CN101477560A - Clothing matching system and method based on scheme database - Google Patents
Clothing matching system and method based on scheme database Download PDFInfo
- Publication number
- CN101477560A CN101477560A CNA2009100370506A CN200910037050A CN101477560A CN 101477560 A CN101477560 A CN 101477560A CN A2009100370506 A CNA2009100370506 A CN A2009100370506A CN 200910037050 A CN200910037050 A CN 200910037050A CN 101477560 A CN101477560 A CN 101477560A
- Authority
- CN
- China
- Prior art keywords
- sample
- scheme
- database
- clothing matching
- derived units
- 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
Images
Landscapes
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The invention discloses a clothing matching system base on a scheme database, which comprises a sample and derivation unit database, a scheme generation module, a clothing matching scheme database and a retrieval module. The sample and derivation unit database is used for storing the data of the sample and derivation unit. The scheme generation module is used for combining samples (or derivation units) in the sample and derivation unit database into a clothing matching scheme according to the image design rules. The clothing matching scheme database is used for storing the data relative to the clothing matching scheme generated by the scheme generation module. The retrieval module is used for comparing and adapting the matching scheme storage in the clothing matching scheme database and retrieving the matching scheme including the samples according to the samples defined by users. The clothing crowd and the clothing accessories are grouped according to the requirements of the image design rules; a representative model sample is arranged for each clothing crowd group, and a corresponding representative clothing sample is arranged for each clothing accessories group; each sample and each possible change form relate to one sample or one derivation unit thereof; and the correlation data of each sample (or each derivation unit) is stored in the sample and derivation unit database. The samples (or derivation units) are combined into the clothing matching scheme in a manual or artificial intelligent manner through the scheme generation module according to the image design rules; and the data relative to the clothing matching scheme generated is stored in the clothing matching scheme database. When users utilize the retrieval module to retrieve the suitable scheme in the clothing matching scheme database under the condition of one, two or more samples, users can very conveniently find the clothing matching scheme containing the samples so as to obtain instructions.
Description
Technical field
The present invention relates to information processing, database retrieval field, especially a kind of clothing matching system and method based on scheme database.
Background technology
Along with rapid development of economy, it is very abundant and complicated that the kind of clothing products, pattern become, and changes also very fast.People also improve greatly to the requirement of dressing, accessories.In the face of numerous and complicated market, people generally wish to obtain specialty and the opinions and suggestions of science.But common people are difficult to obtain the expert and help targetedly.
In paper " intelligent personal image design (CAD) system ", the genetic algorithm of comprehensive personal image design expertise coding has been proposed, end user's artificial neural networks approaches fitness function, derives the arranging scheme that meets the image design requirement.
At publication number is in the patented claim (network dress appraisal and collocation system and method) of CN1908978 and the patented claim (a kind of network dress appraisal and collocation system and method) that publication number is CN101136096, proposed to form the system and method for arranging scheme based on collocation knowledge rule storehouse and expert system.Though this method can be derived complete arranging scheme, derivation is quite complicated, and is not easy to utilize the arranging scheme that generates by artificial neural network and manual type.
In the prior art, provide the multiple mode of using artificial intelligence derivation clothing matching scheme.But ordinary consumer still is difficult to utilize these means easily, immediately obtains the collocation suggestion of professional level.
Summary of the invention
The object of the invention is, a kind of clothing matching system and method based on scheme database is provided.The derivation artificial and artificial intelligence of large amount of complex is fulfiled ahead of schedule on the backstage.The user can use the mode of similar search engine, is oneself selected clothing products easily, finds to meet the matched combined scheme that the image design specialty requires, and can see design sketch intuitively.
For achieving the above object, a kind of clothing matching system based on scheme database of the present invention comprises: sample and derived units database, scheme generation module, clothing matching scheme database, retrieval module.
Sample and derived units database are in order to stored samples and derived units data.
The scheme generation module is used for according to the image design rule, and the sample (or derived units) in sample and the derived units database is combined into the clothing matching scheme.
The clothing matching scheme database is in order to the clothing matching scheme related data of save scheme generation module generation.
Retrieval module in order to the sample of determining according to the user, compares adaptively to the arranging scheme that stores in the clothing matching scheme database, retrieve the arranging scheme that comprises these samples.
According to the requirement of image design rule, dressing crowd, Fashion and Accessories are divided into groups.Representative model's sample is established in each dressing crowd grouping, and a corresponding representative dress ornament sample is also established in each Fashion and Accessories grouping.Each sample and the version that each is possible thereof are exactly a sample or its derived units.The related data of each sample (or derived units) is stored in sample and the derived units database.According to the image design rule, sample (or derived units) is combined into the clothing matching scheme.The clothing matching scheme related data that generates is stored in the clothing matching scheme database.When the user is a condition with one, two or plurality of sample, during retrieval is fit in the clothing matching scheme database scheme, just can find the clothing matching scheme that comprises these samples very expediently, obtain the guidance that dressing is arranged in pairs or groups.
A complete arranging scheme is independently to exist the object integration to form by several, such as a dressing people, and an overcoat, a shirt, a pair of trousers, a pair of leather shoes and socks can also add necktie, glasses etc.Under perfect condition, everyone, each dress ornament all are unique, should as one independently thinking factor design arranging scheme.But under general normal condition, the image design expert does not need like this, and they just note grasping the key link details usually, that is to say according to these crucial details and divide into groups that different groupings has different design rules.When they face a problem, design arranging scheme according to these with the rule that is grouped into the basis exactly.
The aforesaid scheme generation module of the present invention also can comprise an expert system module.In order to go out the clothing matching scheme according to the clothing matching rule induction in the collocation knowledge rule storehouse.
The aforesaid scheme generation module of the present invention also can comprise an artificial neural network module.Derive the clothing matching scheme in order to use genetic algorithm.
A kind of clothing matching system of the present invention based on scheme database, also can comprise a design sketch generation module, picture in order to all samples (or derived units) that will constitute a clothing matching scheme combines, form the whole structure figure of clothing matching scheme, design sketch is stored in the clothing matching scheme database as the part of scheme.
A kind of clothing matching system based on scheme database of the present invention also can comprise a dress ornament sample determination module, determines the dress ornament sample in order to assisted user.In dress ornament sample determination module, also can comprise a picture retrieval module.The picture retrieval module comprises picture searching software, can search out approximate picture in sample and derived units database and dress ornament arranging scheme database.By the sequencing of similarity tabulation, supply the user to select the sample that suits, in these pictures as the substitute of former commodity, conveniently to utilize clothing matching system of the present invention.The user can use digital camera that the dress ornament of oneself is filmed, and with photo upload, also books magazine or other picture can be brought search, to search out identical or close sample.
