JP2010503914A - Matching the fit of clothing to consumers - Google Patents

Abstract

Disclosed are methods and systems for specifying garment fit and matching individual garment fit to individual consumers. In one embodiment, the method classifies a human body shape based on dimensional and demographic anthropometric data sampled from a population, and statistical analysis of the measured data to determine the appropriate Determining a number of distinct body shapes and a range of measurements characteristic of each body shape. The method includes assigning each body shape a unique name and / or symbol representing a body shape, and / or a body shape code that is a curve. In one embodiment, a recommendation engine is used to specify and match the fit of individual clothing to individual consumers.

Description

  The present invention relates generally to the fields of apparel manufacturing and sales, data processing, and more particularly to a method and system for specifying the fit of a garment for an individual.

  A 1999 study by Kurt Salmon Associates reports that 62% of US consumers are very dissatisfied with the apparel fit. Clothes that don't fit well don't sell, and if sold, they are often returned. The majority of the 40% return rate for women's clothing sold in the United States is due to size and fit defects.

  One cause of fit failure is lack of standards. The US Department of Commerce withdrew the commercial standard for sizing women's apparel on January 20, 1983. Since then, garment manufacturers and retailers have continually changed traditional standards and devised their own sizing schemes. Personal clothing sizes often vary from apparel brand to apparel brand. This is commonly seen in women's clothing. A dress labeled "Size 10" from one manufacturer has a different fit than Size 10 from another manufacturer. One fits well and the other does not fit at all. Consumers must try on all clothing before buying. Even within one size of one manufacturer, fit problems can occur due to wide differences in consumer body shapes.

  In this way, the lack of size standards, combined with unreliable markings, leads to apparel fit failure, which on the other hand has very high apparel return rate, sales loss, brand dissatisfaction, time in the fitting room Cause waste and intense consumer frustration.

  Therefore, there is a need for an easy-to-use and effective method and system that helps individuals find apparel that fits their body shape and lifestyle.

BRIEF SUMMARY Accordingly, there is a need for a method and system for more accurately specifying garment fit and matching well-fitted garments with individual consumers, thereby addressing the above shortcomings. The present invention provides a fit specification method for specifying the fit of a garment and matching the fit of an individual garment to an individual consumer based on a recommendation engine. The fit designation method consists of three processes: classification, fitness evaluation, and personal shopping.

  The classification process classifies the human body by shape and assigns a shape code to each general body shape.

  In one embodiment, there are seven body shapes represented by shape codes numbered 1-7. The classification process also classifies individual garments by body shape and assigns a corresponding shape code or multiple shape codes to each garment. Furthermore, a method is provided for collecting garment profiles, which are data describing individual garments. The garment is then assigned a fit code that is a combination of the garment profile and its shape code. The garment fit code can be stored in a database, printed on a garment label, or embedded in a radio frequency identification (RFID) tag or the like.

  The classification process also classifies individual consumers by body shape and assigns a corresponding shape code to each consumer. In addition, means are provided for collecting consumer profiles, which are data describing individual consumers and their consumer preferences. The consumer is then assigned a fit code that is a combination of his consumer profile and his shape code.

  Accordingly, the consumer can predict the probability of correct fit more accurately by comparing his / her shape code with the shape code of clothing with knowledge of the shape code of his / her body. Consumers can further improve the probability of correct fit by using a fit assessment process that compares their fit code with a garment fit code. The conformity assessment process automatically matches individual consumer fit codes with one or many garment fit codes, while the recommendation engine ranks suitable garments for presentation to consumers. Generate a list.

  The fit specification method includes a personalized shopping process that provides consumers with an individually customized shopping environment that can be embodied in online and / or offline retail situations. Consumers can access the classification and conformity assessment process through their own shopping environment and be encouraged to present recommended clothing that fits their body and can be believed to suit their clothing preferences it can.

  The nature and advantages of the present invention may be better understood from the following detailed description and the accompanying drawings.

It is a simplified block diagram of the fit specification method of the aspect of the present invention. 2A-2C are simplified block diagrams illustrating the classification process of aspects of the present invention. 3A-3B illustrate the circumferential and front dimension measurement techniques of an embodiment of the present invention. 3C-3D illustrate height and length dimension measurement techniques of embodiments of the present invention. FIG. 6 is a simplified block diagram of a conformity assessment process according to an aspect of the present invention. 5A-5F are flowcharts illustrating a conformity assessment process for a fit dress according to an embodiment of the present invention. It is a figure of the matching system of the aspect of this invention. FIG. 6 is a diagram of a clothing display interface according to an aspect of the present invention.

DETAILED DESCRIPTION OF EXEMPLARY EMBODIMENTS These and other aspects of the invention are described in further detail below.

  Aspects of the invention relate to systems and methods for matching clothing fit to an individual. In one aspect, a user submits multiple measurement and lifestyle data to an analysis engine to derive a fit profile. The analysis engine processes multiple measurements and lifestyle data to match the fit of multiple garments. The user's fit profile is classified and the user is given a fit profile code that represents the user's fit profile. In one embodiment, a garment manufacturer, retailer, etc. labels the garment according to a fit profile code based on each garment measurement. The consumer compares his fit profile code with the garment fit profile code to find a garment that matches his fit profile.

  Throughout this specification, the embodiments and examples shown should not be construed as limiting the invention but as examples. In describing embodiments of the present invention, female consumers and female apparel are used as examples for clarity. However, the present invention can be used with various types of apparel, including male and children's apparel, and is not intended to be limited to female apparel.

