CN105956780A - Garment fitness evaluation method based on Bayesian discrimination theory - Google Patents
Garment fitness evaluation method based on Bayesian discrimination theory Download PDFInfo
- Publication number
- CN105956780A CN105956780A CN201610309023.XA CN201610309023A CN105956780A CN 105956780 A CN105956780 A CN 105956780A CN 201610309023 A CN201610309023 A CN 201610309023A CN 105956780 A CN105956780 A CN 105956780A
- Authority
- CN
- China
- Prior art keywords
- fit
- data
- garment
- clothing
- pressure
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
- G06Q10/063—Operations research, analysis or management
- G06Q10/0639—Performance analysis of employees; Performance analysis of enterprise or organisation operations
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
- G06F16/24—Querying
- G06F16/245—Query processing
- G06F16/2458—Special types of queries, e.g. statistical queries, fuzzy queries or distributed queries
- G06F16/2462—Approximate or statistical queries
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q30/00—Commerce
- G06Q30/06—Buying, selling or leasing transactions
- G06Q30/0601—Electronic shopping [e-shopping]
- G06Q30/0641—Shopping interfaces
- G06Q30/0643—Graphical representation of items or shoppers
Abstract
The invention relates to a garment fitness evaluation method based on a Bayesian discrimination theory, which comprises the steps of collecting prediction model output training data and dividing garments in a database into two categories, wherein one category is well-fitting garments and the other category is unfitting garments; collecting prediction model input training data so as to measure virtual garment pressure; building a garment fitness prediction model based on the Bayesian discrimination theory, wherein an input item of the model is garment pressure, and an output item of the model is whether the garment fits or not; training the built garment fitness prediction model according to the collected prediction model output training data and the prediction model input training data; and carrying out evaluation on the garment fitness through the trained garment fit prediction model. The garment fitness evaluation method can evaluate the garment fitness accurately without truly trying on the garment.
Description
Technical field
The present invention relates to garment fitting assessment technology field, particularly relate to a kind of clothes based on Bayesian discrimination theory
Dress fitness appraisal procedure.
Background technology
Along with the fast development of ecommerce, increasing people selects online purchase clothing.But, online purchase clothing
Maximum shortcoming is to try on, which results in the highest return of goods and rate of exchanging goods.In order to solve this problem, corresponding 3D is empty
Intending fitting software such as, Clo 3D, Lectra 3D Prototype, OptiTex, V-Stitcher 3D etc. develop in succession,
The fitness of clothing is assessed by virtually trying.But the process of this assessment simply observes the effect of virtually trying with eye,
And then judge whether the style of virtual costume is satisfied with and the size of clothing is the most fit.By observing the effect of virtual fitting
Can quickly judge whether style is satisfied with, but whether zoarium is then difficult to judge accurately to garment dimension.
At present, virtually trying fit assessment mainly according to virtual costume pressure-plotting or virtual stress envelope or
Quantity of margin is assessed, but these appraisal procedures remain and judge in the way of eye observation by evaluator.The method is by observer
The impact of my individual specialized capability is very big, lacks science, it is also difficult to convincing, final result is also inaccurate.
Summary of the invention
The technical problem to be solved is to provide the assessment of a kind of garment fitting based on Bayesian discrimination theory
Method, it is possible to the fitness of accurate evaluation clothing.
The technical solution adopted for the present invention to solve the technical problems is: provide a kind of clothes based on Bayesian discrimination theory
Dress fitness appraisal procedure, comprises the following steps:
(1) collecting forecast model output training data so that the clothing in data base to be divided into two classes, a class is fit clothes
Dress, another kind of is the most fit clothing;
(2) forecast model input training data is collected to measure virtual costume pressure;
(3) garment fitting forecast model is built according to based on Bayesian discrimination theory;
(4) forecast model output training data and forecast model according to collecting input training data to constructed clothing
Fitness forecast model is trained;
(5) fitness of clothing is estimated by the garment fitting forecast model by training.
Described step (1) is particularly as follows: examination wearer tries all of clothing in data base on, and data base is divided into two decimals
According to storehouse, be respectively fit garment data and the most fit garment data, and to the clothing in data base put on " 1 " and
" 0 " is to distinguish fit clothing and the most fit clothing.
