CN109814724A - Design shows methods, devices and systems - Google Patents
Design shows methods, devices and systems Download PDFInfo
- Publication number
- CN109814724A CN109814724A CN201910148113.9A CN201910148113A CN109814724A CN 109814724 A CN109814724 A CN 109814724A CN 201910148113 A CN201910148113 A CN 201910148113A CN 109814724 A CN109814724 A CN 109814724A
- Authority
- CN
- China
- Prior art keywords
- model
- design
- collocation
- attached
- shows
- 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
Landscapes
- Processing Or Creating Images (AREA)
Abstract
Show methods, devices and systems this application discloses a kind of design.Wherein, this method comprises: the figure collocation model obtained based on preparatory training generates the collocation result data of design according to the image information of design；The model obtained based on preparatory training shows model, and according to the collocation result data of design, generation design, which is attached to model, tries master drawing on；Output design, which is attached to model, tries master drawing on.Present application addresses in the prior art, user needs human eye to watch when selecting original design, and imagines that the effect that design is worn in model causes design expressive force poor, and then the technical issues of the purchasing power of influence design.
Description
Technical field
This application involves textile printing field, showing method, apparatus and be in particular to a kind of design
System.
Background technique
Flower pattern refers to the weaving that picture on surface, color, the processing effect of various textile finished products or semi-finished product are shown
The unique designs of product and typesetting.The quality of flower pattern selection and design, directly influences the beauty of textile (for example, clothes)
Sense.
Requirement with people to dressing is higher and higher, also increasing to the challenge of designer.Currently, designer
When selecting design, can only by visually going to see, then imagine design be through the effect with model.This side
Formula greatly limits design applied field so that designer can not can be visually seen the effect that design is through with model
The form of expression of scape leads to the reduction of design purchasing power.
User needs human eye to watch when selecting original design in the prior art for above-mentioned, and imagines pattern figure
Case, which is worn in the effect with model, causes design expressive force poor, and then the problem of the purchase of influence design, at present
Not yet put forward effective solutions.
Summary of the invention
What the embodiment of the present application provided a kind of design shows methods, devices and systems, at least to solve existing skill
In art, user needs human eye to watch when selecting original design, and imagines that design is worn in the effect with model
The technical issues of causing design expressive force poor, and then influencing the purchasing power of design.
According to the one aspect of the embodiment of the present application, provide a kind of design shows method, comprising: acquisition flower pattern
The image information of pattern；The figure collocation model obtained based on preparatory training generates pattern figure according to the image information of design
The collocation result data of case, wherein the input data of figure collocation model is the image information of design, schemes the defeated of collocation model
Data are the collocation result data of design out；The model obtained based on preparatory training shows model, according to design
Collocation result data, generation design, which is attached to model, tries master drawing on, wherein the input data of model's displaying model
For the model of the collocation result data and selection of design, model shows that the output data of model is that design is attached to mould
Master drawing is tried on spy；Output design, which is attached to model, tries master drawing on.
Further, figure collocation model is one of any as follows: the serializing collocation based on shot and long term memory network LSTM
Model, based on depth adaptation network DAN without serializing arrange in pairs or groups model.
Further, model shows that model is official documents and correspondence descriptive model CPGN.
Further, output design, which is attached to model, tries master drawing on, comprising: is showed with three-dimensional scenic mode
Design is attached to model and tries master drawing on.
Further, output design be attached to model try master drawing on after, method further include: delete flower
The image information of type pattern.
Further, output design be attached to model try master drawing on after, method further include: detecting
In the case where model's switching command, the model for showing design in master drawing is tried in switching on.
Further, model's switching command includes at least one following: sliding up and down switching command, the switching that horizontally slips refers to
It enables.
According to the another aspect of the embodiment of the present application, additionally provide a kind of design shows system, comprising: camera shooting
Head, for acquiring the image information of design；Processor is communicated with camera, and the figure for being obtained based on preparatory training is taken
The collocation result data of design is generated according to the image information of design with model, the mould obtained based on preparatory training
Spy shows model, and according to the collocation result data of design, generation design, which is attached to model, tries master drawing on,
In, the input data of figure collocation model is the image information of design, and the output data of figure collocation model is design
Collocation result data, model show that the input data of model is the model of the collocation result data and selection of design, model
It shows that the output data of model is attached to model for design and tries master drawing on；Display screen is used for processor communication
Show design and be attached to model and tries master drawing on.
