claim 4
It is, the selection function in the DIY collocation unit is realized using navigation and human-computer dialogue.
7. provide the system of taking of wearing of 3-D effect, feature for virtual fitting person based on artificial intelligence as claimed in claim 4
It is, the dress ornament that the clothes are worn in taking includes：Clothes, footwear, cap, scarf, tippet, silk scarf, necktie, belt, waistband, gloves,
Scarf, handbag, hosiery, headwear, decorative glasses, decorative jewelry and hair style.
The matching method 8. the virtual fitting person based on artificial intelligence wears the clothes, which is characterized in that this method is applied in such as claim
1-6 it is arbitrary described provide the wearing in the system of taking of 3-D effect for virtual fitting person based on artificial intelligence, this method is including following
Step：
S1. user's registration step：User is registered by the user management module；
S2.3D modeling procedures：After register account number, user is logged in by account number cipher, then utilizes the 3D modeling module
By mobile phone shooting and mobile phone scanning obtain user two-dimentional mug shot and the colour of skin, build, height, human body measurements of the chest, waist and hips, shoulder breadth,
Build data information generation is met the 3D models of type data information, will obtained by brachium, shape of face and hair style build data information
Take mug shot and 3D models store to user management module simultaneously by 3D models be transferred to database, artificial intelligence module and
Expert system module；
S3. wear the clothes collocation step：If target user selects artificial intelligence collocation unit and for first purchase, the artificial intelligence is taken
It gives birth to the specification of recommended user according to the 3D models of target user first with unit, then recommends to meet from expert system module
Expert's clothes of target user's specification, which are worn, takes template for user's selection；
If target user selects artificial intelligence collocation unit and is not first purchase, artificial intelligence arranges in pairs or groups unit based on target use
The purchaser record at family is calculated by Collaborative Filtering Recommendation Algorithm to be recommended with target user's phase purchaser record similarity higher than setting the
Wearing the clothes for other similar users of one threshold value arranges in pairs or groups and is higher than the similar of setting second threshold to target user's preference similarity
Clothes are selected for user；
If user select DIY collocation unit, DIY arrange in pairs or groups unit first with selection function allow user select oneself preference with not
Then the dress ornament style of preference excludes the dress ornament style of user's not preference simultaneously using preferentially from the database and expert automatically
The standard garments for meeting user preference is recommended to wear in system module and take template and expert's clothes are worn and take template for user's selection；
S4. step is selected：User wears the template click examination for taking stencil-chosen satisfaction according to the clothes recommended in collocation step of wearing the clothes
It wears.
S5. 3D fittings step：User clicks try on after, 3D fitting module be used for the specification for selecting user and
Clothes, which are worn, to be taken template and tries on 3D models, forms virtual dynamic 3D clothes effects；
S6. purchase or exit step：For user according to after 3D fits step, user selects satisfied clothes according to clothes effect
The dress ornament taken in template is worn to be bought and paid or exited by the user management module.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810064913.8A CN108256975A (en) | 2018-01-23 | 2018-01-23 | Wearing for 3-D effect is provided for virtual fitting person take system and method based on artificial intelligence |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810064913.8A CN108256975A (en) | 2018-01-23 | 2018-01-23 | Wearing for 3-D effect is provided for virtual fitting person take system and method based on artificial intelligence |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN108256975A true CN108256975A (en) | 2018-07-06 |
