CN110175253A - A kind of user individual garment coordination method and device - Google Patents
A kind of user individual garment coordination method and device Download PDFInfo
- Publication number
- CN110175253A CN110175253A CN201910393633.6A CN201910393633A CN110175253A CN 110175253 A CN110175253 A CN 110175253A CN 201910393633 A CN201910393633 A CN 201910393633A CN 110175253 A CN110175253 A CN 110175253A
- Authority
- CN
- China
- Prior art keywords
- user
- clothes
- modeling
- garment coordination
- compatibility
- 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
- 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
- G06F16/53—Querying
- G06F16/535—Filtering based on additional data, e.g. user or group profiles
-
- 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
- G06F16/50—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
- G06F16/5866—Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually using information manually generated, e.g. tags, keywords, comments, manually generated location and time information
-
- 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
- G06Q30/0271—Personalized advertisement
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Business, Economics & Management (AREA)
- General Engineering & Computer Science (AREA)
- Databases & Information Systems (AREA)
- Data Mining & Analysis (AREA)
- Library & Information Science (AREA)
- Strategic Management (AREA)
- Development Economics (AREA)
- Finance (AREA)
- Accounting & Taxation (AREA)
- Entrepreneurship & Innovation (AREA)
- Game Theory and Decision Science (AREA)
- Economics (AREA)
- Marketing (AREA)
- General Business, Economics & Management (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The invention discloses a kind of user individual garment coordination method and devices, this method comprises: receiving garment coordination positive example is set with data set, carries out feature extraction respectively to its visual information and text information, obtain visual signature and text feature;According to visual signature and text feature, modeled by compatibility of the multi-layer perception (MLP) to complementary clothes;According to visual signature, text feature and the user of storage to the history preference of clothes, multi-modal modeling is carried out to user preference by matrix decomposition;The compatibility modeling and the multi-modal modeling of user preference of complementary clothes are linearly combined, building customized clothing collocation model carries out customized clothing collocation.
Description
Technical field
The disclosure belongs to the technical field of intelligent clothing collocation, is related to a kind of user individual garment coordination method and dress
It sets.
Background technique
Only there is provided background technical informations relevant to the disclosure for the statement of this part, it is not necessary to so constitute first skill
Art.
Garment coordination has become a part indispensable in people's daily life, increasingly numerous with fashion industry
The problem of honor, how the suit of appropriate mix has been increasingly becoming some institute's headaches from dazzling clothes.Have benefited from more
Continuous mature and online fashion community (such as I QON and Ch i ctop i a) fashion fans of media processing techniques mention
The real-world data of confession, there are many research work to be unfolded around garment coordination.Existing method is primarily focused on through depth mind
It is modeled through compatibility of the network to complementary clothes.
However, inventor has found in R&D process, existing method does not account for the preference of user.The aesthetic of user is high
Degree subjectivity, different people may hold different collocation preferences.It needs to solve the problems, such as that customized clothing is arranged in pairs or groups, not only to consider
General aesthetics, it is also contemplated that the personal preference of user;Customized clothing collocation should consider the compatibility between clothes, simultaneous again
User is cared for the preference of clothes.
Summary of the invention
For the deficiencies in the prior art, one or more other embodiments of the present disclosure provide a kind of user individual
Garment coordination method and device, in view of the visual information and text information of clothes all may be comprising about individual subscriber preferences
Important information, the present invention combines vision and the information of text both modalities which models the preference of user, and carries out mutual
Pretend capacitive modeling and user preference modeling are taken, is characterized between jacket and lower clothing respectively and the friendship between user and clothes
Mutually.
According to the one aspect of one or more other embodiments of the present disclosure, a kind of user individual garment coordination side is provided
Method.
A kind of user individual garment coordination method, this method comprises:
It receives garment coordination positive example and is set with data set, feature extraction is carried out to its visual information and text information respectively, is obtained
To visual signature and text feature;
According to visual signature and text feature, modeled by compatibility of the multi-layer perception (MLP) to complementary clothes;
According to visual signature, text feature and the user of storage to the history preference of clothes, by matrix decomposition to user
Preference carries out multi-modal modeling;
The compatibility modeling and the multi-modal modeling of user preference of complementary clothes are linearly combined, personalized clothes are constructed
Dress collocation model, carries out customized clothing collocation.
