KR102773871B1 - Clothes fitting and recommendation System for using 3D avatar - Google Patents

Abstract

The present invention provides a 3D avatar by creating and providing the 3D avatar based on the user's body information (body measurements, body features, facial measurements, facial features, etc.), and when the user requests a clothing fitting, a 3D item, which is a 3D model created based on clothing information (size, material, elasticity, etc.) of the selected clothing, is fitted to the 3D avatar to provide a virtual fit, and if the 3D item for the selected clothing has not yet been built, the most similar 3D item is extracted from the 3D items already built by preset categories (color, design, design, pattern, material, etc.) and the extracted 3D item is fitted to the 3D avatar to provide a virtual fit, thereby providing a virtual fit similar to the appearance fitted by the actual user, thereby dramatically improving the convenience and accessibility of clothing shopping, and also maximizing the service reliability and efficiency by providing a virtual fit for all clothing distributed in reality, and reducing the inconvenience caused by clothing shopping. The purpose is to provide a clothing fitting and recommendation system using 3D avatars that can significantly reduce time consumption.

Description

{Clothes fitting and recommendation system for using 3D avatar}

The present invention relates to a clothing fitting and recommendation system using a 3D avatar, and more specifically, to a system for creating and providing a 3D avatar based on a user's body information (body measurements, body features, facial measurements, facial features, etc.), and when a user requests a clothing fitting, a 3D item, which is a 3D model created based on clothing information (size, material, elasticity, etc.) of the selected clothing, is fitted to the 3D avatar to provide a virtual fit, and when a 3D item for the selected clothing has not yet been built, the most similar 3D item is extracted from among the 3D items already built by preset categories (color, design, design, pattern, material, etc.) and the extracted 3D item is fitted to the 3D avatar to provide a virtual fit, thereby making it possible to provide a virtual fit similar to an actual user's fitted appearance, thereby dramatically improving the convenience and accessibility of clothing shopping, and also providing a virtual fit for all clothing distributed in reality. The present invention relates to a clothing fitting and recommendation system using 3D avatars, which can maximize service reliability and efficiency and significantly reduce time consumption due to clothing shopping.

In recent years, as the quality of life has improved and interest in appearance has increased rapidly, clothing such as clothes, shoes, hats, and accessories have rapidly changed from being used simply for clothing to being used for style and fashion.

The most common methods of purchasing clothing are 1) offline shopping, where the buyer visits an actual shopping mall or clothing store, tries on the clothing on sale, and then purchases it, and 2) online shopping, where the buyer uses e-commerce to purchase clothing by referring to information about the clothing, such as photos, provided by the e-commerce server.

However, the conventional offline shopping method has the disadvantage of requiring buyers to visit the shopping mall in person, try on each piece of clothing, and decide whether to purchase, which leads to unnecessary time consumption and increased buyer fatigue. The conventional online shopping method has the disadvantage of frequently purchasing clothing that does not actually fit buyers, because buyers decide whether to purchase based on limited information provided by the e-commerce server without being able to try the clothing on in person.

Meanwhile, recently, as the Internet infrastructure expands and computer graphic technology becomes more advanced, the digital content paradigm is changing from PCs to smartphones and VR HMDs, and 3D virtual reality (VR) and augmented reality (AR) technologies are attracting a lot of attention and are gaining attention as important next-generation industries.

These VR/ARs refer to a digital world where people live based on avatars. Among their important characteristics, they have the advantage of being able to achieve psychological freedom through anonymity by performing activities through a virtual character called an ‘avatar’ that represents oneself in the real world, and being able to freely engage in activities without worrying about exposing one’s private life or identity.

Accordingly, various studies are being conducted on fitting service systems that provide fitting services using 3D avatars to solve the problems of the conventional online/offline shopping methods.

Figure 1 is a flow chart showing a clothing fitting method disclosed in domestically registered patent No. 10-1775327 (Title of invention: Clothing fitting method and fitting program in virtual reality).

The clothing fitting method (hereinafter referred to as “prior art”) (900) of FIG. 1 comprises a step (S910) in which a computer acquires three-dimensional avatar data reflecting a user’s body information, a step (S920) in which three-dimensional clothing data reflecting dimensional information of specific clothing are acquired, a step (S930) in which at least one first vertex included in the clothing and at least one second vertex included in the avatar are arranged to be close to each other, wherein the first vertex and the second vertex are points corresponding to each other when the avatar wears the clothing, a step (S940) in which the clothing is deformed to conform to the body shape of the avatar, and a step (S950) in which the degree of contact or collision between the clothing and the avatar is calculated.

The step of deforming the clothing (S940) further includes a step of expanding one or more first vertices included in the clothing in a direction perpendicular to the clothing so that the clothing corresponds to a specific body part of an avatar to be worn, and expanding the clothing to a state where the clothing can include the entire specific body part of the avatar to be worn, a step of activating a physical collision function between the avatar and the clothing, and a step of restoring the expanded clothing based on physical characteristics of the material of the clothing.

The prior art (S900) configured in this way has the advantage of allowing users to check the size directly through a 3D avatar online and then purchase clothing without being affected by differences in size depending on the clothing retailer.

Typically, clothing is sensitive to trends, seasons, and fashions, so it is characterized by a large number of types and quantities being produced and distributed every day.

However, the prior art (S900) does not take into account the characteristics of such clothing at all, and in order to provide a fitting service, three-dimensional clothing data for the selected clothing must be acquired in advance, and in order to create such three-dimensional clothing data, the user must prepare construction data such as a 360° omnidirectional image (or video) and drawings in advance and input them into a computer, and at the same time, the computer must process and utilize the input construction data to create three-dimensional clothing data. Therefore, it has the disadvantage of making real-time fitting service for clothing sold on-site at shopping malls, etc. practically impossible.

In other words, since the conventional technology (S900) can only provide fitting services when the user has prepared construction data for 3D clothing data in advance or when 3D clothing data has already been created, not only is the service convenience and accessibility significantly reduced, but it also has a structural limitation of only providing one-way services.

In general, when choosing clothes that suit a user, body structure is very important, but clothes that suit each person differ depending on factors such as hairstyle, face shape, and skin tone.

However, since the prior art (S900) was configured to create an avatar using only the user's body information and then fit 3D clothing data, it has the disadvantage of a high gap between the fitting result in virtual space and the fitting result in reality.

In addition, since the prior art (S900) does not describe any technology or method for recommending clothes that suit the user, the user has to go through the hassle of having to fit the desired style of clothes onto the avatar one by one.

The present invention is to solve such problems, and the solution of the present invention is to provide a virtual fit service for clothes selected by a user based on a 3D avatar based on the user's body information (body measurements, body features, facial measurements, facial features, etc.), thereby providing a virtual fit similar to the actual fit of the user, thereby dramatically improving the convenience and accessibility of clothing shopping, and significantly reducing the difference between fitting in virtual space and actual fitting, thereby significantly reducing the time consumed for clothing shopping, thereby providing a clothing fitting and recommendation system using a 3D avatar.

In addition, another problem of the present invention is to provide a clothing fitting and recommendation system using an avatar, which, when a fitting-service integrated management server receives 3D item request data including a clothing image from a user, analyzes the image included in the received 3D item request data to detect feature information by preset category, and then selects the most similar 3D item among the registered 3D items based on the detected feature information and provides it to the user, thereby enabling real-time virtual fitting for all clothing currently distributed online/offline without the user having to prepare a separate 3D item, thereby maximizing service reliability and efficiency.