Described step (2) particularly as follows: arrange multiple pressure measurement point in garment data on the model of each clothing,
Adjust the size of anthropometric dummy used by virtually trying and make the size of anthropometric dummy with to try wearer identical, successively by data base
Template is through in anthropometric dummy, according to the position measurement virtual costume force value of pressure measurement point, obtains the pressure of clothing
Data.
Described step (3) includes following sub-step:
(31) normality of virtual costume pressure data is checked;
(32) covariance of fit garment virtual garment pressure data place zoarium data base and the most fit clothes are checked
Dress virtual costume pressure data place does not conforms to the equality of the covariance in volume data storehouse;
(33) garment fitting forecast model is built:
If the virtual costume of the fit clothing data base at the virtual costume pressure data place of fit clothing and not fit clothing
The most fit garment data at pressure data place broadly falls into normal distribution, if fit clothing data base and the most fit garment data
Covariance matrix be respectively ΣFitAnd ΣThe most fit, and ΣFit=ΣThe most fit;Then based on the pattra leaves that two normal population misclassification loss are equal
This criterion is:
Wherein,
CPDSample to be sentencedVirtual costume pressure data for sample to be sentenced;It it is virtual costume pressure data in fit clothing data base
Average;It it is the average of virtual costume pressure data in the most fit garment data;Σ and fit clothing data base
Covariance matrix ΣFitThe covariance matrix Σ of the most fit garment dataThe most fitEqual;pFitFor fit clothing, data base goes out
Existing prior probability;pThe most fitThe prior probability occurred for the most fit garment data;
If the virtual costume of the fit clothing data base at the virtual costume pressure data place of fit clothing and not fit clothing
The most fit garment data at pressure data place broadly falls into normal distribution, if fit clothing data base and the most fit garment data
Covariance matrix be respectively ΣFitAnd ΣThe most fit, and ΣFit≠ΣThe most fit;Then based on the pattra leaves that two normal population misclassification loss are equal
This criterion is:
Wherein,,
CPDSample to be sentencedVirtual costume pressure data for sample to be sentenced;It it is virtual costume pressure data in fit clothing data base
Average;It it is the average of virtual costume pressure data in the most fit garment data;ΣFitFor fit clothing data
The covariance matrix in storehouse;ΣThe most fitCovariance matrix for the most fit garment data;pFitOccur for fit clothing data base
Prior probability;pThe most fitThe prior probability occurred for the most fit garment data.
Beneficial effect
Owing to have employed above-mentioned technical scheme, the present invention compared with prior art, has the following advantages that and actively imitates
Really:
The present invention need not try on really, only by virtually trying, then measures virtual costume pressure, obtains measuring
Virtual costume pressure data be input in set up forecast model, it is possible to assessment clothing the most fit.With assessment at present
Essential difference is in that of virtually trying fitness, model proposed by the invention is that (Bayes divides algorithm based on machine learning
Class device) from empirical data, carry out supervised learning, with the regular fitness predicting clothing obtained by machine learning.Whole mistake
Journey is made prediction automatically by machine, is totally different from the fitness the most simply assessing virtually trying by mode soon.This
Invention can well be applicable to the fitness assessment of online purchase clothing, and the result of assessment can be supplied to clothing buyer and make
For the reference whether bought.
The present invention the most truly tries being on the increase of data and virtual costume pressure data collecting amount on along with training sample, should
Model can be constantly from these empirical data learnings so that the prediction accuracy of model constantly promotes, say, that instruction
Practicing sample number the most, the present invention predicts that the accuracy of garment fitting is the highest.
Accompanying drawing explanation
Fig. 1 is the output training sample data flow chart collecting forecast model;
Fig. 2 is the input training sample data flow chart collecting forecast model;
Fig. 3 is virtual costume pressure measurement method schematic diagram;
Fig. 4 is garment fitting forecast model figure;
Fig. 5 is the application schematic diagram in garment fitting forecast model shopping on the web.
Detailed description of the invention
Below in conjunction with specific embodiment, the present invention is expanded on further.Should be understood that these embodiments are merely to illustrate the present invention
Rather than restriction the scope of the present invention.In addition, it is to be understood that after having read the content that the present invention lectures, people in the art
The present invention can be made various changes or modifications by member, and these equivalent form of values fall within the application appended claims equally and limited
Scope.