In dress ornament sample determination module, also can comprise an expert data comparing module.The data in design, production in order to comparison sample and commodity are so that comparatively Zhuan Ye user seeks suitable alternative sample.
A kind of clothing matching system based on scheme database of the present invention also can comprise model's sample determination module, determines the model's sample that is fit in order to assisted user.
In model's sample determination module, also can comprise people's face type identification module.Belong to any people's face type in order to identification people face picture.Can accurately judge people's face type by images such as photo, videos by assisted user, select model's sample to make things convenient for the user.People's face type identification module comprises people's face type identification database and pattern recognition module.The user uploads facial picture, compares by the image data in pattern recognition module and the people's face type identification database, determines people's face type.
In model's sample determination module, also can comprise an information sifting module.The user is by the prompting logging data and click option, the information data that input is necessary.The information sifting module can be judged the model's sample that belongs to this user according to the data that obtain.
On the other hand, a kind of clothing matching method based on scheme database of the present invention is characterized in that: may further comprise the steps:
1.. according to the requirement of image design rule, dressing crowd, Fashion and Accessories are divided into groups.
2.. a sample is established in each grouping.Each sample and the version that each is possible thereof are exactly a sample or its derived units.With the related data of each sample (or derived units), be stored in sample and the derived units database.
3.. according to the image design rule, the sample (or derived units) with in sample and the derived units database is combined into the clothing matching scheme.Arranging scheme related data with forming is stored in the clothing matching scheme database.
4.. the user is a condition with one, two or plurality of sample, the arranging scheme that retrieval is fit to.
5.. in the clothing matching scheme database, retrieval comprises the clothing matching scheme of these samples.
6.. the clothing matching scheme content with retrieving is sent to user interface.
A kind of clothing matching method based on scheme database of the present invention is characterized in that: it is as follows that step is added the back:
1.. according to the requirement of image design rule, dressing crowd, Fashion and Accessories are divided into groups.
2.. a sample is established in each grouping.Each sample and the version that each is possible thereof are exactly a sample or its derived units.With the related data of each sample (or derived units), be stored in sample and the derived units database.
3.. according to the image design rule, the sample (or derived units) with sample and derived units database is combined into the clothing matching scheme.Arranging scheme related data with forming is stored in the clothing matching scheme database.
4.. the user utilizes the picture retrieval module, in sample and derived units database and dress ornament arranging scheme database, searches out approximate picture.By the sequencing of similarity tabulation, assisted user is selected suitable sample, as the substitute of former commodity with these pictures.
5.. the user is a condition with one, two or plurality of sample, the scheme that retrieval is fit to.
6.. in the clothing matching scheme database, retrieval comprises the clothing matching scheme of these samples.
7.. the clothing matching scheme content with retrieving is sent to user interface.
A kind of clothing matching method based on scheme database of the present invention is characterized in that: it is as follows that step is added the back:
1.. according to the requirement of image design rule, dressing crowd, Fashion and Accessories are divided into groups.
2.. a sample is established in each grouping.Each sample and the version that each is possible thereof are exactly a sample or its derived units.With the related data of each sample (or derived units), be stored in sample and the derived units database.
3.. according to the image design rule, the sample (or derived units) with in sample and the derived units database is combined into the clothing matching scheme.Arranging scheme related data with forming is stored in the clothing matching scheme database.
4.. utilize the design sketch generation module, the picture that constitutes all samples (or derived units) of a clothing matching scheme is combined, form the whole structure figure of clothing matching scheme.Design sketch is stored in the clothing matching scheme database as the part of scheme.
5.. the user is a condition with one, two or plurality of sample, the scheme that retrieval is fit to.
6.. in the clothing matching scheme database, retrieval comprises the clothing matching scheme of these samples.
7.. the clothing matching scheme content with retrieving is sent to user interface.
Description of drawings
Fig. 1 is that the scheme of a kind of clothing matching system based on scheme database of the present invention generates synoptic diagram
Fig. 2 is the retrieval synoptic diagram of a kind of clothing matching system based on scheme database of the present invention
Fig. 3 is the retrieval flow figure of a kind of clothing matching system based on scheme database of the present invention
Embodiment
The specific embodiment of the present invention is described with reference to the accompanying drawings.
Scheme generates (as shown in Figure 1)
Grouping
Features of skin colors, face type feature, physical characteristic according to dressing crowd 100 are divided into groups, and form a plurality of crowds and divide into groups 101.
With the male sex is example, and features of skin colors can be divided into four types according to tone, lightness, purity and contrast situation: shallow warm contrast type, shallow cold gradation type, warm deeply gradation type, deep cooling contrast type.Face type feature can be divided into amount sense (containing ratio), the soft a small amount of sense of profile (containing ratio) in the hale and hearty a large amount of senses of profile (containing ratio), profile hale and hearty middle amount sense (containing ratio), the hale and hearty a small amount of sense of profile (containing ratio), the profile centre, profile soft middle amount sense (containing ratio) six types, totally 24 types according to profile, volume sense, ratio.Physical characteristic is divided into tall and thin build, soft-fat type, triangle build, inverted triangle build, narrow build, osculant, totally six types, 144 basic grouped altogether.
With man's occupation easy shirt is example, is divided into standard neck, wide angle collar, collar button neck, stand-up collar, narrow notch lapel, fillet neck, heterochromatic neck (standard or wide angle), hidden discount neck with the neck type.Be divided into real silk, blending, wired, flax, cotton etc. with lining, cotton dress material commonly used divides chambray, oxford, gingham, dimity again.The clothing color of standard is divided into pure hue, light tone, white cold, utmost point light appearance, rich shade accent, middle tone, gentle tone, light gray tone, shade, dull colour accent, middle gray tone, low key tone, dull gray tone, utmost point gray tone, 15 kinds of tones of netrual colour black-white-gray class, 384 kinds of colors, practical application may be more.The adularescent that shirt is commonly used, navy blue, ink blue, light blue, blue, light blue, ash indigo plant, ash, light gray, brown, rust red, brick red, coffee color, bright Huang, light yellow, purple, ink look, pink colour, pink etc., striped then has: slightly nicking, pencil striped, replace nicking, Ta Tesaer lattice, Pei Zili decorative pattern, polychrome grid etc.