According to the another aspect of the embodiment of the present application, additionally provide a kind of design shows system, comprising: terminal is set
It is standby, for obtaining the image information of design；Server, and terminal equipment in communication, the figure for being obtained based on preparatory training
Collocation model generates the collocation result data of design according to the image information of design, is obtained based on preparatory training
Model shows model, and according to the collocation result data of design, generation design, which is attached to model, tries master drawing on,
Wherein, the input data of figure collocation model is the image information of design, and the output data of figure collocation model is design
Collocation result data, model show model input data be design collocation result data and selection model, mould
Spy shows that the output data of model is attached to model for design and tries master drawing on；Wherein, terminal device is also used to connect
It receives the design that server returns and is attached to the master drawing of trying on model, and show design and be attached to model
Try master drawing on.
According to the another aspect of the embodiment of the present application, a kind of display device of design is additionally provided, comprising: acquisition is single
Member, for acquiring the image information of design；First processing units, the figure collocation model for being obtained based on preparatory training,
According to the image information of design, generate the collocation result data of design, wherein figure collocation model input data be
The output data of the image information of design, figure collocation model is the collocation result data of design；The second processing unit,
Model for being obtained based on preparatory training shows model, and according to the collocation result data of design, it is attached to generate design
Try master drawing on model, wherein model show model input data be design collocation result data and
The model of selection, model show that the output data of model is attached to model for design and try master drawing on；Output unit,
It is attached to model for exporting design and tries master drawing on.
In the embodiment of the present application, pass through the image information of acquisition design；The figure collocation obtained based on preparatory training
Model generates the collocation result data of design according to the image information of design, wherein the input of figure collocation model
Data are the image information of design, and the output data of figure collocation model is the collocation result data of design；Based on pre-
The model that first training obtains shows model, according to the collocation result data of design, generates design and is attached to model's body
On try master drawing on, wherein model show model input data be design collocation result data and selection model,
Model shows that the output data of model is attached to model for design and tries master drawing on；Output design is attached to mould
Master drawing is tried on spy, has achieved the purpose that generating model according to collected design tries master drawing on, to realize
The bandwagon effect for improving design, to improve the technical effect of the buying rate of design, and then solves in the prior art,
User needs human eye to watch when selecting original design, and imagines that the effect that design is worn in model causes to spend
Type pattern expressive force is poor, and then the technical issues of the purchasing power of influence design.
Detailed description of the invention
The drawings described herein are used to provide a further understanding of the present application, constitutes part of this application, this Shen
Illustrative embodiments and their description please are not constituted an undue limitation on the present application for explaining the application.In the accompanying drawings:
Fig. 1 is to show method flow diagram according to a kind of design of the embodiment of the present application；
Fig. 2 is to show schematic illustration according to a kind of optional design of the embodiment of the present application；
Fig. 3 is the user's operation flow chart showed according to a kind of optional design of the embodiment of the present application；
Fig. 4 is to show system schematic according to a kind of design of the embodiment of the present application；
Fig. 5 is to show system schematic according to a kind of design of the embodiment of the present application；And
Fig. 6 is the display device schematic diagram according to a kind of design of the embodiment of the present application.
Specific embodiment
In order to make those skilled in the art more fully understand application scheme, below in conjunction in the embodiment of the present application
Attached drawing, the technical scheme in the embodiment of the application is clearly and completely described, it is clear that described embodiment is only
The embodiment of the application a part, instead of all the embodiments.Based on the embodiment in the application, ordinary skill people
Member's every other embodiment obtained without making creative work, all should belong to the model of the application protection
It encloses.
It should be noted that the description and claims of this application and term " first " in above-mentioned attached drawing, "
Two " etc. be to be used to distinguish similar objects, without being used to describe a particular order or precedence order.It should be understood that using in this way
Data be interchangeable under appropriate circumstances, so as to embodiments herein described herein can in addition to illustrating herein or
Sequence other than those of description is implemented.
According to the embodiment of the present application, the embodiment of the method for a kind of design showed is provided, it should be noted that
The step of process of attached drawing illustrates can execute in a computer system such as a set of computer executable instructions, also,
It, in some cases, can be to be different from shown in sequence execution herein although logical order is shown in flow charts
The step of out or describing.
Fig. 1 is to show method flow diagram according to a kind of design of the embodiment of the present application, as shown in Figure 1, this method
Include the following steps:
Step S101 acquires the image information of design.
Specifically, above-mentioned design can be wait the pattern for taking office anticipate a kind of weave finished product or surface of semi-finished of printing and dyeing,
It can be the design gone out by electronic device designs such as computers, be also possible to Freehandhand-drawing design, can also be entity
Design on product.Above-mentioned image information can be still image (for example, photo), be also possible to dynamic image (for example,
Video).