Family
ID=62742164
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810064913.8A Pending CN108256975A (en) | 2018-01-23 | 2018-01-23 | Wearing for 3-D effect is provided for virtual fitting person take system and method based on artificial intelligence |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108256975A (en) |
Cited By (27)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109242593A (en) * | 2018-07-24 | 2019-01-18 | 武汉纺织大学 | A kind of clothes recommended method, device and the storage medium of case-based reasioning |
| CN109360050A (en) * | 2018-09-27 | 2019-02-19 | 东华大学 | Personal care garment management and personalized collocation recommendation intelligence system based on perceptual demand |
| CN109685001A (en) * | 2018-12-24 | 2019-04-26 | 石狮市森科智能科技有限公司 | Human body measurements of the chest, waist and hips data acquisition method and intelligence sell clothing system and Intelligent unattended sells clothing machine |
| CN109816493A (en) * | 2019-01-29 | 2019-05-28 | 济南奥维信息科技有限公司济宁分公司 | The virtual accessories recommender system of new media and method based on liquid crystal light modulation film |
| CN109919735A (en) * | 2019-03-19 | 2019-06-21 | 江苏皓之睿数字科技有限公司 | A kind of 3D simulation dress ornament matching method and system based on mobile terminal |
| CN110059247A (en) * | 2019-03-19 | 2019-07-26 | 江苏皓之睿数字科技有限公司 | A kind of personalization based on artificial intelligence technology, which is worn, takes recommender system |
| CN110287518A (en) * | 2019-05-14 | 2019-09-27 | 洪岩 | Customized clothing quantity of margin intelligent generation method based on 3-D scanning |
| CN110706076A (en) * | 2019-09-29 | 2020-01-17 | 浙江理工大学 | Virtual fitting method and system capable of performing network transaction by combining online and offline |
| CN110796503A (en) * | 2018-08-02 | 2020-02-14 | 青岛海尔洗衣机有限公司 | Clothing management method based on character model |
| CN110825963A (en) * | 2019-10-18 | 2020-02-21 | 山东大学 | Generation-based auxiliary template enhanced clothing matching scheme generation method and system |
| WO2020098982A1 (en) * | 2018-11-16 | 2020-05-22 | Fision Ag | Method and devices for fitting a garment onto a body |
| TWI696138B (en) * | 2019-01-18 | 2020-06-11 | 國立臺北商業大學 | Recommendation system and method for dialogue interface |
| CN111310859A (en) * | 2020-03-26 | 2020-06-19 | 上海景和国际展览有限公司 | Rapid artificial intelligence data training system used in multimedia display |
| CN111709806A (en) * | 2020-06-16 | 2020-09-25 | 贾苏南 | Money-selecting and fitting system capable of customizing service |
| CN111784442A (en) * | 2020-06-05 | 2020-10-16 | 陈庆奇 | 3D virtual intelligent fitting system |
| CN112015983A (en) * | 2020-08-21 | 2020-12-01 | 广州欢网科技有限责任公司 | Crossing and matching recommendation method and device based on weather forecast application |
| CN112163930A (en) * | 2020-09-27 | 2021-01-01 | 深圳莱尔托特科技有限公司 | Matching recommendation method and matching recommendation system |
| CN112633969A (en) * | 2020-12-15 | 2021-04-09 | 中科彭州智慧产业创新中心有限公司 | Intelligent wardrobe system and intelligent fitting recommendation method |
| CN112667889A (en) * | 2020-12-24 | 2021-04-16 | 吴边 | Intelligent wardrobe management system, control method, terminal and medium |
| CN112703286A (en) * | 2018-09-12 | 2021-04-23 | Lg电子株式会社 | Clothes registering device, clothes recommending device and on-line system comprising same |
| CN112750005A (en) * | 2020-11-24 | 2021-05-04 | 深圳市创悦诺鑫科技有限公司 | Method for shopping through mirror makeup |
| CN113190585A (en) * | 2021-04-12 | 2021-07-30 | 郑州轻工业大学 | Big data acquisition and analysis system for clothing design |
| CN113223147A (en) * | 2021-04-26 | 2021-08-06 | 库迪热提江·亚森 | Character decoration and matching system for providing three-dimensional effect for consumers |
| CN113255052A (en) * | 2021-07-09 | 2021-08-13 | 佛山市陶风互联网络科技有限公司 | Home decoration scheme recommendation method and system based on virtual reality and storage medium |
| CN113487373A (en) * | 2020-05-28 | 2021-10-08 | 海信集团有限公司 | Fitting mirror, terminal, clothing recommendation method and storage medium |