Fit Specification Process FIG. 1 is a simplified diagram illustrating the fit specification process 100 described herein and its configuration process described herein, namely, a classification process 110, a conformity assessment process 120, and a personalized shopping process 130. FIG. Processes are interdependent and share inputs and outputs. The classification process 110 classifies the human body shape into a set of shapes (in this embodiment, represented by shape codes 1-7). The classification process 110 also assigns both body shape and fit codes to individual consumers and individual garments. Given an individual consumer's shape and fit code, the fitness assessment process 120 calculates a ranked list of recommended clothing that is likely to fit the consumer. The personalized shopping process 130 provides a means for consumers to access the fit specification system via a customized shopping environment, as further described below. Thus, the designation process 100 positions the person in the N-dimensional person space P based on its shape, measurements, etc., positions the garment in the N-dimensional garment space G, and repeats this for all persons and clothes. Thus, it can be described as generating a mapping f: P-> G between a person and clothing.

Classification Process FIGS. 2A-2C illustrate a classification process 110. In one embodiment, the classification process 110 includes an engine 210 that defines a human body shape as a shape code, an engine 220 that classifies individual consumers by shape and fit code, and an engine 230 that classifies individual clothing by shape and fit code. Including.

Defining Human Body Shape FIG. 2A shows an engine 210 that defines a human body shape. In one aspect, engine 210 determines a list of important measurements of the human body. Table 1 lists 21 such measurements used in one embodiment of the present invention.

(Table 1) Body measurements

3A-3D show the location and technique for obtaining such body measurements. In one embodiment, the engine 210 generates a user profile using a human body shape classification algorithm, which is a series of calculations that establish arithmetic and / or geometric relationships between different body measurements. Such human shape classification algorithms consider front and side profiles in both 2D and 3D for each measurement, and relative proportions of specific points on the torso, such as, but not limited to, the ratio of shoulder to hip, Assess the ratio of the shoulder to the bust, the ratio of the bust to the waist, the ratio of the waist to the hips, and the ratio of the weight in the front half of the body. For example, two calculations of the human body shape classification algorithm are as follows.
Shoulder circumference-hip circumference, ie
1C-5C = value 1
(Bust circumference-bust front) / Bust circumference, ie
(2C-7F) / 2C = value 6

  Table 1A lists 13 similar calculations used in one embodiment of the present invention.

(Table 1A) Calculation of body shape classification algorithm

  Referring to FIG. 2A, engine 210 processes body measurements from representative samples of a population and subpopulation (eg, US women, 40-65 years). In one embodiment, engine 210 processes the sample using a human shape classification algorithm. Engine 210 statistically analyzes the results to recognize clustered subsets within the population that each share a common data value. In one embodiment, there are seven such subsets designated body shapes 1-7. Each body shape is defined by a core set of measurements along with an acceptable deviation range from the average for each value. When so defined, these seven body shapes can be assigned to individual consumers and clothing as shape codes 1-7.

Individual Consumer Classification In one embodiment, the engine 220 receives individual consumer physical measurements as listed in Table 1 and FIGS. 3A-3D and inputs the data to a consumer body classification algorithm. . The consumer shape classification algorithm can be implemented using a variety of techniques. For example, a consumer body classification algorithm can be implemented using a series of calculations that establish arithmetic and / or geometric relationships between different body measurements. These calculations follow the transformation of the human body shape classification algorithm, but also include a best fit analysis to determine which body shape is closest to the consumer. The consumer can then be assigned a corresponding shape code.

  In another embodiment, the shape code can also be determined using a three-dimensional (3D) line of body measurements and the relative proportions of shoulder, bust, waist, high hip and hip and knee and the circumference. Such 3D measurements can be used to determine body shape curves in 3D. The 3D measurement of the body can be correlated to a garment and / or garment combination curve that represents the outfit in 3D. A comparison of two 3D measurements can be used to geometrically determine body shape codes.

  FIG. 2B illustrates one embodiment of an engine 220 for generating a consumer fit code using consumer physical measurements and a consumer profile. In one embodiment, the engine 220 collects consumer profiles that are data describing individual consumers and their clothing preferences. For example, Table 2 lists 33 such data points used in one embodiment of the present invention.

(Table 2) Consumer profile data

  As shown in FIG. 2B, engine 220 processes consumer profile data using a consumer fit classification algorithm. The consumer fit classification algorithm encodes each data point according to the type of data represented. For example, the bust tolerance (1002D) is a numeric value; the homeowner (1029D) is a Boolean value; the “brands you usually buy” (1008D) is an alphanumeric value derived from a popular brands matching table . Once coded, the data points are concatenated and combined with the consumer shape code. In one embodiment, the resulting numbers are essentially consumer measurements and profile records, which can be represented as a single compressed and encrypted alphanumeric sequence that is a consumer fit code. .

Individual Garment Classification FIG. 2C illustrates an embodiment of an engine 230 that is used to process garment shape codes and garment profiles to generate garment fit codes. In one embodiment, engine 230 receives individual garment measurements. Table 3 lists 35 similar measurements used in one embodiment of the present invention.

(Table 3) Clothing measurement

  Referring to FIG. 2C, the engine 230 processes clothing measurement data using a clothing shape classification algorithm. In one embodiment, the garment shape classification algorithm can be a series of calculations that establish arithmetic and / or geometric relationships (represented as curves) between various garment measurements. There are important measurements that vary by clothing type. For example, a dress has different measurement points than a jacket or pants. These measurements can be taken from the pattern guide, imported from the CAD representation in the manufacturer's cutting system, or manually imported from the garment itself. To derive a matching shape code, engine 230 compares the garment curve derived from the measurements to the curve represented by each of the seven body shapes, and the garment is suitable for one or more body shapes. Determine whether or not. The curves are compared with front, side and back profiles. As previously mentioned, the curve can also be compared three-dimensionally (ie, 3D) with the volume of the front half of the body while being compared with the volume of the front half of the garment. Once a matching curve is found, the garment is assigned the matching shape code.

