CN106235525A - A kind of footwear code Forecasting Methodology and system - Google Patents
A kind of footwear code Forecasting Methodology and system Download PDFInfo
- Publication number
- CN106235525A CN106235525A CN201610711957.6A CN201610711957A CN106235525A CN 106235525 A CN106235525 A CN 106235525A CN 201610711957 A CN201610711957 A CN 201610711957A CN 106235525 A CN106235525 A CN 106235525A
- Authority
- CN
- China
- Prior art keywords
- footwear
- dimensional
- code
- measured value
- footwear code
- 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.)
- Granted
Links
Classifications
-
- A—HUMAN NECESSITIES
- A43—FOOTWEAR
- A43D—MACHINES, TOOLS, EQUIPMENT OR METHODS FOR MANUFACTURING OR REPAIRING FOOTWEAR
- A43D1/00—Foot or last measuring devices; Measuring devices for shoe parts
- A43D1/02—Foot-measuring devices
-
- A—HUMAN NECESSITIES
- A43—FOOTWEAR
- A43D—MACHINES, TOOLS, EQUIPMENT OR METHODS FOR MANUFACTURING OR REPAIRING FOOTWEAR
- A43D1/00—Foot or last measuring devices; Measuring devices for shoe parts
- A43D1/02—Foot-measuring devices
- A43D1/027—Shoe fit indicating devices
-
- A—HUMAN NECESSITIES
- A43—FOOTWEAR
- A43D—MACHINES, TOOLS, EQUIPMENT OR METHODS FOR MANUFACTURING OR REPAIRING FOOTWEAR
- A43D1/00—Foot or last measuring devices; Measuring devices for shoe parts
- A43D1/04—Last-measuring devices
-
- A—HUMAN NECESSITIES
- A43—FOOTWEAR
- A43D—MACHINES, TOOLS, EQUIPMENT OR METHODS FOR MANUFACTURING OR REPAIRING FOOTWEAR
- A43D1/00—Foot or last measuring devices; Measuring devices for shoe parts
- A43D1/08—Measuring devices for shoe parts
Landscapes
- Life Sciences & Earth Sciences (AREA)
- Biophysics (AREA)
- Footwear And Its Accessory, Manufacturing Method And Apparatuses (AREA)
Abstract
The present invention relates to footwear code recommended technology field, be specifically related to a kind of footwear code Forecasting Methodology and system.The invention provides a kind of footwear code Forecasting Methodology and system, method therein includes that the measured value according to three-dimensional samples foot type obtains the measured value of three-dimension foot type sample set；Try applicable footwear code on by the foot type of three-dimension foot this concentration of pattern, obtain the footwear code that three-dimension foot type sample set is suitable；By measuring three-dimensional shoe last, obtain three-dimensional footwear type measured value；Footwear code and three-dimensional footwear type measured value that three-dimensional foot type measuring value, three-dimension foot type sample set are suitable pass through the automatic learning model of Prediction Parameters and obtain the Prediction Parameters of three-dimensional footwear type required in footwear code forecast model；Three-dimensional foot type measuring value is obtained according to spatial digitizer；Prediction Parameters, three-dimensional foot type measuring value and three-dimensional footwear type measured value are mutually matched the suitable footwear code of prediction in footwear code forecast model.In the present invention, footwear code Forecasting Methodology and system can speed solve not try on, when purchasing footwear, the problem that can't buy suitable footwear accurately and rapidly.
Description
Technical field
The present invention relates to footwear code recommended technology field, be specifically related to a kind of footwear code Forecasting Methodology and system.
Background technology
Find through market survey, purchase the consumer of footwear on the net in real experiences, be frequently encountered the problem in table 1:
Table 1
| Sequence number | The problem run in real experiences | Impression |
| 1 | The same shoe sizes cause not of uniform size of different brands, to try out applicable size every time | Trouble |
| 2 | Net purchase shoes can not be tried on, it is impossible to experiences size, dare not buy easily | Hesitate |
| 3 | Net purchase shoes size is improper, returns goods and wants prior carriage charges and trouble | Loss |
| 4 | The when of buying footwear to kith and kin, even if knowing that size dare not be bought | Worry |
| 5 | Not at all easy choose favorite style, but inform there is no suitable size | Disappointed |
| 6 | Buy footwear to need to search in Ge great shops or the electricity a large amount of shoes of business's platform and compare every time | Loaded down with trivial details |
As seen from Table 1, " size puzzlement " is the root place causing the problems referred to above.At present, shoes are based on foot length
Distinguishing footwear code, on China's footwear market, existing footwear sizes is formulated according to " China's footwear sizes standard GB/T3293.1-1998 ".
Clear stipulaties in " China's footwear sizes standard GB/T3293.1-1998 ", " interval between length marking " of the various footwear of China's footwear sizes is 10mm, half
Number difference is 5mm.But, reality is, the shoes of a lot of international top-brands and export trade brand in accordance be Europe code, U.S.'s code etc.
International standard；On the other hand, most of footwear enterprise can not design in strict accordance with national standard and produce shoes.Thus cause
Shoes size and size on market can not unified.One people removes to try on the different styles (brand) that indicate identical size
Shoes, are experienced as different.
It addition, foot is an irregular said three-dimensional body, length width and instep height are different.Only in accordance with foot length it is only
Can do nothing to help consumer and find oneself footwear applicable.Therefore, traditional footwear mode of purchasing is that consumer is through for a long time, repeatedly try on
Just can find suitable shoes.
But, can not try on when consumer purchases footwear on the net, cause the low probability of transaction of net purchase user and high goods return and replacement ratio
Rate, invisible adds transaction cost.Whether can provide a kind of new technology: just can be for consumption in the case of need not try on
Person selects suitable shoes, has become as the topic of common relation inside and outside industry.
For the problem that can not try on when solving to purchase on the net footwear, it is long that the method that existing electricity business uses is to provide a foot
With the synopsis of foot width (or bottoms) Yu size, consumer judges the suitable of oneself by synopsis after needing to measure oneself foot
Size.This method causes success rate the lowest because of the uncertainty of consumer's certainty of measurement.Customization business passes through 3-D scanning
The three-dimensional foot type measuring value of technical limit spacing consumer so that certainty of measurement is greatly improved.But follow-up still use comparison
Table determines size.But the relation of size and the foot length (foot width or bottoms) being suitable for not is linear, therefore synopsis
Accuracy rate is the highest.
Have scholar to propose by the surface distance of relatively three-dimensional foot type and three-dimensional shoe last to determine whether properly.This method
Shortcoming be that amount of calculation is very big, thus real-time response can not be realized, little to actual application value.
Therefore, for solving the problems referred to above, the invention provides a kind of footwear code Forecasting Methodology based on weights election and system.