| CN115760319A (en) * | 2023-02-13 | 2023-03-07 | 欧瑞科斯科技产业(集团)有限公司 | Automatic commodity collocation method and device, electronic equipment and storage medium |
| CN117611299A (en) * | 2024-01-22 | 2024-02-27 | 青岛威达体育用品有限公司 | Customization method and system for personalized school uniform, electronic equipment and storage medium |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103440587A (en) * | 2013-08-27 | 2013-12-11 | 刘丽君 | Personal image designing and product recommendation method based on online shopping |
| CN104376160A (en) * | 2014-11-07 | 2015-02-25 | 薛景 | Real person simulation individuality ornament matching system |
-
2018
- 2018-01-23 CN CN201810064913.8A patent/CN108256975A/en active Pending
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103440587A (en) * | 2013-08-27 | 2013-12-11 | 刘丽君 | Personal image designing and product recommendation method based on online shopping |
| CN104376160A (en) * | 2014-11-07 | 2015-02-25 | 薛景 | Real person simulation individuality ornament matching system |
Cited By (31)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109242593B (en) * | 2018-07-24 | 2022-01-28 | 武汉纺织大学 | Case reasoning-based clothing recommendation method and device and storage medium |
| CN109242593A (en) * | 2018-07-24 | 2019-01-18 | 武汉纺织大学 | A kind of clothes recommended method, device and the storage medium of case-based reasioning |
| CN110796503A (en) * | 2018-08-02 | 2020-02-14 | 青岛海尔洗衣机有限公司 | Clothing management method based on character model |
| CN112703286A (en) * | 2018-09-12 | 2021-04-23 | Lg电子株式会社 | Clothes registering device, clothes recommending device and on-line system comprising same |
| US11910920B2 (en) | 2018-09-12 | 2024-02-27 | Lg Electronics Inc. | Clothing treatment apparatus and online system |
| CN109360050A (en) * | 2018-09-27 | 2019-02-19 | 东华大学 | Personal care garment management and personalized collocation recommendation intelligence system based on perceptual demand |
| WO2020098982A1 (en) * | 2018-11-16 | 2020-05-22 | Fision Ag | Method and devices for fitting a garment onto a body |
| CN109685001A (en) * | 2018-12-24 | 2019-04-26 | 石狮市森科智能科技有限公司 | Human body measurements of the chest, waist and hips data acquisition method and intelligence sell clothing system and Intelligent unattended sells clothing machine |
| TWI696138B (en) * | 2019-01-18 | 2020-06-11 | 國立臺北商業大學 | Recommendation system and method for dialogue interface |
| CN109816493A (en) * | 2019-01-29 | 2019-05-28 | 济南奥维信息科技有限公司济宁分公司 | The virtual accessories recommender system of new media and method based on liquid crystal light modulation film |
| CN109919735A (en) * | 2019-03-19 | 2019-06-21 | 江苏皓之睿数字科技有限公司 | A kind of 3D simulation dress ornament matching method and system based on mobile terminal |
| CN110059247A (en) * | 2019-03-19 | 2019-07-26 | 江苏皓之睿数字科技有限公司 | A kind of personalization based on artificial intelligence technology, which is worn, takes recommender system |
| CN110287518A (en) * | 2019-05-14 | 2019-09-27 | 洪岩 | Customized clothing quantity of margin intelligent generation method based on 3-D scanning |
| CN110706076A (en) * | 2019-09-29 | 2020-01-17 | 浙江理工大学 | Virtual fitting method and system capable of performing network transaction by combining online and offline |
| CN110825963A (en) * | 2019-10-18 | 2020-02-21 | 山东大学 | Generation-based auxiliary template enhanced clothing matching scheme generation method and system |
| CN110825963B (en) * | 2019-10-18 | 2022-03-25 | 山东大学 | Generation-based auxiliary template enhanced clothing matching scheme generation method and system |
| CN111310859A (en) * | 2020-03-26 | 2020-06-19 | 上海景和国际展览有限公司 | Rapid artificial intelligence data training system used in multimedia display |
| CN113487373A (en) * | 2020-05-28 | 2021-10-08 | 海信集团有限公司 | Fitting mirror, terminal, clothing recommendation method and storage medium |
| CN111784442A (en) * | 2020-06-05 | 2020-10-16 | 陈庆奇 | 3D virtual intelligent fitting system |
| CN111709806A (en) * | 2020-06-16 | 2020-09-25 | 贾苏南 | Money-selecting and fitting system capable of customizing service |
| CN112015983A (en) * | 2020-08-21 | 2020-12-01 | 广州欢网科技有限责任公司 | Crossing and matching recommendation method and device based on weather forecast application |