  The engine 230 processes the clothing profile. The clothing profile is data describing individual clothing. Table 4 lists examples of 23 such data points used in one embodiment of the present invention.

(Table 4) Clothing profile data

  In one embodiment, engine 230 processes garment profile data using a garment fit classification algorithm. The clothing fit classification algorithm functions in the same manner as the consumer fit classification algorithm described above to derive a result value. The resulting value is a garment measurement and profile record represented as a single compressed, encrypted alphanumeric string that is a garment fit code.

Conformity Assessment Process FIG. 4 shows a conformity assessment process 120. In one embodiment, the fitness assessment process 120 can be used to compare a consumer's fit code with a garment code. As described herein, a fit code is an encapsulation of multiple consumer or clothing attributes, i.e. measurements, profiles, shape codes. In one embodiment, the values of these six attributes (three from the consumer 410, three from the garment 420) are input to the fitness assessment process 120. The data is passed through three corresponding filters: a measurement filter 430, a profile filter 450 and a shape code filter 460. The output of the filter is a ranked, sorted list of clothing to be evaluated, as shown in Table 470. The sort consists of seven “holding bins” 480 (one for each shape in this embodiment) and bin D “prohibit display” (ie, 440 throw away unfit clothing). Furthermore, with each evaluation, the clothing is temporarily assigned a priority code (123D). The priority code determines the rank of the clothing in its holding bin 480. This is very useful when a personalized shopping process is performed online where priority codes are used to determine the order in which matching garments are displayed to consumers as described herein. It is.

  The type of garment (102-104D) necessarily affects which data points are considered in the evaluation and the order of their consideration. For example, the jacket has a shoulder circumference (1C), but not the pants. Therefore, the conformity evaluation rule varies depending on the clothing type.

  As an example of the rules and steps necessary to perform a conformity assessment, consider a fit dress from consumer Jane and designer Leona Edmiston. Table 5 lists the data that makes up Jane's fit code.

(Table 5) Jane's data

  Table 6 lists the data that make up the dress fit code.

(Table 6) Dress data

Measurement Filter As shown in FIG. 4, a measurement filter 430 is used to compare clothing measurements with consumer measurements as part of the measurement filter process.

Circumferential dimension comparison measurement filter 430 compares circumferential dimension data points 1C-5C for clothing and consumers. For example, compare bust circumference (2C), waist circumference (3C), hip circumference (5C), shoulder circumference (1C) and finally high hip circumference (4C). For each data point, the measurement filter 430 calculates the difference between the garment circumferential dimension Cg-the consumer circumferential dimension Cc, as illustrated by the following equation:
x = 1Cg−1Cc

If the result (x) is greater than or equal to 0 and less than or equal to the corresponding garment tolerance (28D-32D), then the measurement filter 430 indicates that the following data: Process points.
If (0 ≦ x ≦ 28 Dg), proceed to the next data point.

  Otherwise, the measurement value filter 430 discards the clothing data in the bin D, and if there is a next clothing, proceeds to the evaluation of that clothing. A flowchart 500A for these calculations is shown in FIG. 5A.

Referring to FIG. 5A and the data in Tables 5 and 6, the bust circumference (2C) of the dress is 34 and Jane's bust is 32. Using the above formula in step 502A
34−32 = 2
It is. In step 504A, 2 is greater than 0 and less than or equal to the dress bust tolerance (29D).
0 ≦ 2 ≦ 2
Accordingly, a match is found at step 504A.

The measurement value filter 430 processes the next data point, ie, waist circumference (3C). In step 506A and 508A, using the above formula
30−29 = 1
0 ≦ 1 ≦ 1.25
Therefore, a match is found at step 508A.

The measurement value filter 430 processes the next data point, namely the hip circumference (5C). Using the above formula in steps 510A and 512A
39−35 = 4
0 ≦ 4 ≦ 4
Therefore, a match is found at step 512A.

The measurement value filter 430 processes the next data point, namely the shoulder circumference (1C). Using the above formula in steps 514A and 516A
37−36.5 = 0.5
0 ≦ 0.5 ≦ 2
Therefore, a match is found at step 516A.

The measurement value filter 430 processes the next data point, namely the high hip circumference (4C). Using the above formula in steps 518A and 520A
34−32 = 2
0 ≦ 2 ≦ 2
Therefore, a match is found at step 520A.

  The measured value filter 430 proceeds to the front comparison calculation. If any of the above steps do not fit, at step 522A the clothing and / or data is discarded.

Front Comparison In one embodiment, the measurement filter 430 compares the front data points 6F-10F for clothing and consumers. For each data point, the measurement filter 430 calculates the front of the garment—the front of the consumer using the following equation:
x = 6Fg-6Fc

If result (x) is greater than or equal to 0 and the corresponding tolerance (also 28D to 32D) x consumer front Fg ÷ consumer circumferential dimension Fc (1C to 5C) or less than or equal to this The measured value filter 430 proceeds to the next data point. This can be represented by the following equation:
If (0 ≦ x ≦ 28Dg ^* (6Fc / 1Cc)), proceed to the next data point.

  Otherwise, the measurement value filter 430 discards the garment in the bin D and proceeds to evaluate the garment if there is a next garment. A flowchart 500B of these calculations is shown in FIG. 5B.