It should be noted that as an alternative embodiment, each embodiment of the application is wait print and dye onto clothes
Design for, to be illustrated.
Optionally, above-mentioned steps S101 can pass through the camera (for example, mobile phone camera) or external built in equipment
The image information of camera (for example, the camera being connect with computer) acquisition design.
Step S102, the figure collocation model obtained based on preparatory training generate flower pattern according to the image information of design
The collocation result data of pattern, wherein the input data of figure collocation model is the image information of design, figure collocation model
Output data is the collocation result data of design.
Specifically, above-mentioned figure collocation model can be through machine learning, and training is obtained for generating various flower pattern in advance
The model of design.Due to same design, different typesetting designs, the difference on effect showed on the textile is very
Greatly, the application uses intelligent algorithm, and training in advance obtains the collocation model of the figure for generating patterning design, will be collected
The image information of design is input to figure collocation model, then can generate the collocation result data of corresponding design.
Step S103, the model obtained based on preparatory training show model, raw according to the collocation result data of design
It is attached to model at design and tries master drawing on, wherein model shows that the input data of model is taking for design
Model with result data and selection, model show that the output data of model is attached to model for design and try sample on
Figure.
Specifically, above-mentioned model shows that model can be through machine learning, and training in advance is obtained for by design
It is attached to mould to generate the model for trying master drawing on.
Step S104, output design, which is attached to model, tries master drawing on.
Optionally, in order to which more intuitive viewing effect is presented, trying on model is attached in output design
When master drawing, design can be showed with three-dimensional scenic mode it is attached to model and try master drawing on.
As a kind of optional embodiment, Fig. 2 is showing according to a kind of optional design of the embodiment of the present application
Schematic illustration, as shown in Fig. 2, the application passes through convolutional neural networks CNN (Convolutional Neural Networks)
Picture pretreatment is carried out, using DSSM (Deep Semantic Similarity Model) as basic framework, on this basis
Using the design collocation algorithm model of two kinds of logics: the first is based on shot and long term memory network LSTM (Long Short
Term Memory) serializing arrange in pairs or groups model, be for second based on depth adaptation network DAN (Deep Aggregated
Network) without serializing arrange in pairs or groups model.After result of being arranged in pairs or groups by any one collocation algorithm output figure, collocation is tied
Fruit data are input to model's model, i.e. official documents and correspondence descriptive model CPGN (Context aware Pointer-Generator
Networks), model's clothes master drawing is adhered to output stamp.It is entire product skill comprising whole master drawing in final result
The integration of art describes.
As seen from Figure 2, the application is based on TensorFlow technology, uses CNN (convolutional neural networks) and DP (depth
Study) algorithm and generate confrontation network (GAN, Generative Adversarial Networks), excavate to the full extent
Compactness and aesthetic feeling when design and model show, expressive force when improving design attachment model, use WebGL skill
Art (WebGL is a kind of 3D drafting standards) solves the bandwagon effect that identification is attached to automatically later on model, uses Three.js
(library based on webgl encapsulates tool method important in some 3D rendering demands and recycles with rendering) reduction is true complete
Orientation design clothes effect.
Wherein, the training process of convolutional neural networks CNN, has references to back-propagation algorithm, is divided into two stages:
First stage, propagation stage forward:
A) sample (X, a Y are taken from sample setp), X is inputted into network；
B) corresponding reality output O is calculatedp。
Output layer is transmitted to by converting step by step from input layer by information.Information comes in input the weight square with every layer
Battle array phase point layer is that network completes the process executed when operating normally after training, is executed by network and calculates output acquisition result:
Op=Fn(…(F2(F1(XpW(1))W(2))…)W(n))；
Second stage, the back-propagation stage:
A) reality output O is calculatedpWith corresponding ideal output YpDifference；
B) weight matrix is adjusted by the method backpropagation of minimization error.
Generating confrontation network G AN is a kind of deep learning model, which includes two modules: generating model G and differentiation
Model D；Wherein, model G is generated for generating a picture, is received a random noise z, is generated picture；Discrimination model D is used for
Differentiate a picture whether be it is true, input as x, export as D (x), indicate that x is the probability of true picture.As a kind of optional
Embodiment, generating the target formula that uses of confrontation network G AN can be with are as follows:
Wherein, x indicates that true picture, z indicate that the noise of input G model, G (z) indicate the picture that G model generates, D (G
(z)) indicate that D model judges G (z) for the probability of true picture.