For dressing crowd grouping, the general male sex's stature is tall and big relatively, and stature factor shared ratio in whole image is bigger, and face's factor is smaller relatively.And the general stature of women is comparatively petite, and it is more important that face's factor just seems.Thereby, being characterized as the grouping of foundation with blee and face type, the women just can be meticulousr than the male sex.On the other hand, face feature belongs to dominant character, and data acquisition is relatively easy.Comparatively speaking, the feature of stature belongs to recessive character, and the collection of data is then much more difficult.General ordinary people lacks the knowledge of specialty, differs and makes accurate judgement surely.The sewer that the specialty level is high and the cost of contactless automatic measuring equipment are all very high.Just the finite data that provides by the user also can only be done comparatively rough grouping.It is contemplated that, improve in technical conditions, when the collection of demographic data becomes easy, adopt meticulousr grouping, so that the present invention can obtain better effect, is very natural thing.
For furnishings, because the manufacturer of specialty, supplier's existence are arranged usually, just can collect complete data at an easy rate, accomplish very fine grouping.The many more thin more samples of grouping are many more in theory, and the scheme of design is also just accurate more.But the variety design of dress ornament commodity is extremely abundant on the one hand.On the other hand, dressing crowd's grouping is not accomplished very meticulous.Actually, grouping is after fineness acquires a certain degree, and further the effect of segmentation also can be more and more not obvious again, the arranging scheme of formation also more and more approaching.Thereby grouping only needs according to the actual needs, and it is just passable to reach a balance.Generally be that the big feature of image influence is got carefully, influence little feature and get slightly that the product of main flow main road is got carefully, the rare product of unexpected winner is got slightly.
Each crowd divides into groups 101, by the feature of this grouping representational model's sample 102 is set.Each dress ornament grouping 111 is provided with a representational dress ornament sample 112 by the feature of this grouping.Also be that this sample has all features of this grouping.Such as, the colour of skin: shallow warm contrast type, the face type: amount sense (containing ratio) during profile is hale and hearty, build is tall and thin, male sex's grouping.Its sample is exactly a colour of skin: shallow warm contrast type, face type: amount sense (containing ratio) during profile is hale and hearty, the male sex model that build is tall and thin.
May there be several different versions in a dress ornament sample 112.Such as, a kind of suit that can split dress, a whole set of dress and the effect that splits dress can be different fully.They are different design cells in image design, in use can realize the conversion easily of different arranging schemes again.Therefore, a whole set of dress can be considered as a sample 112, and splitting dress is exactly its two derived units 113.In more extreme example, have seven or eight kinds of play such as a silk scarf, if all can produce bigger influence, just have a sample 112 and six or seven derived units 113 to image.With sample 112 is condition when retrieving, sample 112 and its derived units 113, and the arranging scheme that is constituted all meets search condition.
The purpose that the dressing crowd is divided into groups is, unfavorable clothing matching scheme is got rid of.And in fact, some people has two or more grouping feature.That is to say, he divide which the group all right, have several model's samples to be applicable to him.In this case, can determine that optimum one is its model's sample 102, and other be made as derived units 103.
With the detailed data of each independent sample and derived units, be stored in the database, just constituted sample and derived units database 1.Whether dress ornament sample (or derived units) data comprise: kind, style, details, color, lining, evaluation speech, current popular, (also can comprise two dimension, tri-dimensional picture) or the like.Model's sample (or derived units) data comprise: eye feature, lip feature, features of skin colors, blush feature, hair feature, contour feature, amount sense feature, physical characteristic, need cover up shortcoming or the like.
Utilize scheme generation module 2, according to the image design rule, the sample (or derived units) with in sample and the derived units database 1 is combined into the clothing matching scheme.
Scheme generates can different modes:
1. by the imagination of image design expert, utilize artificial module 21, directly from sample and derived units database 1, select suitable sample (or derived units), be combined into the clothing matching scheme according to oneself.
2. utilize expert system module 22,, derive the clothing matching scheme according to the collocation rule in the clothing matching knowledge rule storehouse.
A selected model's sample (or derived units) and a dress ornament sample (or derived units) (as above dress) propose the collocation request.These model's sample (or derived units) data and these dress ornament sample (or derived units) data of being stored in sample and the derived units database 1 are accessed, together import clothing matching expert system module 22, with be stored in clothing matching knowledge rule storehouse in advance in data carry out corresponding relatively, rational analysis, infer the collocation condition of next dress ornament sample (or derived units) (following dress).Afterwards, according to the collocation condition, in sample and derived units database 1, the dress ornament sample (or derived units) (following dress) that search is fit to.Then, again with dress ornament sample (or derived units) (dress down) data that search out, input clothing matching expert system module 22, model's sample (or derived units) and dress ornament sample (or derived units) (going up dress) data with the front, with be stored in clothing matching knowledge rule storehouse in advance in data carry out corresponding relatively, rational analysis, infer the collocation condition of next dress ornament sample (or derived units) (as socks).According to new collocation condition, in sample and derived units database 1, search out the next dress ornament sample (or derived units) (as socks) that is fit to again.So circulation is until the complete arranging scheme of last formation.Clothing matching scheme with forming at last is stored in the clothing matching scheme database 4.
3. utilize an artificial neural network module 23, use genetic algorithm to derive the clothing matching scheme.
The clothes arranging scheme has very big difference because of season, occasion, formal dress, leisure.Such as, the clothes number of packages of winter dress, summer clothing is very big with regard to difference.Thereby, be necessary for dissimilar arranging schemes different artificial neural network systems is set.
Be the example explanation now with the comparatively simple male sex between season wear that lies fallow.
The GA coding: definition genetic algorithm chromosome (G1, G2, G3, G4, G5).G1 represents 9 in model's sample (or derived units), and G2 represents dress sample (or derived units) 11, and G3 represents 10 in trousers sample (or derived units), and G4 represents this (or derived units) 6 of the shoe pattern, and G5 represents 4 in accessories sample (or derived units).Totally 40 of whole chromosome string lengths.
Utilize neural network to approach fitness function
Radial basis function neural network (RBFNN) comprises an input layer, a hidden layer and an output layer.Output has only a node, the fitness among the expression GA, and the span of fitness is between 0～1.Hidden layer is made of one group of radial basis function, and the parameter vector relevant with each hidden layer node is A
i(being the center) and B
i(being width), radial basis function has various ways, uses Gaussian function here.
Between network input and the output, can think a kind of mapping relations Y (x): R
n→ R (n is the input node)
In the following formula, nc is the hidden layer node number; ‖ ‖ is an euclideam norm; X, A
i∈ R
n, C
iBe I basis function with output node be connected weights (I=1,2 ...., nc); C
0For adjusting the side-play amount of output.