Referring to FIG. 5B and the data in Tables 5 and 6, the dress has 19 front shoulders (6F) and 18 Jane shoulder fronts. At step 502B, a difference between the front shoulder circumference of the garment and the front shoulder circumference of the consumer is determined. This is expressed by the above equation as follows.
19−18 = 1

In step 504B, 1 is greater than 0 and less than the shoulder tolerance of the dress (28D) x consumer's shoulder front (6F) ÷ consumer shoulder circumference (1C), as shown by the following calculation: To be equal to or
0 ≦ 1 ≦ 2 ^* (19 / 36.5)
In step 504B, a match is found.

The measured value filter 430 proceeds to the processing of the next data point, namely the bust front surface (7F). At steps 506B and 508B, the difference between the bust front circumference dimension of the garment and the consumer's bust front circumference dimension is determined and evaluated. For example, if the above formula is applied,
17−17 = 0
0 ≦ 0 ≦ 2 ^* (17/32)
Therefore, a match is found in step 508B.

The measured value filter 430 proceeds to processing the next data point, ie, the waist front (8F). At steps 510B and 512B, the difference between the garment waist front circumference and the consumer waist front circumference is determined and evaluated. For example, if the above formula is applied,
15.5−15 = 0.5
0 ≦ 0.5 ≦ 1.25 ^* (16/29)
Therefore, a match is found in step 512B.

The measurement value filter 430 proceeds to processing the next data point, that is, the front of the high hip (9F). At steps 514B and 516B, the difference between the garment's high hip front circumference and the consumer high hip front circumference is determined and evaluated. For example, if the above formula is applied,
17.75−17 = 0.75
0 ≦ 0.75 ≦ 2 ^* (17/32)
Therefore, a match is found in step 516B.

The measurement value filter 430 proceeds to processing the next data point, namely the hip front (10F). At steps 518B and 520B, the difference between the garment's hip front circumference and the consumer's hip front circumference is determined and evaluated. For example, if the above formula is applied,
20.5−19 = 0.5
0 ≦ 0.5 ≦ 4 ^* (19/35)
Therefore, a match is found in step 520B.

  The measured value filter 430 proceeds to the height comparison. If any of the above steps do not fit, at step 522B, the clothing and / or data is discarded.

Height Comparison In one embodiment, the measurement filter 430 calculates the height and ensures that the difference is greater than zero. The measurement value filter 430 calculates the consumer's shoulder to hem (12H) -clothes shoulder to hem (15H), which can be expressed as:
x = 12Hc−15Hg

If the result (x) is greater than or equal to 0 and less than or equal to the consumer's knee height (17H) + the desired garment length (33D), the measurement filter 430 Process the next data point as shown by the equation.
If (0 ≦ x ≦ 17Hc−33Dg), proceed to the next data point.

  Otherwise, the measurement value filter 430 discards the garment in the bin D and proceeds to evaluate the garment if there is a next garment. A flowchart 500C for these calculations is shown in FIG. 5C.

Referring to FIG. 5C and the data in Tables 5 and 6, Jane's shoulder to hem is 53 and dress's shoulder to hem is 38.75. At step 502C, the difference between consumer Jane's shoulder to hem and clothing shoulder to hem is determined. For example, the above formula is applied.
53−38.75 = 14.5

In step 504C, the calculated difference is evaluated. For example, when Jane's knee height is 17 and the desired length of the dress is 0,
0 ≦ 14.5 ≦ 17 + 0
It becomes.

  At step 504C, a match is found, and the measurement filter 430 can proceed from shoulder to waist (12H).

In one aspect, at step 506C, the measurement filter 430 calculates the difference between the consumer's shoulder height (12H) and the consumer's waist height (14H), which can be expressed as: To do.
x = 12Hc-14Hc

At step 508C, if the result (x) is greater than or equal to 0 and less than or equal to (12H) from the shoulder to the waist of the garment, the measurement filter 430 determines that the next data point at step 510C Process. Step 508C can be expressed using the following equation:
If (0 ≦ x ≦ 12Hg), proceed to the next data point.

Otherwise, the measurement value filter 430 proceeds to step 522C, discards the garment in the bin D, and proceeds to evaluate the garment if there is a next garment. In steps 506C and 508C, referring to the data in Table 5 and Table 6, and applying the above formula
53-39 = 14
0 ≦ 14 ≦ 16.5
It is. In step 508C, a match is found.

  The measurement value filter 430 can proceed to the sleeve comparison process in step 512C.

If in sleeve comparison step 510C, measurement filter 430 determines that the consumer's armhole circumference (19D) is less than or equal to the garment armhole circumference (21D), then measurement filter 430 processes the next data point. Proceed to Step 510C can be represented by the following equation:
If 19Dc ≦ 21Dg, proceed to the next data point.

  Otherwise, proceed to measurement filter 430 step 522C, discard the garment in bin D, and proceed to processing the garment if there is a next garment.

Referring to the data in Table 5 and Table 6, Jane's armhole circumference is 18 and that of the dress is 20. Applying the above formula in step 510C
18 ≦ 20
Therefore, a match is found.

Next, the measurement value filter 430 proceeds to the sleeve length (23D). At steps 512C and 514C, measurement filter 430 determines that garment sleeve length (23D)-garment sleeve tolerance (34D)-consumer arm length (21D) is less than or equal to zero. If so, the measurement filter 430 proceeds to processing the profile using the profile filter 450, as further described below. Step 514C can be represented by the following equation:
If (23Dg−34Dg−21Dc ≦ 0), go to the profile filter 450.