Embodiments of the present invention relate to a kind of garment fitting appraisal procedure based on Bayesian discrimination theory, including with
Lower step: collecting forecast model output training data so that the clothing in data base to be divided into two classes, a class is fit clothing, separately
One class is the most fit clothing;Collect forecast model input training data to measure virtual costume pressure;According to based on Bayes
Differentiating the Theory Construction garment fitting forecast model, the input item of model is garment pressure, the output item of model be clothing whether
Fit;Forecast model output training data and forecast model according to collecting input training data to constructed garment fitting
Forecast model is trained;By the garment fitting forecast model trained, the fitness of clothing is estimated.
The present invention is further illustrated below as a example by trousers.
(1) collection of model output training data
The main purpose of this step is the trousers in data base to be divided into two classes by truly trying on: fit trousers and not
Fit trousers, its flow chart as shown in Figure 1:
1) comprise m+n bar trousers inside an existing trousers data base G, be g respectively1,g2,g3,…,gm+n。
2) idiotype is examination wearer's all trousers of trying in data base respectively of 160/84A, by trying a data on
Trousers in storehouse are divided into two small database: the data base G of fitness trousersFit=(g1,g2,…,gm) and not fit trousers
Data base GThe most fit=(gm+1,gm+2,…,gm+n)。
3) fit trousers labelling " 1 ", the most fit trousers are labeled as " 0 ", this data conduct being made up of " 1 " and " 0 "
The output data of garment fitting forecast model combine input data to be collected and are jointly trained this model (i.e.
Supervised learning).
(2) collection of mode input training data
The main purpose of this step is by virtually trying, measures virtual costume pressure, its flow chart as shown in Figure 2:
1) k pressure measurement point is arranged respectively on the model of the every a pair of trousers in trousers data base G, this k measurement
Point is mainly distributed on the waist of people, buttocks, crotch and these four positions of thigh because these four positions to the fitness of fitted pants and
Comfortableness impact maximum, for any trousers measured, this k the position measuring point is all changeless such as Fig. 3 (a)
Shown in.
2) adjust the size of anthropometric dummy used by virtually trying, be allowed to complete with the size truly trying wearer (160/84A)
Identical.
3) trousers g1Model by the way of virtually trying through adjust after 3D anthropometric dummy on, such as Fig. 3 (b) institute
Show;Then according to the position measurement virtual costume force value (CP measuring point1 1,...,CP1 j,...,CP1 k), as shown in Fig. 3 (c);
The most successively trousers g2,g3,…,gmModel respectively through in same 3D anthropometric dummy, according to k survey of previous definition
Amount point measures garment pressure.Finally obtain garment pressure dataThese data are closed as clothing
The input data of body forecast model combine the output data that previous step collects and jointly are trained (i.e. supervising to this model
Practise).
(3) structure of garment fitting forecast model
The structure of garment fitting forecast model is broadly divided into three steps: data test of normality, covariance matrix phase
Etc. property inspection and the structure of model.
1) virtual costume pressure data CPD test of normality
1. mean vector is calculated by garment pressure matrix data CPDWith covariance matrix Σ.
2. computation sequence statistic CPD(t)To mean vectorHorse formula squared-distance
3. to above-mentioned horse formula squared-distanceCarry out sequence from small to large.
4. calculateAndWhereinMeet
5. with horse formula distance as abscissa,Quantile is that vertical coordinate makees m+n pointPlanar point set figure,
The QQ figure being i.e. distributed.
6. investigate scatterplot whether at one by initial point and straight line that slope is 1, the most then accepting data CPD
From k unit normal population it is assumed that otherwise refusal normal distribution assume.
2) fit trousers virtual costume pressure data CPDFitThe overall G at placeFitThe most fit trousers virtual costume
Pressure data CPDThe most fitThe overall G at placeThe most fitCovariance matrix Test of Equality.
1. null hypothesis H0: Si=S;Alternative hvpothesis H0: Si≠ S, (i=is fit, the most fit), statistical test amount QFitWith
QThe most fit。
QFit=(m-1) [ln | S |-ln | SFit|-k+tr(S-1SFit)];
QThe most fit=(n-1) [ln | S |-ln | SThe most fit|-k+tr(S-1SThe most fit)]。
Wherein:
SFitIt is fit trousers virtual costume pressure data CPDFitCovariance matrix;
SThe most fitIt is the most fit trousers virtual costume pressure data CPDThe most fitCovariance matrix;
M is fit trousers virtual costume pressure data CPDFitIn sample size;
N is the most fit trousers virtual costume pressure data CPDThe most fitIn sample size;
K is the quantity of every trousers pressure measurement point.