D can first be optimized, re-optimization G is as follows after dismantling:
A) optimize D:
When i.e. optimization judges other network, network is not produced, subsequent G (z) is dummy copy.Optimize the formula first item of D,
True sample x input prediction result is better closer to 1, and the result of dummy copy D (G (z)) optimization is the smaller the better, because label is 0.But
It is that first item is bigger, then Section 2 is smaller, it may appear that contradiction, thus, Section 2 is revised as 1-D (G (z)), is thus got over
It is big better.
B) optimize G:
It is unrelated with true sample when optimizing G, so first item is directly removed, only dummy copy, but this when
Time wishes that the label of dummy copy is 1, and then D (G (z)) is the bigger the better, and purpose is unified into the form of 1-D (G (z)), then can only
It is minimum 1-D (G (z)), is substantially not different, is intended merely to unity of form.The two Optimized models can merge later
To write, the minimax objective function most started is reformed into.
From the foregoing, it will be observed that in the above embodiments of the present application, by acquiring the image information of design, and based on instruction in advance
The figure collocation model got generates the collocation result data of design according to the image information of collected design,
And then the collocation result data of the design of generation is input to the model that training obtains in advance and shows model, to generate flower pattern
Pattern is attached to model and tries master drawing on, has reached and has generated the mesh that model tries master drawing on according to collected design
, to realize the bandwagon effect for improving design, to improve the technical effect of the buying rate of design, and then solve
In the prior art, user needs human eye to watch when selecting original design, and imagine that design is worn in model's body
On effect cause design expressive force poor, and then influence design purchasing power the technical issues of.
In order to ensure original design is not lost, as an alternative embodiment, in output design attachment
With the model try master drawing on after, the above method further include: delete the image information of design.Master drawing is tried in displaying
Afterwards, the problem of being automatically deleted collected design, design being prevented to be leaked.
Since the clothes fashion with different models is different, more show effect to realize, as a kind of optional
Embodiment, output design be attached to model try master drawing on after, the above method can also include: to detect
In the case where model's switching command, the model for showing design in master drawing is tried in switching on.
Optionally, above-mentioned model's switching command includes at least one following: sliding up and down switching command, horizontally slip switching
Instruction.
As a preferred embodiment, showing method based on design provided by the present application, one can be provided
Kind client application or the application based on Web, optionally, which can have but be not limited to following module: design is adopted
Collection module, model's selecting module re-recognize module and display output module, acquire pattern figure by design acquisition module
Case selects model by model's selecting module, and the model of selection is matched with the design of acquisition, generates pattern figure
Case is attached to the design with model, finally, passing through display output module output display.Preferably, the application is improved
Application server end may include Back Administration Module, wherein Back Administration Module may include advertising management module, Mo Teguan
Manage module etc..
By taking the application of mobile phone as an example, Fig. 3 is the use showed according to a kind of optional design of the embodiment of the present application
Family operational flowchart, as shown in figure 3, including following operating procedure:
1. user, which opens cell phone software, enters the design identification page, stability contorting mobile phone camera focuses and keeps colored
Type is in frame choosing；
2. clicking " intelligent recognition ", background system acquires information, and design can be attached to model automatically after completing acquisition
With, show effect (master drawing) of the design on clothes；
3. can click when recognition effect, design and attachment model's master drawing show dissatisfied and re-recognize (repetition
1., 2. process) identified again, to reach optimum efficiency；
4. be attached to the design with model, can by horizontally slip replace model realize show pattern application
Scene different-effect；
5. selecting different types of model and clothes according to different design styles, setting is clicked, corresponding mould is selected
Spy, clicking application can be completed setting.
By above embodiment, user opens software and enters the identification page, clicks intelligent recognition, and software can be according to just
Certain specific information streams are grabbed inside camera automatically in the scene of identification to upload onto the server, server is according to the spy of identification
Determine scene and the message stream data of crawl and match and generate with artificial intelligence technology to be taken in identification flower pattern to clothes according to model
It fills style and shows pattern effect.Wherein, the design being attached to model can also be instructed by slide handover to switch
Model realizes the design bandwagon effects of more scenes.It, can be in the case that user is unsatisfied with current design
Click re-recognizes, to identify other designs.After end of identification, it is automatically deleted design, the original flower pattern of effective protection
The problem of design of author is lost.
As can be seen that the application, which can help original flower pattern author to provide a variety of designs for user, shows scene, no
Only patterned display with a model, realize that multiple models switch comparison design scene effect, after use,
The application Product Logic can be automatically deleted effective protection design by technology and not be lost, and improve safety, improve purchase
Buy rate.This software is that designer provides more dress designing inspirations on design application direction simultaneously.