Further, in the method, garment coordination positive example suit data set include user's set, jacket set,
Lower clothing set and the corresponding history garment coordination set of user, the clothes in the garment coordination data set include clothes vision letter
Breath and clothes text information;
The visual information is image of clothing, and the text information is the description of clothes type.
Further, in the method, the visual information carries out feature extraction by trained deep neural network,
Obtain visual signature.
Further, in the method, the text information carries out feature extraction by text convolutional network, obtains text
Feature.
Further, in the method, it is modeled by semantic relation of the multi-layer perception (MLP) to different modalities, it is described mutual
The compatibility for taking dress is the potential expression of lower clothing visual signature, the potential expression of jacket text feature, lower clothing text feature
The function of the non-negative parameter of potential expression and regulation both modalities which.
Further, in the method, the compatibility modeling and the multi-modal modeling of user preference of complementary clothes are carried out
It is linear to combine, customized clothing collocation model is constructed by Bayes's personalized ordering algorithm.
Further, in the method, the multi-modal modeling of the compatibility modeling and user preference by complementary clothes
The specific steps linearly combined include:
The compatibility modeling and the multi-modal modeling of user preference of complementary clothes are linearly combined, based on regulation jacket
Compatibility and user between lower clothing is to the non-negative parameter of lower clothing preference weight, and modeling is for given jacket specific user to lower clothing
Preference pattern;
For giving jacket to be matched, compared to certain, once clothing user is more prone to the four of another lower clothing and jacket collocation for building
Tuple constructs loss function, building customized clothing collocation model according to Bayes's personalized ordering algorithm;
The model convergence until customized clothing is arranged in pairs or groups by repetitive exercise.
According to the one aspect of one or more other embodiments of the present disclosure, a kind of computer readable storage medium is provided.
A kind of computer readable storage medium, wherein being stored with a plurality of instruction, described instruction is suitable for by terminal device
Reason device loads and executes a kind of user individual garment coordination method.
According to the one aspect of one or more other embodiments of the present disclosure, a kind of terminal device is provided.
A kind of terminal device comprising processor and computer readable storage medium, processor is for realizing each instruction;Meter
Calculation machine readable storage medium storing program for executing is suitable for being loaded by processor and being executed a kind of user for storing a plurality of instruction, described instruction
Customized clothing matching method.
According to the one aspect of one or more other embodiments of the present disclosure, a kind of user individual garment coordination dress is provided
It sets.
A kind of user individual garment coordination device, based on a kind of user individual garment coordination method, comprising:
Characteristic extracting module is configured as receiving garment coordination positive example suit data set, to its visual information and text envelope
Breath carries out feature extraction respectively, obtains visual signature and text feature;
Complementary clothes compatibility carries out modeling module, is configured as passing through multilayer sense according to visual signature and text feature
Know that machine models the compatibility of complementary clothes;
The multi-modal modeling module of user preference is configured as according to visual signature, text feature and the user of storage to clothes
The history preference of dress carries out multi-modal modeling to user preference by matrix decomposition;
Models coupling module is configured as carrying out the compatibility modeling and the multi-modal modeling of user preference of complementary clothes
Linear to combine, building customized clothing collocation model carries out customized clothing collocation.
The disclosure the utility model has the advantages that
A kind of user individual garment coordination method and device that the disclosure provides, from the preference of clothes compatibility and user
Two angles are set out, it is contemplated that the visual information and information of clothes may all reflect the relevant information about user preference, comprehensive
The information of vision and text both modalities which comprehensively models the preference of user, has better solved customized clothing and has taken
With problem.
Detailed description of the invention
The accompanying drawings constituting a part of this application is used to provide further understanding of the present application, and the application's shows
Meaning property embodiment and its explanation are not constituted an undue limitation on the present application for explaining the application.
Fig. 1 is a kind of user individual garment coordination method flow diagram according to one or more embodiments;
Fig. 2 is specific a kind of user individual garment coordination method flow diagram according to one or more embodiments.
Specific embodiment:
Below in conjunction with the attached drawing in one or more other embodiments of the present disclosure, to one or more other embodiments of the present disclosure
In technical solution be clearly and completely described, it is clear that described embodiments are only a part of the embodiments of the present invention,
Instead of all the embodiments.Based on one or more other embodiments of the present disclosure, those of ordinary skill in the art are not being made
Every other embodiment obtained, shall fall within the protection scope of the present invention under the premise of creative work.