Otherwise, the measurement value filter 430 proceeds to step 522C, discards the garment in the bin D, and proceeds to evaluate the garment if there is a next garment. By applying the above formula with reference to the data in Table 5 and Table 6, a match was found.
(22.75−3−20) ≦ 0
The conformity assessment process 120 can proceed to processing the profile using the profile filter 450.

Profile Filter Referring now to FIG. 5D, in one embodiment, the garment priority code is almost always zero. However, during the conformity assessment process 110, the priority code may be temporarily given a numerical value. If a garment fails the comparison, the garment is penalized by adding a number to its priority code. The priority code determines the order in which clothing is recommended and displayed to the consumer. In one aspect, the higher the priority code of a garment, the more unsuitable the garment is for the consumer and the later the order in which it is displayed to the consumer (e.g., given a lower priority). Clothing with priority code “1” is recommended and displayed prior to clothing with priority code “5” (eg, given higher priority). In one aspect, a lower priority code has a higher priority. During this example, for simplicity, if the comparison fails, “1” is added to the priority code. Note that the value of this penalty is variable and may be weighted for a particular comparison. For example, a failure to conform to a consumer's color preference may impose a penalty of 3 on clothing, while a failure to conform to a consumer's fabric preference may only impose a penalty of 2.

  In one aspect, each consumer profile data point may be assigned a secondary value called an “importance value” to indicate its relative importance to the consumer. The importance value can be used to modify the corresponding penalty value to make the penalty value higher or lower depending on how important that particular aspect of the garment is to the consumer. For example, Jane may feel that a clothing fabric is more important than its color. If so, Jane can give an importance value of 2 to the fabric and an importance value of 1 to the color. Using these importance values to modify the previous example, the clothing color penalty remains 3 (3 * 1 = 3), but the fabric penalty increases from 2 to 4 (2 * 2 = 4). It is clear that it jumps up. In the following example, for simplicity and clarity, it is assumed that all consumer profile data is equally important, no importance value is assigned, and no penalty value correction is calculated.

Desired fit comparison profile filter 450 compares the results of the above circumferential dimension comparison for consumer desired shoulder tolerances, bust, waist and hips (1001D-1004D). If the profile filter 450 determines that the consumer's desired tolerance is less than the garment tolerance, then the profile filter 450 will determine its circumferential dimension by using the consumer tolerance instead of the garment tolerance. Correct the calculation. A flowchart 500D of these calculations is shown in FIG. 5D.

For example, in step 502D, if the consumer's shoulder tolerance (1001D) is less than the clothing shoulder tolerance (28D), in step 504D, the profile filter 450 replaces the clothing shoulder tolerance (28D). Reprocess shoulder circumference calculations by using consumer shoulder tolerance (1001D). Therefore, if the above formula is applied,
x = 1Cg−1Cc
It is. In step 506D, if (0 ≦ x ≦ 28Dg), proceed to the next data point, where x is
x = 1Cg−1Cc
become. In step 506D, if (0 ≦ x ≦ 1001Dc), profile filter 450 processes the next data point.

Referring to FIG. 5D and the data in Tables 5 and 6, Jane prefers a close fit around the shoulder. The desired shoulder tolerance is only 1, which is less than the garment shoulder tolerance 2 used in the unmodified shoulder circumference comparison above. Accordingly, the profile filter 450 substitutes Jane's value and recalculates around the shoulder.
37−36.5 = 0.5
0 ≦ 0.5 ≦ 1

  After steps 510D-522D, Jane's bust, waist and hip tolerances (1002D-1004D) are greater than or equal to the corresponding garment tolerances (29D, 30D and 32D), so these circumferential dimensions need to be recalculated Because there is no, the conformity remains. However, if any of the tolerances do not pass, they will be recalculated as in step 504D and step 506D above, and "1" will be added to the priority code for each recalculation failure. .

  For example, if, at step 506D, the garment fails the circumference calculation, “1” is added to the garment priority code (123D) and the conformity assessment process 120 proceeds to profile comparison. In this example, with the dress passing around the shoulder and the priority code still being equal to 0, the recalculation fit evaluation process 120 proceeds to profile comparison.

A flow chart 500E for profile comparison profile comparison calculation is shown in FIG. 5E. The conformity assessment process 120 compares these four consumer and clothing data points as follows:
In step 502E, if the clothing color (115d) is included in the numerical array of the consumer's color palette (1005d), and in step 506E, the clothing style (118d) is the consumer's desired style (1006d). If included in the numerical array, and in step 508E, the clothing fabric (119d) is included in the numerical array of the consumer desired fabric (1007d), and in step 510E, the retail price of the clothing (107d ) Is less than or equal to the consumer's "I usually use" (1019d), this garment is fit and its priority code is not changed. Otherwise, the conformity assessment process 120 proceeds to step 504E and adds 1 to the clothing priority code each time the comparison fails.

  With reference to the data in Tables 5 and 6, Jane's color, style, fabric and price preferences all fit. The conformity assessment process 120 proceeds to size comparison, still with the priority code 0.

At step 514, conformity assessment process 120 compares the garment manufacturer size (121D) to the consumer's normal size (1012D). This is an array of size values depending on the clothing type. As noted above, it is well known that manufacturer sizes vary among manufacturers and are not consistent within the same manufacturer. Manufacturers often have their own sizing scheme, eg “A” for “10”. Thus, a separate size matching table (not shown) is used to normalize the garment manufacturer size (121D) for use in the size comparison. Referring to our exemplary data in Tables 5 and 6, the garment manufacturer size (121D) is 1. The size comparison table shows that a dress of Leona Edmiston size 1 corresponds to size 8. At step 512E, the fitness assessment process 120 subtracts the normalized garment manufacturer size from the consumer's normal size. In step 514E, if the difference is greater than ± 4, conformity assessment process 120 adds 1 to the priority code. Step 514E can be expressed by the following equation.
If ((1012Dc−121Dg)> ± 4), priority code = priority code + 1

In this example, Jane's normal dress size is 10, and the normalized manufacturer size of the dress is 8.
((10−8)> ± 4) is false, so the dress in this example is still a perfect match and its priority code remains 0.