According to the another aspect of the embodiment of the present application, additionally provide a kind of design shows system, and Fig. 4 is according to this
That applies for a kind of design of embodiment shows system schematic, as shown in Figure 4, comprising:
Camera 401, for acquiring the image information of design；
Processor 402 is communicated with camera 401, the figure collocation model for being obtained based on preparatory training, according to pattern figure
The image information of case generates the collocation result data of design, and the model obtained based on preparatory training shows model, according to flower
The collocation result data of type pattern, generation design, which is attached to model, tries master drawing on, wherein schemes the defeated of collocation model
Enter the image information that data are design, the output data of figure collocation model is the collocation result data of design, model
The input data for showing model is the model of the collocation result data and selection of design, and model shows the output data of model
It is attached to model for design and tries master drawing on；
Display screen 403 is communicated with processor 402, is attached to model for showing design and is tried master drawing on.
It should be noted that above-mentioned camera 401, processor 402 and display screen can be an equipment (for example, mobile phone)
On component, the component being also possible on multiple devices (for example, external camera and computer).
From the foregoing, it will be observed that the image information of design is acquired by camera 401 in the above embodiments of the present application, and
It is generated by the figure collocation model that processor 402 is obtained based on preparatory training according to the image information of collected design
The collocation result data of design, and then the collocation result data of the design of generation is input to what training in advance obtained
Model shows model, is attached to model with to generate design and tries master drawing on, finally by what is communicated with processor 402
Display screen 403 shows design and is attached to model and tries master drawing on, has reached and has been generated according to collected design
Model tries the purpose of master drawing on, so that the bandwagon effect for improving design is realized, to improve the buying rate of design
Technical effect, and then solve in the prior art, user needs human eye to watch when selecting original design, and imagines flower pattern
Pattern, which is worn in the effect with model, causes design expressive force poor, and then the technology for influencing the purchasing power of design is asked
Topic.
According to the another aspect of the embodiment of the present application, additionally provide a kind of design shows system, and Fig. 5 is according to this
That applies for a kind of design of embodiment shows system schematic, as shown in Figure 5, comprising: terminal device 501 and server
502。
Terminal device 501, for obtaining the image information of design；
Server 502 is communicated with terminal device 501, the figure collocation model for being obtained based on preparatory training, according to flower pattern
The image information of pattern generates the collocation result data of design, and the model obtained based on preparatory training shows model, according to
The collocation result data of design, generation design, which is attached to model, tries master drawing on, wherein figure collocation model
Input data is the image information of design, and the output data of figure collocation model is the collocation result data of design, mould
Spy shows that the input data of model is the model of the collocation result data and selection of design, and model shows the output number of model
Master drawing is tried on according to being attached to model for design；
Wherein, the design that terminal device 501 is also used to receive server return, which is attached to model, tries sample on
Figure, and show design and be attached to model and try master drawing on.
As a kind of optional embodiment, above-mentioned terminal device 501 can be computer, mobile phone, tablet computer, notebook
Any one smart machine such as computer.Wherein, in the case that terminal device 501 is computer, computer passes through external camera shooting
The image information of head acquisition design；When terminal device 501 is the equipment such as mobile phone, tablet computer, laptop the case where
Under, the image information of design is collected by the camera built in equipment.Above-mentioned server 502 can be application service
Device is connect with terminal device by internet.
From the foregoing, it will be observed that terminal device 501 obtains the image information of design, concurrently in the above embodiments of the present application
Give server 502, the figure collocation model that server 502 is obtained based on preparatory training, according to the figure of collected design
As information, the collocation result data of design is generated, and then the collocation result data of the design of generation is input to pre-
The model that first training obtains shows model, to generate the master drawing of trying on that design is attached to model, and is sent to terminal
Equipment 501 shows that design is attached to model by the terminal device 501 and tries master drawing on, reached according to acquisition
To design generate model try the purpose of master drawing on, thus realize improve design bandwagon effect, to improve flower
The technical effect of the buying rate of type pattern, and then solve in the prior art, user needs people when selecting original design
Eye viewing, and imagine that the effect that design is worn in model causes design expressive force poor, and then influence pattern figure
The technical issues of purchasing power of case.
According to the another aspect of the embodiment of the present application, a kind of display device of design is additionally provided, Fig. 6 is according to this
Apply for a kind of display device schematic diagram of design of embodiment, as shown in Figure 6, comprising: the processing of acquisition unit 601, first
Unit 602, the second processing unit 603 and output unit 604.