It is noted that following detailed description is all illustrative, it is intended to provide further instruction to the application.Unless another
It indicates, all technical and scientific terms that the present embodiment uses have and the application person of an ordinary skill in the technical field
Normally understood identical meanings.
It should be noted that term used herein above is merely to describe specific embodiment, and be not intended to restricted root
According to the illustrative embodiments of the application.As used herein, unless the context clearly indicates otherwise, otherwise singular
Also it is intended to include plural form, additionally, it should be understood that, when in the present specification using term "comprising" and/or " packet
Include " when, indicate existing characteristics, step, operation, device, component and/or their combination.
It should be noted that flowcharts and block diagrams in the drawings show according to various embodiments of the present disclosure method and
The architecture, function and operation in the cards of system.It should be noted that each box in flowchart or block diagram can represent
A part of one module, program segment or code, a part of the module, program segment or code may include one or more
A executable instruction for realizing the logic function of defined in each embodiment.It should also be noted that some alternately
Realization in, function marked in the box can also occur according to the sequence that is marked in attached drawing is different from.For example, two connect
The box even indicated can actually be basically executed in parallel or they can also be executed in a reverse order sometimes,
This depends on related function.It should also be noted that each box and flow chart in flowchart and or block diagram
And/or the combination of the box in block diagram, the dedicated hardware based system that functions or operations as defined in executing can be used are come
It realizes, or the combination of specialized hardware and computer instruction can be used to realize.
In the absence of conflict, the feature in the embodiment and embodiment in the disclosure can be combined with each other, and tie below
It closes attached drawing and embodiment is described further the disclosure.
Technical term is explained: deep neural network, multi-layer perception (MLP) (MLP), matrix decomposition (MF), Bayes's personalization row
Sequence algorithm.
Deep neural network: deep neural network is the artificial mind between input layer and output layer with multiple hidden layers
Through network.Neural network is made of the be mutually related artificial neuron of generalities of many, can be between these artificial neurons
Transmitting data mutually, and associated weight is adjusted according to network.
Multi-layer perception (MLP) (MLP): multi-layer perception (MLP) is the artificial neural network before one kind to structure, one group of input of mapping to
It measures to one group of output vector.Multi-layer perception (MLP) can be counted as a digraph, be made of multiple node layers, each layer
It is all connected to next layer entirely.In addition to input node, each node be one with nonlinear activation function neuron (or
Processing unit).
Matrix decomposition (MF): matrix decomposition is one kind collaborative filtering used in recommender system.Matrix decomposition algorithm
User and project are mapped to the united hidden factor space an of low dimensional, user and item are characterized by the inner product in space
Interaction between mesh.
Bayes's personalized ordering algorithm: Bayes's personalized ordering algorithm is based on bayesian theory pole under priori knowledge
The corresponding article of any user u is marked in bigization posterior probability, if user u has article i and j when point at the same time
I is hit, then we have just obtained a triple<u,i,j>, it indicates that for user u, the sequence ratio j of i is forward.
Embodiment one
According to the one aspect of one or more other embodiments of the present disclosure, a kind of user individual garment coordination side is provided
Method.
As shown in Figs. 1-2, a kind of user individual garment coordination method, this method comprises:
S1: are carried out by feature respectively and is mentioned for the visual information and text information of the collocation positive example suit in fashion community website
It takes.
S2: it using the visual signature and text feature of clothes, is carried out by compatibility of the multi-layer perception (MLP) to complementary clothes
Modeling.
S3: the history preference of clothes is recorded using the visual signature and text feature and user of clothes, passes through matrix
The method of decomposition is modeled from preference of the multi-modal angle to user.
S4: the model of S2 and S3 is linearly combined, and completes model structure by Bayes's personalized ordering algorithm (BPR)
It builds, model is continued to optimize by repetitive exercise, realize personalized garment coordination.
The step S1 characteristic extraction procedure further include:
S11: the present invention constructs user using history preference record of the user to clothes and gathers
Jacket setLower clothing set With user umHistory garment coordination collection
It closesWherein M, NtAnd NbRespectively indicate user, jacket and
The sum of lower clothing, NmIndicate user umGarment coordination total tricks.The present invention is using the image data of clothes as the vision of clothes
The type of clothes is described the text information as clothes by information.
S12: the present invention usesIndicate the visual information of jacket (lower clothing) by trained depth nerve
The visual signature that network obtains is usedIndicate that the text information of jacket (lower clothing) passes through text convolutional network (TextCNN)
Obtained text feature.
