KR20170016578A - Clothes Fitting System And Operation Method of Threof - Google Patents

Abstract

The present invention relates to a clothes fitting system and an operating method thereof. The clothes fitting system comprises: a crawler to periodically collect an external clothes image and external clothes information of an external site; a repository to collect a user clothes image and user clothes information from a user; a feature point mapper to extract feature points from the external clothes image and the user clothes image; a learning device to use the feature points to generate first pattern information of the external clothes image and second pattern information of the user clothes image; and an external interface to receive a target clothes image from the user to receive third pattern information for the target clothes image from the learning device, and compare the third pattern information with the first and the second pattern information to derive a fitting prediction result for the target clothes image.

Description

[0001] Clothes Fitting System and Operation Method [0002]

The following embodiments relate to a garment fitting system and method of operation thereof.

One of the biggest operational issues for online apparel retailers or offline apparel retailers is related to customer return processing. The main reason for the return is that the customer did not get the expected fit when they actually wore it after the purchase. The fit here refers not to the degree of matching in terms of design, but rather to the fit of the wearer when he wears his or her body, for example, when clothes are tight or loose, or when arms are short or legs are long.

In the case of the seller, the processing cost due to the return means a decrease in the operating profit. On the other hand, the return from the customer means that the desired product can not be used in the near future, and sometimes the return is not troublesome, There are a lot of things that happen frequently when you pay the amount without satisfying the unwanted goods.

However, in order to solve these problems, some of the virtual fitting solutions utilizing camera-based augmented reality (AR) technology have been introduced along with the recent technology development, The bar is designed to wear a virtual image that is modeled in computer graphics (CG) in advance to my due diligence taken by the camera. It is the main purpose to guess the feeling of. However, this technique has limitations that can not accurately determine whether the garment fits well with my physical characteristics (eg, when the stomach is longer than the arm, if it is out of the stomach). In other words, even if the existing augmented reality-based virtual fitting solution is used, the main cause of the return can not be solved.

In addition, the online apparel retailer may provide various lengths (arm length, chest circumference, etc.) of the apparel by describing the numerical information in the corresponding online shopping site, but usually stores the numerical data about the self- It is insufficient to judge the fit.

In addition, among the existing fitting solutions, in order to accurately recognize the physical characteristics of the customer, it is necessary to put on clothes such as tights that can be attached to the body as possible, or to reveal the maximum figure such as sleeveless clothes and then to use the motion capture technology 3D reconstruction technology that takes several postures to capture a person's body and then takes several pictures while rotating 360 degrees and combines them is also introduced, but it is still used for general use There is a limit in technical completeness, and since it requires a lot of preparation process for the customer, it may cause inconvenience to be used universally for general customers.

According to the embodiment, it is possible to solve the above-mentioned problems and reduce unnecessary returns to the customers and reduce unnecessary returns to the customers through a clothing fitting method different from the existing ones in the case of companies and customers in online shopping malls and the like, A clothing fitting system is provided that reduces the occurrence of situations in which unwanted goods are used.

A crawler that periodically collects external clothing images of the external site and external clothing information; A repository for collecting user apparel images and user apparel information from users; A feature point mapper for extracting feature points for the outerwear image and the user's clothing image; A learning device for generating first pattern information of the outer garment image and second pattern information of the user garment image using the feature points; And a third pattern information acquiring unit for receiving third pattern information on the target clothes image from the learning unit, receiving the target clothes image from the user, comparing the third pattern information with the first pattern information, A garment fitting system may be provided, which includes an external interface to derive a garment fitting system.

On one side, the minutiae mapper may extract minutiae for the outer garment image and the user garment image, and categorize each garment image.

In yet another aspect, the number of feature points of each garment image and the spacing or angle between feature points can be used to categorize each garment image.

According to another aspect of the present invention, the learning device may receive feature points extracted from the feature point mapper and map the image based on the external clothing image and the user clothing image, respectively.

In another aspect of the present invention, the learning device may generate the first pattern information and the second pattern information of each garment image by calculating a relative distance between the external clothing image and the feature points of the user garment image.

In another aspect of the present invention, the repository may receive the second pattern information for the user's clothing image from the learning device, and generate the user's body shape information and clothing preference information using the second pattern information .

According to another aspect of the present invention, the external interface can derive the fitting expectation result by comparing relative distances between the minutiae points with respect to the first pattern information, the second pattern information, and the third pattern information.