Summary of the invention
For above-mentioned weak point of the prior art, the invention provides a kind of footwear code Forecasting Methodology, including:
Measured value according to three-dimensional samples foot type obtains the measured value of three-dimension foot type sample set；
Try applicable footwear code on by the foot type of three-dimension foot this concentration of pattern, obtain the footwear that three-dimension foot type sample set is suitable
Code；
By measuring three-dimensional shoe last or product shoes, obtain three-dimensional footwear type measured value；
Footwear code and three-dimensional footwear type measured value that described three-dimensional foot type measuring value, described three-dimension foot type sample set are suitable pass through pre-
Survey the automatic learning model of parameter and obtain the Prediction Parameters of three-dimensional footwear type required in footwear code forecast model, wherein, described prediction ginseng
Number includes that weighting parameter, described weighting parameter are obtained by statistical；
Three-dimensional foot type measuring value is obtained according to spatial digitizer；
Described Prediction Parameters, described three-dimensional foot type measuring value and described three-dimensional footwear type measured value are at described footwear code forecast model
In be mutually matched prediction suitable footwear code；Wherein,
It is mutually matched acquisition pre-between described Prediction Parameters, described three-dimensional foot type measuring value and described three-dimensional footwear type measured value
The footwear code set surveyed, predicts suitable footwear code according to the weighting parameter that the footwear code set of described prediction is corresponding.
The present invention moves not try on when footwear code Forecasting Methodology can accurately, quickly solve to purchase footwear and can't buy suitable footwear
Problem.
Further, during described weighting parameter is obtained by statistical,
Initialize weighting parameter；
Described in the measured value of described three-dimension foot type sample set, described three-dimensional footwear type measured value, the input of described Prediction Parameters
Footwear code forecast model, it was predicted that each of obtain in described footwear code forecast model the footwear code collection of matching process；
According to the footwear code that described three-dimension foot type sample set is suitable, obtain the suitable footwear code of described three-dimension foot type sample set with pre-
The difference of the footwear code recorded；
Meansigma methods and the standard side of the difference of footwear code is obtained according to matching process each of in described footwear code forecast model
Difference；
Matching process is ranked up and adjusts the weights of its correspondence by the meansigma methods of the difference according to footwear code and standard variance,
And the weighting parameter after adjusting is normalized；
New weighting parameter is substituted into the checking of described footwear code forecast model, if predictablity rate rises, applies, otherwise
Applicating history weighting parameter.
Further, at described Prediction Parameters, described three-dimensional foot type measuring value and described three-dimensional footwear type measured value described
During footwear code forecast model is mutually matched the suitable footwear code of prediction,
It is mutually matched between described Prediction Parameters, described three-dimensional foot type measuring value and described three-dimensional footwear type measured value and dopes
Different footwear codes；
A corresponding footwear code is obtained according to the weighting parameter that each footwear code doped is corresponding；
The real footwear code being suitable for according to the corresponding footwear code obtained with feedback compares, it was predicted that suitably footwear code.
Further, described Prediction Parameters bag also includes that redundancy parameter, linear compensation parameter and linear segmented compensate ginseng
Number.
Further, footwear code and described three-dimensional footwear type measured value that described three-dimension foot type sample set is suitable are mutually matched, and carry
Take out the three-dimensional footwear type measured value that in the footwear code suitable with described three-dimension foot type sample set, each suitable footwear code-phase is corresponding；
Measured value according to the three-dimensional footwear type measured value extracted and described three-dimension foot type sample set obtains redundancy parameter
Collection；
Described redundancy parameter is obtained according to described redundancy parameter set.
Further, according to the measured value of described three-dimension foot type sample set, described three-dimensional footwear type measured value and described redundancy
The yardage that the every three-dimensional footwear type measured value of degree parameter acquiring is corresponding；
According to the yardage that the footwear code that described three-dimension foot type sample set is suitable is corresponding with the every three-dimensional footwear type measured value of acquisition,
Obtain the difference of the suitable yardage of yardage and the actual feedback of sundry item, and the difference of foot long code and suitable yardage；
The coefficient that the linear relationship of two kinds of differences that matching obtains obtains is described linear compensation parameter.
Further, according to the measured value of described three-dimension foot type sample set, described three-dimensional footwear type measured value and described redundancy
The yardage that the every three-dimensional footwear type measured value of degree parameter acquiring is corresponding；
According to the yardage that the footwear code that described three-dimension foot type sample set is suitable is corresponding with the every three-dimensional footwear type measured value of acquisition,
Obtain difference d1 of the suitable yardage of yardage and the actual feedback of sundry item, and difference d2 of foot long code and suitable yardage；
The segmentation of scope respectively according to described difference d1 and ask for the meansigma methods of described difference d2 in corresponding segments, its
In, the boundary value of segmentation is described linear segmented compensating parameter with the meansigma methods of difference d2 in each segmentation.
Present invention also offers a kind of footwear code prognoses system, including:
The measured value acquiring unit of three-dimension foot type sample set, obtains three-dimension foot for the measured value according to three-dimensional samples foot type
The measured value of type sample set；
The footwear code acquiring unit that three-dimension foot type sample set is suitable, for trying on suitable by the foot type of three-dimension foot this concentration of pattern
The footwear code closed, obtains the footwear code that three-dimension foot type sample set is suitable；
Three-dimensional footwear type measured value acquiring unit, for by measuring three-dimensional shoe last or product shoes, obtains three-dimensional footwear type and surveys
Value；
Prediction Parameters acquiring unit, for by the measured value of described three-dimension foot type sample set, described three-dimension foot type sample set
Suitable footwear code and three-dimensional footwear type measured value pass through three needed for the automatic learning model of Prediction Parameters obtains in footwear code forecast model
Wherein, described Prediction Parameters includes that weighting parameter, described weighting parameter are obtained by statistical to the Prediction Parameters of dimension footwear type；
Three-dimensional foot type measuring value acquiring unit, for obtaining three-dimensional foot type measuring value according to spatial digitizer；
Footwear code predicting unit, for measuring described Prediction Parameters, described three-dimensional foot type measuring value and described three-dimensional footwear type
Value is mutually matched the suitable footwear code of prediction in described footwear code forecast model；Wherein,
It is mutually matched acquisition pre-between described Prediction Parameters, described three-dimensional foot type measuring value and described three-dimensional footwear type measured value
The footwear code set surveyed, predicts suitable footwear code according to the weighting parameter that the footwear code set of described prediction is corresponding.
Further, weighting parameter module acquiring unit, it is used for initializing weighting parameter；
Described in the measured value of described three-dimension foot type sample set, described three-dimensional footwear type measured value, the input of described Prediction Parameters
Footwear code forecast model, it was predicted that each of obtain in described footwear code forecast model the footwear code collection of matching process；
According to the footwear code that described three-dimension foot type sample set is suitable, obtain the suitable footwear code of described three-dimension foot type sample set with pre-
The difference of the footwear code recorded；
Meansigma methods and the standard side of the difference of footwear code is obtained according to matching process each of in described footwear code forecast model
Difference；
Matching process is ranked up and adjusts the weights of its correspondence by the meansigma methods of the difference according to footwear code and standard variance,
And the weighting parameter after adjusting is normalized；
New weighting parameter is substituted into the checking of described footwear code forecast model, if predictablity rate rises, applies, otherwise
Applicating history weighting parameter.