2. for given significant level α, ifThen accept H0;No
Then refuse H0。
3) garment fit evaluation model construction
If the 1. virtual costume pressure data CPD of fit trousersFitThe overall G at placeFitFit trousers is not virtual
Garment pressure data CPDThe most fitThe overall G at placeThe most fitBroadly fall into normal distribution.If overall GFitAnd GThe most fitCovariance matrix divide
Wei ΣFitAnd ΣThe most fit, and ΣFit=ΣThe most fit=Σ;Then based on the Bayesian Decision that two normal population misclassification loss are equal
Criterion is:
Wherein
CPDSample to be sentencedIt it is the virtual costume pressure data of sample to be sentenced;
It is overall GFitAverage, with training sample virtual costume pressure data CPD in actual operationFitEqual
Value is estimated
It is overall GThe most fitAverage, with training sample virtual costume pressure data CPD in actual operationThe most fit's
Average is estimated
Σ and overall GFitWith overall GThe most fitCovariance matrix ΣFitAnd ΣThe most fitEqual, actual operation is used training sample
Virtual costume pressure data CPDFitAnd CPDThe most fitThe covariance matrix of mixing sample CPD estimate Σ;
pFitIt is overall GFitThe prior probability occurred, estimates p with m/ (m+n) in actual operationFit;
pThe most fitIt is overall GThe most fitThe prior probability occurred, estimates p with n/ (m+n) in actual operationThe most fit。
If the 2. virtual costume pressure data CPD of fit trousersFitThe overall G at placeFitFit trousers is not virtual
Garment pressure data CPDThe most fitThe overall G at placeThe most fitBroadly fall into normal distribution.If overall GFitAnd GThe most fitCovariance matrix divide
Wei ΣFitAnd ΣThe most fit, and ΣFit≠ΣThe most fit;Then based on the Bayesian principle that two normal population misclassification loss are equal
For:
Wherein,
CPDSample to be sentencedIt is the garment pressure data of sample to be sentenced;
It is overall GFitAverage, with training sample virtual costume pressure data CPD in actual operationFitEqual
Value is estimated
It is overall GThe most fitAverage, with training sample virtual costume pressure data CPD in actual operationThe most fit's
Average is estimated
ΣFitIt is overall GFitCovariance matrix, with training sample virtual costume pressure data CPD in actual operationFit's
Covariance matrix estimates ΣFit;
ΣThe most fitIt is overall GThe most fitCovariance matrix, with training sample virtual costume pressure data in actual operation
CPDThe most fitCovariance matrix estimate ΣThe most fit;
pFitIt is overall GFitThe prior probability occurred, estimates p with m/ (m+n) in actual operationFit;
pThe most fitIt is overall GThe most fitThe prior probability occurred, estimates p with n/ (m+n) in actual operationThe most fit。
Garment fitting forecast model constructed above, for the assessment of trousers fitness, as shown in Figure 4.As long as surveying
The virtual costume pressure CPD of one trousers to be assessed of amountSample to be sentenced=(CP1 Sample to be sentenced, CP2 Sample to be sentenced..., CPk Sample to be sentenced), Clothing Pressure
Force data is input in model, and this model just can predict the fitness situation of clothing automatically.
4. the application of model
According to the above model built, as it is shown in figure 5,
1) after shopping online client selectes a pair of trousers, it is provided that the size data of he or she.
2) according to the size data point reuse 3D anthropometric dummy provided, the size of 3D model and the complete of real human body are made
Equal.
3) in template data storehouse, search for corresponding model according to the selected trousers of client, and model is tried on 3D human body
On model.
4) according to the measurement point defined in Fig. 3 (a), the pressure of virtual costume is measured respectively.
5) pressure data measured is input in the forecast model trained, according to the algorithm output trousers in model
The most fit.If output result display zoarium, then recommend enough to buy;If output result shows the most fit, then recommend again to select
Select a pair of trousers.
It is seen that, the present invention need not try on really, only by virtually trying, then measures virtual costume pressure,
It is input in set up forecast model measuring the virtual costume pressure data obtained, it is possible to assessment clothing are the most fit.