For given training set P
AB={ (X
(P), f (X
(P))) | 1≤p≤q, (X)
(P)∈ R
n, f (X)
(P)) ∈ R}, the funtcional relationship of using the RBF neural network can approach training set, its approximate error can be expressed as
According to this formula, use the training of gradient descent method, can obtain each parameter of RBF neural network, comprise center, width and the output weights of radial basis function (latent node), obtain the funtcional relationship in the training set.
In the starting stage of system's operation, system becomes the dress ornament arranging scheme with the symphysis of five kinds of samples (or derived units) random groups, and every kind of scheme is corresponding to the individual model of the chromosome in the genetic algorithm.Evaluation to various schemes is provided by Specialty Design teacher, and such evaluation of estimate is this chromosomal fitness.The fitness function of corresponding relation GA between the coding of clothing matching scheme and the evaluation, the set of organizing such relation composition can be used to train the RBF neural network more.Use 240 kinds of arranging scheme structure training sets, obtain each parameter of RBF network, just obtain the fitness function of GA, use this function can calculate each individual fitness in the genetic algorithm.
Decoding technique: set a threshold value, module is saved in clothing matching scheme database 4 with fitness all individual pairing clothing matching schemes greater than this threshold value.At set intervals, the height of user satisfaction is added up.On this basis, the scheme fitness value is adjusted, module is with the fitness function of this adjustment as feedback modifiers RBF neural network.It also is the function that module has study.Also tenant group can be added up respectively different user group's satisfaction, on this basis the scheme fitness value is adjusted respectively.So just can form the RBF neural network that satisfies different user group needs gradually.
Different scheme constructs modes respectively have its relative merits, and different modes is used in combination, and both can keep enough flexible elasticity, again higher efficient can be arranged.
The arranging scheme relevant information data that scheme generation module 2 is generated is kept in the database, constitutes clothing matching scheme database 4.
After a clothing matching scheme forms, from sample and derived units database 1, the picture that constitutes each sample (or derived units) of this scheme is accessed, utilize design sketch generation module 3, it is combined, form the whole structure figure of clothing matching scheme, design sketch is stored in clothing matching scheme database 4 as the part of scheme.
Retrieval
2,3 structure and the flow process that the present invention retrieve is described with reference to the accompanying drawings
The user is connected to the clothing matching system that the present invention is based on scheme database by user interface 10.The user utilizes dress ornament sample determination module 6, selected dress ornament sample.(step S1).
Three kinds of means can be arranged.
1. the user can directly utilize dress ornament sample determination module 6,, finds and the selected retrieval dress ornament sample that needs in sample and derived units database 1 by category index.
2. the user can import relevant information data.From sample and derived units database 1, filter out possible sample by expert data comparing module 62, selected by the user again.A dress ornament sample is the representative of these all concrete furnishingss that divide into groups, thus at color, lining, style, the expert data of each side such as cut out, be not concrete value, but a scope.The precise information that the user of specialty may very clear own product relatively, which scope unclear is.Whether the expert data of expert data comparing module 62 comparison users input is among which range of the sample.Because data might be not exclusively, can be after a plurality of data of comparison, find out immediate generally several, for user oneself selection.
3. the user also can import relevant picture, searches out same or analogous picture by picture retrieval module 61, and is selected by the user again.The picture retrieval module comprises picture searching software, can search out approximate picture in sample and derived units database 1 and dress ornament arranging scheme database 4.By the sequencing of similarity tabulation, supply the user to select the sample that suits, in these pictures as the substitute of former commodity.
Various means can be used alone, but also also mix together.
After the dress ornament sample was determined, the user proposed the collocation request.The system requirements registered user logins (step S2), and registered users is inputed user name, password confirming registered user identity, execution in step S12.Non-registered users execution in step S3.
The non-registered users registered user name, reserving cipher (step S3).System prompt, the user determines one's own model's sample (step S4).If the user wishes to determine execution in step S5 at once.If determine execution in step S14 after the user wishes again.
The user utilizes model's sample determination module 5, determines one's own model's sample (step S5).
Selected model's sample can use 3 kinds of different means.
1. through the clear own affiliated grouping of user of professional image consulting, can judge one's own model's sample.They can directly utilize model's sample determination module 5, in sample and derived units database 1, find one's own model's sample by category index, selected getting final product.But most of user does not also know and oneself belongs to the sort of type.
2. the user can utilize the information sifting module 51 in model's sample determination module 5, by pointing out logging data and click option, the information data that input is necessary.The user presses prompt window, and input age, height are by the option of prompting input various features.
Eye feature
The white of the eye (acid blue, gentle white, cold white)
Eyeball (pale brown look, dark-brown, rose are brown, grey black, light amber, dark brown, oil look, bottle green, pitchy)
Expression in the eyes (bright, active, full of water, soft, sedate, sharp, contrast, have god)
Lip feature (coral, pink, hair powder, blue, whiting, higher red of the purity of purple) partially
Features of skin colors (transparent shallow ivory white, pink colour, rice white, wheat, healthy look, brown, ivory white, drabon look, crineous)
Blush feature (have or do not have)
Hair feature color development (dark brown, yellow, light brown, grey black, dark-brown, brown, Dark grey, black) gloss (soft, glossy)
Contour feature (hale and hearty, middle, soft)
Amount sense feature (large, medium and small)
.........
3. the user also can upload the head portrait that meets certain standard, utilizes people's face type identification module 52 to determine people's face type.People's face type identification module comprises people's face type identification database and pattern recognition module.The user uploads facial picture, compares by the image data in pattern recognition module and the people's face type identification database, lists the high type of similarity.Then, selected by the user.In the present embodiment, do not design special people's face identification system.Just a ready-made two-dimensional static face identification system some little modifications have been carried out.Be provided with people's face type identification database, each type comprises a plurality of individual standard header pictures.Need the head portrait of comparison, compare one by one with all head portraits in people's face type identification database.Calculate with each type between mean difference, the difference minimum be exactly affiliated type.If there is plural type difference to be lower than threshold value, then prompting has multiple suitable type.(, certain distance is still arranged with the requirement of type identification because feature extracting method that uses and algorithm all are directly at recognition of face.It is contemplated that, be purpose with people's face type identification, and specialized designs or improved face identification system should be obtained better effect.) in addition, the head portrait that the user uploads, at least one Zhang Yaoqiu is with Standard colour board as a setting.Photo blee and Standard colour board are as a setting compared, just can assist definite color character.
Various means can be used alone, but also also mix together.
After model's sample was determined, system was saved in customer data base 7 (step S6) with related data.So that when this user logins once more, just can the invoke user data, determine model's sample.System automatically performs step S7 after preserving user data.