The step S2 clothes compatibility modeling process further include:
S21: the present invention models the semantic relation of different modalities by multi-layer perception (MLP) (MLP), gives input x:
h1=s (W1x+b1),
Wherein WkAnd bkIt is the parameter of multi-layer perception (MLP) (MLP), k is the number of plies, and s is sigmoid function,It is upper
The potential expression of clothing visual signature, the present invention use
To characterize jacket tiWith lower clothing bjBetween compatibility, whereinFor the potential expression of lower clothing visual signature,For
The potential expression of jacket text feature,For the potential expression of lower clothing text feature, π is the non-negative parameter for regulating and controlling both modalities which.
The preference modeling process of the step S3 user further include:
S31: the present invention records the history preference of clothes using the visual signature and text feature and user of clothes,
By matrix decomposition (MF), modeled from preference of the multi-modal angle to user, formula is as follows:
Wherein cmjIndicate user umTo lower clothing bjPreference, wherein α be global offset amount to be learned, βm,βjFor user um
With lower clothing bjOffset, γm,γjFor user umWith lower clothing bjHidden vector,For user umWith lower clothing bjVision it is hidden
Vector,WithDot product characterize user umTo lower clothing bjThe preference of visual information,For user umWith lower clothing bjText
This hidden vector,WithDot product characterize user umTo lower clothing bjThe preference of text information, wherein
η is the non-negative parameter for regulating and controlling both modalities which weight.
The step 4 linearly combines and training process further include:
S41: the present invention is linearly combined S2 with the model of S3, and modeling is for giving jacket tj, user umTo lower clothing
Preference is as follows:
Wherein μ is to regulate and control compatibility under jacket between clothing and user to the non-negative parameter of lower clothing preference weight.
S42: building four-tuple It should
Quadruple notation is for giving jacket t to be matchedi, compare lower clothing bk, user umIt is more prone to lower clothing bjWith jacket tiCollocation.According to
Bayes's personalized ordering algorithm constructs loss function:
Wherein ΘFIt is the set of parameter, last is used to prevent model over-fitting.
S43: by repetitive exercise until model is restrained, Θ is savedF, personalized garment coordination can be realized.
Embodiment two
According to the one aspect of one or more other embodiments of the present disclosure, a kind of computer readable storage medium is provided.
A kind of computer readable storage medium, wherein being stored with a plurality of instruction, described instruction is suitable for by terminal device
Reason device loads and executes a kind of user individual garment coordination method.
Embodiment three
According to the one aspect of one or more other embodiments of the present disclosure, a kind of terminal device is provided.
A kind of terminal device comprising processor and computer readable storage medium, processor is for realizing each instruction;Meter
Calculation machine readable storage medium storing program for executing is suitable for being loaded by processor and being executed a kind of user for storing a plurality of instruction, described instruction
Customized clothing matching method.
These computer executable instructions execute the equipment according to each reality in the disclosure
Apply method or process described in example.
In the present embodiment, computer program product may include computer readable storage medium, containing for holding
The computer-readable program instructions of row various aspects of the disclosure.Computer readable storage medium, which can be, can keep and store
By the tangible device for the instruction that instruction execution equipment uses.Computer readable storage medium for example can be-- but it is unlimited
In-- storage device electric, magnetic storage apparatus, light storage device, electric magnetic storage apparatus, semiconductor memory apparatus or above-mentioned
Any appropriate combination.The more specific example (non exhaustive list) of computer readable storage medium includes: portable computing
Machine disk, hard disk, random access memory (RAM), read-only memory (ROM), erasable programmable read only memory (EPROM or
Flash memory), static random access memory (SRAM), Portable compressed disk read-only memory (CD-ROM), digital versatile disc
(DVD), memory stick, floppy disk, mechanical coding equipment, the punch card for being for example stored thereon with instruction or groove internal projection structure, with
And above-mentioned any appropriate combination.Computer readable storage medium used herein above is not interpreted instantaneous signal itself,
The electromagnetic wave of such as radio wave or other Free propagations, the electromagnetic wave propagated by waveguide or other transmission mediums (for example,
Pass through the light pulse of fiber optic cables) or pass through electric wire transmit electric signal.