Further, described footwear code predicting unit is in described Prediction Parameters, described three-dimensional foot type measuring value and described three-dimensional
During footwear type measured value is mutually matched the suitable footwear code of prediction in described footwear code forecast model,
It is mutually matched between described Prediction Parameters, described three-dimensional foot type measuring value and described three-dimensional footwear type measured value and dopes
Different footwear codes；
A corresponding footwear code is obtained according to the weighting parameter that each footwear code doped is corresponding；
The real footwear code being suitable for according to the corresponding footwear code obtained with feedback compares, it was predicted that suitably footwear code.
Further, described Prediction Parameters bag also includes that redundancy parameter, linear compensation parameter and linear segmented compensate ginseng
Number.
Further, redundancy parameter acquisition module, for by footwear code suitable for described three-dimension foot type sample set and described
Three-dimensional footwear type measured value is mutually matched, and extracts each suitable footwear code-phase in the footwear code suitable with described three-dimension foot type sample set
Corresponding three-dimensional footwear type measured value；
Measured value according to the three-dimensional footwear type measured value extracted and described three-dimension foot type sample set obtains redundancy parameter
Collection；
Described redundancy parameter is obtained according to described redundancy parameter set.
Further, linear compensation parameter acquisition module, for according to the measured value of described three-dimension foot type sample set, described
The yardage that three-dimensional footwear type measured value three-dimensional footwear type measured value every with described redundancy parameter acquiring is corresponding；
According to the yardage that the footwear code that described three-dimension foot type sample set is suitable is corresponding with the every three-dimensional footwear type measured value of acquisition,
Obtain the difference of the suitable yardage of yardage and the actual feedback of sundry item, and the difference of foot long code and suitable yardage；
The coefficient that the linear relationship of two kinds of differences that matching obtains obtains is described linear compensation parameter.
Further, linear segmented compensating parameter acquisition module, for according to the measured value of described three-dimension foot type sample set,
The yardage that described three-dimensional footwear type measured value three-dimensional footwear type measured value every with described redundancy parameter acquiring is corresponding；
According to the yardage that the footwear code that described three-dimension foot type sample set is suitable is corresponding with the every three-dimensional footwear type measured value of acquisition,
Obtain difference d1 of the suitable yardage of yardage and the actual feedback of sundry item, and difference d2 of foot long code and suitable yardage；
The segmentation of scope respectively according to described difference d1 and ask for the meansigma methods of described difference d2 in corresponding segments, its
In, the boundary value of segmentation is described linear segmented compensating parameter with the meansigma methods of difference d2 in each segmentation.
Use technique scheme, including following Advantageous Effects:
1) problem of " size puzzlement " when the present invention can solve to buy shoes from root；
2) present invention just can select suitable shoes for consumer in the case of need not try on.
Accompanying drawing explanation
Fig. 1 schematically illustrates footwear code Forecasting Methodology schematic flow sheet of the present invention；
Fig. 2 schematically illustrates the present invention and moves footwear code forecast model main logic schematic diagram；
Fig. 3 schematically illustrates redundancy parameter of the present invention automatic learning process schematic diagram；
Fig. 4 schematically illustrates linear compensation parameter of the present invention automatic learning process schematic diagram；
Fig. 5 schematically illustrates linear segmented compensating parameter of the present invention automatic learning process schematic diagram；
Fig. 6 schematically illustrates weighting parameter of the present invention automatic learning process schematic diagram；
Fig. 7 schematically illustrates footwear code prognoses system structural representation of the present invention；
Fig. 8 schematically illustrates Prediction Parameters acquiring unit structural representation of the present invention.
Detailed description of the invention
Below by specific embodiment and combine accompanying drawing the present invention is described in further detail.
As it is shown in figure 1, the invention provides a kind of footwear code Forecasting Methodology, including a kind of footwear code Forecasting Methodology, including: the
One step: obtain the measured value of three-dimension foot type sample set according to the measured value of three-dimensional samples foot type；
Second step: try applicable footwear code on by the foot type of three-dimension foot this concentration of pattern, obtains three-dimension foot type sample set
Suitably footwear code；
Third step: by measuring three-dimensional shoe last or product shoes, obtains three-dimensional footwear type measured value；
4th step: footwear code and three-dimensional footwear type that described three-dimensional foot type measuring value, described three-dimension foot type sample set are suitable are surveyed
Value is by the Prediction Parameters of three-dimensional footwear type required in Prediction Parameters automatic learning model acquisition footwear code forecast model, wherein,
Described Prediction Parameters includes that weighting parameter, described weighting parameter are obtained by statistical；
5th step: obtain three-dimensional foot type measuring value according to spatial digitizer；
6th step: described Prediction Parameters, described three-dimensional foot type measuring value and described three-dimensional footwear type measured value are at described footwear
Code forecast model is mutually matched the suitable footwear code of prediction；Wherein,
It is mutually matched acquisition pre-between described Prediction Parameters, described three-dimensional foot type measuring value and described three-dimensional footwear type measured value
The footwear code set surveyed, predicts suitable footwear code according to the weighting parameter that the footwear code set of described prediction is corresponding.
Footwear code Forecasting Methodology of the present invention does not try on when can accurately, quickly solve to purchase footwear and can't buy asking of suitable footwear
Topic.
Wherein, in the present invention, the measured value of three-dimensional samples foot type is by measuring what three-dimensional scanning device obtained；By surveying
Amount three-dimensional shoe last or product shoes, thus obtain the measured value of key position, it is three-dimensional footwear type measured value.Need explanation
It is that three-dimensional shoe last is measured content and included that original pattern shape, original pattern divide heel line, original pattern axis, original pattern heel endpoint location, original pattern head
Position, original pattern width points outside position, original pattern heel is put inside portion's endpoint location, original pattern little toe position, original pattern big toe position, original pattern width
Heart outer fix, original pattern call in person intracardiac side position, shoe tree system mouth line, system mouth before put position, system staphylion position, head thickness point position,
Sole of the foot toe encloses high point, waistline high point position, shank enclose high point position, pocket with enclose high point position, pocket with enclose with a position, shoe last or hat block after one's death with
Wide outer fix, shoe last or hat block after one's death with wide inner side, shoe tree back curve, shoe tree heel arc bump position, with high point position, shoe tree before
Lift up, shoe tree touchdown point position.