In another aspect, the external interface may include a mapping table for comparing relative distances between minutiae points of the first pattern information, the second pattern information, and the third pattern information.

Obtaining first pattern information of an outer garment image; Obtaining second pattern information of the user's clothing image; Receiving a target apparel image from a user; Generating third pattern information for the target garment image; And deriving a fitting expectation result for the target garment image based on the first pattern information, the second pattern information, and the third pattern information. have.

According to the embodiment, it is possible to solve the above-mentioned problems and reduce unnecessary returns to the customers and reduce unnecessary returns to the customers through a clothing fitting method different from the existing ones in the case of companies and customers in online shopping malls and the like, It is possible to provide a garment fitting system that reduces the occurrence of situations in which an unwanted article is used.

The clothing fitting system according to the embodiment can reduce unnecessary operating expenses by reducing the return due to the fit error even to the online shopping clothing salesman, prevent the sales damages caused by negative word of mouth of returning customers, From the standpoint, it is possible to improve the experience of online clothing shopping more than before and to increase the confidence in online clothing shopping.

1 is a block diagram for explaining the configuration of a garment fitting system, in an embodiment.
2 is an example of extracting feature points from a garment image, in one embodiment.
3 is an example of generating pattern information using minutiae of a clothing image in one embodiment.
4 is a flow chart for explaining a method of operating a garment fitting system, in one embodiment.

Hereinafter, embodiments of the present invention will be described in detail with reference to the accompanying drawings.

Various modifications may be made to the embodiments described below. It is to be understood that the embodiments described below are not intended to limit the embodiments, but include all modifications, equivalents, and alternatives to them.

The terms used in the examples are used only to illustrate specific embodiments and are not intended to limit the embodiments. The singular expressions include plural expressions unless the context clearly dictates otherwise. In this specification, the terms "comprises" or "having" and the like refer to the presence of stated features, integers, steps, operations, elements, components, or combinations thereof, But do not preclude the presence or addition of one or more other features, integers, steps, operations, elements, components, or combinations thereof.

Unless defined otherwise, all terms used herein, including technical or scientific terms, have the same meaning as commonly understood by one of ordinary skill in the art to which this embodiment belongs. Terms such as those defined in commonly used dictionaries are to be interpreted as having a meaning consistent with the contextual meaning of the related art and are to be interpreted as either ideal or overly formal in the sense of the present application Do not.

In the following description of the present invention with reference to the accompanying drawings, the same components are denoted by the same reference numerals regardless of the reference numerals, and redundant explanations thereof will be omitted. In the following description of the embodiments, a detailed description of related arts will be omitted if it is determined that the gist of the embodiments may be unnecessarily blurred.

In the present invention, image recognition technology and machine learning technology are applied in a completely different way from the conventional technology, so that it is possible to preserve the suitability of the user's body for shopping online without actually checking the clothing We present a platform technology that enables us to judge.

Specifically, when a photograph of a garment (top and bottom) owned by the user is registered with a camera, feature-points in the garment are recognized and the user's body shape is predicted from the characteristic point analysis. If you want to buy clothing by other online shopping, you can input the clothing photo provided by the apparel retailer and automatically analyze whether the fit is fit to the user's body type, have. It is possible to reduce the inconvenience of the use because it is not to photograph a person, and it is possible to make an accurate judgment on the fit because the user's body shape is quantitatively considered and compared.

1 is a block diagram for explaining the configuration of a garment fitting system, in an embodiment. A clothing fitting system 100 according to an embodiment may include a crawler 110, a storage 120, a feature point mapper 130, a learning device 140, and an external interface 150 have.

The crawler 110 periodically collects clothing information for the outer garment image and the outer garment image in the background at an external site.

The external site means an online site such as an open market on the Internet, a clothing company site such as an SPA (Specialty store retailer of a private label Apparel Brand), an SNS, a cafe or a blog, You can collect garment images along with clothing type (top, t-shirt, southern, etc.) and size information. Or, it may have been built through this process.

The store 120 collects user apparel images and user apparel information from the user.

For example, it is possible to receive a photograph taken on a piece of clothing actually owned by the user, and to receive size information and the like for each piece of clothing. In this way, the storage 120 can be provided as a kind of database storing user's clothing image and user's clothing information owned by the user.