Computer-readable program instructions described herein can be downloaded to from computer readable storage medium it is each calculate/
Processing equipment, or outer computer or outer is downloaded to by network, such as internet, local area network, wide area network and/or wireless network
Portion stores equipment.Network may include copper transmission cable, optical fiber transmission, wireless transmission, router, firewall, interchanger, gateway
Computer and/or Edge Server.Adapter or network interface in each calculating/processing equipment are received from network to be counted
Calculation machine readable program instructions, and the computer-readable program instructions are forwarded, for the meter being stored in each calculating/processing equipment
In calculation machine readable storage medium storing program for executing.
Computer program instructions for executing present disclosure operation can be assembly instruction, instruction set architecture (ISA)
Instruction, machine instruction, machine-dependent instructions, microcode, firmware instructions, condition setup data or with one or more programmings
The source code or object code that any combination of language is write, the programming language include the programming language-of object-oriented such as
C++ etc., and conventional procedural programming languages-such as " C " language or similar programming language.Computer-readable program refers to
Order can be executed fully on the user computer, partly be executed on the user computer, as an independent software package
Execute, part on the user computer part on the remote computer execute or completely on a remote computer or server
It executes.In situations involving remote computers, remote computer can include local area network by the network-of any kind
(LAN) or wide area network (WAN)-is connected to subscriber computer, or, it may be connected to outer computer (such as utilize internet
Service provider is connected by internet).In some embodiments, by being believed using the state of computer-readable program instructions
Breath comes personalized customization electronic circuit, such as programmable logic circuit, field programmable gate array (FPGA) or programmable logic
Array (PLA), the electronic circuit can execute computer-readable program instructions, to realize the various aspects of present disclosure.
Example IV
According to the one aspect of one or more other embodiments of the present disclosure, a kind of user individual garment coordination dress is provided
It sets.
A kind of user individual garment coordination device, based on a kind of user individual garment coordination method, comprising:
Characteristic extracting module is configured as receiving garment coordination positive example suit data set, to its visual information and text envelope
Breath carries out feature extraction respectively, obtains visual signature and text feature;
Complementary clothes compatibility carries out modeling module, is configured as passing through multilayer sense according to visual signature and text feature
Know that machine models the compatibility of complementary clothes;
The multi-modal modeling module of user preference is configured as according to visual signature, text feature and the user of storage to clothes
The history preference of dress carries out multi-modal modeling to user preference by matrix decomposition;
Models coupling module is configured as carrying out the compatibility modeling and the multi-modal modeling of user preference of complementary clothes
Linear to combine, building customized clothing collocation model carries out customized clothing collocation.
It should be noted that although being referred to several modules or submodule of equipment in the detailed description above, it is this
Division is only exemplary rather than enforceable.In fact, in accordance with an embodiment of the present disclosure, two or more above-described moulds
The feature and function of block can embody in a module.Conversely, the feature and function of an above-described module can be with
Further division is to be embodied by multiple modules.
The disclosure the utility model has the advantages that
(1) a kind of user individual garment coordination method and device that the disclosure provides, a kind of user that the disclosure provides
Customized clothing matching method and device, from two angles of clothes compatibility and the preference of user, it is contemplated that the view of clothes
Feel relevant information of all possible reflection of information and information about user preference, combines the information pair of vision and text both modalities which
The preference of user is comprehensively modeled, and customized clothing collocation problem has been better solved.
(2) a kind of user individual garment coordination method and device that the disclosure provides, customized clothing collocation should examine
Consider the compatibility between clothes, takes into account user again to the preference of clothes, the disclosure is seamless by the preference of garment coordination and user
Ground combines, so that model not only can satisfy garment coordination, can more cater to the personal taste of user.
(3) a kind of user individual garment coordination method and device that the disclosure provides, by multi-layer perception (MLP) (MLP),
Model the compatibility of complementary clothes.
(4) a kind of user individual garment coordination method and device that the disclosure provides, passes through the side of matrix decomposition (MF)
Formula models user to the preference of clothes, and the image of clothes and text data is made full use of to characterize user to the preference of clothes.
The foregoing is merely preferred embodiment of the present application, are not intended to limit this application, for the skill of this field
For art personnel, various changes and changes are possible in this application.Within the spirit and principles of this application, made any to repair
Change, equivalent replacement, improvement etc., should be included within the scope of protection of this application.Therefore, the present invention is not intended to be limited to this
These embodiments shown in text, and it is to fit to the widest scope consistent with the principles and novel features disclosed herein.