In the embodiment shown in fig. 1, three-dimensional foot type measuring value is that the participation obtained by spatial digitizer scanning is predicted
Three-dimensional foot type on the measured value of one group of key position that gets, three-dimensional foot type measuring content includes sole maximum shape, foot
Outside impression shape, sole aft terminal position, Metatarsophalangeal joint points outside position, the 5th metatarsophalangeal joints points outside position, the first toe
Position outside some position, side, the 5th toe points outside position, the first toe point position, the 5th toe point position, intracardiac outer fix of calling in person, the heel heart
Put, sole touchdown point position, the position that the first toe is high, inside bump position, front shank, bump position, heel arc are dashed forward on front shank
Point position, pocket are with enclosing high point position, pocket with enclosing with a position, ankle inner side, ankle outer fix, and sole of the foot toe encloses long high point position
Put；Its site location is corresponding with three-dimensional footwear type measured value position.If three-dimensional foot type measuring value set is vector F{fi, i=1,2,
L, N}, wherein, N is the quantity of the measured value participating in prediction.
In the embodiment shown in fig. 1, the measured value of three-dimension foot type sample set is all measured values of one group of sample foot type
Set, this sample set suitable footwear code corresponding with the most each sample is as automatic learning model defeated of Prediction Parameters
Enter；If this matrix isWherein M number of samples, is the measurement participating in prediction for N
The quantity of value.
In the embodiment shown in fig. 1, the footwear code that three-dimension foot type sample set is suitable is that the foot type in sample set is by reality
Try the shoes yardage being suitable for the dress of this foot type of acquisition on；If this yardage collection is combined into vector G{gi, i=1,2, L, M}, its
In be number of samples for M.
Above-mentioned Fig. 1 shows the concrete grammar flow process of footwear code prediction according to embodiments of the present invention, in the 5th step, described
Prediction Parameters, described three-dimensional foot type measuring value and described three-dimensional footwear type measured value are mutually matched the shoes yardage collection obtaining prediction
Close；The corresponding weights of shoes yardage set according to described prediction predict suitable footwear code.
Fig. 2 shows footwear code forecast model main logic according to embodiments of the present invention；As in figure 2 it is shown, predict at footwear code
Matching process 1 to matching process K in model is represented and is come this foot type by three-dimensional foot type measuring value and three-dimensional footwear type measured value
All methods that the yardage of these footwear being suitable for dress is predicted.
Wherein, it was predicted that parameter has been each term coefficient, threshold value and the weights etc. needed for matching process 1 to matching process K
Quantize number.The Prediction Parameters used in the present invention includes that redundancy parameter, linear compensation parameter, linear segmented compensate ginseng
Number and weighting parameter.
The process of the weighting election footwear code shown in Fig. 2 is:
S=S × WT=s1×w1+s2×w2+L+sn×wn
Wherein, s is the shoes yardage of prediction；Vector S is the collection of the shoes yardage of matching process 1 to matching process N prediction
Close；Vector W is the set of the weights corresponding to matching process 1 to matching process N shown in Fig. 2, itself and be 1, i.e. have
Therefore, the footwear code of recommending in Fig. 2 is the shoes size being suitable for input foot type dress, predicts the outcome for final.
In the 5th step of Fig. 1, described Prediction Parameters, described three-dimensional foot type measuring value and described three-dimensional footwear type measured value
Between be mutually matched and dope different footwear codes；A phase is obtained according to the weighting parameter that each footwear code doped is corresponding
The footwear code answered；The real footwear code being suitable for according to the corresponding footwear code obtained with feedback compares, it was predicted that suitably footwear code.
In an embodiment of the present invention, it was predicted that parameter is obtained by the flow and method shown in Fig. 3, Fig. 4, Fig. 5 and Fig. 6,
Wherein, Fig. 3 is redundancy parameter of the present invention automatic learning process schematic diagram；Fig. 4 is that linear compensation parameter of the present invention learns automatically
Schematic flow sheet；Fig. 5 is linear segmented compensating parameter of the present invention automatic learning process schematic diagram；Fig. 6 is weighting parameter of the present invention
Automatically learning process schematic diagram.
As it is shown on figure 3, the suitable footwear code of described three-dimension foot type sample set and described three-dimensional footwear type measured value are mutually matched, carry
Take out the three-dimensional footwear type measured value that in the footwear code suitable with described three-dimension foot type sample set, suitably footwear code-phase is corresponding；According to extraction
The three-dimensional footwear type measured value gone out and the measured value of described three-dimension foot type sample set obtain redundancy parameter set；According to described redundancy
Parameter set obtains described redundancy parameter.
Specifically, total in the embodiment shown in Fig. 3, the automatic learning process of redundancy parameter comprises the following steps that
First step: extract the three-dimensional footwear type measured value corresponding to each applicable shoes size g
Second step: trying to achieve redundancy parameter matrix is
Third step: for each measured value, substantially will reject beyond a range of value in R；
4th step: average each measured value (or averaging by yardage segmentation), obtains redundancy ginseng
Number is R{ri, j=1,2, L, N}；If segmentation is averaged (segmentation in the case of standard variance is relatively big), then record point
Section point yardage.
As shown in Figure 4, the automatic learning process of linear compensation parameter comprises the following steps: according to described three-dimension foot type sample set
Measured value, yardage that described three-dimensional footwear type measured value three-dimensional footwear type measured value every with described redundancy parameter acquiring is corresponding；
According to the yardage that the footwear code that described three-dimension foot type sample set is suitable is corresponding with the every three-dimensional footwear type measured value of acquisition,
Obtain the difference of the suitable yardage of yardage and the actual feedback of sundry item, and the difference of foot long code and suitable yardage；
The coefficient that the linear relationship of two kinds of differences that matching obtains obtains is described linear compensation parameter.
Specifically, linear compensation parameter automatically learns idiographic flow and is:
First step: each measured value of three-dimension foot type sample set does interpolation in three-dimensional footwear type measured value matrix H, asks
Each sample in yardage corresponding to this measured value；The yardage corresponding to length of each sample that such as interpolation obtains
Do length code L；
Second step: the difference of the footwear code G that computational length code L is suitable with three-dimension foot type sample set, as the benefit of length code
Repay value
Third step: calculate the footwear code that the yardage of other project in addition to length code is fitted with three-dimension foot type sample set
The difference of G, is set to gather C{C1,C2,L,CN-1}；
4th step: willFit to linear relationship with C, have:
5th step: coefficient set A{a of this linear relationship1,a2,L,aN-1It is required linear compensation parameter.
As it is shown in figure 5, the idiographic flow obtaining linear segmented compensating parameter is as follows:
Measured value according to described three-dimension foot type sample set, described three-dimensional footwear type measured value and described redundancy parameter acquiring
The yardage that every three-dimensional footwear type measured value is corresponding；
According to the yardage that the footwear code that described three-dimension foot type sample set is suitable is corresponding with the every three-dimensional footwear type measured value of acquisition,
Obtain difference d1 of the suitable yardage of yardage and the actual feedback of sundry item, and difference d2 of foot long code and suitable yardage；
The segmentation of scope respectively according to described difference d1 and ask for the meansigma methods of described difference d2 in corresponding segments, its
In, the boundary value of segmentation is described linear segmented compensating parameter with the meansigma methods of difference d2 in each segmentation.