The store 120 may analyze the user's body shape information and clothing preference information by receiving the user's clothing image and the user's clothing information. Through the embodiment, the user can obtain analysis information on the clothes he owns. Alternatively, the analyzed information may be constructed by receiving the user's clothing image and user's clothing information from the user.

The minutiae mapper 130 extracts minutiae for the outer garment image and the user garment image.

In an embodiment, feature points may be extracted for the outer garment image collected at the crawler 110 and for the user garment image collected from the store 120. The extracted feature points may be extracted from the garment fitting system 100 using metadata Data.

The feature point mapper 130 may classify the category of the garment for the outer garment image and the user garment image. In an embodiment, the category may be classified using the clothing information input together with the external clothing image and the user's clothing image, or the category may be classified using the extracted feature points. The categories for the outer garment image and the user garment image can be classified as shown in Table 1 below.

The category shown in Table 1 corresponds to the embodiment, and is not limited to the embodiment in classifying the category. Further, more kinds can be listed by normal classification, and can be subdivided.

As shown in the figure, the top and bottom of the image can be classified into the top and bottom in step 1. In step 2, the type of the top and bottom of the image can be classified.

In an embodiment, when information about a category is entered together with apparel information together with an outer apparel image and a user apparel image, the manner of defining the feature points for each category may be different, and in another embodiment, And category of the user's clothing image can be classified as shown in Table 1 using the result of extracting the feature points.

A method of mapping feature points in the feature point mapper 130 according to the embodiment will be described with reference to FIG.

2 is an example of extracting feature points from a garment image, in one embodiment.

Fig. 2 (a) shows characteristic points for a garment image corresponding to a long sleeve shirt / southern area, and Fig. 2 (b) shows characteristic points for a garment image corresponding to a semi-sleeve shirt / Respectively.

As shown in the figure, the feature points extract feature points for a specific point of each garment image, and the feature points are extracted differently according to the category by comparing FIG. 2 (a) and FIG. 2 (b).

Distribution of minutiae by category in the clothing image according to the embodiment can be explained with reference to Table 2. [

Each feature point may have a feature point number as shown. The number of the minutiae can be extracted differently according to the category of the garment image, the minutiae point number can be specified in advance for the predetermined position of the garment image, and the extracted minutiae can be numbered.

2 (a) and 2 (b) correspond to a garment with a color, the feature points 1, 2 and 3 can be extracted. For example, in a clothing image such as a round shirt without a collar, the feature points 1 to 3 may not be extracted.

It can be defined as No. 4 and No. 19 at the point where the angles are changed by overlapping the shoulders and arms, and can be defined as No. 9 and No. 14 at the point where the angle changes to the intersection between the arms and the torso. For the body part, the lower end of the body can be defined as 11 and 12, and the midpoint between 9 and 14, 11 and 12 can be defined as 10 and 13, respectively. The point at which the button is recognized can be displayed after 20th, and the number of the minutiae points can be changed according to the number of the buttons. For example, between a long sleeve shirt / south and a long sleeve T-shirt, it is possible to distinguish whether or not the characteristic points after the characteristic points 3 and 20 extend straight to the bottom of the garment.

As shown in the figure, a minutiae point number can be defined differently for each garment image, and in the case of a lower side, a minutiae point number can be defined in another manner.

Figures 2 (a) and 2 (b) can be categorized in sleeve length. FIG. 2 (b) is a semi-retail garment image. As shown in FIG. 2 (a), feature points 6, 7, 16 and 17 are not found unlike the long-sleeve clothing image.

The number of feature points may be the same between a garment category and a shirt / southern region, but when the feature points 1 to 3 and the feature points 2 and 3 are connected by a straight line, 2, (a) and (b) of FIG. 2, the categories may be classified as different number of feature points, or may be classified into attribute differences such as intervals or angles between feature points.

In the same manner, the feature points can be clearly extracted without overlapping according to each garment image with respect to the step 3 of Table 1, and the feature point mapper 130 can store such metadata in advance.

1, the learning device 140 may generate the first pattern information of the outer garment image and the second pattern information of the user garment image using the minutiae points.

In the embodiment, the outer clothing image and the outer clothing information input from the crawler 110 are extracted from the minutiae mapper 130 in correspondence with the corresponding clothing category, and the learning device 140 extracts the extracted minutiae points from the outer garment image And the attributes between the minutiae points are acquired, so that the pattern can be stored for various clothes in the category by the machine learning technique.