Claims (10)
1. a kind of user individual garment coordination method, which is characterized in that this method comprises:
It receives garment coordination positive example and is set with data set, feature extraction is carried out respectively to its visual information and text information, depending on
Feel feature and text feature;
According to visual signature and text feature, modeled by compatibility of the multi-layer perception (MLP) to complementary clothes;
According to visual signature, text feature and the user of storage to the history preference of clothes, by matrix decomposition to user preference
Carry out multi-modal modeling;
The compatibility modeling and the multi-modal modeling of user preference of complementary clothes are linearly combined, building customized clothing is taken
With model, customized clothing collocation is carried out.
2. a kind of user individual garment coordination method as described in claim 1, which is characterized in that in the method,
The garment coordination positive example suit data set includes user's set, jacket set, lower clothing set and the corresponding history of user
Garment coordination set, the clothes in the garment coordination data set include clothes visual information and clothes text information;
The visual information is image of clothing, and the text information is the description of clothes type.
3. a kind of user individual garment coordination method as described in claim 1, which is characterized in that in the method, described
Visual information carries out feature extraction by trained deep neural network, obtains visual signature.
4. a kind of user individual garment coordination method as described in claim 1, which is characterized in that in the method, described
Text information carries out feature extraction by text convolutional network, obtains text feature.
5. a kind of user individual garment coordination method as described in claim 1, which is characterized in that in the method, pass through
Multi-layer perception (MLP) models the semantic relation of different modalities, and the compatibility of the complementation clothes is the latent of lower clothing visual signature
In expression, the potential expression of jacket text feature, the potential expression of lower clothing text feature and the non-negative parameter for regulating and controlling both modalities which
Function.
6. a kind of user individual garment coordination method as described in claim 1, which is characterized in that in the method, will be mutual
The compatibility modeling and the multi-modal modeling of user preference for taking dress are linearly combined, and Bayes's personalized ordering algorithm is passed through
Construct customized clothing collocation model.
7. a kind of user individual garment coordination method as described in claim 1, which is characterized in that in the method, described
Include: by the specific steps that the compatibility modeling and the multi-modal modeling of user preference of complementary clothes are linearly combined
The compatibility modeling and the multi-modal modeling of user preference of complementary clothes are linearly combined, based on clothing under regulation jacket
Between compatibility and user to the non-negative parameter of lower clothing preference weight, modeling is for given jacket specific user to the inclined of lower clothing
Good model;
For giving jacket to be matched, compared to certain, once clothing user is more prone to the four-tuple that another lower clothing and jacket are arranged in pairs or groups for building,
Loss function, building customized clothing collocation model are constructed according to Bayes's personalized ordering algorithm;
The model convergence until customized clothing is arranged in pairs or groups by repetitive exercise.
8. a kind of computer readable storage medium, wherein being stored with a plurality of instruction, which is characterized in that described instruction is suitable for by terminal
The processor of equipment is loaded and is executed such as a kind of described in any item user individual garment coordination methods of claim 1-7.
9. a kind of terminal device comprising processor and computer readable storage medium, processor is for realizing each instruction;It calculates
Machine readable storage medium storing program for executing is for storing a plurality of instruction, which is characterized in that described instruction is suitable for being loaded by processor and being executed such as power
Benefit requires a kind of described in any item user individual garment coordination methods of 1-7.