Retrieval module 8 is a condition with dress ornament sample and the model's sample of having confirmed, the arranging scheme that stores in the clothing matching scheme database 4 is compared adaptive, retrieves the scheme (step S7) that comprises these samples.Output module 9 will be retrieved qualified scheme content, export user interface 10 to.(step S8).The user can select whether to continue retrieval (step S9), if no longer continue, execution in step S11 finishes retrieval.If continue retrieval, execution in step S10, the user can be in the scheme that has retrieved, and selected interested dress ornament sample can determine that also new dress ornament sample restarts new round retrieval (step S10).After the dress ornament sample is determined, get back to step S7, retrieval module 8 is a condition with dress ornament sample and the model's sample of having confirmed, the scheme that stores in the clothing matching scheme database 4 is compared adaptive, retrieves the scheme (step S7) that comprises these samples.
In step S2, the user who had registered inputs user name, password confirming registered user identity, execution in step S12 on the other hand.This user data (step S12) calls from customer data base 7 in system.Confirm whether this user has confirmed model's sample (step S13).If confirm, execution in step S7 is a conditional information retrieval with dress ornament sample and model's sample.Do not determine model's sample, execution in step S4 as the user.
In step S4, user's inconvenience is determined under the situation of model's sample at once, execution in step S14.Retrieval module 8 is a condition with the dress ornament sample of having confirmed, to the arranging scheme that stores in the clothing matching scheme database 4, compares adaptively, retrieves the scheme that comprises these samples.Output module 9 will be retrieved qualified scheme content, export user interface 10 (step S15) to.The user can select whether to continue retrieval (step S16), if no longer continue, execution in step S11 finishes retrieval.If continue retrieval, execution in step S17.The user can be in the scheme that has retrieved, and selected interested dress ornament sample also can be determined new dress ornament sample, restarts new round retrieval (step S17).After the dress ornament sample is determined, get back to step S14, retrieval module 8 is a condition with the dress ornament sample of having confirmed, to the scheme that stores in the clothing matching scheme database 4, compares adaptively, retrieves the scheme (step S14) that comprises these samples.
It is contemplated that,, show the Commdity advertisement of same grouping at the arranging scheme page empty place that is shown to the user.Equally, because sample is a representative of its place grouping, use real commodity in the same grouping not have what difference as representative.Like this, the picture of a sample also can be used as an advertisement position in fact, offers the commercial user.
Application example 1
Consumer B uses the computer terminal, and by the internet login, the personal account of registering exclusive use becomes registered user B.User B imports the personal information of oneself by prompting: sex, birthdate, educational background, income, hobby, height, build etc., and upload the standard background pictures.System call people face type identification module 52 judges that the facial characteristics of B belongs to the face type: amount sense (containing ratio) during profile is hale and hearty, the colour of skin: shallow warm contrast type.Add the build data of user oneself input: build is tall and thin, sex: the man.Model's sample of determining B is the colour of skin: shallow warm contrast type, face type: amount sense (containing ratio), the tall and thin male sex model of build during profile is hale and hearty.Model's sample and other data of B are set up user information profile, be kept at customer data base 7.
User B browses on shopping website, sees a shirt a1, but uncertain whether being fit to.He gets off picture reproduction, signs in to system with registered user's identity, in picture retrieval window uploading pictures.Picture retrieval module 61 retrieves a plurality of similar pictures in sample and derived units database 1, show.B clicks therein and confirms that a for being fit to sample, clicks the collocation button.According to the information in the customer data base 7, confirmed model's sample of B, retrieval module 8 is a condition with model's sample and the dress ornament sample a of B, and match retrieval scheme in clothing matching scheme database 4 searches out model's sample of comprising and four arranging schemes of dress ornament sample a.Two western-style clothes suit schemes, an easy dress scheme and a summer clothing scheme.With the scheme content that retrieves, export the user interface 10 of user B to through output module 9.
Application example 2
A overcoat f1 has been produced in the F of clothes manufacturer design.The F of manufacturer utilizes expert data comparing module 62, and confirmatory sample f is the representative sample of this grouping.The F of manufacturer has bought the advertisement position of sample f, uses the former figure of the picture replacement sample f of f1.System generates unique cognizance code, sends the F of manufacturer to.F is printed as bar code with cognizance code and is attached on the f1 list marketing.
A leather shoes d1 has been produced in the D of leather shoes manufacturer design.The D of manufacturer utilizes expert data comparing module 62, and confirmatory sample d is the representative sample of this grouping.The D of manufacturer has bought the advertisement position of sample d, uses the former figure of the picture replacement sample d of d1.System generates unique cognizance code and sends the D of manufacturer to.D is printed as bar code with cognizance code and is attached on the d1 list marketing.
Consumer G sees overcoat f1 in the market.The market is provided with computer terminal demonstration inquiry system.She scans the special-purpose bar-code label of having on the clothes f1 with bar-code scanner, computer terminal demonstration inquiry system is connected to clothing matching system automatically, from sample and derived units database 1, access the information of sample f, picture and the advertising message of commodity f1, be shown in user interface 10.G clicks the collocation button and proposes the collocation request, system prompt input username and password.G also is not the registered user, and the personal account that G registers exclusive use becomes registered user G, but temporary transient inconvenient typing personal data directly propose the collocation request.System validation G does not also determine model's sample.Retrieval module 8 is a condition with dress ornament sample f, and match retrieval scheme in clothing matching scheme database 4 searches out the arranging scheme of the dress ornament sample f that comprises.Scheme content with retrieving exports user interface 10 to through output module 9.G has taken a fancy to leather shoes sample d again in the scheme that feeds back, so select the data entries of sample d.System accepts the request of user G, and the information data of sample d and the advertising message of commodity d1 (comprising business informations such as picture, brand, price, purchase approach off the net, net purchase link) are sent to user interface 10.G clicks the collocation button once more and proposes the collocation request, and retrieval module 8 is a condition with leather shoes sample d again, and match retrieval scheme in clothing matching scheme database 4 searches out the arranging scheme of the leather shoes sample d that comprises.Scheme content with retrieving exports user interface 10 to through output module 9.
The principle of native system is applicable to that any machine word developing instrument of making peace develops.Can use in this locality, be more suitable for remote application in the Intranet of the public's internet, in-house network and extranets Extrant etc.The tangible product of other individual differenceizations such as cosmetics, cosmetology all can suitably revised back use the present invention.
Above description and embodiment not delimit the scope of the invention just in order to make people's easy to understand.Any those skilled in the art under the premise without departing from the spirit and scope of the present invention, can easily make numerous modifications and variations.