Wherein, acquisition unit 601, for acquiring the image information of design；
First processing units 602, the figure collocation model for being obtained based on preparatory training, are believed according to the image of design
Breath, generates the collocation result data of design, wherein and the input data of figure collocation model is the image information of design,
The output data of figure collocation model is the collocation result data of design；
The second processing unit 603, the model for being obtained based on preparatory training shows model, according to the collocation of design
Result data, generation design, which is attached to model, tries master drawing on, wherein model shows that the input data of model is flower
The model of the collocation result data and selection of type pattern, model show that the output data of model is that design is attached to model's body
On try master drawing on；
Output unit 604 is attached to model for exporting design and tries master drawing on.
Herein it should be noted that above-mentioned acquisition unit 601, first processing units 602, the second processing unit 603 and defeated
Unit 604 corresponds to the step S101 to S104 in embodiment of the method, the example that above-mentioned module is realized with corresponding step out
It is identical with application scenarios, but it is not limited to above method embodiment disclosure of that.It should be noted that above-mentioned module is as dress
That sets a part of can execute in a computer system such as a set of computer executable instructions.
From the foregoing, it will be observed that the image information of design is acquired by acquisition unit 601 in the above embodiments of the present application,
And the figure collocation model obtained by first processing units 602 based on preparatory training, according to the image of collected design
Information generates the collocation result data of design, and then passes through the second processing unit 603 for the collocation of the design of generation
Result data is input to the obtained model of training in advance and shows model, is attached to model with to generate design and tries sample on
Figure is attached to model finally by the output design of output unit 604 and tries master drawing on, reached according to collected
Design generates model and tries the purpose of master drawing on, so that the bandwagon effect for improving design is realized, to improve pattern figure
The technical effect of the buying rate of case, and then solve in the prior art, user needs human eye to see when selecting original design
It sees, and imagines that the effect that design is worn in model causes design expressive force poor, and then influence design
The technical issues of purchasing power.
Optionally, above-mentioned first processing units 602 use figure collocation model can be to be following one of any: based on length
Phase memory network LSTM serializing collocation model, based on depth adaptation network DAN without serializing arrange in pairs or groups model.
Optionally, the model that above-mentioned the second processing unit 603 uses shows model for official documents and correspondence descriptive model CPGN.
In an alternative embodiment, above-mentioned output unit 604 is also used to show design with three-dimensional scenic mode
It is attached to model and tries master drawing on.
In an alternative embodiment, above-mentioned apparatus further include: delete unit, the image for deleting design is believed
Breath.
In an alternative embodiment, above-mentioned apparatus further include: switch unit, in the feelings for detecting model's switching command
Under condition, the model for showing design in master drawing is tried in switching on.
Optionally, above-mentioned model's switching command may include at least one following: sliding up and down switching command, horizontally slips
Switching command.
Above-mentioned the embodiment of the present application serial number is for illustration only, does not represent the advantages or disadvantages of the embodiments.
In above-described embodiment of the application, all emphasizes particularly on different fields to the description of each embodiment, do not have in some embodiment
The part of detailed description, reference can be made to the related descriptions of other embodiments.
In several embodiments provided herein, it should be understood that disclosed technology contents can pass through others
Mode is realized.Wherein, the apparatus embodiments described above are merely exemplary, such as the division of the unit, Ke Yiwei
A kind of logical function partition, there may be another division manner in actual implementation, for example, multiple units or components can combine or
Person is desirably integrated into another system, or some features can be ignored or not executed.Another point, shown or discussed is mutual
Between coupling, direct-coupling or communication connection can be through some interfaces, the INDIRECT COUPLING or communication link of unit or module
It connects, can be electrical or other forms.
The unit as illustrated by the separation member may or may not be physically separated, aobvious as unit
The component shown may or may not be physical unit, it can and it is in one place, or may be distributed over multiple
On unit.It can some or all of the units may be selected to achieve the purpose of the solution of this embodiment according to the actual needs.
It, can also be in addition, each functional unit in each embodiment of the application can integrate in one processing unit
It is that each unit physically exists alone, can also be integrated in one unit with two or more units.Above-mentioned integrated list
Member both can take the form of hardware realization, can also realize in the form of software functional units.
If the integrated unit is realized in the form of SFU software functional unit and sells or use as independent product
When, it can store in a computer readable storage medium.Based on this understanding, the technical solution of the application is substantially
The all or part of the part that contributes to existing technology or the technical solution can be in the form of software products in other words
It embodies, which is stored in a storage medium, including some instructions are used so that a computer
Equipment (can for personal computer, server or network equipment etc.) execute each embodiment the method for the application whole or
Part steps.And storage medium above-mentioned includes: that USB flash disk, read-only memory (ROM, Read-Only Memory), arbitrary access are deposited
Reservoir (RAM, Random Access Memory), mobile hard disk, magnetic or disk etc. be various to can store program code
Medium.