Shape Code Filter In one embodiment, a holding bin is used for each body shape. As shown, there are seven holding bins 1-7. Based on the garment shape code (101), the garment is placed in its corresponding holding bin. For example, the shape code of the garment can be a number, for example an array of 3, 5, 7. In this case, it is placed in bins 3, 5 and 7. Once the shape of the garment is evaluated, the fitness assessment process 120 proceeds to the fitness assessment for the next garment, if any. A flowchart of these calculations is shown in FIG.

  Referring to the exemplary data in Tables 5 and 6, at step 502F, the shape code of the dress is “1, 5”. Therefore, it is placed in the holding bins 1 and 5. Since this dress shares Jane's shape code 5 and has a priority code 0, it can be recommended to Jane as the best fit. When finished, the conformity assessment process 120 proceeds to step 504F and ends.

  In some situations, a large number of garments are evaluated in a single shopping session. As shown in FIG. 5G, when all garments have been subjected to the conformity assessment process 120, they are sorted in ascending numerical order, and the contents of each holding bin 544 depends on the priority code of each garment in each bin. The garments are then separated by shape codes and given an order from highest to lowest suitability. As described, the garment that is the best fit can be displayed and recommended to the consumer. From the holding bin 544 corresponding to the consumer shape code (100c). In one process, a garment having a priority code of 0 can be selected and displayed to the consumer. In this situation, a garment with a priority code of 0, which is the garment that fits and is most likely to fit the consumer, is the best fit, but those skilled in the art are discouraged to indicate the best fit garment. It will be appreciated that priority codes, symbols, flags, etc. can be used.

  In one embodiment, the consumer may wish to consider clothing that is the best fit, less than perfect for him. If so, the conformity assessment process 120 can also retrieve clothing having a priority code greater than 0 from its shape code holding bin, and retrieve and display in order of suitability according to the priority code. Consumers can also look around clothing of various body shapes. In fact, clothing can be removed, sorted and displayed in a number of ways, as described below.

Personalized Shopping Process FIG. 6 shows a matching system 600 that is used to enable the shopping process. The shopping process allows the consumer to access the classification process 110 and the conformity assessment process 120.

Consumer Classification Process In one aspect, the personal shopping process 130 and the matching system 600 are as follows.
1. Measure the consumer's body,
2. Provide measurements to the classification process 110;
3. Receive and process consumer shape code,
4. Receive consumer profile,
5. Provide profile data to classification process 110;
6. Receive a consumer fit code.

Shopping with Shape Codes Once a consumer knows his shape code, the consumer can use the shape code to help select appropriate clothing. There are several options when choosing appropriate clothing. For example, clothing can be selected at a “conventional” retail store, or an online shopping situation can be used. Simply look for the shape code of the clothing printed on the “hang tag” or label and compare the shape code to your shape code. Of course, you must make sure that the clothing is the correct size, but the shape code gives you confidence that the clothing is appropriate for your body shape.

If a shopping consumer with a fit code finds that he is in an environment where he can access the matching system 600 (an aspect of which is described below), he can use his fit code to Can be used. For example, you can check how much a particular garment meets your needs, shop for a particular type of garment, or browse all types of garments. Because consumer body measurements and preferences can and will change over time, it is likely that they will want and need new clothing over time. Thus, the consumer is likely to visit the personal shopping process 130 repeatedly.

Matching system Matching system 600 is an implementation of fit specification process 100. In one embodiment, the matching system 600 is a computerized system, ie, a collection of hardware and software for data processing and distribution. The system hardware can include or be one or more computers or a combination of multiple computing devices including PCs, PDAs, cell phones, servers, firewalls and routers. The term software as used herein includes any instruction that can be executed on any type of computer processor.

  The system software can be executed in any computer language and can be executed as object code, assembly or machine code, or a combination of these and others. The software can include one or more modules, software, programs, and combinations thereof. The software can be in the form of one or more applications and integrated software, and can include lower level drivers, object code, and other lower level software.

  The software may be any local or remote machine readable medium such as, but not limited to, magnetic media (e.g. hard disk, tape, floppy disk, card media), optical media (e.g. CD, DVD), flash memory products (e.g. memory Sticks, compact flash and others), radio frequency identification tags (RFID), SmartCardsTM and volatile and non-volatile silicon memory products (e.g. random access memory (RAM), programmable read only memory (PROM), Electrically extinguished / programmable read-only memory (EEPROM) and others), as well as stored on paper (eg, printed UPC barcodes) and can be executed from there.

  Matching system 600 may include a web server 644, a file and application server, and a database server 648, each operating in a networked environment that may include a local area network (LAN) 660, a wide area network (WAN), and the Internet 620. Available network. Data is transferred to the matching system 600 and transferred between its components. This can be accomplished in a conventional manner using standard programs of the TCP / IP protocol, including Hypertext Transfer Protocol (HTTP) and File Transfer Protocol (FTP), for transferring various data. Data exchange and interprocess communication can also be facilitated using Extensible Markup Language (XML), a compatible format for data exchange on the Internet and between different vendor databases and different operating systems . Additional and fewer components, units, modules or software, hardware, and other configurations of data structures may also be used to achieve the invention described herein.