For example, the first pattern information of each garment image may be generated by calculating a relative distance between minutiae points of clothing images input as an outer garment image.

An embodiment of storing a pattern for various clothes in a category by calculating a relative distance between minutiae points will now be described with reference to FIG.

3 is an example of generating pattern information using minutiae of a clothing image in one embodiment.

Fig. 3 (a) is an example in which feature points are extracted and mapped to a clothing image, and Fig. 3 (b) shows a relative distance calculated for the example of Fig. 3 (a).

The relative distance between the major feature points can be obtained by defining the length of the straight line connecting the feature points 1 to 2 and the straight line connecting the feature points 11 to 12 as the relative length = 1. Here, the distance between the minutiae can mean the main attributes that ultimately affect the fit (e.g., neck loose, tighten the chest, etc.) when a person wears clothes. FIG. 3 (b) shows the relative distance between the calculated feature points for predicting the fit according to the example of FIG. 3 (a).

In the embodiment, the relative distances to be calculated for the comfort feeling correspond to the neck circumference, the shoulder width, the arm length, the arm circumference, the chest circumference, the circumference of the abdomen, and so forth. For example, the relative distance between the first to third feature points can be calculated to estimate the fit of the neckline, and the relative distances between the feature points 9 and 14 can be calculated to predict the fit of the chest.

The learning device 140 acquires the attributes of the minutiae and minutiae extracted from the minutiae mapper 130 according to each category for a large number of external clothing images inputted by the crawler, maps them on the clothing image, Can be repeated. FIG. 3 (c) can correspond to first pattern information, for example, a distribution, statistically processing data on relative distances obtained through repetitive operations on a plurality of outer clothing images in the embodiment .

In the embodiment, Fig. 3 (c) shows a distribution diagram for the relative distances 1 to 2 of the minutiae points. For example, when the distribution diagram of FIG. 3 (c) is a distribution diagram for the category of the embodiment of FIG. 3 (a), the relative distances 1 and 2 of the feature points 1 and 2 are 0.159, Can be recognized as a clothing image having a relative distance of 0.186 or less.

The learning device 140 not only searches the external clothing image inputted by the crawler 110 but also the feature points extracted from the feature point mapper 130 according to each category, The attributes of the minutiae points are acquired, the minutiae are mapped on the clothes image, and the task of finding the relative distance between the main minutiae points can be repeated. In addition, the distribution of the second pattern information can be obtained by statistically processing data on the relative distance obtained through the repetitive operation.

In this embodiment, the first pattern information and the second pattern information can be obtained in this way, or distribution information obtained by this method can be stored.

1, the external interface 150 receives the target apparel image from the user, receives the third pattern information about the target apparel image through the learning device 140, and compares the third pattern information with the first pattern information and the second pattern information To obtain fit prediction results for the target garment image.

In the embodiment, the first pattern information acquired for the outer garment image and the second pattern information acquired for the user garment image through the operation of the learning device 140 are used to perform fitting for the target garment image to be purchased by the user The expected result can be obtained.

In an embodiment, as the user increases the user's clothing image input to the store 120, the store 120 may analyze the body information and clothing preference information for the user, e.g., The relative distance of the feature points 1 and 4 (the feature points 2 and 19) is the same, while the shoulder is 10 times smaller than the average % Wide shirt can be obtained together with the second pattern information.

The external interface 150 may play a role of interlocking when the user wants to utilize the clothing fitting system 100. For example, an open API (Application Programming Interface) is provided in the external interface 150, and an external site uses this API to judge whether the user's selected clothing matches with the body when purchasing clothes in the form of a plug-in When a customer presses a button to determine suitability of a garment to a body shape of a customer, the corresponding external site displays basic information of the customer, clothing image currently being watched by the customer, and basic information of the clothes To the clothing fitting system 100 via the API. Among the information thus transmitted, the clothing image is transmitted to the learning device 140, and the third pattern information can be generated by analyzing based on the attribute between the minutiae and the minutiae belonging to the apparel category. In another embodiment, the body shape information and the clothing preference information of the user are retrieved and extracted from the storage 120, and then the extracted information is specifically compared among the minutiae points, . ≪ / RTI >

Details will be described with reference to Tables 3 to 5 for the embodiment.