With essential difference is in that of assessment virtually trying fitness at present, model proposed by the invention is algorithm based on machine learning
(Bayes classifier) carries out supervised learning from empirical data, with the regular zoarium predicting clothing obtained by machine learning
Property.Whole process is made prediction automatically by machine, is totally different from and the most simply assesses virtually trying by mode soon
Fitness.The present invention can well be applicable to the fitness assessment of online purchase clothing, and the result of assessment can be supplied to clothes
Dress buyer is as the reference whether bought.
Claims (4)
1. a garment fitting appraisal procedure based on Bayesian discrimination theory, it is characterised in that comprise the following steps:
(1) collecting forecast model output training data so that the clothing in data base to be divided into two classes, a class is fit clothing, separately
One class is the most fit clothing;
(2) forecast model input training data is collected to measure virtual costume pressure;
(3) garment fitting forecast model is built according to based on Bayesian discrimination theory;
(4) fit to constructed clothing according to the forecast model output training data collected and forecast model input training data
Property forecast model is trained;
(5) fitness of clothing is estimated by the garment fitting forecast model by training.
Garment fitting appraisal procedure based on Bayesian discrimination theory the most according to claim 1, it is characterised in that institute
State step (1) particularly as follows: examination wearer tries all of clothing in data base on, and data base is divided into two small database, respectively
For fit garment data and the most fit garment data, and put on " 1 " and " 0 " to distinguish conjunction to the clothing in data base
The clothing of body and the most fit clothing.
Garment fitting appraisal procedure based on Bayesian discrimination theory the most according to claim 2, it is characterised in that institute
State step (2) particularly as follows: arrange multiple pressure measurement point in garment data on the model of each clothing, adjust virtual examination
The size wearing anthropometric dummy used makes the size of anthropometric dummy identical with trying wearer, is worn by the template in data base successively
In anthropometric dummy, according to the position measurement virtual costume force value of pressure measurement point, obtain the pressure data of clothing.
Garment fitting appraisal procedure based on Bayesian discrimination theory the most according to claim 1, it is characterised in that institute
State step (3) and include following sub-step:
(31) normality of virtual costume pressure data is checked;
(32) covariance and the most fit clothing of the garment virtual garment pressure data place zoarium data base that inspection is fit are empty
Intend the equality that garment pressure data place does not conforms to the covariance in volume data storehouse;
(33) garment fitting forecast model is built:
If the virtual costume pressure of the fit clothing data base at the virtual costume pressure data place of fit clothing and not fit clothing
The most fit garment data at force data place broadly falls into normal distribution, if fit clothing data base and the most fit garment data
Covariance matrix is respectively ΣFitAnd ΣThe most fit, and ΣFit=ΣThe most fit;Then based on the pattra leaves that two normal population misclassification loss are equal
This criterion is:
Wherein,,
CPDSample to be sentencedVirtual costume pressure data for sample to be sentenced;It it is virtual costume pressure data in fit clothing data base
Average;It it is the average of virtual costume pressure data in the most fit garment data;Σ and fit clothing data base
Covariance matrix ΣFitThe covariance matrix Σ of the most fit garment dataThe most fitEqual;pFitFor fit clothing, data base goes out
Existing prior probability;pThe most fitThe prior probability occurred for the most fit garment data;
If the virtual costume pressure of the fit clothing data base at the virtual costume pressure data place of fit clothing and not fit clothing
The most fit garment data at force data place broadly falls into normal distribution, if fit clothing data base and the most fit garment data
Covariance matrix is respectively ΣFitAnd ΣThe most fit, and ΣFit≠ΣThe most fit;Then based on the Bayes that two normal population misclassification loss are equal
Criterion is:
Wherein,,
CPDSample to be sentencedVirtual costume pressure data for sample to be sentenced;It it is virtual costume pressure data in fit clothing data base
Average;It it is the average of virtual costume pressure data in the most fit garment data;ΣFitFor fit clothing data
The covariance matrix in storehouse;ΣThe most fitCovariance matrix for the most fit garment data;pFitOccur for fit clothing data base