  The matching system 600 includes a number of interconnected areas: a fit-specified back end 640, a manufacturer's module 630, and a retail module 610.

  In one embodiment, the fit specification backend 640 includes a web server 644, a file server 650, a database 648, and an application server 652 that house primary storage locations for classification and conformity assessment software. The matching system 600 also includes a master database of individual consumer shapes and fit codes 642 and individual garment shapes and fit codes 654.

  The manufacturer module 630 is software / hardware that enables a manufacturer to obtain the shape and fit code of his / her item. For example, the manufacturer enters garment dimensional measurements and profile data into the manufacturer's module for each garment of a particular size or SKU. This data may be entered manually or automatically through the workstation 634 while interacting with the manufacturer's own internal system, such as the CAD system 632, for example. The data is subjected to a clothing classification engine 230 as described herein. This can be done locally or remotely via the fit specified back end 640. The manufacturer then uses the resulting shape and fit code in the manufacturing process, for example, printing and / or electronically embedding the shape code and fit code on clothing labels, sales tags, RFID tags, etc. Can be drunk.

  The retail module 610 is found in two general areas: “traditional” retail stores and online stores. Typically, retail stores are located in malls, shopping centers, etc., while online stores are typically accessed via a home or office personal computer. Retail module 610 can also be accessed via kiosks, cell phones, PDAs and other stand-alone or networked devices. Through this retail module 610, consumers can use their own shopping process. For example, Jane enters a women's apparel section at a local department store. So Jane takes advantage of the opportunity to receive her own shape and fit code. Jane's body measurements can be made automatically, for example, by a whole body scanner. Jane's profile data is also collected and entered into the retail module 610, which may be a software / hardware package residing on the store or network. The resulting shape and fit code can be returned to Jane in a variety of forms, such as printed receipts, magnetic cards or SmartCard ™.

  In one embodiment, Jane obtains a printed card that is stamped with a unique curve that represents her shape code. Jane can use her shape code to physically browse the store's inventory to find clothing. You can also use your fit code and conformity assessment process to find in-store clothing that fits you. Jane can do this by using a store kiosk, her cell phone, PDA or other portable computing device device. For example, Jane can embed her shape and fit code in a SmartCard ™. Similarly, a blouse shape code and fit code can be embedded in an RFID tag attached to the blouse. A handheld computing device, such as a PDA or mobile phone, reads Jane's SmartCard (TM) and RFID tags and has conformity assessment software embedded in the computing device, or fit code data fit The conformity assessment can be performed on the fly, either by sending it to the backend 640 and displaying the returned results. Thus, using such a method, Jane can either sample individual clothing found on the store shelves or quickly find clothing that fits herself in the store.

  In another embodiment, the shape and fit code may be associated with another number or code, such as an item ID. For example, the manufacturer can use the item ID to look up clothing information stored in a central database. Using this association, for example, any garment identifier stored in an RFID attached to the garment can be linked with a shape and fit code stored in a database, such as database 648.

  On the other hand, Jill prefers to shop from home or office. Jill accesses the retail module 140 via a PC and online store. Jill also takes advantage of the opportunity to receive his own shape and fit code. For example, Jill collects his body measurements using a tape measure and fills in an online form according to the instructions on the screen. Also enter your profile information. This data is sent to the fit specification backend 640 for processing. Jill's shape code returned is displayed to her. Jill can also receive an email containing his shape code and fit code in a printable machine-readable format such as a barcode. As with Jane, the resulting shape and fit code can be physically or electronically sent to Jill in various forms, such as printed receipts, magnetic cards or SmartCard ™. It can also be transferred to Jill's mobile phone as a data file or executable program. When shopping online, Jill can have direct access to the retail module 610 or it can be presented to Jill via an online store that subscribes to the retail module as a world wide web service. By pulling a conformity assessment process for many garments, retail module 610 acts as a garment search engine.

  In one embodiment, the results of multiple garment conformity assessments can be displayed in Jill using a graphical user interface (GUI) 700 as shown in FIG. This GUI interface 700 allows Jill to quickly view and filter the results of conformity assessment queries. Referring to FIG. 7, shape control is the top row 710 showing the seven body shapes of this embodiment. Clicking on a body shape selects the shape, and the rest of page 712 is updated with clothing that matches that body shape. When the interface 700 is first displayed, the consumer's body shape is automatically selected and a matching garment may be displayed on the page 712. Based on the contents of the holding bin obtained from the conformity assessment described herein, the garment can be displayed in the garment area 720. In one aspect, a priority code assigned to each garment can be used to determine the display order of the garments. For example, the best fit garment, i.e., the garment with priority code 0 may be displayed first. The consumer can “look through” clothing by selecting page control 760. Clothing can be displayed with photos, descriptions, order information, shopping cart buttons, and the like. The results of conformity assessment can also be delivered by email to Jane, delivered via cell phone, PDA, physically mailed in the form of a personalized print catalog, or by any other means.

  Additional filter controls 770 can also be displayed. For example, the garment type (102Dg) filter lists various types of matching garments, such as "dress", and the brand (106Dg) filter lists brands and designers, such as Leona Edmiston, and style (118Dg) The filter lists clothing styles such as “romantic”. In one embodiment, a filter can be displayed for any garment profile data point, such as color (115 Dg), fabric (119 Dg), sleeve style (112 Dg), and the like. For example, if the user selects a filter option, such as “jacket”, interface 700 shows all matching garments that are jackets.

  In another aspect, a “check box” style selection interface may be used to account for a number of discrete selections. For example, Jill can click on “Skirt, Pants, Brand A, Romantic and Artistic”. The clothing area can be updated with clothing that satisfies all these filter options.