Wherein, specifically, the idiographic flow that linear segmented compensating parameter learns automatically is:
First step: each measured value of three-dimension foot type sample set does interpolation in three-dimensional footwear type measured value matrix H, asks
Each sample in yardage corresponding to this measured value；The yardage corresponding to length of each sample that such as interpolation obtains
Do length code L；
Second step: the difference of the footwear code G that computational length code L is suitable with three-dimension foot type sample set, as the benefit of length code
Repay value
Third step: calculate the footwear code that the yardage of other project in addition to length code is fitted with three-dimension foot type sample set
The difference of G, is set to gather C{C1,C2,L,CN-1}；
4th step: the numerical range of C is divided into n section；
5th step: by corresponding in each segment limitAveraged obtains n offset；
6th step: the offset in all segment end points and corresponding section is required linear segmented compensating parameter.
As shown in Figure 6, the flow process obtaining weighting parameter is: initialize weighting parameter；
By the measured value of described three-dimension foot type sample set, described three-dimensional footwear type measured value, described redundancy parameter, described line
Property compensating parameter and described linear segmented compensating parameter input described footwear code forecast model, it was predicted that obtain footwear code collection；
According to the footwear code that described three-dimension foot type sample set is suitable, obtain the suitable footwear code of described three-dimension foot type sample set with pre-
The difference of the footwear code recorded；
Meansigma methods and the standard side of the difference of footwear code is obtained according to matching process each of in described footwear code forecast model
Difference；
Matching process is ranked up and adjusts the weights of its correspondence by the meansigma methods of the difference according to footwear code and standard variance,
And the weighting parameter after adjusting is normalized；
New weighting parameter is substituted into the checking of described footwear code forecast model, if predictablity rate rises, applies, otherwise
Applicating history weighting parameter.
Specifically, in an embodiment of the present invention, the idiographic flow that weighting parameter learns automatically is:
First step: initialize weighting parameter W{w1,w2,L,wK(or continuing to use history weighting parameter) so that
Second step: every measured value described above and parameter are inputted footwear code forecast model, applies all K couplings
Method prediction obtains footwear code collection S{s1,s2,L,sK}；
Third step: ask for difference J of the footwear code that the suitable footwear code of three-dimension foot type sample set obtains with matching process prediction
=G-S；
4th step: ask for meansigma methods and the standard variance of J by each matching process；
5th step: matching process is ranked up according to the meansigma methods of J and standard variance and adjusts the weights of its correspondence,
Meansigma methods and standard variance are the least, and weights increase is the most；
6th step: the weighting parameter after adjusting is normalized so that
7th step: new weighting parameter is substituted into the checking of footwear code forecast model, if predictablity rate rises, applies,
Otherwise continue to use history weighting parameter.
Complex chart 3 to Fig. 6, obtains the Prediction Parameters needed in the present invention, by pre-for the Prediction Parameters input footwear code got
Survey in model, finally give suitable footwear code.In the present invention, it is assumed that a kind of specific matching process can predict footwear
Code.Then N kind matching process obtains N number of footwear code.Each matching process corresponding weights again, represent this matching process can
Reliability, that is the footwear code of this method prediction has much degree to be believable, and N kind method just has N number of weights；Then by N number of
The footwear code predicted combines the weight computing of its correspondence and goes out a final footwear code as output, and certainly, the footwear code that weights are big is right
Final result contribution is the biggest.
Wherein, it is embodied in the real footwear code being suitable for fed back by contrast and each method predicts the footwear obtained
Code, closer to and closer to the quantity of sample the most, the weight setting of the most this method the highest, real reaction be then
This matching process is more accurate to more people.
It should be noted that each of which kind matching process is all based on the measurement data of foot and footwear or the measurement of shoe tree
The mutual relation design of data, can be any method, illustrate the most one by one.
The above-mentioned footwear code Forecasting Methodology provided for the present invention, corresponding with footwear code Forecasting Methodology, the present invention also provides for one
Footwear code prognoses system.
Fig. 7 shows footwear code prognoses system structure according to embodiments of the present invention, as it is shown in fig. 7, the footwear that the present invention provides
Code prognoses system 700 includes that the footwear code that the measured value acquiring unit 710 of three-dimension foot type sample set, three-dimension foot type sample set are fitted obtains
Take unit 720, three-dimensional footwear type measured value acquiring unit 730, Prediction Parameters acquiring unit 740, three-dimensional foot type measuring value acquisition list
Unit 750 and footwear code predicting unit 760.
Wherein, the measured value acquiring unit 710 of three-dimension foot type sample set, for obtaining according to the measured value of three-dimensional samples foot type
Take the measured value of three-dimension foot type sample set；
The footwear code acquiring unit 720 that three-dimension foot type sample set is suitable, for trying by the foot type of three-dimension foot this concentration of pattern
Wear applicable footwear code, obtain the footwear code that three-dimension foot type sample set is suitable；
Three-dimensional footwear type measured value acquiring unit 730, for by measuring three-dimensional shoe last or product shoes, obtains three-dimensional footwear type
Measured value；
Prediction Parameters acquiring unit 740, for by the measured value of described three-dimension foot type sample set, described three-dimension foot pattern originally
Collect suitable footwear code and described three-dimensional footwear type measured value obtains institute in footwear code forecast model by the automatic learning model of Prediction Parameters
The Prediction Parameters of the three-dimensional footwear type needed, wherein, described Prediction Parameters includes that weighting parameter, described weighting parameter pass through statistical
Obtain；
Three-dimensional foot type measuring value acquiring unit 750, for obtaining three-dimensional foot type measuring value according to spatial digitizer；
Footwear code predicting unit 760, for surveying described Prediction Parameters, described three-dimensional foot type measuring value and described three-dimensional footwear type
Value is mutually matched the suitable footwear code of prediction in described footwear code forecast model；Wherein, described Prediction Parameters, described three-dimensional foot type
It is mutually matched the footwear code set obtaining prediction, according to the footwear code collection of described prediction between measured value and described three-dimensional footwear type measured value
Close corresponding weighting parameter and predict suitable footwear code.
Further, described footwear code predicting unit 760 is in described Prediction Parameters, described three-dimensional foot type measuring value and described three
During dimension footwear type measured value is mutually matched the suitable footwear code of prediction in described footwear code forecast model,
Described redundancy parameter, described linear compensation parameter, described linear segmented compensating parameter, described three-dimensional foot type measuring
Value and described three-dimensional footwear type measured value are mutually matched the shoes yardage set obtaining prediction；
The corresponding weights of shoes yardage set according to described prediction predict suitable footwear code.
Wherein, Fig. 8 shows the structure of Prediction Parameters acquiring unit, as shown in Figure 8, it was predicted that parameter acquiring unit 740 is wrapped
Include redundancy parameter acquisition module 741, linear compensation parameter acquisition module 742, linear segmented compensating parameter acquisition module 743 and
Weighting parameter acquisition module 744.