The solution of the present invention for solving the above problem comprises: at least one user terminal having an application for obtaining body information from a user installed; at least one affiliate server for generating 3D modeling data for clothing sold through an online shopping mall and then sending it to the outside to request registration of the 3D modeling data; When a request for registration of a 3D item, which is 3D modeling data, is received from the above-mentioned affiliate server, identification information of the received 3D item is provided, and 3D item registration information matching the provided identification information and the 3D item is stored in a database, and when a request for membership registration is received from the above-mentioned application, an avatar is created based on the received body information, and the created avatar is assigned to the corresponding user account and then transmitted to the above-mentioned application, and when a fitting service for a specific item is requested from the above-mentioned application, an integrated management server is included which searches for the registered 3D item registration information and transmits the requested 3D item to the above-mentioned application, and when the identification information of the 3D item is generated in response to the request for 3D item registration from the above-mentioned affiliate server, the integrated management server transmits the generated identification information to the above-mentioned affiliate server, and when the above-mentioned affiliate server receives identification information from the above-mentioned integrated management server, a thumbnail photo and identification information of the corresponding clothing are displayed when creating an e-commerce interface, which is a graphical user interface (GUI) for selling clothing. An e-commerce interface is produced, and the application includes a fitting-service request module that is executed according to a user's request, and when a 3D item is received from the integrated management server, the 3D item is fitted to the user's avatar and then displayed on the monitor of the user's terminal, and the fitting-service request module is executed when a fitting-service for an online clothing image exposed including the e-commerce interface is requested by the user, and according to the user's control, an image loading module that downloads or captures the online clothing image exposed including the e-commerce interface; an image acquisition and identification information reading module that analyzes the image downloaded by the image loading module and reads identification information included in the image; a camera/GPS activation module that activates the camera and GPS module of the user's terminal; a shooting module that takes pictures of offline clothing using the camera activated by the camera/GPS activation module; a location information acquisition module that acquires GPS location information detected by the GPS module activated by the camera/GPS activation module; An identification information detection determination module which analyzes an image downloaded by the image loading module to determine whether the downloaded image includes identification information, and executes the image acquisition and identification information reading modules when it is determined that the identification information is included; An URL address input module which is executed when it is determined by the identification information detection determination module that the downloaded image does not include identification information, and which receives from the user a URL address of an online shopping mall where the downloaded image is displayed; An image collection module which is executed after the location information acquisition module or the URL address input module, and 1) collects an image acquired through shooting by the shooting module when it is executed after the location information acquisition module, or 2) collects an image downloaded by the image loading module when it is executed after the URL address input module; A physical characteristic information input module which receives from the user physical characteristic information including at least one of chest circumference, torso length, sleeve length, sleeve circumference, shoulder length, material, and elasticity for clothing which is a target of the fitting service; 1) When executed after the physical characteristic information input module via the location information acquisition module, the image collected by the image collection module, the physical characteristic information input by the physical characteristic information input module, and the location information acquired by the location information acquisition module are matched to generate item request data, 2) When executed after the physical characteristic information input module via the URL address input module, the image collected by the image collection module, the physical characteristic information input by the physical characteristic information input module, and the URL location information input through the URL address input module are matched to generate item request data, 3) When executed after the image acquisition and identification information reading module, the 3D item request module generates 3D item request data including the identification information read by the image acquisition and identification information reading module, and transmits the generated 3D item request data to the integrated management server, and the database section of the integrated management server includes the features of the 3D item. A feature detection algorithm, which is an algorithm that outputs feature information by category using preset category-specific information for classification, 3D modeling data, and physical specific information as input values, is stored, and each of the category-specific information is set as an ‘identical category’ in which a selection process is performed based on whether or not the items are identical when selecting 3D items, or a ‘similar category’ in which a selection process is performed based on similarity, and the integrated management server includes a 3D item request/selection unit that is executed when 3D item request data is transmitted from the application, and the 3D item request/selection unit includes an identification information presence/absence judgment module that determines whether identification information is included in the 3D item request data transmitted from the application; a search and extraction module that is executed when it is determined in the identification information presence/absence judgment module that the identification information is included, and searches 3D item registration information registered in the database section to extract 3D item registration information having identification information included in the transmitted 3D item request data; An AI-based data analysis module which is executed when it is determined that the 3D item request data transmitted from the above identification information presence/absence judgment module does not include identification information, and analyzes the image and physical characteristic information of the transmitted 3D item request data using a pre-designed feature detection algorithm, and outputs feature information by preset category; A category-specific feature information generation module which matches the data output by the AI-based data analysis module to generate feature information by category; A first selection module which extracts feature information belonging to a ‘fixed category’ among the feature information by category generated by the above category-specific feature information generation module, searches the registered 3D item registration information, compares the extracted feature information by each fixed category with the feature information by each fixed category included in the registered 3D item registration information, and then selects 3D item registration information having the same feature information as the extracted feature information by each fixed category; An AI-based similarity calculation module which is executed when 3D items are selected by the first selection module, and extracts feature information belonging to the ‘similar category’ from among feature information of each category of the requested items, which are the 3D item request data received, and then extracts feature information belonging to the ‘similar category’ from among feature information of each category included in each of the selected items, which are the 3D item registration information selected by the first selection module, and detects the similarity between the requested items and each selected item by comparing the feature information of the requested items and the feature information of each selected item with respect to the ‘similar category’ using a pre-produced similarity diagnosis algorithm; A selection item sorting module which sorts the selection items in order of the similarities calculated by the AI-based similarity calculation module from high to low; A similar item determination module which determines, among the selection items sorted by the above selection item sorting module, a similar item which is a selection item arranged first, as the 3D item registration information most similar to the corresponding 3D item request data; A 3D item determination module which transmits a 3D item of the 3D item registration information extracted by the above search and extraction module to the above application or extracts a 3D item from the 3D item registration information of the similar item determined by the above similar item determination module and transmits the 3D item to the above application, wherein the integrated management server further includes an item history information generation unit which is executed when a 3D item is determined by the similarity-based final selection module in the 3D item request/selection unit, and the item history information generation unit includes a 3D item request data input module which receives the 3D item request data transmitted from the above application; A similar item identification information extraction module that extracts identification information of a 3D item determined as a similar item by the similar item determination module; A location/URL presence determination module that determines whether the 3D item request data input through the 3D item request data input module includes location information or URL address information, and if it is determined that the input 3D item request data does not include both location information and URL address information, terminates the operation of the item detail information generation unit; A location information extraction module that is executed when the location/URL presence determination module determines that the input 3D item request data includes location information and extracts location information from the input 3D item request data; A URL address information extraction module that is executed when the location/URL presence determination module determines that the input 3D item request data includes URL address information and extracts URL address information from the input 3D item request data; A feature information extraction module that extracts feature information belonging to the ‘same category’ from feature information of each category of 3D item request data input through the above 3D item request data input module; A similarity extraction module that extracts similarity data between similar items and 3D item request data, which are produced by the above similarity production module; 1) An item history information generation module which matches identification information of a similar item extracted by the similar item identification information extraction module, 2) location information or URL address information extracted by the location information extraction module or the URL address information extraction module, 3) feature information belonging to the ‘same category’ extracted by the feature information extraction module, and 4) similarity data extracted by the similarity extraction module to generate item history information and then stores the same in the database unit, and the 3D item request/selection unit is an item history-based selection module which is executed when the identification information presence/absence judgment module determines that the received 3D item request data does not include identification information and determines a similar item by utilizing the previously registered item history information. The method further comprises: determining whether a similar item has been selected by the above item history-based selection module; executing the 3D item determination module if it is determined that a similar item has been selected; and executing the AI-based data analysis module if it is determined that a similar item has not been selected; wherein the 3D item determination module extracts a 3D item from the 3D item registration information of the similar item selected by the item history-based selection module when it is determined in the above selection-based selection module that a similar item has been selected; and transmitting the extracted 3D item to the application; wherein the item history-based selection module comprises: an item history information extraction module that extracts item history information stored in the database section; a location/URL address extraction module that extracts location information or URL address information included in 3D item request data transmitted from the application; 1) When the data extracted by the above location/URL address extraction module is location information, a location/URL address-based primary filtering module that primarily selects, among the item history information extracted by the above item history information extraction module, item history information having a location within a critical range of the extracted location information, or 2) When the data extracted by the above location/URL address extraction module is URL address information, a location/URL address-based primary filtering module that primarily selects, among the item history information extracted by the above item history information extraction module, item history information having the same URL address information; A same-category based filtering module which extracts feature information belonging to a ‘fixed category’ from among feature information of each category included in the 3D item request data transmitted from the above application, and then compares the extracted feature information with feature information belonging to the ‘same category’ included in the item history information first selected by the location/URL address based first filtering module, and secondarily selects item history information having the same data as the feature information of each fixed category of the 3D item request data; A second similarity calculation module which compares the feature information included in the 3D item request data with the feature information of each second-selected item history information for ‘similar categories’ using the above similarity diagnosis algorithm, and calculates the similarity between the 3D item request data and each of the second-selected item history information; A similarity-based final filtering module which calculates a difference value between the similarity calculated by the second similarity calculation module and the similarity included in each of the second-selected item history information, and then selects item history information whose calculated difference value is less than a threshold value; An item history-based final decision module which detects item history information with the smallest difference value among the item history information finally selected by the similarity-based final filtering module, and then finally determines a similar item included in the detected item history information as a 3D item corresponding to 3D item request data, wherein the 3D item request/selection unit refers to the 3D item registration information registered according to the request of the affiliate server and the item history information of the 3D item selected by the similarity-based final selection module to select a 3D item, thereby increasing a DB to be utilized for 3D item selection regardless of the 3D item registration of the affiliate server.

In addition, the application in the present invention includes a body information registration module, and the body information registration module includes: a camera activation module that activates a camera of the user terminal; a body image acquisition module that acquires body images of a user through shooting with a camera activated by the camera activation module; a body image acquisition module that acquires facial images of a user through shooting with a camera activated by the camera activation module; a body size input module that receives body size information from a user; a body information generation module that generates body information by matching body images acquired by the body image acquisition module, facial images acquired by the facial image acquisition module, and body size information input through the body size input module; A control module for transmitting body information generated by the body information generation module to the integrated management server and requesting body information registration is included, wherein the integrated management server includes a user appearance characteristic information registration unit that is executed when a request for body information registration is received from the application, and an avatar generation/allocation unit, wherein the user appearance characteristic information registration unit comprises: an input module for receiving body information transmitted from the application; a body image analysis module for analyzing each of body images of the body information input through the input module using a pre-designed image analysis algorithm; a user body part dimension detection module for detecting the dimensions of each body part of the corresponding user with reference to the analysis data detected by the body image analysis module; a body characteristic information detection module for detecting the features of each body part of the corresponding user with reference to the analysis data detected by the body image analysis module; a facial image analysis module for analyzing facial images of the body information input through the input module using the image analysis algorithm; A facial dimension detection module for detecting the dimensions of each facial part of a user with reference to the analysis data detected by the facial image analysis module; A facial feature information detection module for detecting the facial features of the corresponding user with reference to the analysis data detected by the facial image analysis module; A user appearance feature information generation module for matching the user body part-specific dimension information detected by the user body part-specific dimension detection module, the body feature information detected by the body feature information detection module, the facial part-specific dimension information detected by the facial dimension detection module, the facial feature information detected by the facial feature information detection module, and the corresponding user identification information to generate user appearance feature information, and then storing the generated user appearance feature information in the database unit, wherein the avatar generation/allocation unit preferably generates an avatar of the corresponding user based on the user appearance feature information generated by the user appearance feature information registration unit.

delete

delete

delete

delete

In addition, in the present invention, it is preferable that the user appearance characteristic information generated by the user appearance characteristic information registration unit includes at least one of the following: dimensions of each body part, turtle neck, degree of back bending, degree of straight legs, face shape, facial cross-sectional area, head circumference, forehead width, eyebrow direction and length, hairstyle, skin tone, presence or absence of double eyelids, and ear shape.

According to the present invention having the above-mentioned tasks and solutions, a virtual fit service for clothing selected by a user is provided based on a 3D avatar based on the user's body information (body measurements, body features, facial measurements, facial features, etc.), thereby providing a virtual fit similar to the actual appearance fitted by the user, thereby dramatically improving the convenience and accessibility of clothing shopping, and significantly reducing the difference between fitting in virtual space and actual fitting, thereby significantly reducing the time consumed due to clothing shopping.

In addition, according to the present invention, when the fitting-service integrated management server receives 3D item request data including a clothing image from a user, the server analyzes the image included in the received 3D item request data to detect feature information by preset category, and then selects the most similar 3D item among the registered 3D items based on the detected feature information and provides it to the user, thereby enabling real-time virtual fitting for all clothing currently distributed online/offline even without the user preparing a separate 3D item, thereby maximizing service reliability and efficiency.

Figure 1 is a flow chart showing a clothing fitting method disclosed in domestically registered patent No. 10-1775327 (Title of invention: Clothing fitting method and fitting program in virtual reality).
FIG. 2 is a diagram showing a clothing fitting and recommendation system using a 3D avatar, which is an embodiment of the present invention.
Figure 3 is a block diagram showing the user terminal of Figure 2.
Figure 4 is a flowchart showing the process of registering 3D modeling data from the online shopping mall affiliate servers of Figure 2 to the fitting-service integrated management server.
Figure 5 is a block diagram showing the fitting-service application of Figure 2.
Figure 6 is a block diagram for explaining GUIs provided by the interface operating module of Figure 5.
Figure 7 is an example diagram showing the VR costume-room interface of Figure 6.
Figure 8 is a block diagram showing the body information registration module of Figure 5.
Figure 9 is a block diagram showing the fitting-service request module of Figure 5.
Figure 10 is a block diagram showing the VR costume-room management module of Figure 5.
Figure 11 is a block diagram showing the clothing recommendation module of Figure 5.
Figure 12 is a block diagram showing the pseudo-avatar recommendation module of Figure 5.
Figure 13 is a block diagram showing the fitting-service integrated management server of Figure 2.
Figure 14 is a block diagram showing the 3D item registration section of Figure 13.
Figure 15 is a block diagram showing the user appearance feature information registration section of Figure 13.
Figure 16 is a block diagram showing the 3D item request/selection unit of Figure 13.
Figure 17 is a block diagram showing the first selection module of Figure 16.
Figure 18 is a block diagram showing the AI-based similarity calculation module of Figure 16.
Figure 19 is a block diagram showing the item history information generation unit of Figure 13.
FIG. 20 is a block diagram showing a second embodiment of the 3D item request/selection unit of FIG. 16.
Figure 21 is a block diagram showing the item history-based selection module of Figure 20.
Figure 22 is a block diagram showing the similar avatar recommendation service unit of Figure 13.

Hereinafter, an embodiment of the present invention will be described with reference to the attached drawings.

FIG. 2 is a diagram showing a clothing fitting and recommendation system using a 3D avatar, which is an embodiment of the present invention.