The above is only the preferred embodiment of the application, it is noted that for the ordinary skill people of the art
For member, under the premise of not departing from the application principle, several improvements and modifications can also be made, these improvements and modifications are also answered
It is considered as the protection scope of the application.
Claims (10)
1. a kind of design shows method characterized by comprising
Acquire the image information of design；
The figure collocation model obtained based on preparatory training generates the design according to the image information of the design
Collocation result data, wherein the input data of figure collocation model is the image information of the design, and the figure is taken
Output data with model is the collocation result data of the design；
The model obtained based on preparatory training shows that model generates the flower according to the collocation result data of the design
Type pattern is attached to model and tries master drawing on, wherein the model shows that the input data of model is the design
Collocation result data and selection model, the model shows that the output data of model is that the design is attached to model
With it try master drawing on；
It exports the design and is attached to model and try master drawing on.
2. design according to claim 1 shows method, which is characterized in that the figure collocation model is following appoints
One of meaning: serializing collocation model, taking without serializing based on depth adaptation network DAN based on shot and long term memory network LSTM
With model.
3. design according to claim 1 shows method, which is characterized in that the model shows that model is official documents and correspondence
Descriptive model CPGN.
4. design according to claim 1 shows method, which is characterized in that export the design and be attached to
Master drawing is tried on model, comprising:
Show the design with three-dimensional scenic mode and be attached to model and tries master drawing on.
5. design according to claim 1 shows method, which is characterized in that exporting the design attachment
With the model try master drawing on after, the method also includes:
Delete the image information of the design.
6. design according to claim 1 shows method, which is characterized in that exporting the design attachment
With the model try master drawing on after, the method also includes:
In the case where detecting model's switching command, the model for showing the design in master drawing is tried described in switching on.
7. design according to claim 6 shows method, which is characterized in that model's switching command includes such as
It is at least one lower: to slide up and down switching command, horizontally slip switching command.
8. a kind of design shows system characterized by comprising
Camera, for acquiring the image information of design；
Processor is communicated with the camera, the figure collocation model for being obtained based on preparatory training, according to the design
Image information, generate the collocation result data of the design, the model obtained based on preparatory training shows model, according to
The collocation result data of the design generates the design and is attached to model and try master drawing on, wherein is described
The input data of figure collocation model is the image information of the design, and the output data of the figure collocation model is the flower
The collocation result data of type pattern, the model show model input data be the design collocation result data and
The model of selection, the model show that the output data of model is attached to model for the design and try master drawing on；
Display screen, and the processor communication are attached to model for showing the design and try master drawing on.
9. a kind of design shows system characterized by comprising
Terminal device, for obtaining the image information of design；
Server, and the terminal equipment in communication, the figure for being obtained based on preparatory training are arranged in pairs or groups model, according to the pattern figure
The image information of case generates the collocation result data of the design, and the model obtained based on preparatory training shows model, root
According to the collocation result data of the design, generates the design and be attached to model and try master drawing on, wherein institute
The input data for stating figure collocation model is the image information of the design, and the output data of the figure collocation model is described
The collocation result data of design, the model show that the input data of model is the collocation result data of the design
With the model of selection, the model shows that the output data of model is attached to model for the design and tries sample on
Figure；
Wherein, the design that the terminal device is also used to receive that the server returns is attached to the examination with model
Master drawing is worn, and shows the design and is attached to model and try master drawing on.