Further, redundancy parameter acquisition module 741, for the footwear code that described three-dimension foot type sample set is suitable and institute
State three-dimensional footwear type measured value to be mutually matched, extract each suitable footwear code in the footwear code suitable with described three-dimension foot type sample set
Corresponding three-dimensional footwear type measured value；
Measured value according to the three-dimensional footwear type measured value extracted and described three-dimension foot type sample set obtains redundancy parameter
Collection；
Described redundancy parameter is obtained according to described redundancy parameter set.
Further, linear compensation parameter acquisition module 742, for according to the measured value of described three-dimension foot type sample set,
The yardage that described three-dimensional footwear type measured value three-dimensional footwear type measured value every with described redundancy parameter acquiring is corresponding；
According to the yardage that the footwear code that described three-dimension foot type sample set is suitable is corresponding with the every three-dimensional footwear type measured value of acquisition,
Obtain the difference of the suitable yardage of yardage and the actual feedback of sundry item, and the difference of foot long code and suitable yardage；
The coefficient that the linear relationship of two kinds of differences that matching obtains obtains is described linear compensation parameter.
Further, linear segmented compensating parameter acquisition module 743, for the measurement according to described three-dimension foot type sample set
Value, the yardage that described three-dimensional footwear type measured value three-dimensional footwear type measured value every with described redundancy parameter acquiring is corresponding；
According to the yardage that the footwear code that described three-dimension foot type sample set is suitable is corresponding with the every three-dimensional footwear type measured value of acquisition,
Obtain difference d1 of the suitable yardage of yardage and the actual feedback of sundry item, and difference d2 of foot long code and suitable yardage；
The segmentation of scope respectively according to described difference d1 and ask for the meansigma methods of described difference d2 in corresponding segments, its
In, the boundary value of segmentation is described linear segmented compensating parameter with the meansigma methods of difference d2 in each segmentation.
Further, weighting parameter acquisition module 744, it is used for initializing weighting parameter；
By the measured value of described three-dimension foot type sample set, described three-dimensional footwear type measured value, described redundancy parameter, described line
Property compensating parameter and described linear segmented compensating parameter input described footwear code forecast model, it was predicted that obtain footwear code collection；
According to the footwear code that described three-dimension foot type sample set is suitable, obtain the suitable footwear code of described three-dimension foot type sample set with pre-
The difference of the footwear code recorded；
Meansigma methods and the standard side of the difference of footwear code is obtained according to matching process each of in described footwear code forecast model
Difference；
Matching process is ranked up and adjusts the weights of its correspondence by the meansigma methods of the difference according to footwear code and standard variance,
And the weighting parameter after adjusting is normalized；
New weighting parameter is substituted into the checking of described footwear code forecast model, if predictablity rate rises, applies, otherwise
Applicating history weighting parameter.
The present invention just can select suitable shoes for consumer in the case of need not try on；Solve from root
The problem of " size puzzlement " when buying shoes.
The foregoing is only the preferred embodiments of the present invention, be not limited to the present invention, for the skill of this area
For art personnel, the present invention can have various modifications and variations.All within the spirit and principles in the present invention, that is made any repaiies
Change, equivalent, improvement etc., should be included within the scope of the present invention.
Claims (10)
1. a footwear code Forecasting Methodology, it is characterised in that including:
Measured value according to three-dimensional samples foot type obtains the measured value of three-dimension foot type sample set；
Try applicable footwear code on by the foot type of three-dimension foot this concentration of pattern, obtain the footwear code that three-dimension foot type sample set is suitable；
By measuring three-dimensional shoe last or product shoes, obtain three-dimensional footwear type measured value；
Footwear code and described three-dimensional footwear type that the measured value of described three-dimension foot type sample set, described three-dimension foot type sample set are suitable are measured
It is worth the Prediction Parameters being obtained three-dimensional footwear type required in footwear code forecast model by the automatic learning model of Prediction Parameters, wherein, institute
State Prediction Parameters and include that weighting parameter, described weighting parameter are obtained by statistical；
Three-dimensional foot type measuring value is obtained according to spatial digitizer；
Described Prediction Parameters, described three-dimensional foot type measuring value and described three-dimensional footwear type measured value phase in described footwear code forecast model
Coupling predicts suitable footwear code mutually；Wherein,
It is mutually matched between described Prediction Parameters, described three-dimensional foot type measuring value and described three-dimensional footwear type measured value and obtains prediction
Footwear code set, predicts suitable footwear code according to the weighting parameter that the footwear code set of described prediction is corresponding.
Footwear code Forecasting Methodology the most according to claim 1, it is characterised in that
During described weighting parameter is obtained by statistical,
Initialize weighting parameter；
The measured value of described three-dimension foot type sample set, described three-dimensional footwear type measured value, described Prediction Parameters are inputted described footwear code
Forecast model, it was predicted that each of obtain in described footwear code forecast model the footwear code collection of matching process；
According to the footwear code that described three-dimension foot type sample set is suitable, obtain the suitable footwear code of described three-dimension foot type sample set and record in advance
The difference of the footwear code arrived；
Meansigma methods and the standard variance of the difference of footwear code is obtained according to matching process each of in described footwear code forecast model；
Matching process is ranked up and adjusts the weights of its correspondence by the meansigma methods of the difference according to footwear code and standard variance, and right
Weighting parameter after adjustment is normalized；
New weighting parameter is substituted into the checking of described footwear code forecast model, if predictablity rate rises, applies, otherwise apply
History weighting parameter.
Footwear code Forecasting Methodology the most according to claim 1, it is characterised in that
At described Prediction Parameters, described three-dimensional foot type measuring value and described three-dimensional footwear type measured value in described footwear code forecast model
During being mutually matched the suitable footwear code of prediction,
It is mutually matched between described Prediction Parameters, described three-dimensional foot type measuring value and described three-dimensional footwear type measured value and dopes difference
Footwear code；
A corresponding footwear code is obtained according to the weighting parameter that each footwear code doped is corresponding；
The real footwear code being suitable for according to the corresponding footwear code obtained with feedback compares, it was predicted that suitably footwear code.
Footwear code Forecasting Methodology the most according to claim 1, it is characterised in that
Described Prediction Parameters bag also includes redundancy parameter, linear compensation parameter and linear segmented compensating parameter.
Footwear code Forecasting Methodology the most according to claim 4, it is characterised in that
Footwear code and described three-dimensional footwear type measured value that described three-dimension foot type sample set is suitable are mutually matched, and extract and described three-dimensional
The three-dimensional footwear type measured value that in foot type sample set suitable footwear code, each suitable footwear code-phase is corresponding；
Measured value according to the three-dimensional footwear type measured value extracted and described three-dimension foot type sample set obtains redundancy parameter set；
Described redundancy parameter is obtained according to described redundancy parameter set.