  The foregoing descriptions of specific embodiments of the present invention have been presented for purposes of illustration and description. It is not intended to be exhaustive or intended to limit the invention to the precise form described, and many modifications and variations are possible in light of the above teaching. The embodiments best explain the principles of the invention and its practical application so that those skilled in the art will make various modifications to suit the invention in a variety of ways and for the specific applications contemplated. It has been selected and described to enable use.

Claims (34)

A method of fitting a garment to an individual comprising the following steps:
Receiving consumer body dimensions;
Determining a first garment classification for the consumer based on the body dimensions of the consumer; and comparing the first garment classification with at least one garment classified by the second garment classification; A step of determining whether or not there is.   The method of claim 1, further comprising providing the consumer with one or more garments having a matching second garment classification upon request from the consumer.   The method of claim 1, further comprising measuring a plurality of dimensions of at least one garment to generate a second garment classification.   The method of claim 1, wherein receiving a consumer's body dimensions includes receiving a plurality of upper body dimensions.   The method of claim 1, wherein receiving a consumer body dimension comprises receiving a plurality of lower body dimensions.   The method of claim 1, wherein determining a first garment classification comprises determining a difference between at least some body dimensions and at least one garment dimension.   The method of claim 1, wherein determining the first garment classification comprises determining a difference between a three-dimensional curve representation of the consumer's body and a three-dimensional curve representation of the at least one garment.   The method of claim 1, further comprising receiving consumer profile data, wherein the first garment classification is determined using the consumer body dimensions and the consumer profile data.   The consumer profile data is at least one garment fit tolerance, or garment designer tolerance, or color tolerance, or garment style tolerance, or garment material tolerance, or application tolerance, or garment country of origin, or 9. The method of claim 8, comprising clothing handling or clothing brand.   The method of claim 1, further comprising determining a priority for each of the plurality of garments using a second garment classification that matches the first garment classification with a tolerance within a predetermined range.   The method of claim 1, further comprising providing an output identifying at least one garment having a second clothing classification that matches the consumer's first clothing classification.   The method of claim 1, wherein the first garment classification includes a body fit code associated with the consumer's body shape.   The method of claim 1, wherein the second garment classification includes at least one body fit code associated with a plurality of consumer body shapes. A method for specifying the fit of a garment comprising the following steps:
Receiving the dimensions of the garment;
Receiving clothing profile data identifying a clothing style or manufacturer;
Determining a first clothing classification based on the clothing dimensions and the clothing profile data; and comparing the first clothing classification with at least one consumer classified in a second clothing classification. A step of determining whether or not there is.   15. The method of claim 14, wherein the garment profile data includes data describing the garment.   The data describing the garment is at least one garment fit tolerance, or garment designer tolerance, or color tolerance, or garment style tolerance, or garment material tolerance, or application tolerance, or garment country of origin, or 16. The method of claim 15, comprising clothing handling or clothing branding.   15. The method of claim 14, wherein the garment dimensions are based on computer design coordinates.   The method of claim 17, wherein the coordinates comprise three dimensional coordinates defining one or more curves of the garment.   15. The method of claim 14, wherein determining the first garment classification includes classifying the garment according to a difference between the anthropometric data and the garment dimensions according to a predetermined dimensional variation.   15. The method of claim 14, wherein the comparing includes filtering the first clothing category and the second clothing category to determine clothing presentation priority.   15. The method of claim 14, wherein the first garment classification includes at least one body fit code associated with at least one consumer figure.   15. The method of claim 14, wherein the second garment classification includes a body fit code associated with the consumer. A system for matching one or more garments to at least one consumer, including:
A body classification engine capable of determining a consumer's body category based on the consumer's multiple body dimensions and the consumer's one or more clothing preferences;
A clothing classification engine capable of determining a body category of clothing for one or more clothing based on at least one clothing attribute associated with clothing dimensions and one or more clothing preferences; and A clothing matching engine that can process the body category and one or more of the body categories of clothing to determine which of the clothing fits a consumer.   24. The system of claim 23, wherein the body classification engine is configured to classify a plurality of consumer body shapes into a plurality of different body fit categories.   The consumer's clothing preference is at least one garment fit tolerance, or garment designer tolerance, or color tolerance, or garment style tolerance, or garment material tolerance, or application tolerance, or garment country of origin, 24. The system of claim 23, comprising handling of clothing.   24. The system of claim 23, wherein the garment classification engine is configured to classify the plurality of garments into a plurality of different body fit categories.   24. The system of claim 23, wherein the consumer's body category includes a body fit code.   24. The system of claim 23, wherein the clothing body category includes at least one body fit code. A method of presenting clothing that fits a consumer's body to the consumer, the method comprising the following steps:
Classifying a plurality of human body shapes into a plurality of different body categories generated from a plurality of body dimensions and a consumer's plurality of clothing preferences;
Classifying a plurality of garments into a plurality of garment categories generated from the garment dimensions and a plurality of garment profiles; and identifying a garment category within a tolerance range of each body category.   30. The method of claim 29, wherein classifying a plurality of human body shapes includes providing a tolerance range for each body category.   30. The method of claim 29, wherein classifying the plurality of garments includes comparing the garment dimensions to the plurality of body dimensions to determine one or more body fit categories to which the plurality of garments belong.   30. The method of claim 29, further comprising the step of presenting the consumer with clothing that fits the consumer's body, or consumer preferences, or the consumer's body shape.   30. The method of claim 29, wherein the body category includes a plurality of different body fit codes.   30. The method of claim 29, wherein the clothing category includes a plurality of different body fit codes.

Images