Prior probability;pThe most fitThe prior probability occurred for the most fit garment data.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201610309023.XA CN105956780A (en) | 2016-05-11 | 2016-05-11 | Garment fitness evaluation method based on Bayesian discrimination theory |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201610309023.XA CN105956780A (en) | 2016-05-11 | 2016-05-11 | Garment fitness evaluation method based on Bayesian discrimination theory |
Publications (1)
Publication Number | Publication Date |
---|---|
CN105956780A true CN105956780A (en) | 2016-09-21 |
Family
ID=56912268
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201610309023.XA Pending CN105956780A (en) | 2016-05-11 | 2016-05-11 | Garment fitness evaluation method based on Bayesian discrimination theory |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN105956780A (en) |
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN109523367A (en) * | 2018-11-29 | 2019-03-26 | 上海与德通讯技术有限公司 | A kind of data analysing method, device, server and storage medium |
CN110879914A (en) * | 2019-10-10 | 2020-03-13 | 西安工程大学 | Trousers virtual try-on fitness evaluation method based on ANFIS |
Citations (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2015155550A1 (en) * | 2014-04-11 | 2015-10-15 | Metail Limited | Garment size recommendation and fit analysis system and method |
-
2016
- 2016-05-11 CN CN201610309023.XA patent/CN105956780A/en active Pending
Patent Citations (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2015155550A1 (en) * | 2014-04-11 | 2015-10-15 | Metail Limited | Garment size recommendation and fit analysis system and method |
Non-Patent Citations (2)
Title |
---|
何慧 等: "基于裤装合体性的青年女性臀部分类", 《东华大学学报(自然科学版)》 * |
齐行祥: "基于个性化虚拟人台的服装合体性评价模型研究", 《中国博士学位论文全文数据库(电子期刊)》 * |
Cited By (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN109523367A (en) * | 2018-11-29 | 2019-03-26 | 上海与德通讯技术有限公司 | A kind of data analysing method, device, server and storage medium |
CN110879914A (en) * | 2019-10-10 | 2020-03-13 | 西安工程大学 | Trousers virtual try-on fitness evaluation method based on ANFIS |
CN110879914B (en) * | 2019-10-10 | 2023-11-24 | 西安工程大学 | ANFIS-based trousers virtual fit evaluation method |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
Devarajan et al. | Validation of female figure identification technique (FFIT) for apparel software | |
CN110367991A (en) | A kind of Falls in Old People methods of risk assessment | |
Byrne | A note on the use of empirical AUC for evaluating probabilistic forecasts | |
JP2016097228A5 (en) | ||
Espitia-Contreras et al. | Development of a Kinect-based anthropometric measurement application | |
CN105956780A (en) | Garment fitness evaluation method based on Bayesian discrimination theory | |
CN106595788A (en) | Large pump station flow monitoring method based on multi-parameter coupling correction | |
McMaster et al. | Validity of vertical jump measurement systems | |
Abtew et al. | Determinations of 3D ease allowance in a virtual environment for customized garment design using fuzzy modelling | |
CN102937645A (en) | Evaluating method for pressure of tight seamless underwear based on double logarithmic model | |
CN106529562A (en) | OSS (Open Source software) project developer prediction method based on Email networks | |
CN105224808B (en) | Projecting integral's function skin condition integrated evaluating method based on three-dimensional coordinate | |
CN112131711B (en) | Sports close-fitting garment fit level prediction method based on probabilistic neural network | |
Ranger et al. | Testing fit of latent trait models for responses and response times in tests | |
Galada et al. | Developing a prediction model for improving bifurcated garment fit for mass customization | |
Olaru et al. | SOFTWARE SOLUTION TO ASSESS MORPHOLOGICAL BODY THROUGH 3D SCANNING RESULTS. | |
Hu et al. | Research on the evaluation method of virtual clothing pressure comfort based on fuzzy clustering | |
CN106128098B (en) | A kind of multi-display apparatus that can carry out traffic flow forecasting | |
Liu et al. | A novel evaluation technique for human body perception of clothing fit | |
KR20240014341A (en) | Fabric characteristic prediction system and method for a smoother prediction of wearing sensation | |
Zhou et al. | The range of uncertainty on the property market pricing: The case of the city of Shanghai | |
LIU et al. | Garment fit evaluation based on bayesian discriminant | |
Xia et al. | Algorithms behind a body measurement app built on smartphones from supplied color-coded garments for online apparel purchasing platforms | |
Hatami et al. | Predicting US Drought Monitor Drought Categories with multiple land surface models and machine learning | |
Devarajan | Validation of'Female Figure Identification Technique (FFIT) for Apparel'methodology |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
C06 | Publication | ||
PB01 | Publication | ||
C10 | Entry into substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
RJ01 | Rejection of invention patent application after publication |
Application publication date: 20160921 |
|
RJ01 | Rejection of invention patent application after publication |