Claims (10)
1. a kind of clothing matching system based on scheme database of the present invention is characterized in that, comprising: sample and derived units database, scheme generation module, clothing matching scheme database, retrieval module;
Sample and derived units database are in order to stored samples and derived units data;
The scheme generation module is used for according to the image design rule, and the sample (or derived units) in sample and the derived units database is combined into the clothing matching scheme;
The clothing matching scheme database is in order to store the clothing matching scheme related data that described scheme generation module generates;
Retrieval module in order to the sample of determining according to the user, compares adaptively to the scheme that stores in the clothing matching scheme database, retrieve the scheme that comprises these samples;
According to the requirement of image design rule, dressing crowd, Fashion and Accessories are divided into groups; Representative model's sample is established in each dressing crowd grouping; A corresponding representative dress ornament sample is also established in each Fashion and Accessories grouping; Each sample and the version that each is possible thereof are exactly a sample or its derived units; The related data of each sample (or derived units) is stored in sample and the derived units database; According to the image design rule, the scheme generation module is combined into the clothing matching scheme with sample (or derived units); Described clothing matching scheme related data is stored in the clothing matching scheme database; It is condition with one, two or plurality of sample that the user utilizes retrieval module, the scheme in the clothing matching scheme database is compared adaptive, retrieves the clothing matching scheme that comprises these samples.
2. the described scheme generation module of claim 1 is characterized in that: comprise an expert system module, in order to according to the collocation rule in the clothing matching knowledge rule storehouse, derive the clothing matching scheme.
3. the described scheme generation module of claim 1 is characterized in that: comprise an artificial neural network module, derive the clothing matching scheme in order to use genetic algorithm.
4. the described a kind of clothing matching system of claim 1 based on scheme database, it is characterized in that: comprise a design sketch generation module, picture in order to all samples (or derived units) that will constitute a clothing matching scheme combines, and forms the whole structure figure of clothing matching scheme; Design sketch is stored in the described clothing matching scheme database as the part of scheme.
5. the described a kind of clothing matching system of claim 1 based on scheme database, also can comprise a dress ornament sample determination module, it is characterized in that: aforementioned dress ornament sample determination module comprises a picture retrieval module, in order in described sample and derived units database and dress ornament arranging scheme database, find approximate picture; By the sequencing of similarity tabulation, assisted user is selected suitable sample, as the substitute of former commodity, conveniently to utilize described clothing matching system with these pictures.
6. the described dress ornament sample of claim 5 determination module is characterized in that: comprise an expert data comparing module, in order to compare sample and the commodity data in design, production, so that the user seeks suitable alternative sample.
7. the described a kind of clothing matching system of claim 1 based on scheme database, also can comprise model's sample determination module, it is characterized in that: aforementioned model's sample determination module comprises people's face type identification module, belongs to any people's face type in order to identification people face picture; Can accurately judge people's face type by images such as photo, videos by assisted user, select model's sample to make things convenient for the user.
8. clothing matching method based on scheme database is characterized in that: may further comprise the steps:
1.. according to the requirement of image design rule, dressing crowd, Fashion and Accessories are divided into groups;
2.. a sample is established in each grouping, and each sample and the version that each is possible thereof are exactly a sample or its derived units; With the related data of each sample (or derived units), be stored in sample and the derived units database;
3.. according to the image design rule, the sample (or derived units) with in sample and the derived units database is combined into the clothing matching scheme, and the arranging scheme related data with forming is stored in the clothing matching scheme database;
4.. the user is a condition with one, two or plurality of sample, the scheme that retrieval is fit to;
5.. retrieval comprises the clothing matching scheme of these samples in the clothing matching scheme database;
6.. the clothing matching scheme content with retrieving is sent to user interface.
9. the described a kind of clothing matching method based on scheme database of claim 8 is characterized in that: it is as follows that step is added the back:
1.. according to the requirement of image design rule, dressing crowd, Fashion and Accessories are divided into groups;
2.. a sample is established in each grouping; Each sample and the version that each is possible thereof are exactly a sample or its derived units; With the related data of each sample (or derived units), be stored in sample and the derived units database;
3.. according to the image design rule, the sample (or derived units) in sample and the derived units database is combined into the clothing matching scheme, the arranging scheme related data with forming is stored in the clothing matching scheme database;
4.. the user utilizes the picture retrieval module, in sample and derived units database and dress ornament arranging scheme database, searches out approximate picture; By sequencing of similarity tabulation, the user selects the sample that suits, as the substitute of former commodity with these pictures;
5.. the user is a condition with one, two or plurality of sample, the scheme that retrieval is fit to;
6.. in the clothing matching scheme database, retrieval comprises the clothing matching scheme of these samples;
7.. the clothing matching scheme content with retrieving is sent to user interface.
10. the described a kind of clothing matching method based on scheme database of claim 8 is characterized in that: it is as follows that step is added the back:
1.. according to the requirement of image design rule, dressing crowd, Fashion and Accessories are divided into groups;
2.. a sample is established in each grouping, and each sample and the version that each is possible thereof are exactly a sample or its derived units; With the related data of each sample (or derived units), be stored in sample and the derived units database;