A clothing fitting and recommendation system (1) using a 3D avatar, which is an embodiment of the present invention, creates and provides a 3D avatar based on a user's body information (body size, body features, facial size, facial features, etc.), and when a user requests a clothing fitting, a 3D item, which is a 3D model created based on clothing information (size, material, elasticity, etc.) of the selected clothing, is fitted to the 3D avatar to provide a virtual fit, and if a 3D item for the selected clothing has not yet been built, the most similar 3D item is extracted from among the 3D items already built by preset categories (color, design, design, pattern, material, etc.) and the extracted 3D item is fitted to the 3D avatar to provide a virtual fit, thereby making it possible to provide a virtual fit similar to the appearance fitted by an actual user, thereby dramatically improving the convenience and accessibility of clothing shopping, as well as providing a virtual fit for all clothing distributed in reality. The aim is to maximize service reliability and efficiency by providing virtual fittings and to significantly reduce time spent on clothing shopping.

In addition, the clothing fitting and recommendation system (1) using a 3D avatar of the present invention, as illustrated in FIG. 2, is composed of a fitting-service integrated management server (3) that creates and assigns an avatar with an account of a connected user and provides a fitting platform service to the user at the request of the connected user, user terminals (6-1), ..., (6-N), which are digital devices carried by the user, fitting-service applications (7) that are installed in each of the user terminals (6-1), ..., (6-N) and provide fitting services to the user in conjunction with the fitting-service integrated management server (3), online shopping mall affiliate servers (9-1), ..., (9-N) that sell clothing online and register a 3D model of the clothing to be sold with the fitting-service integrated management server (3), and a communication network (10) that provides a data transfer path between the fitting-service integrated management server (3), the user terminals (6-1), ..., (6-N), and the online shopping mall affiliate servers (9-1), ..., (9-N).

The communication network (10) is a network that supports data communication between the fitting-service integrated management server (3), user terminals (6-1), ..., (6-N) and online shopping mall affiliate servers (9-1), ..., (9-N), and can be implemented in detail as a wide area network (WAN), a local area network (LAN) network, a value added network (VAN) network, a wired communication network, etc.

Figure 3 is a block diagram showing the user terminal of Figure 2.

The user terminals (6-1), ..., (6-N) of Fig. 3 are digital devices possessed or carried by users who have registered as members of the fitting-service integrated management server (3), and in detail, they may be composed of a desktop PC, a notebook, a smart phone, a tablet PC, etc.

Additionally, a fitting service application (7) is installed on user terminals (6-1), ..., (6-N).

In addition, the user terminal (6), as illustrated in FIG. 3, is composed of a communication module (61) for transmitting and receiving data with a server or node connected to a communication network (10), a memory (62) in which a fitting-service application (7) is installed, a keypad (63) for receiving commands from a user, a monitor (64) for displaying graphical user interfaces (GUIs) provided by the fitting-service application (7), an application management module (65) for linking with the fitting-service application (6), a control module (66) which is an O.S (Operating System) of the user terminal (6), a camera (67) for capturing a subject to obtain a video or image, and a GPS module (68) for linking with a GPS satellite to obtain a GPS location.

Online shopping mall affiliate servers (9-1), ..., (9-N) are e-commerce servers that sell clothing through pre-designed online shopping malls and at the same time register 3D modeling data on the clothing being sold and information on the physical characteristics (body measurements, material, elasticity, etc.) of the clothing in question to the fitting-service integrated management server (3).

That is, the online shopping mall affiliate servers (9-1), ..., (9-N) refer to typical online shopping mall servers, but after creating 3D modeling data for clothing to be sold, the created 3D modeling data and physical characteristic information are registered in the fitting-service integrated management server (3).

At this time, for the convenience of explanation, the present invention has been described as an example in which online shopping mall affiliate servers (9-1), ..., (9-N) design and create 3D modeling data, and then transmit the created 3D modeling data and physical characteristic information to the fitting-service integrated management server (3) for registration. However, the online shopping mall affiliate servers (9-1), ..., (9-N) may transmit construction data necessary for constructing 3D modeling data to the fitting-service integrated management server (3), and the fitting-service integrated management server (3) may utilize a dedicated engine constructed in-house or a well-known VR engine such as Unity or Unreal to create 3D modeling data based on the construction data.

Figure 4 is a flowchart showing the process of registering 3D modeling data from the online shopping mall affiliate servers of Figure 2 to the fitting-service integrated management server.

The 3D modeling data registration step (S10) is composed of a sales target determination step (S11), a 3D modeling creation step (S12), a physical characteristic information creation step (S13), a registration request step (S14), an identification information assignment step (S15), a registration step (S16), and an e-commerce interface creation step (S17), as illustrated in FIG. 4.

The sales target determination step (S11) is the step where the online shopping mall affiliate server (9) determines the clothing to be sold through the online shopping mall.

The 3D modeling generation step (S12) is the step in which the online shopping mall affiliate server (9) generates 3D modeling data for clothing determined as a sales target in the sales target determination step (S11).

The physical characteristic information generation step (S13) is a step in which the online shopping mall affiliate server (9) receives physical characteristic information about clothing determined as a sales target in the sales target determination step (S11) from an administrator and matches the input data to generate physical characteristic information.

At this time, physical characteristic information includes body measurements (chest circumference, torso length, sleeve length, sleeve circumference, shoulder length, etc.), material, elasticity, etc.

The registration request step (S14) is the step in which the online shopping mall affiliate server (9) transmits the 3D modeling data created by the 3D modeling creation step (S12) and the physical characteristic information created by the physical characteristic information creation step (S13) to the fitting-service integrated management server (3).

The identification information assignment step (S15) is the step in which the fitting-service integrated management server (3) assigns identification information to the 3D modeling data for which registration has been requested.

The registration step (S16) is the step where the fitting-service integrated management server (3) matches the 3D modeling data received through the registration request step (S14) with the physical characteristic information, identification information, and the URL address of the corresponding online shopping mall to create registration data, and then stores the created registration data in the database.

The e-commerce interface production stage (S17) is the stage where the online shopping mall affiliate server (9) produces an e-commerce interface, which is a graphic user interface (GUI) for selling clothing for which 3D modeling data has been registered.

At this time, the e-commerce interface displays photos (thumbnails) of the clothing being sold, the selling price, function, material, detailed information, etc., and on one side of each photo, the identification information assigned through the identification information assignment step (S15) is displayed in the form of watermarking, holograms, text, etc.

Figure 5 is a block diagram showing the fitting-service application of Figure 2.

The fitting-service application (7) of Fig. 5 is an application program, software or application that is installed on each of the user terminals (6-1), ..., (6-N) and provides fitting-service to users by linking with the fitting-service integrated management server (3).

In addition, the fitting-service application (7) is composed of a control module (70), a data storage module (71), a server connection module (72), an interface operation module (73), a member registration module (74), a body information registration module (75), a fitting-service request module (76), an avatar fitting module (77), a VR clothing-room management module (78), a clothing recommendation module (79), and a pseudo-avatar recommendation module (80), as shown in FIG. 5.

The control module (70) manages and controls the operation of the fitting-service application (7).

The data storage module (71) stores and extracts data in the memory (62) of the user terminal (9).

The server linkage module (72) transmits and receives data with the fitting-service integrated management server (3) through the communication module (61) of the user terminal (9).

The interface operation module (73) displays pre-made graphic user interfaces (GUIs) on the monitor (64) of the user terminal (6), and at the same time, receives data from the user, executes a processor corresponding to the input data, and then displays an interface on which response data is exposed on the monitor (64) of the user terminal (6).

Figure 6 is a block diagram for explaining GUIs provided by the interface operating module of Figure 5.

The interface operation module (73), as illustrated in FIG. 6, displays a pre-fabricated graphical user interface (GUI) on the monitor (62) of the user terminal (6) according to the user's request, and the pre-fabricated graphical user interface (GUI) includes a membership registration interface (731), a body information registration interface (732), a VR clothing room interface (733), a fitting-service request interface (734), a recommended clothing list interface (735), and a pseudo-avatar list interface (736).

At this time, the membership registration interface (731) is provided when a membership registration request is received from a user and is a GUI for receiving membership registration information from the user, the body information registration interface (732) is provided when a body information registration request is received from a user and is a GUI for obtaining the user's body information, the fitting-service request interface (734) is executed when a user requests avatar fitting for specific clothing and is a GUI for collecting information about the clothing to be fitted, the recommended clothing list interface (735) is executed when a user requests recommended clothing and is a GUI for displaying lists of clothing recommended by the clothing recommendation module (79), and the similar avatar list interface (736) is executed when a user requests a similar avatar list and is a GUI for displaying lists of similar avatars recommended by the similar avatar recommendation module (80).

In addition, in the present invention, for the convenience of explanation, the graphical user interface (GUI) utilized in the interface operation module (73) is illustrated as a member registration interface (731), a body information registration interface (732), a VR clothing room interface (733), a fitting-service request interface (734), a recommended clothing list interface (735), and a pseudo-avatar list interface (736) as shown in FIG. 6, but the type of the graphical user interface is not limited thereto, and it is obvious that it can be produced with various functions and types that can interface between a user and a fitting-service platform.

Figure 7 is an example diagram showing the VR costume-room interface of Figure 6.

When the interface operation module (73) receives a request from a user to visit a VR costume room, it displays the VR costume room interface (733) of Fig. 7 on the monitor (64) of the user terminal (6).

At this time, the VR costume-room interface (733) is composed of an avatar exposure column (7331) where an avatar (A) created under the user's account is displayed, a product group display column (7332) where icons of each product group to which 3D modeling data stored in the user's VR costume-room belong are displayed, and a 3D modeling data display column (7333) where 3D modeling data included in the product group of an icon touched (clicked) by the user among the icons in the product group display column (7332) are displayed.

This VR costume-room interface (733) is utilized in the VR costume room management module (78) described later.

Returning to Figure 5 again, looking at the membership registration module (74), the membership registration module (74) is executed when a request for membership registration for the fitting-service integrated management server (3) is received from a user.

In addition, the membership registration module (74) receives user ID, password, email address, contact information, personal information, etc. through the membership registration interface (731) provided to the user by the interface operation module (73).

In addition, the membership registration module (74) requests membership registration of the user by transmitting membership registration information that matches the data entered by the user through the membership registration interface (731) to the fitting-service integrated management server (3).

Figure 8 is a block diagram showing the body information registration module of Figure 5.

The body information registration module (75) of Fig. 8 is executed according to a user's request and obtains body information from the user by providing the body information registration interface (732) of Fig. 6 described above.

In addition, the body information registration module (75) is composed of a camera activation module (751), a body image acquisition module (752), a facial image acquisition module (753), a body size input module (754), and a body information generation module (755), as illustrated in FIG. 8.

When the body information registration module (75) is executed, the camera activation module (751) activates the camera (67) of the user terminal (6).

The body image acquisition module (752) acquires the user's body image by taking a picture with the camera (67) activated by the camera activation module (751).

At this time, the user can take pictures from the front, back, left side, right side, with arms outstretched, or take pictures of the upper and lower body so that the user's body size measurement can be analyzed through image analysis.