10. a kind of user individual garment coordination device, which is characterized in that based on such as claim 1-7 described in any item one
Kind user individual garment coordination method, comprising:
Characteristic extracting module is configured as receiving garment coordination positive example suit data set, to its visual information and text information point
Not carry out feature extraction, obtain visual signature and text feature;
Complementary clothes compatibility carries out modeling module, is configured as passing through multi-layer perception (MLP) according to visual signature and text feature
The compatibility of complementary clothes is modeled;
The multi-modal modeling module of user preference is configured as according to visual signature, text feature and the user of storage to clothes
History preference carries out multi-modal modeling to user preference by matrix decomposition;
Models coupling module is configured as carrying out the compatibility modeling and the multi-modal modeling of user preference of complementary clothes linear
In conjunction with building customized clothing collocation model carries out customized clothing collocation.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910393633.6A CN110175253A (en) | 2019-05-13 | 2019-05-13 | A kind of user individual garment coordination method and device |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910393633.6A CN110175253A (en) | 2019-05-13 | 2019-05-13 | A kind of user individual garment coordination method and device |
Publications (1)
Publication Number | Publication Date |
---|---|
CN110175253A true CN110175253A (en) | 2019-08-27 |
Family
ID=67690833
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201910393633.6A Pending CN110175253A (en) | 2019-05-13 | 2019-05-13 | A kind of user individual garment coordination method and device |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN110175253A (en) |
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN112862538A (en) * | 2021-03-02 | 2021-05-28 | 中国工商银行股份有限公司 | Method, apparatus, electronic device, and medium for predicting user preference |
CN113034237A (en) * | 2020-12-28 | 2021-06-25 | 武汉纺织大学 | Dress suit recommendation system and method |
Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2014074072A1 (en) * | 2012-11-12 | 2014-05-15 | Singapore University Of Technology And Design | Clothing matching system and method |
CN108875910A (en) * | 2018-05-23 | 2018-11-23 | 山东大学 | Garment coordination method, system and the storage medium extracted based on attention knowledge |
CN109741112A (en) * | 2019-01-10 | 2019-05-10 | 博拉网络股份有限公司 | A kind of user's purchase intention prediction technique based on mobile big data |
-
2019
- 2019-05-13 CN CN201910393633.6A patent/CN110175253A/en active Pending
Patent Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2014074072A1 (en) * | 2012-11-12 | 2014-05-15 | Singapore University Of Technology And Design | Clothing matching system and method |
CN108875910A (en) * | 2018-05-23 | 2018-11-23 | 山东大学 | Garment coordination method, system and the storage medium extracted based on attention knowledge |
CN109741112A (en) * | 2019-01-10 | 2019-05-10 | 博拉网络股份有限公司 | A kind of user's purchase intention prediction technique based on mobile big data |
Non-Patent Citations (1)
Title |
---|
赵光明: ""基于深度学习的服装搭配推荐算法研究与实现"", 《中国优秀硕士学位论文全文数据库 信息科技辑》 * |
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN113034237A (en) * | 2020-12-28 | 2021-06-25 | 武汉纺织大学 | Dress suit recommendation system and method |
CN112862538A (en) * | 2021-03-02 | 2021-05-28 | 中国工商银行股份有限公司 | Method, apparatus, electronic device, and medium for predicting user preference |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN109800789A (en) | Diabetic retinopathy classification method and device based on figure network | |
JP6788264B2 (en) | Facial expression recognition method, facial expression recognition device, computer program and advertisement management system | |
CN107169573A (en) | Using composite machine learning model come the method and system of perform prediction | |
US10719693B2 (en) | Method and apparatus for outputting information of object relationship | |
CN108734338A (en) | Credit risk forecast method and device based on LSTM models | |
Shen et al. | A privacy-preserving social computing framework for health management using federated learning | |
CN107066628A (en) | Wear the clothes recommendation method and device | |
CN108664999A (en) | A kind of training method and its device, computer server of disaggregated model | |
CN110334245A (en) | A kind of short video recommendation method and device of the figure neural network based on Temporal Order | |
CN108121995A (en) | For identifying the method and apparatus of object | |
CN108961369A (en) | The method and apparatus for generating 3D animation | |
CN110489582A (en) | Personalization shows the generation method and device, electronic equipment of image | |
CN108509892A (en) | Method and apparatus for generating near-infrared image | |
CN109754317A (en) | Merge interpretation clothes recommended method, system, equipment and the medium of comment | |
CN110321761A (en) | A kind of Activity recognition method, terminal device and computer readable storage medium | |
CN113191479A (en) | Method, system, node and storage medium for joint learning | |
CN110175253A (en) | A kind of user individual garment coordination method and device | |
CN113767398A (en) | Apparatus for machine learning based visual device selection | |
CN110263236A (en) | Social network user multi-tag classification method based on dynamic multi-view learning model | |
CN110414550A (en) | Training method, device, system and the computer-readable medium of human face recognition model | |
CN107767152A (en) | Product purchase intention analysis method and server | |
CN110111620A (en) | A kind of virtual reality learning platform | |
CN108737491A (en) | Information-pushing method and device and storage medium, electronic device | |
CN110209860A (en) | A kind of interpretable garment coordination method and device based on clothes attribute of template-directed | |
Kim | Artificial intelligence for 6G |
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: 20190827 |
|
RJ01 | Rejection of invention patent application after publication |