The customer characteristics in Table 3 correspond to the second pattern information statistically processed on the user clothes images inputted through the storage 120, and the average corresponds to the statistical processing on the external clothing images inputted through the crawler 110 May correspond to the first pattern information. Accordingly, product characteristics can be provided by calculating the relative distance to each feature point through the learning device 140 with respect to the target clothing image input through an external site or the like.

As shown in Table 3, the external interface module 150 has a mapping rule table that can generate a response message to the fitting prediction result in a form that the user can easily understand about the result input to the external interface module 150 , Which can be provided in the form of Table 4. < tb > < TABLE >

According to Table 4, the calculated relative distances for the feature points 1 to 4 can be used to predict the fitting prediction result with respect to the neckline. Condition 1 is a condition for obtaining the user's body shape information or preferred clothing information using the average and customer characteristics, and Condition 2 means a condition for whether the inputted target clothing image matches the user's body information or the preferred clothing information . For example, if customer characteristics are less than 10% below average, the user may be judged to prefer a garment that has a neckline that is less than average or has a slightly tight neck, The prediction result can be obtained.

In the embodiment, a message finally delivered to the user may be provided in the form of a message as shown in Table 5, and the user may select whether the final purchase is to be made.

4 is a flow chart for explaining a method of operating a garment fitting system, in one embodiment.

At step 410, the clothing fitting system 100 acquires first pattern information of the outer garment image.

In an embodiment, the apparel fitting system 100 may periodically collect apparel information for an outer apparel image and an apparel apparel image in the background of the system at an external site.

The external site means an online site such as an open market on the Internet, a clothing company site such as an SPA (Specialty store retailer of a private label Apparel Brand), an SNS, a cafe or a blog, You can collect garment images along with clothing type (top, t-shirt, southern, etc.) and size information.

The garment fitting system 100 may classify categories of garments for the collected outer garment images. In an embodiment, the category may be classified using the clothing information input together with the external clothing image, or the category may be classified using the extracted feature points.

In addition, the extracted minutiae are mapped on the basis of image recognition based on the external garment image, and the attributes between the minutiae are acquired, so that a pattern can be stored for various clothes in the category by a machine learning technique.

For example, the first pattern information of each garment image may be generated by calculating a relative distance between minutiae points of clothing images input as an outer garment image.

The attributes of the minutiae and the minutiae extracted for each category can be acquired for the inputted external clothing image, the minutiae may be mapped onto the outer garment image, and then the relative distance between the main minutiae can be determined. In addition, a distribution corresponding to the first pattern information can be obtained by statistically processing data on the relative distance obtained through the repetitive operation.

Alternatively, the first pattern information generated through this method can be obtained.

Referring to FIG. 3 described above, an embodiment of storing a pattern for various clothes in a category by calculating a relative distance between minutiae can be referred to.

In step 420, the clothing fitting system 100 acquires second pattern information of the user's clothing image.

In an embodiment, the apparel fitting system 100 may collect user apparel images and user apparel information from a user. For example, it is possible to receive a photograph taken on a piece of clothing actually owned by the user, and to receive size information and the like for each piece of clothing. In this way, a kind of database storing the user's clothing image and the user's clothing information owned by the user can be provided.

The clothing fitting system 100 may analyze the user's body shape information and clothing preference information by receiving the user's clothing image and user's clothing information. Through the embodiment, the user can obtain analysis information on the clothes he owns. Alternatively, the analyzed information may be constructed by receiving the user's clothing image and user's clothing information from the user.

The garment fitting system 100 may classify categories of garments for the collected user garment images, in a similar embodiment that collects first pattern information for the outer garment images previously. In the embodiment, categories may be classified using the clothing information input together with the user's clothing image, or the category may be classified using the extracted feature points.

Also, the extracted minutiae points are mapped on the basis of image recognition based on the user's clothing image, and the attributes between the minutiae points are acquired, so that the pattern can be stored for various clothes in the category by the machine learning technique.

For example, the second pattern information of each garment image may be generated by calculating a relative distance between the minutiae points of each of the garment images input as the user garment image.

The garment fitting system 100 acquires not only attributes of the input external clothing image but also the attributes of the feature points and the feature points extracted for each user's clothing image inputted from the user in accordance with each category and maps the attributes on the clothing image, It is possible to repeat the operation of obtaining the relative distance between the main feature points. In addition, the distribution of the second pattern information can be obtained by statistically processing data on the relative distance obtained through the repetitive operation.

Alternatively, the second pattern information generated through this method can be obtained.