Footwear code Forecasting Methodology the most according to claim 4, it is characterised in that
Measured value according to described three-dimension foot type sample set, described three-dimensional footwear type measured value and described redundancy parameter acquiring are every
The yardage that three-dimensional footwear type measured value is corresponding；
According to the yardage that the footwear code that described three-dimension foot type sample set is suitable is corresponding with the every three-dimensional footwear type measured value of acquisition, obtain
The difference of the suitable yardage of the yardage of sundry item and actual feedback, and the difference of foot long code and suitable yardage；
The coefficient that the linear relationship of two kinds of differences that matching obtains obtains is described linear compensation parameter.
Footwear code Forecasting Methodology the most according to claim 4, it is characterised in that
Measured value according to described three-dimension foot type sample set, described three-dimensional footwear type measured value and described redundancy parameter acquiring are every
The yardage that three-dimensional footwear type measured value is corresponding；
According to the yardage that the footwear code that described three-dimension foot type sample set is suitable is corresponding with the every three-dimensional footwear type measured value of acquisition, obtain
Difference d1 of the suitable yardage of the yardage of sundry item and actual feedback, and difference d2 of foot long code and suitable yardage；
The segmentation of scope respectively according to described difference d1 and ask for the meansigma methods of described difference d2 in corresponding segments, wherein,
The boundary value of segmentation is described linear segmented compensating parameter with the meansigma methods of difference d2 in each segmentation.
8. a footwear code prognoses system, it is characterised in that including:
The measured value acquiring unit of three-dimension foot type sample set, obtains three-dimension foot pattern for the measured value according to three-dimensional samples foot type
The measured value of this collection；
The footwear code acquiring unit that three-dimension foot type sample set is suitable, for trying on applicable by the foot type of three-dimension foot this concentration of pattern
Footwear code, obtains the footwear code that three-dimension foot type sample set is suitable；
Three-dimensional footwear type measured value acquiring unit, for by measuring three-dimensional shoe last or product shoes, obtains three-dimensional footwear type measured value；
Prediction Parameters acquiring unit, for the footwear measured value of described three-dimension foot type sample set, described three-dimension foot type sample set fitted
Code and described three-dimensional footwear type measured value obtain three-dimensional footwear required in footwear code forecast model by the automatic learning model of Prediction Parameters
The Prediction Parameters of type, wherein, described Prediction Parameters includes that weighting parameter, described weighting parameter are obtained by statistical；
Three-dimensional foot type measuring value acquiring unit, for obtaining three-dimensional foot type measuring value according to spatial digitizer；Footwear code predicting unit,
For by described Prediction Parameters, described three-dimensional foot type measuring value and described three-dimensional footwear type measured value in described footwear code forecast model
It is mutually matched the suitable footwear code of prediction；Wherein,
It is mutually matched between described Prediction Parameters, described three-dimensional foot type measuring value and described three-dimensional footwear type measured value and obtains prediction
Footwear code set, predicts suitable footwear code according to the weighting parameter that the footwear code set of described prediction is corresponding.
Footwear code prognoses system the most according to claim 8, it is characterised in that
Weighting parameter module acquiring unit, is used for initializing weighting parameter；
The measured value of described three-dimension foot type sample set, described three-dimensional footwear type measured value, described Prediction Parameters are inputted described footwear code
Forecast model, it was predicted that each of obtain in described footwear code forecast model the footwear code collection of matching process；
According to the footwear code that described three-dimension foot type sample set is suitable, obtain the suitable footwear code of described three-dimension foot type sample set and record in advance
The difference of the footwear code arrived；
Meansigma methods and the standard variance of the difference of footwear code is obtained according to matching process each of in described footwear code forecast model；
Matching process is ranked up and adjusts the weights of its correspondence by the meansigma methods of the difference according to footwear code and standard variance, and right
Weighting parameter after adjustment is normalized；
New weighting parameter is substituted into the checking of described footwear code forecast model, if predictablity rate rises, applies, otherwise apply
History weighting parameter.
Footwear code prognoses system the most according to claim 8, it is characterised in that
Described footwear code predicting unit at described Prediction Parameters, described three-dimensional foot type measuring value and described three-dimensional footwear type measured value in institute
State during footwear code forecast model is mutually matched the suitable footwear code of prediction,
It is mutually matched between described Prediction Parameters, described three-dimensional foot type measuring value and described three-dimensional footwear type measured value and dopes difference
Footwear code；
A corresponding footwear code is obtained according to the weighting parameter that each footwear code doped is corresponding；
The real footwear code being suitable for according to the corresponding footwear code obtained with feedback compares, it was predicted that suitably footwear code；
Described Prediction Parameters bag also includes redundancy parameter, linear compensation parameter and linear segmented compensating parameter；
Redundancy parameter acquisition module, for the footwear code that described three-dimension foot type sample set is suitable and described three-dimensional footwear type measured value
It is mutually matched, extracts the three-dimensional footwear type that in the footwear code suitable with described three-dimension foot type sample set, each suitable footwear code-phase is corresponding
Measured value；
Measured value according to the three-dimensional footwear type measured value extracted and described three-dimension foot type sample set obtains redundancy parameter set；
Described redundancy parameter is obtained according to described redundancy parameter set；
Linear compensation parameter acquisition module, measures for the measured value according to described three-dimension foot type sample set, described three-dimensional footwear type
It is worth the yardage that three-dimensional footwear type measured value every with described redundancy parameter acquiring is corresponding；
According to the yardage that the footwear code that described three-dimension foot type sample set is suitable is corresponding with the every three-dimensional footwear type measured value of acquisition, obtain
The difference of the suitable yardage of the yardage of sundry item and actual feedback, and the difference of foot long code and suitable yardage；
The coefficient that the linear relationship of two kinds of differences that matching obtains obtains is described linear compensation parameter；
Linear segmented compensating parameter acquisition module, for the measured value according to described three-dimension foot type sample set, described three-dimensional footwear type
The yardage that measured value three-dimensional footwear type measured value every with described redundancy parameter acquiring is corresponding；