The facial image acquisition module (753) acquires the user's facial image by taking a picture with the camera (67) activated by the camera activation module (751).

At this time, the user can take pictures from various angles to enable recognition of their face shape, facial features, skin tone, hairstyle, etc. through image analysis.

The body size input module (754) is executed according to the user's selection, and receives body size information obtained by the user by visiting a specialized hospital, etc. from the user through the body information registration interface (732).

At this time, body measurement information includes body part measurements such as height, weight, waist circumference, hip circumference, chest circumference, arm length, leg length, shoulder width, and InBody information, as well as facial part measurements such as face size, forehead width, eye-nose-mouth ratio, and head circumference.

The body information generation module (755) generates body information by matching body images acquired by the body image acquisition module (752) and facial images acquired by the facial image acquisition module (753).

At this time, the body information generation module (755) adds the entered body size information to the body information when body size information is input through the body size input module (754) according to the user's selection.

In addition, the body size information generated by the body information generation module (755) is transmitted to the fitting-service integrated management server (3) through the server linkage module (72) under the control of the control module (70), and the fitting-service integrated management server (3) creates and assigns an avatar and VR clothing room to the account of the user who has registered as a member through the member registration module (74). However, when creating the avatar, the avatar is created based on the body information transmitted from the fitting-service application (7), thereby enabling the creation of an avatar having similar body size, hairstyle, skin tone, and face shape to the user, thereby significantly reducing the difference between fitting in virtual space and actual fitting, and dramatically increasing service convenience and reliability.

Figure 9 is a block diagram showing the fitting-service request module of Figure 5.

The fitting service request module (76) of Fig. 9 is executed when a fitting service for clothing is requested from a user online/offline. In this case, the online clothing refers to clothing for which the user wishes to receive a fitting service when shopping at an online shopping mall, and the offline clothing refers to actual clothing for which the user wishes to receive a fitting service when shopping offline.

In addition, the fitting-service request module (76) is composed of a camera/GPS activation module (761), a shooting module (762), a location information acquisition module (763), an image loading module (764), an identification information detection determination module (765), a URL address input module (766), an image collection module (767), a physical characteristic information input module (768), an image acquisition and identification information reading module (769), and a 3D item request module (770), as illustrated in FIG. 9.

The camera/GPS activation module (761) is executed when a user requests a fitting service for clothing offline, and activates the camera (67) and GPS module (68) of the user terminal (6) of FIG. 3 described above.

The shooting module (762) takes pictures of clothing offline using a camera (67) activated by the camera/GPS activation module (761).

The location information acquisition module (763) acquires GPS location information detected by the GPS module (68) activated by the camera/GPS activation module (761).

The image loading module (764) is executed when a user requests a fitting service for clothing online, and downloads or captures clothing images exposed online according to the user's control.

The identification information detection determination module (765) analyzes the image downloaded by the image loading module (764) to determine whether the downloaded image contains identification information.

At this time, in the present invention, when selling clothing, the online shopping mall affiliate server (9) registers the 3D item in the fitting-service integrated management server (3) and, at the same time, when creating the e-commerce interface, the identification information assigned to the clothing is displayed together with the photo. Therefore, when a user downloads an image through the e-commerce interface provided by the online shopping mall affiliate server (9), the image includes the identification information, and when the user downloads the image through the online shopping mall and not the online shopping mall affiliate server (9), the image does not include the identification information.

That is, the identification information detection determination module (765) determines whether identification information is included in the image downloaded through the image loading module (764), thereby enabling the determination of whether a 3D item for the corresponding image is registered in the fitting-service integrated management server (3).

At this time, if the control module (70) determines in the identification information detection determination module (765) that the downloaded image does not contain identification information, it executes the URL address input module (766) as the next processor, and if it determines that the downloaded image contains identification information, it executes the image acquisition and identification information reading module (769) as the next processor.

The URL address input module (766) is executed when the identification information detection determination module (765) determines that the downloaded image does not contain identification information, and receives the URL address of the online shopping mall where the image downloaded from the user is displayed.

The image collection module (767) is executed after the location information acquisition module (763) or the URL address input module (766).

In addition, the image collection module (767) collects images acquired through shooting by the shooting module (762) when 1) it is executed after the location information acquisition module (763), or collects images downloaded by the image loading module (764) when it is executed after the URL address input module (766).

The physical characteristic information input module (768) receives physical characteristic information about clothing that is the subject of fitting service from the user.

At this time, physical characteristic information includes body measurements (chest circumference, torso length, sleeve length, sleeve circumference, shoulder length, etc.), material, elasticity, etc. of the clothing.

The image acquisition and identification information reading module (769) is executed when the identification information detection judgment module (764) determines that the downloaded image contains identification information.

Additionally, the image acquisition and identification information reading module (769) analyzes the image downloaded by the image loading module (764) and reads the identification information included in the image.

At this time, the technology and method for analyzing images and deciphering identification information such as watermarks, holograms, and texts are known technologies, so a detailed description will be omitted.

The 3D item request module (770) is executed after 1) the physical characteristic information input module (768) or 2) the image acquisition and identification information reading module (769).

In addition, the 3D item request module (770) matches the image collected by the image collection module (766), the physical characteristic information input by the physical characteristic information input module (768), and the location information acquired by the location information acquisition module (763) when executed after the physical characteristic information input module (768) via the location information acquisition module (763), thereby generating item request data.

In addition, the 3D item request module (770) matches the image collected by the image collection module (766), the physical characteristic information input by the physical characteristic information input module (768), and the URL location information input through the URL address input module (766) when executed after the physical characteristic information input module (768) via the 2) URL address input module (766), thereby generating item request data.

At this time, the 3D item request data generated by the 3D item request module (770) is transmitted to the fitting-service integrated management server (3) through the server linkage module (72) under the control of the control module (70), and when the fitting-service integrated management server (3) receives the 3D item request data from the fitting-service application (7), it analyzes the images, physical characteristics, and location information (or URL address information) included in the transmitted 3D item request data by preset categories, selects the 3D item most similar to the corresponding clothing among the registered 3D items, and then transmits the selected 3D item data to the corresponding fitting-service application (7).

In addition, when the 3D item request module (770) is executed after the 3) image acquisition and identification information reading module (769), it generates 3D item request data including the identification information read by the image acquisition and identification information reading module (769).

At this time, the 3D item request data generated by the 3D item request module (770) is transmitted to the integrated management server (3) under the control of the control module (70), and when the fitting-service integrated management server (3) receives the 3D item request data from the fitting-service application (7), it extracts the 3D item of the corresponding clothing based on the identification information included in the transmitted 3D item request data, and then transmits the extracted 3D item data to the corresponding fitting-service application (7).

The avatar fitting module (77) uses a pre-designed fitting engine to fit a 3D item received from the fitting-service integrated management server (3) to an avatar assigned to the user in response to a request from the fitting-service request module (76) of FIG. 9 described above.

At this time, the fitting engine performs a simulation of wearing the input 3D item on the user's avatar's body, and positions the 3D item, modeled as a triangular polygon by giving it physical properties, according to the location data based on a specific reference point.

In addition, since the configuration and technology of the fitting engine of the avatar fitting module (77) of the present invention are already known technologies, it is natural that various known technologies or technologies to be developed in the future can be applied to the fitting engine of the present invention.

Additionally, the avatar fitting module (77) displays an avatar with a 3D item fitted on the monitor (64) of the user terminal (6).

In addition, when an avatar with a 3D item fitted is displayed, the avatar fitting module (77) receives a user's selection as to whether to store the 3D item in the VR costume room.

At this time, the control module (70) does not perform a separate operation in the avatar fitting module (77) if 1) the VR costume-room storage of the 3D item currently fitted by the user is not selected, and 2) if the VR costume-room storage of the 3D item currently fitted by the user is selected, it executes the VR costume-room management module (78).

Figure 10 is a block diagram showing the VR costume-room management module of Figure 5.

The VR costume-room management module (78) of Fig. 10 is executed when the avatar fitting module (77) requests storage of a new 3D item or when the user requests editing of a VR costume-room, and interfaces with the user through the VR costume-room interface (733) of Figs. 6 and 7 described above.

In addition, the VR costume-room management module (78) is composed of a registration list display module (781), a new 3D item registration module (782), an existing 3D item deletion module (783), a VR costume-room storage information update module (784), a 3D item selection module (785), an avatar fitting module (786), and an display module (787), as illustrated in FIG. 10.

The registration list display module (781) displays the VR costume room interface (733) that exposes the avatar assigned to the user and 3D items stored in the VR costume room on the monitor (64) of the user terminal (6).

The new 3D item registration module (782) is executed at the user's request, and registers a 3D item that the user wishes to store, i.e., a 3D item input through the 3D item input module (769) of FIG. 9 described above, as a storage target in the VR costume room of the corresponding avatar.

The 3D item deletion module (783) deletes a 3D item selected by the user from among the 3D items already registered in the VR costume room from the storage target at the user's request.

The VR costume-room storage information update module (784) updates the existing VR costume-room storage information so that the 3D item registered by the new 3D item registration module (782) is included in the registration list and stores it in the memory (62) of the user terminal (6).

In addition, the VR costume-room storage information update module (784) updates the existing VR costume-room storage information and stores it in the memory (62) of the user terminal (6) so that the 3D items deleted by the 3D item deletion module (783) are deleted from the registration list.

The 3D item selection module (785) is executed when a fitting service for one of the 3D items stored in the VR costume room is requested from a user, and the 3D item for which the fitting service is to be received is selected by the user.

The avatar fitting module (786) uses a pre-designed fitting engine to fit a 3D item selected by the 3D item selection module (785) to an avatar.

The exhibition module (787) displays an avatar fitted with a 3D item by the avatar fitting module (786) on the monitor (64) of the user terminal (6) through the VR costume-room interface (733).

Figure 11 is a block diagram showing the clothing recommendation module of Figure 5.

The clothing recommendation module (79) of Fig. 11 is executed when a request for clothing recommendation is received from a user.

In addition, the clothing recommendation module (79) is composed of the first, ..., n recommendation-category information input modules (791-1), ..., (791-n), a recommendation request data generation module (792), a recommendation item input module (793), a recommendation item selection module (794), an avatar fitting module (795), and a display module (796), as illustrated in FIG. 11.

The first, ..., n recommendation-category information input modules (791-1), ..., (791-n) receive information about each of the preset recommendation categories from the user.

At this time, the recommended category refers to the standard information for selecting the clothing to be recommended, and can be composed of, for example, ‘gender’, ‘product group (top, bottom, skirt, jeans, etc.)’, ‘color’, ‘season’, ‘purpose (event, vacation, commuting, etc.)’, and ‘material’.

The recommendation request data generation module (792) matches the first, ..., n recommendation category information input by the user through the first, ..., n recommendation category information input modules (791-1), ..., (791-n) to generate recommendation request data.

At this time, the recommendation request data generated by the recommendation request data generation module (792) is transmitted to the fitting-service integrated management server (3) through the server linkage module (72) under the control of the control module (70), and when the fitting-service integrated management server (3) receives the recommendation request data from the fitting-service application (7), it first selects 3D items based on the recommendation-category information included in the transmitted recommendation request data, and then finally extracts the top 3D items in order of the number of purchases among the first-selected 3D items, and transmits the extracted recommendation 3D item list and the recommendation 3D item data to the corresponding fitting-service application (7).

The recommended item input module (793) receives a recommended 3D item list and recommended 3D item data transmitted from the fitting-service integrated management server (3) and provides the input recommended 3D item list to the user.

The recommended item selection module (794) receives a user's selection of one of a list of recommended 3D items.

The avatar fitting module (795) uses a pre-designed fitting engine to fit a 3D item selected by the recommended item selection module (794) to an avatar.

The display module (796) displays an avatar fitted with a 3D item by the avatar fitting module (795) on the monitor (64) of the user terminal (6) through a GUI.

Figure 12 is a block diagram showing the pseudo-avatar recommendation module of Figure 5.

The similar avatar recommendation module (80) of Fig. 12 is executed when a list of similar avatars having similar body information to the corresponding avatar is requested.

In addition, the pseudo-avatar recommendation module (80) is composed of a pseudo-avatar recommendation request data generation module (801), a pseudo-avatar list input/display module (802), a pseudo-avatar selection module (803), a selected pseudo-avatar VR costume-room display module (804), a 3D item selection module (805), an avatar fitting module (806), and an display module (807).

The similar avatar recommendation request data generation module (801) is executed at the user's request and generates similar avatar recommendation request data to request a list of similar avatars having similar body information to the user.

At this time, the similar avatar recommendation request data generated by the similar avatar recommendation request data generation module (801) is transmitted to the fitting-service integrated management server (3) through the server linkage module (72) under the control of the control module (70), and when the fitting-service integrated management server (3) receives the similar avatar recommendation request data, it selects similar avatars having similar body information to the user based on the user's body information, and then transmits the similar avatar list information, which is a list of the selected similar avatars, to the fitting-service application (7).

The pseudo-avatar list input/display module (802) receives a pseudo-avatar list transmitted from the fitting-service integrated management server (3) and displays the input pseudo-avatar list on the monitor (64) of the user terminal (6) through the GUI.

The pseudo-avatar selection module (803) allows the user to select one of the pseudo-avatars on display.

In addition, the pseudo-avatar selection module (803) requests 3D items stored in the VR costume room of the selected pseudo-avatar to the fitting-service integrated management server (3). At this time, the fitting-service integrated management server (3) searches for previously stored VR costume room registration information and transmits the 3D item list and 3D item data stored in the corresponding pseudo-avatar to the corresponding fitting-service application (7).

The selection pseudo-avatar VR costume-room display module (804) creates a VR costume room so that the user's own avatar and the 3D item list and 3D items received from the fitting-service integrated management server (3) through the pseudo-avatar selection module (803) are displayed, and then the created VR costume room is displayed on the monitor (64) of the user terminal (6) through the GUI.

The 3D item selection module (805) allows the user to select one of the 3D items of the pseudo-avatar displayed through the selection pseudo-avatar VR costume-room display module (804).

The avatar fitting module (806) uses a pre-designed fitting engine to fit a 3D item selected by the 3D item selection module (805) to the avatar.

The display module (807) displays an avatar fitted with a 3D item by the avatar fitting module (806) on the monitor (64) of the user terminal (6) through a GUI.

The fitting service application (7) of FIGS. 5 to 12 configured in this manner generates and assigns an avatar most similar to the user's body information, and at the same time, when clothing is selected, fits and displays a 3D item of the selected clothing to the avatar, and enables immediate fitting service for clothing online or offline, and not only recommends similar avatars to the user's body information (age, gender, size for each body, etc.) but also provides fitting service for 3D items stored by each similar avatar, thereby reducing platform accessibility and maximizing user participation and interest, and also enabling virtual fitting for all clothing distributed in reality, thereby maximizing service reliability and efficiency, and significantly reducing time consumption due to clothing shopping.

Figure 13 is a block diagram showing the fitting-service integrated management server of Figure 2.

The fitting-service integrated management server (3) of Fig. 13 is a server that provides a fitting-service platform to connected clients and manages and controls the operation of the fitting-service application (7).

In addition, the fitting-service integrated management server (3) is composed of a control unit (30), a database unit (31), a data transmission/reception unit (32), an application management unit (33), an online affiliate server registration unit (34), a 3D item registration unit (35), a member registration processing unit (36), a user appearance characteristic information registration unit (37), an avatar creation/allocation unit (38), a VR-costume room creation/allocation unit (39), a 3D item request/selection unit (40), an item history information creation unit (41), a VR-costume room storage information registration/update unit (42), a clothing recommendation service unit (43), and a similar avatar recommendation service unit (44), as illustrated in FIG. 13.

The control unit (30) is an O.S (Operating System) of the fitting-service integrated management server (3), and manages and controls the operations of the control targets (31), (32), (33), (34), (35), (36), (37), (38), (39), (40), (41), (42), (43), and (44).

In addition, the control unit (30) receives 3D modeling data and physical characteristic information from online shopping mall affiliate servers (9-1), ..., (9-N), and when a request for registration of an online affiliate server is received, it executes the online affiliate server registration unit (34) and the 3D item registration unit (35).

In addition, when the control unit (30) receives a request for membership registration from the fitting service application (7), it executes the membership registration processing unit (36).

In addition, when the control unit (30) receives body information from the fitting service application (7), it inputs the received body information into the user appearance characteristic information registration unit (37), and when the user appearance characteristic information is registered by the user appearance characteristic information registration unit (37), it executes the avatar creation/allocation unit (38) and the VR costume-room creation/allocation unit (39).

In addition, when the control unit (30) receives 3D item request data from the fitting-service application (7), it inputs the received 3D item request data into the 3D item request/selection unit (40) to execute the 3D item request/selection unit (40), and when the 3D item is selected by the 3D item request/selection unit (40), it executes the item history information generation unit (41) and simultaneously controls the data transmission/reception unit (32) so that the selected 3D item is transmitted to the corresponding fitting-service application (7).

In addition, when the control unit (30) receives VR costume-room storage information from the fitting-service application (7), it inputs the received VR costume-room storage information into the VR costume-room storage information registration/update unit (42).

In addition, when the control unit (30) receives recommendation request data from the fitting service application (7), it executes the clothing recommendation service unit (43), and when it receives similar avatar recommendation request data, it executes the similar avatar recommendation service unit (44).

The data transmission and reception unit (32) transmits and receives data with user terminals (6-1), ..., (6-N) and online shopping mall affiliate servers (9-1), ..., (9-N).

The database section (31) stores the member information of registered users and the affiliated store information of online shopping mall affiliate servers.

Additionally, each user's appearance characteristic information, avatar information, and VR costume room storage information are stored in the database section (31).

Additionally, the database section (31) stores 3D item registration information registered by the 3D item registration section (35).

Additionally, the database section (31) stores item history information generated by the item history information generation section (41).

The online affiliate server registration book (34) is executed when an affiliate request is received from an online shopping mall affiliate server (9-1), ..., (9-N).

In addition, the online affiliate server registration section (34) receives identification information, sales information (URL address, etc.), and sales list information of the online shopping mall that requested the affiliate, matches these to create online shopping mall information, and then registers the created online shopping mall information in the database section (31).

Figure 14 is a block diagram showing the 3D item registration section of Figure 13.

The 3D item registration section (35) of Fig. 14 is executed when a request for 3D item registration is received from an affiliated online shopping mall affiliate server (9).

In addition, the 3D item registration unit (35) is composed of a data input module (351), an identification information assignment module (352), an AI-based input data analysis module (353), a feature information generation module for each category (354), a URL address extraction module (355), and a 3D item registration information generation module (356), as illustrated in FIG. 14.

The data input module (351) receives 3D modeling data and physical characteristic information transmitted from the online shopping mall affiliate server (9).

The identification information assignment module (353) assigns identification information to the 3D item to be registered.

At this time, when identification information is assigned by the identification information assignment module (353), the control unit (30) controls the data transmission/reception unit (32) so that the assigned identification information is transmitted to the corresponding online shopping mall affiliate server (9).

The AI-based input data analysis module (353) uses a feature detection algorithm, which is a pre-learned machine learning algorithm, to analyze 3D modeling data and physical characteristic information input through the data input module (351) and outputs feature information by preset category.

At this time, the category is composed of gender, product group, color, size by body part for each size, design, design, pattern, material, elasticity, etc., and the feature detection algorithm is a machine learning algorithm that takes 3D modeling data and physical characteristic information as input values and outputs feature information by preset category.

For example, when an online shopping mall affiliate server (9) requests 3D item registration by transmitting 3D modeling data and physical characteristic information on a men's short-sleeved T-shirt with a star shape in the center to a fitting-service integrated management server (3), the AI-based input data analysis module (353) of the 3D item registration section (35) of the fitting-service integrated management server (3) analyzes the 3D modeling data and physical characteristic information transmitted from the online shopping mall affiliate server (9) using a pre-learned feature detection algorithm, and outputs the characteristic information of 'gender-category', 'product group-category', 'color-category', 'design-category', 'pattern-category', and 'material-category', namely 'men', 'short-sleeved T-shirt', 'white', 'chest pocket', 'star characteristic value', '80% cotton, 20% nylon'. there is.

In other words, the AI-based input data analysis module (353) analyzes the 3D modeling data and physical characteristic information received when requesting an affiliation from an online shopping mall affiliation server (9) and outputs characteristic information by preset category.

The feature information generation module (354) for each category matches the data output by the AI-based input data analysis module (353) to generate feature information for each category.

The URL address extraction module (355) extracts the URL address of the online shopping mall.

The 3D item registration information generation module (356) matches 1) 3D modeling data and physical characteristic information input through the data input module (351), 2) identification information provided by the identification information providing module (352), 3) characteristic information for each category generated by the characteristic information generation module for each category (354), and 4) the URL address of the corresponding online shopping mall extracted by the URL address extraction module (355) to generate 3D item registration information.

In other words, 3D item registration information includes 3D modeling data, physical characteristic information, identification information, characteristic information for each category, and URL address information.

At this time, the 3D item registration information generated by the 3D item registration information generation module (355) is stored in the database unit (31) under the control of the control unit (30).

The membership registration processing unit (36) receives membership registration information including ID, password, email address, contact information, and personal information from the user terminal (6) and is executed when a membership registration request is received.

In addition, the member registration processing unit (36) registers the member registration information received from the fitting service application (7) in the database unit (31).

Figure 15 is a block diagram showing the user appearance feature information registration section of Figure 13.

The user appearance characteristic information registration section (37) of Fig. 15 is executed when the member registration of a user who requested member registration is processed by the member registration processing section (36).

In addition, the user appearance characteristic information registration unit (37) is composed of a body information input module (371), a body image analysis module (372), a user body part-by-body dimension detection module (373), a body characteristic information detection module (374), a facial image analysis module (375), a facial dimension detection module (376), a facial characteristic information detection module (377), and a user appearance characteristic information generation module (378), as illustrated in FIG. 15.

The body information input module (371) receives body information transmitted from the fitting service application (7) of a user who has requested membership registration.

The body image analysis module (372) analyzes each body image of the body information input through the body information input module (371) using a pre-designed image analysis algorithm.

The user body part size detection module (373) detects the size of each body part of the user by referring to the analysis data detected by the body image analysis module (372).

The body feature information detection module (374) detects features of each body part of the user by referring to the analysis data detected by the body image analysis module (372). The features may include turtle neck, degree of back bending, degree of straightness of legs, etc.

The facial image analysis module (375) analyzes facial images of body information input through the body information input module (371) using a pre-designed image analysis algorithm.

The facial dimension detection module (376) detects the dimensions of each facial area of the user by referring to the analysis data detected by the facial image analysis module (375). At this time, the dimensions of each facial area include face shape, facial cross-sectional area, head circumference, forehead width, eyebrow direction and length, etc.

The facial feature information detection module (377) detects the facial features of the user by referring to the analysis data detected by the facial image analysis module (375). The features here include hairstyle, skin tone, presence or absence of double eyelids, ear shape, etc.

The user appearance feature information generation module (378) matches the user body part-specific size information detected by the user body part-specific size detection module (373), the body feature information detected by the body feature information detection module (374), the facial part-specific size information detected by the facial feature information detection module (376), the facial feature information detected by the facial feature information detection module (377), and the corresponding user identification information to generate the user appearance feature information.

That is, user appearance feature information includes user body part-specific dimension information, body feature information, facial part-specific dimension information, and facial feature information.

At this time, the user appearance characteristic information generated by the user appearance characteristic information generation module (378) is stored in the database section (31) and simultaneously input into the avatar generation/allocation section (38).

The avatar creation/allocation unit (38) creates an avatar assigned to a registered user, and by creating the avatar based on the user appearance characteristic information created by the user appearance characteristic information registration unit (37), it is possible to create an avatar similar to the actual user's body part size, facial part size, body characteristics, and facial characteristics, and accordingly, when fitting a 3D item to the avatar, it is possible to provide a virtual fit most similar to when the actual user fits the corresponding clothing.

The VR costume room creation/allocation section (39) creates a VR costume room that is assigned to a registered user.

At this time, the VR costume room refers to a virtual space where the items (3D modeling data) requested for storage by the user are stored. When first created, it is blank because there are no items being stored.

Figure 16 is a block diagram showing the 3D item request/selection unit of Figure 13.

The 3D item request/selection unit (40) of Fig. 16 receives 3D item request data from the fitting service application (7) and is executed under the control of the control unit (30) when a fitting service is requested.

In addition, the 3D item request/selection unit (40) is composed of a 3D item request data input module (401), an identification information presence/absence judgment module (402), an AI-based data analysis module (403), a feature information generation module for each category (404), a first selection module (405), an AI-based similarity calculation module (406), a similarity-based final selection module (407), a search and extraction module (408), and a 3D item decision module (409), as illustrated in FIG. 16.

The 3D item request data input module (401) receives 3D item request data transmitted from the fitting service application (7).

At this time, the 3D item request data includes 1) an image, physical characteristic information, and location information, 2) an image, physical characteristic information, and a URL address, or 3) identification information.

The identification information presence/absence judgment module (402) determines whether identification information is included in the 3D item request data input through the 3D item request data input module (401).

At this time, the control unit (30) executes the search and extraction module (408) if it is determined in the identification information presence/absence judgment module (402) that identification information is included, but executes the AI-based data analysis module (403) if it is determined that identification information is not included.

The AI-based data analysis module (403) is executed when it is determined in the identification information presence/absence judgment module (402) that the input 3D item request data does not include identification information.

In addition, the AI-based data analysis module (403) analyzes the image and physical characteristic information of the 3D item request data input through the 3D item request data input module (401) using the feature detection algorithm described above, and outputs feature information by preset category.

The feature information generation module (404) for each category matches the data output by the AI-based data analysis module (403) to generate feature information for each category.

At this time, the categories applied to the present invention are classified into ‘same category’ and ‘similar category’, and ‘same category’ means a category in which the selection process is performed based on whether or not they are identical when selecting 3D items, and ‘similar category’ means a category in which the selection process is performed based on similarity when selecting 3D items.

For example, assuming that the categories are ‘Gender-Category’, ‘Product-Group-Category’, ‘Color-Category’, ‘Design-Category’, ‘Pattern-Category’, and ‘Material-Category’, ‘Gender-Category’, ‘Product-Group-Category’, and ‘Color-Category’ can be classified as ‘Fixed-Category’, and ‘Design-Category’, ‘Pattern-Category’, and ‘Material-Category’ can be classified as ‘Similar-Category’.

Figure 17 is a block diagram showing the first selection module of Figure 16.

The first selection module (405) is composed of a feature information input module (451) for each category, a fixed-category feature information extraction module (452), a search and comparison module (453), and a data selection module (454), as illustrated in FIG. 17.

The feature information input module (451) for each category receives feature information for each category generated by the feature information generation module (404) for each category.

The fixed-category feature information extraction module (452) extracts feature information belonging to the ‘fixed category’ from among the feature information of each category input through the feature information input module (451) of each category.

That is, the fixed-category feature information extraction module (452) filters feature information belonging to the ‘similar category’ among the feature information of each input category.

The search and comparison module (453) searches for 3D item registration information already registered in the database section (31) and compares each fixed-category feature information extracted by the fixed-category feature information extraction module (452) with the fixed-category feature information included in the already registered 3D item registration information.

The data selection module (454) selects 3D item registration information having the same feature information as the feature information for each fixed category extracted by the fixed-category feature information extraction module (452) through the search and comparison module (453).

Figure 18 is a block diagram showing the AI-based similarity calculation module of Figure 16.

The AI-based similarity calculation module (406) of Fig. 18 is executed under the control of the control unit (30) when the 3D item registration information selected by the first selection module (405) of Fig. 17 described above is extracted.

In addition, the AI-based similarity calculation module (406) is composed of a first data input module (461), a second data input module (462), a similar-category feature information extraction module (463), a similarity detection module (464), a selection item sorting module (465), and a similar item determination module (466).

The first data input module (461) receives 3D item request data transmitted from the fitting service application (7).

The second data input module (462) receives 3D item registration information selected by the data selection module (457) of FIG. 17 described above.

The pseudo-category feature information extraction module (463) extracts feature information belonging to the ‘pseudo-category’ from among feature information belonging to each category included in the 3D item request data (hereinafter referred to as ‘requested items’) inputted through the first data input module, and then extracts feature information belonging to the ‘pseudo-category’ from among feature information belonging to each category included in each of the selected 3D item registration information (hereinafter referred to as ‘selected items’) inputted through the second data input module (462).

The similarity detection module (464) uses a pre-made similarity diagnosis algorithm to compare the characteristic information of the requested item and the characteristic information of each selected item with respect to the ‘similar category’, thereby detecting the similarity between the requested item and each selected item.

At this time, since the similarity diagnosis algorithm is a technology that is already widely known in object tracking, learning algorithms, analysis systems, etc., a detailed description will be omitted. As an example, the similarity diagnosis algorithm can calculate the similarity between the feature information of the requested item and the feature information of the selected item by the same similar category, and then calculate the average value of these as the similarity.

The selection item sorting module (465) sorts the selection items in order from high to low based on the similarity calculated by the similarity detection module (464).

The similar item determination module (466) determines that the similar item, which is the first selected item among the selected items sorted by the selected item sorting module (465), is the 3D item registration information most similar to the corresponding 3D item request data.

Returning to Figure 16 again, the search and extraction module (408) is executed when the identification information presence/absence judgment module (402) determines that identification information is included.

In addition, the search and extraction module (408) searches for 3D item registration information registered in the database section (31) and extracts 3D item registration information having identification information included in the 3D item request data.

The 3D item determination module (409) 1) extracts a 3D item (3D modeling data) from the 3D item registration information of the similar item when the similar item is determined as the final similar item by the similarity-based final selection module (466) of FIG. 18 described above, and 2) extracts a 3D item from the extracted 3D item registration information when the 3D item registration information is extracted by the search and extraction module (408).

At this time, the control unit (30) transmits the 3D item extracted by the 3D item determination module (409) to the fitting service application (7) that transmitted the 3D item request data.

Figure 19 is a block diagram showing the item history information generation unit of Figure 13.

The item history information generation unit (41) of Fig. 19 is executed under the control of the control unit (30) when a 3D item is determined by the similarity-based final selection module (407) in the 3D item request/selection unit (40) of Fig. 18 described above.

In addition, the item history information generation unit (41) is composed of a 3D item request data input module (411), a similar item identification information extraction module (412), a location/URL determination module (413), a location information extraction module (414), a URL address information extraction module (415), a feature information extraction module (416), a similarity extraction module (417), and an item history information generation module (418), as illustrated in FIG. 19.

The 3D item request data input module (411) receives 3D item request data transmitted from the fitting service application (7) that requested the 3D item.

The similar item identification information extraction module (412) receives identification information of a 3D item determined as a similar item by the similar item determination module (466) of FIG. 18 described above.

The location/URL determination module (413) determines whether the 3D item request data input through the 3D item request data input module (411) includes location information or URL address information.

At this time, the control unit (30) determines in the location/URL determination module (413) that 1) if the input 3D item request data does not include both location information and URL address information, it terminates the operation of the item history information generation unit (41), 2) if the input 3D item request data determines that the location information is included, it executes the location information extraction module (414), and 3) if the input 3D item request data determines that the URL address information is included, it executes the URL address information extraction module (415).

The location information extraction module (414) is executed when the location/URL determination module (413) determines that the input 3D item request data includes location information, and extracts location information from the input 3D item request data.

The URL address information extraction module (415) is executed when the location/URL determination module (413) determines that the input 3D item request data includes URL address information, and extracts URL address information from the input 3D item request data.

The feature information extraction module (416) extracts feature information belonging to the ‘same category’ from among the feature information of each category of the 3D item request data input through the 3D item request data input module (411).

The similarity extraction module (417) extracts similarity data between similar items and 3D item request data, which are produced by the similarity production module (406) of FIG. 18 described above.

The item history information generation module (418) generates item history information by matching 1) identification information of a similar item extracted by the similar item identification information extraction module (412), 2) location information or URL address information extracted by the location information extraction module (414) or URL address information extraction module (415), 3) feature information belonging to the ‘same category’ extracted by the feature information extraction module (416), and 4) similarity data extracted by the similarity extraction module (417).

At this time, the item history information generated by the item history information generation module (418) is stored in the database unit (31) under the control of the control unit (30) and is utilized in the 3D item selection work of the 3D item request/selection unit (40) of FIG. 13 described above.

FIG. 20 is a block diagram showing a second embodiment of the 3D item request/selection unit of FIG. 16.

The second 3D item request/selection unit (24) of Fig. 20 is a second embodiment of the 3D item request/selection unit (40) of Fig. 16 described above.

In addition, the second 3D item request/selection unit (24) includes the 3D item request data input module (401) of FIG. 16 described above, an identification information presence/absence judgment module (402), a feature information generation module (404) for each category, a first selection module (405), an AI-based similarity calculation module (406), a similarity-based final selection module (407), and a search and extraction module (408).

In addition, the second 3D item request/selection unit (24) further includes an item history-based selection module (241), a selection decision module (242), a second AI-based data analysis module (243), and a second 3D item decision module (249).

The 3D item request data input module (401) receives 3D item request data transmitted from the fitting service application (7).