10. a kind of display device of design characterized by comprising
Acquisition unit, for acquiring the image information of design；
First processing units, the figure for being obtained based on preparatory training are arranged in pairs or groups model, according to the image information of the design,
Generate the collocation result data of the design, wherein the input data of the figure collocation model is the design
The output data of image information, the figure collocation model is the collocation result data of the design；
The second processing unit, the model for being obtained based on preparatory training shows model, according to the collocation knot of the design
Fruit data generate the design and are attached to model and try master drawing on, wherein the model shows the input number of model
According to the model of collocation result data and selection for the design, the model shows that the output data of model is the flower
Type pattern is attached to model and tries master drawing on；
Output unit is attached to model for exporting the design and tries master drawing on.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910148113.9A CN109814724A (en) | 2019-02-28 | 2019-02-28 | Design shows methods, devices and systems |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910148113.9A CN109814724A (en) | 2019-02-28 | 2019-02-28 | Design shows methods, devices and systems |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN109814724A true CN109814724A (en) | 2019-05-28 |
Family
ID=66607706
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910148113.9A Pending CN109814724A (en) | 2019-02-28 | 2019-02-28 | Design shows methods, devices and systems |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109814724A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111221988A (en) * | 2020-02-27 | 2020-06-02 | 苏李明洲 | Searching method for model and photography base |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20100097375A1 (en) * | 2008-10-17 | 2010-04-22 | Kabushiki Kaisha Square Enix (Also Trading As Square Enix Co., Ltd.) | Three-dimensional design support apparatus and three-dimensional model display system |
| CN104599313A (en) * | 2013-10-31 | 2015-05-06 | 镇江华扬信息科技有限公司 | 3D virtual simulation tailoring technology of clothes |
| CN105678566A (en) * | 2015-12-25 | 2016-06-15 | 上海真旗电子商务有限公司 | Clothing style online customization system integrating pattern design and collocation design functions |
| CN106651550A (en) * | 2017-01-10 | 2017-05-10 | 广州中橙优布网络科技有限公司 | 3D try-on system |
-
2019
- 2019-02-28 CN CN201910148113.9A patent/CN109814724A/en active Pending
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20100097375A1 (en) * | 2008-10-17 | 2010-04-22 | Kabushiki Kaisha Square Enix (Also Trading As Square Enix Co., Ltd.) | Three-dimensional design support apparatus and three-dimensional model display system |
| CN104599313A (en) * | 2013-10-31 | 2015-05-06 | 镇江华扬信息科技有限公司 | 3D virtual simulation tailoring technology of clothes |
| CN105678566A (en) * | 2015-12-25 | 2016-06-15 | 上海真旗电子商务有限公司 | Clothing style online customization system integrating pattern design and collocation design functions |
| CN106651550A (en) * | 2017-01-10 | 2017-05-10 | 广州中橙优布网络科技有限公司 | 3D try-on system |
Non-Patent Citations (1)
| Title |
|---|
| 技术小能手: "工程师如何解决穿衣搭配烦恼？——滴搭平台与算法", 《HTTPS://DEVELOPER.ALIYUN.COM/ARTICLE/619198》 * |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111221988A (en) * | 2020-02-27 | 2020-06-02 | 苏李明洲 | Searching method for model and photography base |
| CN111221988B (en) * | 2020-02-27 | 2023-08-22 | 苏李明洲 | Searching method for model and photographic base |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Zhou et al. | Photorealistic facial expression synthesis by the conditional difference adversarial autoencoder | |
| US10679044B2 (en) | Human action data set generation in a machine learning system | |
| CN111833418B (en) | Animation interaction method, device, equipment and storage medium | |
| US9626788B2 (en) | Systems and methods for creating animations using human faces | |
| US20200151963A1 (en) | Training data set generation apparatus and method for machine learning | |
| KR20210119438A (en) | Systems and methods for face reproduction | |
| JP7342062B2 (en) | Image processing methods, devices, electronic devices and storage media | |
| US20220318946A1 (en) | Method for image shape transformation based on generative adversarial network | |
| CN109871736A (en) | The generation method and device of natural language description information | |
| RU2764144C1 (en) | Rapid two-layer neural network synthesis of realistic images of a neural avatar based on a single image | |
| CN111476241A (en) | Character clothing conversion method and system | |
| CN111539897A (en) | Method and apparatus for generating image conversion model | |
| CN110163052A (en) | Video actions recognition methods, device and machinery equipment | |
| CN110427864B (en) | Image processing method and device and electronic equipment | |
| CN114187392B (en) | Virtual even image generation method and device and electronic equipment | |
| CN113657272B (en) | Micro video classification method and system based on missing data completion | |
| CN112562045B (en) | Method, apparatus, device and storage medium for generating model and generating 3D animation | |
| CN109814724A (en) | Design shows methods, devices and systems | |
| Li et al. | SPGAN: face forgery using spoofing generative adversarial networks | |
| WO2015033431A1 (en) | Annotation device, and annotation system | |
| Sun et al. | Local facial makeup transfer via disentangled representation | |
| US20230154088A1 (en) | Disentangling latent representations for image reenactment | |
| CN111277871A (en) | Method and device for recommending content in Internet Protocol Television (IPTV) | |
| Serra et al. | Easy generation of facial animation using motion graphs | |
| WO2022221042A1 (en) | Producing a digital image representation of a body |
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 | ||
| WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20190528 |
|
| WD01 | Invention patent application deemed withdrawn after publication |