3.. according to the image design rule, the sample (or derived units) in sample and the derived units database is combined into the clothing matching scheme, the arranging scheme related data with forming is stored in the clothing matching scheme database;
4.. utilize the design sketch generation module, the picture that constitutes all samples (or derived units) of a clothing matching scheme is combined, form the whole structure figure of clothing matching scheme, design sketch is stored in the described clothing matching scheme database as the part of scheme;
5.. the user is a condition with one, two or plurality of sample, the scheme that retrieval is fit to;
6.. in the clothing matching scheme database, retrieval comprises the clothing matching scheme of these samples;
7.. the clothing matching scheme content with retrieving is sent to user interface.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CNA2009100370506A CN101477560A (en) | 2009-02-02 | 2009-02-02 | Clothing matching system and method based on scheme database |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CNA2009100370506A CN101477560A (en) | 2009-02-02 | 2009-02-02 | Clothing matching system and method based on scheme database |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN101477560A true CN101477560A (en) | 2009-07-08 |
Family
ID=40838276
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CNA2009100370506A Pending CN101477560A (en) | 2009-02-02 | 2009-02-02 | Clothing matching system and method based on scheme database |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN101477560A (en) |
Cited By (29)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103246878A (en) * | 2013-05-13 | 2013-08-14 | 苏州福丰科技有限公司 | Facial-recognition-based trial makeup system and trial makeup method thereof |
| CN103246816A (en) * | 2013-05-13 | 2013-08-14 | 苏州福丰科技有限公司 | Fitting system based on face recognition |
| CN103456032A (en) * | 2013-08-19 | 2013-12-18 | 北京京东尚科信息技术有限公司 | Method and device for processing virtual fitting model images |
| CN104200249A (en) * | 2014-08-26 | 2014-12-10 | 重庆邮电大学 | Automatic clothes matching method, device and system |
| CN104834524A (en) * | 2015-05-04 | 2015-08-12 | 小米科技有限责任公司 | Information prompting method and device |
| CN105117935A (en) * | 2015-08-03 | 2015-12-02 | 武汉中研福世科技有限公司 | Realization method of intelligent dress matching and apparatus thereof |
| CN105224775A (en) * | 2015-11-12 | 2016-01-06 | 中国科学院重庆绿色智能技术研究院 | Based on the method and apparatus that picture processing is arranged in pairs or groups to clothes |
| CN105354411A (en) * | 2015-10-19 | 2016-02-24 | 百度在线网络技术（北京）有限公司 | Information processing method and apparatus |
| CN105765574A (en) * | 2013-10-29 | 2016-07-13 | 皮卡佰公司 | System and method for item and item set matching |
| CN106055893A (en) * | 2016-05-27 | 2016-10-26 | 杭州土网络科技有限公司 | Clothes matching scheme generation method based on fashion template database and automatic matching |
| WO2016188277A1 (en) * | 2015-05-25 | 2016-12-01 | 阿里巴巴集团控股有限公司 | Method and device for providing business object collocation information |
| WO2016188278A1 (en) * | 2015-05-25 | 2016-12-01 | 阿里巴巴集团控股有限公司 | Method and device for collecting information about service object |
| CN106529040A (en) * | 2016-11-12 | 2017-03-22 | 温州大学 | Overall personal image determining and diagnosis system |
| CN106570714A (en) * | 2015-10-08 | 2017-04-19 | 阿里巴巴集团控股有限公司 | Method and apparatus for recommending matching object picture, and method and apparatus for establishing mapping relation |
| CN106651501A (en) * | 2016-10-13 | 2017-05-10 | 雷升庆 | Clothing matching method and system based on human body feature information |
| CN106649692A (en) * | 2016-12-19 | 2017-05-10 | 惠州Tcl移动通信有限公司 | Pushing method and system of costume collocation based on AR equipment |
| CN106776865A (en) * | 2016-11-29 | 2017-05-31 | 雷升庆 | A kind of garment coordination method and system based on user's dynamic need |
| CN106779791A (en) * | 2015-11-25 | 2017-05-31 | 阿里巴巴集团控股有限公司 | A kind of generation method and device of object picture combination of arranging in pairs or groups |
| CN107665238A (en) * | 2017-08-24 | 2018-02-06 | 北京搜狗科技发展有限公司 | Image processing method and device, the device for picture processing |
| CN108256648A (en) * | 2018-04-09 | 2018-07-06 | 重庆科技学院 | A kind of image design robot |
| CN108520060A (en) * | 2018-04-09 | 2018-09-11 | 内蒙古工业大学 | A kind of Mongolian Costumes Digital Display method and system |
| CN108985657A (en) * | 2018-08-06 | 2018-12-11 | 百度在线网络技术（北京）有限公司 | Evaluation method, device and the server of personage's clothing collocation |
| CN109416806A (en) * | 2016-05-10 | 2019-03-01 | 发普有限责任公司 | The system and method for data base entries for linked network platform |
| CN109859537A (en) * | 2019-03-22 | 2019-06-07 | 吉首大学 | A kind of picture-weaving in silk tutoring system and method, information data processing terminal |
| CN110147854A (en) * | 2018-07-19 | 2019-08-20 | 深圳码隆科技有限公司 | Clothes recognition methods, computer equipment and storage medium |
| CN110276360A (en) * | 2019-06-20 | 2019-09-24 | 杭州时趣信息技术有限公司 | Generate computer installation, equipment, storage medium and the method for clothing matching scheme |
| CN111445551A (en) * | 2020-03-26 | 2020-07-24 | 魔立数码科技（上海）有限公司 | Image design method and device, electronic equipment and storage medium |
| CN111767420A (en) * | 2019-09-26 | 2020-10-13 | 北京沃东天骏信息技术有限公司 | Method and device for generating clothing matching data |
| TWI760299B (en) * | 2015-08-06 | 2022-04-11 | 香港商阿里巴巴集團服務有限公司 | Image display method and device for selling object information |
-
2009
- 2009-02-02 CN CNA2009100370506A patent/CN101477560A/en active Pending
Cited By (41)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103246878A (en) * | 2013-05-13 | 2013-08-14 | 苏州福丰科技有限公司 | Facial-recognition-based trial makeup system and trial makeup method thereof |
| CN103246816A (en) * | 2013-05-13 | 2013-08-14 | 苏州福丰科技有限公司 | Fitting system based on face recognition |
| CN103456032B (en) * | 2013-08-19 | 2015-10-07 | 北京京东尚科信息技术有限公司 | A kind of method and apparatus processing virtual fitting model image |
| CN103456032A (en) * | 2013-08-19 | 2013-12-18 | 北京京东尚科信息技术有限公司 | Method and device for processing virtual fitting model images |
| CN105765574A (en) * | 2013-10-29 | 2016-07-13 | 皮卡佰公司 | System and method for item and item set matching |
| CN104200249A (en) * | 2014-08-26 | 2014-12-10 | 重庆邮电大学 | Automatic clothes matching method, device and system |
| CN104200249B (en) * | 2014-08-26 | 2017-07-21 | 重庆邮电大学 | A kind of method of clothing automatic collocation, apparatus and system |
| CN104834524A (en) * | 2015-05-04 | 2015-08-12 | 小米科技有限责任公司 | Information prompting method and device |
| WO2016188278A1 (en) * | 2015-05-25 | 2016-12-01 | 阿里巴巴集团控股有限公司 | Method and device for collecting information about service object |
| CN106294420B (en) * | 2015-05-25 | 2019-11-05 | 阿里巴巴集团控股有限公司 | The method and device of business object collocation information is provided |