In step 430, the apparel fitting system 100 receives the apparel apparel image from the user.

The target garment image may correspond to an image of the garment the user wants to purchase.

In an embodiment, the apparel fitting system 100 provides an open API (Application Programming Interface) to an external site, and in an external site, the apparel selected by the user when purchasing clothes in the form of a plug- The button site may provide a button for determining whether the clothing selected by the user is fit to the body type by purchasing the clothes in the form of a plug-in by utilizing the API. In the button site, The customer can transmit the customer basic information, the target clothes image currently being watched by the customer, and the basic information of the clothes to the clothing fitting system 100 through the API.

In step 440, the clothing fitting system 100 generates third pattern information for the target garment image.

Among the transmitted information, the apparel image can generate the third pattern information by analyzing based on the attribute between the minutiae and the minutiae belonging to the apparel category. In generating the third pattern information, it is possible to refer to the embodiments of steps 410 and 420 for generating the first pattern information and the second pattern information.

At step 450, the clothing fitting system 100 derives a fit prediction result for the target garment image, based on the first pattern information, the second pattern information, and the third pattern information.

Receiving the target apparel image from the user, receiving the third pattern information on the target apparel image, comparing the first pattern information and the second pattern information, and deriving the fitting expectation result for the target apparel image that the user intends to purchase have.

In an embodiment, a format for searching for and extracting the user's body type information and clothing preference information, and then comparing the extracted information with the minutiae point to specifically answer the user's suitability of fit to the external site may be included. For example, a fit prediction result of a target garment image based on user's clothing preference information can be obtained by comparing the product characteristics with the customer characteristics and the average.

The embodiments described above may be implemented in hardware components, software components, and / or a combination of hardware components and software components. For example, the devices, methods, and components described in the embodiments may be implemented within a computer system, such as, for example, a processor, a controller, an arithmetic logic unit (ALU), a digital signal processor, such as an array, a programmable logic unit (PLU), a microprocessor, or any other device capable of executing and responding to instructions. The processing device may execute an operating system (OS) and one or more software applications running on the operating system. The processing device may also access, store, manipulate, process, and generate data in response to execution of the software. For ease of understanding, the processing apparatus may be described as being used singly, but those skilled in the art will recognize that the processing apparatus may have a plurality of processing elements and / As shown in FIG. For example, the processing unit may comprise a plurality of processors or one processor and one controller. Other processing configurations are also possible, such as a parallel processor.

The software may include a computer program, code, instructions, or a combination of one or more of the foregoing, and may be configured to configure the processing device to operate as desired or to process it collectively or collectively Device can be commanded. The software and / or data may be in the form of any type of machine, component, physical device, virtual equipment, computer storage media, or device , Or may be permanently or temporarily embodied in a transmitted signal wave. The software may be distributed over a networked computer system and stored or executed in a distributed manner. The software and data may be stored on one or more computer readable recording media.

The method according to an embodiment may be implemented in the form of a program command that can be executed through various computer means and recorded in a computer-readable medium. The computer-readable medium may include program instructions, data files, data structures, and the like, alone or in combination. The program instructions to be recorded on the medium may be those specially designed and configured for the embodiments or may be available to those skilled in the art of computer software. Examples of computer-readable media include magnetic media such as hard disks, floppy disks and magnetic tape; optical media such as CD-ROMs and DVDs; magnetic media such as floppy disks; Magneto-optical media, and hardware devices specifically configured to store and execute program instructions such as ROM, RAM, flash memory, and the like. Examples of program instructions include machine language code such as those produced by a compiler, as well as high-level language code that can be executed by a computer using an interpreter or the like. The hardware devices described above may be configured to operate as one or more software modules to perform the operations of the embodiments, and vice versa.

Although the preferred embodiments of the present invention have been disclosed for illustrative purposes, those skilled in the art will appreciate that various modifications, additions and substitutions are possible, without departing from the scope and spirit of the invention as disclosed in the accompanying claims. For example, it is to be understood that the techniques described may be performed in a different order than the described methods, and / or that components of the described systems, structures, devices, circuits, Lt; / RTI > or equivalents, even if it is replaced or replaced. Therefore, other implementations, other embodiments, and equivalents to the claims are also within the scope of the following claims.