At this time, the 3D item request data includes 1) an image, physical characteristic information, and location information, 2) an image, physical characteristic information, and a URL address, or 3) identification information.

The identification information presence/absence judgment module (402) determines whether identification information is included in the 3D item request data input through the 3D item request data input module (401).

At this time, the control unit (30) executes the search and extraction module (408) if it is determined in the identification information presence/absence judgment module (402) that identification information is included, but executes the item history-based selection module (241) if it is determined that identification information is not included.

Figure 21 is a block diagram showing the item history-based selection module of Figure 20.

The item history-based selection module (241) of Fig. 21 is executed when it is determined in the identification information presence/absence judgment module (402) that the input 3D item request data does not include identification information.

In addition, the item history-based selection module (241) is composed of an item history information extraction module (2411), a 3D item request data input module (2412), a location/URL address extraction module (2413), a location/URL address-based first filtering module (2414), a same-category-based second filtering module (2415), a second similarity calculation module (2416), a similarity-based final filtering module (2417), and an item history-based final decision module (2418), as illustrated in FIG. 21.

The item history information extraction module (2411) extracts item history information generated by the item history information generation unit (41) of FIG. 19 described above and stored in the database unit (31).

The 3D item request data input module (2412) receives 3D item request data transmitted from the fitting service application (7).

The location/URL address extraction module (2413) extracts location information or URL address information included in 3D item request data input through the 3D item request data input module (2412).

The location/URL address-based primary filtering module (2414) 1) first selects, among the item history information extracted by the item history information extraction module (2411), when the data extracted by the location/URL address extraction module (2413) is location information, item history information that has location information but has a location within a critical range of the extracted location information.

In addition, the location/URL address-based primary filtering module (2414) first selects item history information having the same URL address information among the item history information extracted by the item history information extraction module (2411) when the data extracted by the location/URL address extraction module (2413) is URL address information.

The same-category based filtering module (2415) extracts feature information belonging to the ‘fixed category’ from among feature information of each category included in the 3D item request data.

In addition, the same-category based filtering module (2415) compares the extracted feature information with the feature information belonging to the ‘same-category’ included in the item detail information first selected by the location/URL address based first filtering module (2414), and secondarily selects item detail information having the same data as the feature information for each fixed category of the 3D item request data.

The second similarity calculation module (2416) uses a pre-produced similarity diagnosis algorithm to compare the characteristic information included in the 3D item request data with the characteristic information of each second-selected item history information for the ‘similar category’, and calculates the similarity between the 3D item request data and each of the second-selected item history information.

The similarity-based final filtering module (2417) calculates the difference between the similarity calculated by the second similarity calculation module (2416) and the similarity included in the corresponding item history information, and then selects item history information for which the calculated difference is less than a threshold.

The final decision module (2418) based on item history detects the item history information with the smallest difference value among the item history information finally selected by the similarity-based final filtering module (2417), and then determines the similar item included in the detected item history information as the 3D item corresponding to the 3D item request data.

Returning to Figure 20 again, looking at the selection judgment module (242), the selection judgment module (242) determines whether a 3D item has been determined by the item history-based selection module (241).

At this time, the control unit (30) executes the 3D item determination module (249) if it is determined in the selection determination module (242) that the 3D item has been determined by the 1) item history-based selection module (241), and executes the 2nd AI-based data analysis module (243) if it is determined that the 3D item has not been determined by the item history-based selection module (241).

The second AI-based data analysis module (243) is executed when the selection judgment module (242) determines that the 3D item has not yet been determined, and performs the same operation as the AI-based data analysis module (403) of FIG. 16 described above.

The second 3D item determination module (249) performs the same operation as the 3D item determination module (409) of FIG. 16 described above, but when the selection determination module (242) determines that a 3D item has been determined and executes the module, the 3D item determined by the item history-based selection module (241) is finally determined as the 3D item corresponding to the 3D item request data.

The fixed-category feature information extraction module (452) extracts feature information belonging to the ‘fixed category’ from among the feature information of each category input through the feature information input module (451) of each category.

That is, the fixed-category feature information extraction module (452) filters feature information belonging to the ‘similar category’ among the feature information of each input category.

The search and comparison module (453) searches for 3D item registration information already registered in the database section (31) and compares the feature information of each category extracted by the fixed-category feature information extraction module (452) with the feature information of each of the already registered 3D items.

The data selection module (454) selects 3D items (hereinafter referred to as selection items) having the same feature information as the feature information for each category extracted by the fixed-category feature information extraction module (452) through the search and comparison module (453).

Returning to Fig. 13 again, the VR costume-room storage information registration/update unit (42) is executed when the VR costume-room storage information is received from the fitting-service application (7), and the storage information of the corresponding VR costume-room stored in the database unit (31) is deleted, replaced with the received data, and stored in the database unit (31).

Returning to Figure 13, looking at the clothing recommendation service unit (43), the clothing recommendation service unit (43) is executed when recommendation request data is transmitted from the fitting service application (7) and inputs the transmitted recommendation request data.

At this time, the recommendation request data includes 1, ..., n recommendation category information, and the recommendation category is composed of ‘clothing product group (top, bottom, skirt, jeans, etc.)’, ‘color’, ‘season’, ‘purpose (event, vacation, commuting, etc.)’, ‘material’, etc.

In addition, when recommendation request data is input, the clothing recommendation service unit (43) searches the 3D item registration information stored in the database unit (31), selects 3D items matching the 1st, ..., nth recommendation category information of the input recommendation request data, sorts the selected 3D items in order of the number of times they have been displayed, and finally extracts the 3D items in the highest order, and extracts the recommendation list and 3D items, which are lists of the finally extracted 3D items.

At this time, the control unit (30) controls the data transmission/reception unit (32) so that the recommendation list and 3D item data extracted by the clothing recommendation service unit (43) are transmitted to the corresponding fitting service application (7).

Figure 22 is a block diagram showing the similar avatar recommendation service unit of Figure 13.

The pseudo-avatar recommendation service unit (44) of Fig. 22 is executed under the control of the control unit (30) when pseudo-avatar recommendation request data is transmitted from the fitting service application (7).

In addition, the similar avatar recommendation service unit (44) is composed of a similar avatar recommendation request data input module (441), a user appearance feature information extraction module (442), a gender/age/height/weight-based primary selection module (443), an AI-based appearance feature similarity calculation module (444), a sorting module (445), a similar avatar selection module (446), and a similar avatar list generation module (447), as illustrated in FIG. 22.

The similar avatar recommendation request data input module (441) receives similar avatar recommendation request data transmitted from the fitting service application (7).

At this time, the similar avatar recommendation request data includes user identification information.

The user appearance characteristic information extraction module (442) searches for and extracts the appearance characteristic information of the corresponding user (hereinafter referred to as the requested avatar) registered in the database section (31) by referencing the user identification information of the similar avatar recommendation request data input through the similar avatar recommendation request data input module (441).

The first selection module (443) based on gender/age/height/weight searches the user appearance characteristic information registered in the database section (31) and primarily selects avatars having the same gender, age, height and weight range as the requested avatar that requested a similar avatar, and extracts appearance characteristic information of the avatars selected in the first stage.

The AI-based appearance feature similarity calculation module (444) uses a pre-produced second similarity diagnosis algorithm to analyze the appearance feature information of the requested avatar and the appearance feature information of each avatar selected in the first round, and calculates the appearance feature similarity for each of the requested avatar and the avatars selected in the first round.

At this time, since the second similarity diagnosis algorithm is a technology commonly used in object tracking, learning algorithms, analysis systems, etc., a detailed description will be omitted. As an example, the second similarity diagnosis algorithm can calculate the similarity between the requested avatar and the first-selected avatar by category constituting the user appearance feature information, and then calculate the average or sum of these values as the similarity.

The sorting module (445) sorts the first-selected avatars in order of high to low similarity by utilizing the similarity data produced by the AI-based appearance feature similarity production module (444).

The pseudo-avatar selection module (446) selects the top m’ avatars among the avatars sorted by the sorting module (445) as pseudo-avatars.

The pseudo-avatar list generation module (447) generates a list (pseudo-avatar list) of pseudo-avatars selected by the pseudo-avatar selection module (446).

At this time, the control unit (30) controls the data transmission/reception unit (31) so that when a pseudo-avatar list is generated by the pseudo-avatar list generation module (447), the generated pseudo-avatar list is transmitted to the corresponding fitting-service application (7).

In this way, a clothing fitting and recommendation system (1) using a 3D avatar, which is an embodiment of the present invention, provides a virtual fit service for clothing selected by the user based on a 3D avatar based on the user's body information (body measurements, body features, facial measurements, facial features, etc.), thereby enabling the provision of a virtual fit similar to the actual appearance fitted by the user, thereby dramatically improving the convenience and accessibility of clothing shopping, and significantly reducing the difference between fitting in virtual space and actual fitting, thereby significantly reducing the time consumed due to clothing shopping.

In addition, the clothing fitting and recommendation system (1) using a 3D avatar of the present invention, when the fitting-service integrated management server receives 3D item request data including a clothing image from a user, analyzes the image included in the received 3D item request data, detects feature information by preset category, and then selects the most similar 3D item among the registered 3D items based on the detected feature information and provides it to the user, thereby enabling real-time virtual fitting for all clothing currently distributed online/offline even without the user preparing a separate 3D item, thereby maximizing service reliability and efficiency.

1: Clothing fitting and recommendation system using 3D avatars
3:Fitting-Service Integrated Management Server 6:User Terminal
7:Fitting-service application 9:Online shopping mall affiliate server
10:Communication network 30:Control unit
31:Database section 32:Data transmission/reception section
33: Application Management Department 34: Online Affiliate Server Registration Department
35:3D item registration 36:Membership registration processing
37: User appearance feature information registration section 38: Avatar creation/allocation section
39:VR-Costume Room Creation/Assignment Section 40:3D Item Request/Selection Section
41: Item history information creation section 42: VR costume-room storage information registration/update section
43: Clothing Recommendation Service Department 44: Avatar Recommendation Service Department
70: Control module 71: Data storage module
72: Server connection module 73: Interface operation module
74: Membership registration module 75: Body information registration module
76:Fitting-Service Request Module 77:Avatar Fitting Module
78:VR Clothing-Room Management Module 79:Clothing Recommendation Module
80: Recommended module for pseudo-avatars

Claims (7)

At least one user terminal having an application installed to obtain body information from the user;
At least one affiliate server that creates 3D modeling data for clothing sold through an online shopping mall and then sends it externally to request registration of the 3D modeling data;
When a request for registration of a 3D item, which is 3D modeling data, is received from the above-mentioned affiliate server, the system includes an integrated management server that provides identification information for the received 3D item, stores 3D item registration information matching the provided identification information and the 3D item in a database, creates an avatar based on the received body information when a request for membership registration is received from the above-mentioned application, assigns the created avatar to the corresponding user account, and transmits it to the above-mentioned application, and searches for previously registered 3D item registration information and transmits the requested 3D item to the above-mentioned application when a request for fitting service for a specific item is received from the above-mentioned application.
The above integrated management server
When the 3D item identification information is generated in response to the 3D item registration request from the above affiliate server, the generated identification information is transmitted to the above affiliate server.
The above affiliate server is
When receiving identification information from the above integrated management server, when creating an e-commerce interface, which is a graphic user interface (GUI) for selling clothing, the e-commerce interface is created so that the thumbnail photo and identification information of the clothing are displayed.
The above application is
It includes a fitting service request module that is executed upon the user's request, and when a 3D item is received from the integrated management server, it fits the 3D item to the user's avatar and then displays it on the monitor of the user's terminal.
The above fitting-service request module
An image loading module that is executed when a fitting service for an online clothing image including the e-commerce interface is requested by a user, and downloads or captures the online clothing image including the e-commerce interface according to the user's control;
An image acquisition and identification information reading module that analyzes the image downloaded by the image loading module and reads identification information contained in the image;
A camera/GPS activation module that activates the camera and GPS module of the above user terminal;
A shooting module that takes pictures of clothes offline using a camera activated by the above camera/GPS activation module;
A location information acquisition module that acquires GPS location information detected by a GPS module activated by the above camera/GPS activation module;
An identification information detection determination module that analyzes an image downloaded by the image loading module to determine whether the downloaded image contains identification information, and executes the image acquisition and identification information reading module when it is determined that identification information is contained;
A URL address input module that is executed when it is determined in the above identification information detection determination module that the downloaded image does not contain identification information, and receives the URL address of the online shopping mall where the image downloaded from the user is displayed;
An image collection module that is executed after the above location information acquisition module or the URL address input module, and 1) collects images acquired through shooting by the shooting module when executed after the above location information acquisition module, or 2) collects images downloaded by the image loading module when executed after the above URL address input module;
A physical characteristic information input module that receives physical characteristic information including at least one of chest circumference, body length, sleeve length, sleeve circumference, shoulder length, material, and elasticity for clothing that is a fitting service target from a user;
1) When executed after the physical characteristic information input module via the location information acquisition module, the image collected by the image collection module, the physical characteristic information input by the physical characteristic information input module, and the location information acquired by the location information acquisition module are matched to generate item request data, 2) When executed after the physical characteristic information input module via the URL address input module, the image collected by the image collection module, the physical characteristic information input by the physical characteristic information input module, and the URL location information input through the URL address input module are matched to generate item request data, 3) When executed after the image acquisition and identification information reading module, the 3D item request module generates 3D item request data including the identification information read by the image acquisition and identification information reading module, and transmits the generated 3D item request data to the integrated management server.
In the database section of the above integrated management server,
A feature detection algorithm, which is an algorithm that outputs feature information by category using preset category-specific information for classifying the features of the above 3D item and 3D modeling data and physical specific information as input values, is stored.
Each of the above category-specific information is set as a 'same category', which is a category where selection is made based on whether or not they are the same, or a 'similar category', which is a category where selection is made based on similarity when selecting 3D items.
The above integrated management server
Includes a 3D item request/selection unit that is executed when receiving 3D item request data from the above application,
The above 3D item request/selection section
An identification information presence/absence judgment module that determines whether identification information is included in the 3D item request data transmitted from the above application;
In the above identification information presence/absence judgment module, when it is determined that identification information is included, a search and extraction module is executed to search 3D item registration information registered in the database section and extract 3D item registration information having identification information included in the received 3D item request data;
An AI-based data analysis module that is executed when it is determined that the 3D item request data transmitted from the above identification information presence/absence judgment module does not contain identification information, and analyzes the image and physical characteristic information of the transmitted 3D item request data using a pre-designed feature detection algorithm, and outputs feature information by preset category;
A category-specific feature information generation module that matches data output by the above AI-based data analysis module to generate feature information for each category;
A first selection module extracts feature information belonging to a 'fixed category' from among feature information generated by the feature information generation module for each category above, searches for already registered 3D item registration information, compares each extracted feature information for each fixed category with the feature information for each fixed category included in the already registered 3D item registration information, and then selects 3D item registration information having the same feature information as the extracted feature information for each fixed category;
An AI-based similarity calculation module which is executed when 3D items are selected by the first selection module, extracts feature information belonging to the 'similar category' from among feature information of each category of the requested items, which are the transmitted 3D item request data, and then extracts feature information belonging to the 'similar category' from among feature information of each category included in each of the selected items, which are the 3D item registration information selected by the first selection module, and compares the feature information of the requested item and the feature information of each selected item with respect to the 'similar category' using a pre-produced similarity diagnosis algorithm to detect the similarity between the requested item and each selected item;
A selection item sorting module that sorts selection items in order of high to low similarity calculated by the AI-based similarity calculation module;
A similar item determination module that determines that the similar item, which is the first selected item among the selected items sorted by the above-mentioned selected item sorting module, is the 3D item registration information most similar to the corresponding 3D item request data;
A 3D item determination module is included that transmits a 3D item of 3D item registration information extracted by the above search and extraction module to the above application or extracts a 3D item from 3D item registration information of a similar item determined by the above similar item determination module and transmits the extracted 3D item to the above application.
The above integrated management server
In the above 3D item request/selection unit, an item history information generation unit is further included that is executed when a 3D item is determined by the similarity-based final selection module.
The above item details information generation section
A 3D item request data input module that receives 3D item request data transmitted from the above application;
A similar item identification information extraction module that extracts identification information of a 3D item determined as a similar item by the similar item determination module;
A location/URL determination module that determines whether the 3D item request data input through the above 3D item request data input module includes location information or URL address information, and if it is determined that the input 3D item request data does not include either location information or URL address information, terminates the operation of the item history information generation unit;
A location information extraction module that is executed when it is determined in the above location/URL determination module that the input 3D item request data includes location information and extracts location information from the input 3D item request data;
In the above location/URL determination module, when it is determined that the input 3D item request data includes URL address information, a URL address information extraction module is executed and extracts URL address information from the input 3D item request data;
A feature information extraction module that extracts feature information belonging to the 'same category' from among feature information of each category of 3D item request data input through the above 3D item request data input module;
A similarity extraction module that extracts similarity data between similar items and 3D item request data produced by the above similarity production module;
1) The identification information of the similar item extracted by the similar item identification information extraction module, 2) The location information or URL address information extracted by the location information extraction module or the URL address information extraction module, 3) The feature information belonging to the 'same category' extracted by the feature information extraction module, and 4) The similarity data extracted by the similarity extraction module are matched to create the item detail information, and then the item detail information creation module is included to store the created item detail information in the database section.
The above 3D item request/selection section
In the above identification information presence/absence judgment module, when it is determined that the received 3D item request data does not include identification information, an item history-based selection module that determines similar items by utilizing previously registered item history information is executed;
It determines whether a similar item has been selected by the selection module based on the above item details, and if it is determined that a similar item has been selected, it executes the 3D item determination module, but if it is determined that a similar item has not been selected, it further includes a selection determination module that executes the AI-based data analysis module.
The above 3D item decision module
When it is determined that a similar item has been selected in the above selection judgment module, the 3D item is extracted from the 3D item registration information of the similar item selected by the item history-based selection module, and the extracted 3D item is transmitted to the above application.
The above item details-based selection module
An item history information extraction module that extracts item history information stored in the above database section;
A location/URL address extraction module that extracts location information or URL address information included in 3D item request data transmitted from the above application;
1) When the data extracted by the above location/URL address extraction module is location information, a location/URL address-based primary filtering module that primarily selects, among the item history information extracted by the above item history information extraction module, item history information having a location within a critical range of the extracted location information, or 2) When the data extracted by the above location/URL address extraction module is URL address information, a location/URL address-based primary filtering module that primarily selects, among the item history information extracted by the above item history information extraction module, item history information having the same URL address information;
A same-category based filtering module that extracts feature information belonging to 'fixed categories' from among feature information of each category included in the 3D item request data transmitted from the above application, and then compares the extracted feature information with feature information belonging to 'same-category' included in item detail information first selected by the location/URL address based first filtering module, and secondarily selects item detail information having the same data as the feature information of each fixed category of the 3D item request data;
A second similarity calculation module that uses the above similarity diagnosis algorithm to compare the characteristic information included in the 3D item request data with the characteristic information of each of the second-selected item details information for the 'similar category', and calculates the similarity between the 3D item request data and each of the second-selected item details information;
A final filtering module based on similarity that calculates the difference between the similarity calculated by the second similarity calculation module and the similarity included in each second-selected item history information, and then selects item history information whose calculated difference value is less than a threshold value;
Among the item details information finally selected by the similarity-based final filtering module, the item details information with the smallest difference value is detected, and an item details-based final decision module is included that finally determines the similar item included in the detected item details information as a 3D item corresponding to the 3D item request data.
The above 3D item request/selection section
A clothing fitting and recommendation system using a 3D avatar, characterized in that the DB to be utilized for 3D item selection increases regardless of the 3D item registration of the affiliate server by selecting 3D items by referring to the 3D item registration information registered at the request of the affiliate server and the item description information of the 3D items selected by the similarity-based final selection module. In the first paragraph, the application includes a body information registration module,
The above physical information registration module
A camera activation module that activates the camera of the user terminal;
A body image acquisition module that acquires images of a user's body by taking pictures with a camera activated by the camera activation module;
A body image acquisition module that acquires facial images of the user by taking pictures with a camera activated by the camera activation module;
A body size input module that receives body size information from a user;
A body information generation module that generates body information by matching body images acquired by the body image acquisition module, facial images acquired by the facial image acquisition module, and body dimension information input through the body dimension input module;
Includes a control module that transmits the body information generated by the above body information generation module to the integrated management server to request registration of the body information,
The above integrated management server
Includes a user appearance characteristic information registration section that is executed when a request for registration of body information is received from the above application, and an avatar creation/allocation section.
The above user appearance characteristic information registration section
An input module that receives body information transmitted from the above application;
A body image analysis module that analyzes each body image of the body information input through the input module using a pre-designed image analysis algorithm;
A user body part size detection module that detects the size of each body part of the user by referring to the analysis data detected by the above body image analysis module;
A body feature information detection module that detects the features of each body part of the user by referring to the analysis data detected by the above body image analysis module;
A facial image analysis module that analyzes facial images of body information input through the input module using the above image analysis algorithm;
A facial dimension detection module that detects the dimensions of each facial area of the user by referring to the analysis data detected by the above facial image analysis module;
A facial feature information detection module that detects facial features of a user by referring to analysis data detected by the above facial image analysis module;
A user appearance feature information generation module is included that matches the user body part-specific size information detected by the user body part-specific size detection module, the body feature information detected by the body feature information detection module, the facial part-specific size information detected by the facial feature information detection module, the facial feature information detected by the facial feature information detection module, and the corresponding user identification information to generate user appearance feature information, and then stores the user appearance feature information in the database.
The above avatar creation/allocation section
A clothing fitting and recommendation system using a 3D avatar, characterized in that it creates an avatar of the corresponding user based on the user appearance characteristic information generated by the user appearance characteristic information registration unit. delete delete delete delete In the second paragraph, the user appearance characteristic information generated by the user appearance characteristic information registration unit is characterized in that it includes at least one or more of the following: the size of each body part, turtle neck, degree of back bending, degree of straight legs, face shape, facial cross-sectional area, head circumference, forehead width, eyebrow direction and length, hairstyle, skin tone, presence or absence of double eyelids, and ear shape.