| CN112163930A (en) * | 2020-09-27 | 2021-01-01 | 深圳莱尔托特科技有限公司 | Matching recommendation method and matching recommendation system |
| CN112750005A (en) * | 2020-11-24 | 2021-05-04 | 深圳市创悦诺鑫科技有限公司 | Method for shopping through mirror makeup |
| CN112633969A (en) * | 2020-12-15 | 2021-04-09 | 中科彭州智慧产业创新中心有限公司 | Intelligent wardrobe system and intelligent fitting recommendation method |
| CN112667889A (en) * | 2020-12-24 | 2021-04-16 | 吴边 | Intelligent wardrobe management system, control method, terminal and medium |
| CN113190585A (en) * | 2021-04-12 | 2021-07-30 | 郑州轻工业大学 | Big data acquisition and analysis system for clothing design |
| CN113223147A (en) * | 2021-04-26 | 2021-08-06 | 库迪热提江·亚森 | Character decoration and matching system for providing three-dimensional effect for consumers |
| CN113255052A (en) * | 2021-07-09 | 2021-08-13 | 佛山市陶风互联网络科技有限公司 | Home decoration scheme recommendation method and system based on virtual reality and storage medium |
| CN115760319A (en) * | 2023-02-13 | 2023-03-07 | 欧瑞科斯科技产业(集团)有限公司 | Automatic commodity collocation method and device, electronic equipment and storage medium |
| CN117611299A (en) * | 2024-01-22 | 2024-02-27 | 青岛威达体育用品有限公司 | Customization method and system for personalized school uniform, electronic equipment and storage medium |
| CN117611299B (en) * | 2024-01-22 | 2024-04-26 | 青岛威达体育用品有限公司 | Customization method and system for personalized school uniform, electronic equipment and storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108256975A (en) | Wearing for 3-D effect is provided for virtual fitting person take system and method based on artificial intelligence | |
| CN106709781A (en) | Personal image design and collocation purchasing device and method | |
| CN106779977B (en) | Clothing matching method and system based on intelligent mobile terminal | |
| CN107240007A (en) | A kind of AR three-dimensional virtual fitting systems combined with 3D manikins | |
| CN104376160A (en) | Real person simulation individuality ornament matching system | |
| CN110264299A (en) | Clothes recommended method, device and computer equipment based on recognition of face | |
| CN101582143A (en) | Terminal try-on simulation system and method for generating try-on image | |
| CN106097067A (en) | A kind of sweater based on the Internet customizes design system and method | |
| CN103597519A (en) | Computer implemented methods and systems for generating virtual body models for garment fit visualization | |
| TW200828043A (en) | Terminal try-on simulation system and operating and applying method thereof | |
| CN103324860A (en) | Intelligent integrated costume designing and fitting system | |
| CN110298719A (en) | The dress designing method of dress designing platform and the application platform | |
| das Neves et al. | Fashion and Ergonomic Design: Aspects that influence the perception of clothing usability | |
| CN106776865A (en) | A kind of garment coordination method and system based on user's dynamic need | |
| CN110298720A (en) | A kind of custom made clothing design method and platform | |
| CN111612584A (en) | AI intelligent clothing recommendation method based on wearing and putting-on theory | |
| CN108876930A (en) | A kind of personal image management application method and system | |
| CN109410344A (en) | A kind of dress designing method based on three-dimensional structure | |
| CN105167290A (en) | Method for obtaining clothing matching index based on human body external image characteristic analysis | |
| CN107977882A (en) | The online clothes zoarium degree reference method for trying picture library on is formed based on scapegoat's generation examination | |
| Almond | Fashionably voluptuous: Repackaging the fuller-sized figure | |
| CN106651501A (en) | Clothing matching method and system based on human body feature information | |
| CN103455919A (en) | Fitting model forming method and fitting model forming system | |
| CN103258271A (en) | System and method of natural person digitalized image design | |
| Alemany et al. | Anthropometric survey of the Spanish female population aimed at the apparel industry |
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 | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20180706 |