Claims (19)

A crawler that periodically collects external clothing images of the external site and external clothing information;
A repository for collecting user apparel images and user apparel information from users;
A feature point mapper for extracting feature points for the outerwear image and the user's clothing image;
A learning device for generating first pattern information of the outer garment image and second pattern information of the user garment image using the feature points; And
Receiving a target apparel image from the user, receiving third pattern information on the target apparel image in the learning device, and comparing the first pattern information and the second pattern information to obtain a fitting expectation result for the target apparel image Derived External Interface
/ RTI >
Clothes fitting system.
The method according to claim 1,
The minutiae mapper
Extracting feature points for the outer garment image and the user garment image, categorizing each garment image,
Clothes fitting system.
3. The method of claim 2,
And categorizing each of the clothing images using the number of feature points of each of the clothing images and the interval or angle between the feature points,
Clothes fitting system.
The method according to claim 1,
The learning device includes:
The method comprising: receiving feature points extracted from the feature point mapper and mapping the extracted feature points to an image based on the outer garment image and the user garment image;
Clothes fitting system.
The method according to claim 1,
The learning device includes:
Calculating a relative distance between the outer clothing image and each of the minutiae points of the user garment image to generate the first pattern information and the second pattern information of each garment image,
Clothes fitting system.
The method according to claim 1,
Wherein the storage comprises:
Wherein the second pattern information is provided from the learning device and the body shape information and the clothing preference information of the user are generated using the second pattern information,
Clothes fitting system.
The method according to claim 1,
Wherein the external interface comprises:
And comparing the relative distances between the minutiae points with respect to the first pattern information, the second pattern information, and the third pattern information to derive the fitting expectation result,
Clothes fitting system.
The method according to claim 1,
Wherein the external interface comprises:
And a mapping table for comparing relative distances between minutiae points of the first pattern information, the second pattern information, and the third pattern information.
Clothes fitting system.
Obtaining first pattern information of an outer garment image;
Obtaining second pattern information of the user's clothing image;
Receiving a target apparel image from a user;
Generating third pattern information for the target garment image; And
Deriving a fitting expectation result for the target garment image based on the first pattern information, the second pattern information, and the third pattern information
/ RTI >
A method of operating a garment fitting system.
10. The method of claim 9,
The step of obtaining pattern information of the outer garment image
Periodically collecting an outer garment image of the outer site and outer garment information of the outer garment image; And
Extracting feature points of the outer garment image and generating pattern information of the outer garment image
/ RTI >
A method of operating a garment fitting system.
10. The method of claim 9,
Wherein obtaining the second pattern information of the user's clothing image comprises:
Collecting the user apparel image and user apparel information from a user; And
Extracting feature points for the user's clothing image and generating pattern information of the user's clothing image
/ RTI >
A method of operating a garment fitting system.
10. The method of claim 9,
Wherein obtaining the first pattern information of the outer garment image comprises:
Extracting feature points for the outer garment image, and categorizing each garment image
/ RTI >
A method of operating a garment fitting system.
13. The method of claim 12,
Extracting feature points for the outer garment image, and categorizing each garment image,
Categorizing the outer garment image using at least one of the number of extracted feature points of the outer garment image, the spacing between feature points, and the angle between the feature points
/ RTI >
A method of operating a garment fitting system.
11. The method of claim 10,
Wherein obtaining the first pattern information of the outer garment image comprises:
Mapping the outer garment image to an image based on the extracted minutiae;
/ RTI >
A method of operating a garment fitting system.
11. The method of claim 10,
Wherein obtaining the first pattern information of the outer garment image comprises:
Calculating the relative distance between the minutiae points of the outer garment image to generate the first pattern information of the outer garment image
/ RTI >
A method of operating a garment fitting system.
10. The method of claim 9,
Wherein obtaining the second pattern information of the user's clothing image comprises:
Generating the body shape information and the clothing preference information of the user using the second pattern information on the user's clothing image
/ RTI >
A method of operating a garment fitting system.
10. The method of claim 9,
Wherein deriving a fitting expectation result for the target garment image comprises:
Comparing the relative distances between the minutiae points with respect to the first pattern information, the second pattern information, and the third pattern information to derive the fitting expectation result
/ RTI >
A method of operating a garment fitting system.
10. The method of claim 9,
The fitting expectation result may include:
And a mapping table for comparing relative distances between minutiae points of the outer garment image, the user garment image, and the target garment image,
A method of operating a garment fitting system.
18. A computer program stored in a medium for executing the method of any one of claims 9-18 in combination with hardware.

Images