| WO2016188277A1 (en) * | 2015-05-25 | 2016-12-01 | 阿里巴巴集团控股有限公司 | Method and device for providing business object collocation information |
| CN106294420A (en) * | 2015-05-25 | 2017-01-04 | 阿里巴巴集团控股有限公司 | The method and device of business object collocation information is provided |
| CN106294419A (en) * | 2015-05-25 | 2017-01-04 | 阿里巴巴集团控股有限公司 | The collecting method of business object information and device |
| CN105117935A (en) * | 2015-08-03 | 2015-12-02 | 武汉中研福世科技有限公司 | Realization method of intelligent dress matching and apparatus thereof |
| CN105117935B (en) * | 2015-08-03 | 2019-02-26 | 武汉中研福世科技有限公司 | The implementation method and device of intelligent collocation dress ornament |
| TWI760299B (en) * | 2015-08-06 | 2022-04-11 | 香港商阿里巴巴集團服務有限公司 | Image display method and device for selling object information |
| CN106570714B (en) * | 2015-10-08 | 2021-01-15 | 阿里巴巴集团控股有限公司 | Recommendation method for matching object picture, and mapping relation establishment method and device |
| CN106570714A (en) * | 2015-10-08 | 2017-04-19 | 阿里巴巴集团控股有限公司 | Method and apparatus for recommending matching object picture, and method and apparatus for establishing mapping relation |
| CN105354411A (en) * | 2015-10-19 | 2016-02-24 | 百度在线网络技术（北京）有限公司 | Information processing method and apparatus |
| CN105224775A (en) * | 2015-11-12 | 2016-01-06 | 中国科学院重庆绿色智能技术研究院 | Based on the method and apparatus that picture processing is arranged in pairs or groups to clothes |
| CN106779791B (en) * | 2015-11-25 | 2021-01-15 | 阿里巴巴集团控股有限公司 | Generation method and device for collocation object picture combination |
| CN106779791A (en) * | 2015-11-25 | 2017-05-31 | 阿里巴巴集团控股有限公司 | A kind of generation method and device of object picture combination of arranging in pairs or groups |
| CN109416806A (en) * | 2016-05-10 | 2019-03-01 | 发普有限责任公司 | The system and method for data base entries for linked network platform |
| CN106055893A (en) * | 2016-05-27 | 2016-10-26 | 杭州土网络科技有限公司 | Clothes matching scheme generation method based on fashion template database and automatic matching |
| CN106055893B (en) * | 2016-05-27 | 2018-08-31 | 杭州一土网络科技有限公司 | Garment coordination scheme generation method based on fashion template library and Auto-matching |
| CN106651501A (en) * | 2016-10-13 | 2017-05-10 | 雷升庆 | Clothing matching method and system based on human body feature information |
| CN106529040A (en) * | 2016-11-12 | 2017-03-22 | 温州大学 | Overall personal image determining and diagnosis system |
| CN106776865A (en) * | 2016-11-29 | 2017-05-31 | 雷升庆 | A kind of garment coordination method and system based on user's dynamic need |
| CN106649692A (en) * | 2016-12-19 | 2017-05-10 | 惠州Tcl移动通信有限公司 | Pushing method and system of costume collocation based on AR equipment |
| CN107665238A (en) * | 2017-08-24 | 2018-02-06 | 北京搜狗科技发展有限公司 | Image processing method and device, the device for picture processing |
| CN108256648A (en) * | 2018-04-09 | 2018-07-06 | 重庆科技学院 | A kind of image design robot |
| CN108520060A (en) * | 2018-04-09 | 2018-09-11 | 内蒙古工业大学 | A kind of Mongolian Costumes Digital Display method and system |
| CN110147854A (en) * | 2018-07-19 | 2019-08-20 | 深圳码隆科技有限公司 | Clothes recognition methods, computer equipment and storage medium |
| CN108985657A (en) * | 2018-08-06 | 2018-12-11 | 百度在线网络技术（北京）有限公司 | Evaluation method, device and the server of personage's clothing collocation |
| CN109859537A (en) * | 2019-03-22 | 2019-06-07 | 吉首大学 | A kind of picture-weaving in silk tutoring system and method, information data processing terminal |
| CN109859537B (en) * | 2019-03-22 | 2022-10-18 | 吉首大学 | Brocade teaching system and method and information data processing terminal |
| CN110276360B (en) * | 2019-06-20 | 2021-07-06 | 杭州时趣信息技术有限公司 | Computer device, equipment, storage medium and method for generating clothing matching scheme |
| CN110276360A (en) * | 2019-06-20 | 2019-09-24 | 杭州时趣信息技术有限公司 | Generate computer installation, equipment, storage medium and the method for clothing matching scheme |
| CN111767420A (en) * | 2019-09-26 | 2020-10-13 | 北京沃东天骏信息技术有限公司 | Method and device for generating clothing matching data |
| CN111445551B (en) * | 2020-03-26 | 2021-03-19 | 魔立数码科技（上海）有限公司 | Image design method and device, electronic equipment and storage medium |
| CN111445551A (en) * | 2020-03-26 | 2020-07-24 | 魔立数码科技（上海）有限公司 | Image design method and device, electronic equipment and storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN101477560A (en) | Clothing matching system and method based on scheme database | |
| CN106202352B (en) | The method of indoor furniture style and colour match design based on Bayesian network | |
| EP4184414A1 (en) | Product recommendation device and method based on image database analysis | |
| US10799010B2 (en) | Makeup application assist device and makeup application assist method | |
| CN104203042B (en) | Makeup auxiliary device, cosmetic auxiliary method and recording medium | |
| JP7489921B2 (en) | Apparatus and method for generating image/text-based designs | |
| US20140089781A1 (en) | Color-based designs | |
| CN108197574B (en) | Character style recognition method, terminal and computer readable storage medium | |
| US20160026926A1 (en) | Clothing matching system and method | |
| CN104951966A (en) | Clothes commodity recommending method and device | |
| CN106354768B (en) | Color-based user and commodity matching method and commodity matching recommendation method | |
| CN109360050A (en) | Personal care garment management and personalized collocation recommendation intelligence system based on perceptual demand | |
| KR101074691B1 (en) | Apparatus and method for providing personal image in on-line interface | |
| Furukawa et al. | Visualisation for analysing evolutionary dynamics of fashion trends | |
| CN117112915B (en) | Intelligent design method and system based on user characteristics and big data training | |
| CN110458292A (en) | A kind of clothes recommended method based on expertise | |
| JP7133177B2 (en) | Evaluation server device for coordination, control method for evaluation server device for coordination, and program and recording medium used therefor | |
| JP6178487B1 (en) | Coordinate system, coordinate device, and program | |
| KR102208690B1 (en) | Apparatus and method for developing style analysis model based on data augmentation | |
| CN112508114A (en) | Intelligent clothing recommendation system and method | |
| US20220343393A1 (en) | System and Method for Fashion Recommendations | |
| Slavinska et al. | Evaluation of color harmony on the scale of psychological perception in family look clothes | |
| CN117689450B (en) | Digital marketing system based on big data | |
| CN110930294A (en) | Intelligent hair style recommendation and presentation method | |
| Tian et al. | User-depth customized men’s shirt design framework based on BI-LSTM |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C12 | Rejection of a patent application after its publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20090708 |