According to the yardage that the footwear code that described three-dimension foot type sample set is suitable is corresponding with the every three-dimensional footwear type measured value of acquisition, obtain
Difference d1 of the suitable yardage of the yardage of sundry item and actual feedback, and difference d2 of foot long code and suitable yardage；
The segmentation of scope respectively according to described difference d1 and ask for the meansigma methods of described difference d2 in corresponding segments, wherein,
The boundary value of segmentation is described linear segmented compensating parameter with the meansigma methods of difference d2 in each segmentation.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610711957.6A CN106235525B (en) | 2016-08-23 | 2016-08-23 | A kind of shoes code prediction technique and system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610711957.6A CN106235525B (en) | 2016-08-23 | 2016-08-23 | A kind of shoes code prediction technique and system |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN106235525A true CN106235525A (en) | 2016-12-21 |
| CN106235525B CN106235525B (en) | 2018-11-09 |
Family
ID=57596116
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201610711957.6A Active CN106235525B (en) | 2016-08-23 | 2016-08-23 | A kind of shoes code prediction technique and system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN106235525B (en) |
Cited By (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106937774A (en) * | 2017-03-30 | 2017-07-11 | 李文谦 | A kind of shoe tree size Forecasting Methodology and prediction meanss based on deep learning |
| CN107316221A (en) * | 2017-05-26 | 2017-11-03 | 深圳市云智数据服务有限公司 | A kind of personalized online shopping platform analyzed based on big data and implementation method |
| CN107609928A (en) * | 2017-07-24 | 2018-01-19 | 北京知足科技有限公司 | A kind of children's shoes code recommends method and device |
| CN108960262A (en) * | 2017-05-19 | 2018-12-07 | 意礴科技有限公司 | A kind of methods, devices and systems and computer readable storage medium for predicting shoes code |
| CN109064260A (en) * | 2018-07-11 | 2018-12-21 | 北京知足科技有限公司 | A kind of shoe style data capture method and device |
| CN109102017A (en) * | 2018-08-09 | 2018-12-28 | 百度在线网络技术（北京）有限公司 | Neural network model processing method, device, equipment and readable storage medium storing program for executing |
| CN111175973A (en) * | 2019-12-31 | 2020-05-19 | Oppo广东移动通信有限公司 | Head band adjusting method and adjusting device, computer storage medium and head-mounted equipment |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| KR20010068019A (en) * | 2001-04-14 | 2001-07-13 | 최우성 | The sales system of shoes on internet |
| WO2009072000A2 (en) * | 2007-09-14 | 2009-06-11 | Wei Shi | Fit prediction methods for virtual fitting of footwear to a customer |
| US8463657B1 (en) * | 2010-04-01 | 2013-06-11 | Joe Bentvelzen | Self-help system and method for selling footwear |
| WO2014185979A1 (en) * | 2013-05-14 | 2014-11-20 | Parkes Ashley | Method for systematic design and production of footwear components |
| CN104778599A (en) * | 2014-10-29 | 2015-07-15 | 刘莎 | Three-dimensional scanning and comparison-based online shoe and boot trying-on system and method |
| CN105243547A (en) * | 2015-09-17 | 2016-01-13 | 王刚 | Footwear customization service platform and customization method thereof |
-
2016
- 2016-08-23 CN CN201610711957.6A patent/CN106235525B/en active Active
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| KR20010068019A (en) * | 2001-04-14 | 2001-07-13 | 최우성 | The sales system of shoes on internet |
| WO2009072000A2 (en) * | 2007-09-14 | 2009-06-11 | Wei Shi | Fit prediction methods for virtual fitting of footwear to a customer |
| US8463657B1 (en) * | 2010-04-01 | 2013-06-11 | Joe Bentvelzen | Self-help system and method for selling footwear |
| WO2014185979A1 (en) * | 2013-05-14 | 2014-11-20 | Parkes Ashley | Method for systematic design and production of footwear components |
| CN104778599A (en) * | 2014-10-29 | 2015-07-15 | 刘莎 | Three-dimensional scanning and comparison-based online shoe and boot trying-on system and method |
| CN105243547A (en) * | 2015-09-17 | 2016-01-13 | 王刚 | Footwear customization service platform and customization method thereof |
Cited By (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106937774A (en) * | 2017-03-30 | 2017-07-11 | 李文谦 | A kind of shoe tree size Forecasting Methodology and prediction meanss based on deep learning |
| CN106937774B (en) * | 2017-03-30 | 2022-05-20 | 李文谦 | Shoe tree size prediction method and device based on deep learning |
| CN108960262A (en) * | 2017-05-19 | 2018-12-07 | 意礴科技有限公司 | A kind of methods, devices and systems and computer readable storage medium for predicting shoes code |
| CN108960262B (en) * | 2017-05-19 | 2024-02-09 | 意礴数字科技(深圳)有限公司 | Method, device and system for predicting shoe codes and computer readable storage medium |
| CN107316221A (en) * | 2017-05-26 | 2017-11-03 | 深圳市云智数据服务有限公司 | A kind of personalized online shopping platform analyzed based on big data and implementation method |
| CN107609928A (en) * | 2017-07-24 | 2018-01-19 | 北京知足科技有限公司 | A kind of children's shoes code recommends method and device |
| CN109064260A (en) * | 2018-07-11 | 2018-12-21 | 北京知足科技有限公司 | A kind of shoe style data capture method and device |
| CN109102017A (en) * | 2018-08-09 | 2018-12-28 | 百度在线网络技术（北京）有限公司 | Neural network model processing method, device, equipment and readable storage medium storing program for executing |
| CN109102017B (en) * | 2018-08-09 | 2021-08-03 | 百度在线网络技术（北京）有限公司 | Neural network model processing method, device, equipment and readable storage medium |
| CN111175973A (en) * | 2019-12-31 | 2020-05-19 | Oppo广东移动通信有限公司 | Head band adjusting method and adjusting device, computer storage medium and head-mounted equipment |
Also Published As
| Publication number | Publication date |
|---|---|
| CN106235525B (en) | 2018-11-09 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN106235525B (en) | A kind of shoes code prediction technique and system | |
| US10119814B2 (en) | Determining a size of an item based on comparisons of dimensional and stretch data | |
| US20150242929A1 (en) | Method and system for improving size-based product recommendations using aggregated review data | |
| CN107358505A (en) | A kind of size for purchase clothing online recommends method and system | |
| US7346421B2 (en) | Production of made to order clothing | |
| US20170039622A1 (en) | Garment size recommendation and fit analysis system and method | |
| CN105869040A (en) | Matching system, matching device and foot and shoe matching method and system | |
| US10108160B2 (en) | Re-creating an internal shape of a clothing item | |
| CN107103514A (en) | Commodity distinguishing label determines method and apparatus | |
| US20090234489A1 (en) | Fitting Systems | |
| JP7463455B2 (en) | Information processing device, information processing method, and program | |
| CN105243547A (en) | Footwear customization service platform and customization method thereof | |
| WO2019074594A1 (en) | Using artificial intelligence to determine a value for a variable size component | |
| CN109034973A (en) | Method of Commodity Recommendation, device, system and computer readable storage medium | |
| CN108960985A (en) | Body parameter generation method and online shopping item recommendation method based on image or video | |
| CN105354718B (en) | Intelligent Selection shoes method and device | |
| CN103793826A (en) | Clothes size recommendation method and system | |
| CN104268761A (en) | Background product recommendation decision-making assisting method and system based on consumption features | |
| US10453251B2 (en) | Systems and methods for automated measurement of foot size using electronic sensors | |
| CN108829847A (en) | Commodity search method and system based on multi-modal shopping preferences | |
| GB2567061A (en) | System integration for design and production of clothing | |
| CN105469301A (en) | Shopping method, system and user terminal equipment | |
| US20160350819A1 (en) | Comfort-based garment management | |
| WO2019033963A1 (en) | Transaction method and system based on physical store | |
| CN107609933A (en) | A kind of foot-measuring method for customizing based on shops's conclusion of the business record, system |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |