KR20210130420A - System for smart three dimensional garment fitting and the method for providing garment fitting service using there of - Google Patents

Abstract

The present invention relates to a smart three-dimensional (3D) clothes fitting system and a clothes fitting service providing method using the same and, more specifically, to a smart 3D clothes fitting system by combination between a 3D avatar file and a 3D texture file and a clothes fitting service providing method using the same. To this end, a 3D avatar wearing a 3D costume is generated by using a user's 3D avatar generated by a 3D avatar generation system and a 3D texture of clothes generated by a 3D texture generation system are generated, and an image in which the user is wearing clothes in reality can be provided by using the user's 3D avatar. Accordingly, a user can view various user's wearing appearances autonomously.

Description

A smart three-dimensional garment fitting system and a method for providing a garment fitting service using the same {System for smart three dimensional garment fitting and the method for providing garment fitting service using there of}

The present invention provides a three-dimensional fitting system and a method for providing a clothing fitting service using the same, and more particularly, to a smart three-dimensional fitting system by combining a three-dimensional avatar file and a three-dimensional texture file, and a method for providing a clothing fitting service using the same.

Consumers traditionally visit a clothing store to purchase clothing, check the clothes they like, and then enter a fitting room to check if they fit by wearing them. Recently, with the development of Internet technology, there has been an increase in the tendency to purchase products by viewing images or videos uploaded by sellers from a shopping mall site instead of directly visiting a clothing store using a mobile terminal such as a smart phone. The advantage of buying clothes online compared to offline is that you can easily purchase clothes without visiting a store. There was a problem that it was found late or that it was not suitable, and it was the cause of the return. In order to solve this problem, a service that can be worn or demonstrated by coordinating a product using a virtual model is provided.

Republic of Korea Patent No. 10-1775327

The conventional virtual wearing service only provides a demonstration of whether the user is suitable as an image, and even in the case of a three-dimensional avatar image developed after that, the degree of collision between the avatar and clothing is calculated by calculating the number of collisions. In the minimum case, the focus is only on the fitting itself by calculating the placement position, and how the clothing is fitted, and if the clothing fits the body or the clothing is small, how can it be expressed to the user? Could not provide the appearance of wearing.

Therefore, through the embodiment of the present invention, the wearer can realize the wearing according to various types of clothes by satisfying the needs that the wearer can exhibit the same effect as the actual wearing regardless of the size of the clothes through the embodiment of the present invention. Through this, we want to provide a business model for selling clothes between the site and the server.

In the case of existing prior inventions, three-dimensional clothing is interposed on the avatar by composing a three-dimensional avatar, and the contact point between the clothing and the body is minimized or the entire clothing is expanded to the extent that it can cover the entire body in the absence of such harm. How the clothes fit to the 3D avatar for the user, and if the clothes fit the body or appear tight to the user if the clothes are small It doesn't show a lot of variety. The present invention provides a wearing appearance to a user in order to solve the above problems, and through this, a business model for selling clothes between a site and a server can also be provided.

In order to solve the problems of the prior art, a smart three-dimensional avatar generating system according to an embodiment of the present invention is an apparatus for generating a three-dimensional avatar photo of a user by receiving a user's two-dimensional picture as an input. a camera that produces a two-dimensional picture; and

and an avatar generating module configured to receive the two-dimensional picture and generate a three-dimensional avatar.

In addition, the input module may further include an input module capable of receiving and designating specific parts of the user's body on the two-dimensional photo by touch, or receiving a user's key.

In addition, the avatar generating module may include a numerical calculation module for calculating values of predetermined body parts on the two-dimensional picture of the user.

Also, the input module may include receiving new numerical values capable of updating the numerical values automatically calculated by the numerical calculation module.

In addition, the avatar generating module generates a skeleton and a surface texture using the two-dimensional picture and the numerical values or the new numerical values, and uses the head and/or face part of the user on the two-dimensional image. This may include applying the head texture to generate the final three-dimensional avatar.

In addition, the skeleton may include implementing various postures including an upright posture.

A three-dimensional texture generating system according to another embodiment of the present invention is a system for receiving a two-dimensional picture of clothes and generating a three-dimensional texture picture of the clothes, the system comprising: a camera for photographing the clothes to generate a two-dimensional picture; and a texture generating module configured to receive the two-dimensional photo and generate a three-dimensional texture photo of the garment.

The apparatus may further include an input module configured to receive input of predetermined specific portions on the two-dimensional photo of the garment by touch, and input the upper and lower lengths of the garment and the elastic modulus of the garment.

In addition, the texture generation module calculates numerical values of predetermined parts on the two-dimensional photo of the garment, and calculates expansion values corresponding to the expandable size of the garment using the calculated numerical value and the elastic modulus It may include a calculation module.

Also, the input module may include receiving new numerical values capable of updating the numerical values automatically calculated by the numerical calculation module and the expanded numerical values.

In addition, the texture generation module is configured to generate a 3D texture picture of the garment by using the 2D picture of the garment and the numerical values, and generate an extended 3D texture picture of the garment by using the 2D picture of the garment and the expanded numeric values. may include doing

A three-dimensional fitting generating system according to another embodiment of the present invention includes a receiving module that receives a three-dimensional avatar of a human body generated by the three-dimensional avatar generating system and a three-dimensional texture of clothes generated by the three-dimensional texture generating system, and the input and a three-dimensional fitting module for generating a three-dimensional avatar wearing a three-dimensional costume by performing rendering by applying the input three-dimensional texture to the three-dimensional avatar.

In addition, the three-dimensional fitting module is configured to provide perimeter values (avatar values) for designated parts of the three-dimensional avatar, perimeter figures for designated parts of the three-dimensional texture (texture values), and designated parts of the extended three-dimensional texture. Comparing the perimeter values (extended texture values) for , if the difference between the avatar value and the texture value is within a predetermined error range, the 3D texture is applied to the 3D avatar for rendering and a signal that the clothes fit the body and, if the texture value is larger than the avatar value and the predetermined error range, after adding a dummy of a certain size to the three-dimensional avatar, the three-dimensional texture is applied to render, and a signal that the costume is larger than the body is generated and if the avatar value is greater than the texture value but greater than the predetermined error range but smaller than the extended texture value, the extended 3D texture is applied to the 3D avatar for rendering, and a signal indicating that the clothes fit the body is generated and, if the avatar value is greater than the texture value and greater than the predetermined error range and greater than the extended texture value, the 3D avatar is rendered by applying the extended 3D texture in a differential form, and a signal that the clothes are smaller than the body can be implemented to create

According to another embodiment of the present invention, a method of providing a clothing service through the Internet is a method of providing a clothing service through the Internet, wherein a service request of a user who wants to use a virtual fitting service for displayed clothing is requested through a site or an app. receiving input; receiving the ID and password of the user who requested the service and allowing login; retrieving the information about the user and the information about the clothes from a DB of a server, and then configuring a 3D avatar for the user and a 3D texture or an extended 3D texture for the clothes; and rendering the configured 3D avatar and the 3D texture or the extended 3D texture and displaying a result thereof.

Also, the method of providing the clothes service through the Internet may further include, when receiving an order request, transmitting the body size data of the user to the server or a third server.

In addition, the method may further include; displaying, on the site or the app, a signal of completion of garment production transmitted from the server or the third server.

The effect of the present invention is as follows with reference to the above.

According to an embodiment of the present invention, it is possible to implement wearing according to various types of clothes so that the wearer can exhibit the same effect as the actual wearer regardless of the size of the clothes.

Using the above embodiment, it is possible to provide a business model for selling clothes between the site and the server.

The effect according to the present invention is not limited by the contents exemplified above, and more various effects are included in the present specification.

1 is a module layout diagram for a clothing fitting generating system according to an embodiment of the present invention.
2 is an overall schematic flowchart of a smart clothing fitting method according to an embodiment of the present invention.
3 is a flowchart of a method for generating a smart avatar according to an embodiment of the present invention.
4 is a flowchart of a method for generating a smart avatar according to an embodiment of the present invention.
5A is an exemplary diagram for creating a smart avatar according to an embodiment of the present invention.
5B is an exemplary diagram for creating a smart avatar according to an embodiment of the present invention.
6 is a flowchart of a method for generating a three-dimensional texture image according to an embodiment of the present invention.
7 is a flowchart of a method for generating a three-dimensional texture image according to an embodiment of the present invention.
8A is an exemplary diagram for generating a three-dimensional texture image according to an embodiment of the present invention.
8B is an exemplary diagram for generating a three-dimensional texture image according to an embodiment of the present invention.
9 is an exemplary diagram of a three-dimensional fitting generator according to an embodiment of the present invention.
10 is a method of improving a clothing size calculation formula according to an embodiment of the present invention.
11 is a method for improving a clothing size calculation formula according to an embodiment of the present invention.
12 is a learning diagram for fitting fitness using AI according to an embodiment of the present invention.

Hereinafter, preferred embodiments of the present invention will be described in detail with reference to the accompanying drawings. Advantages and features of the present invention, and a method of achieving them will become apparent with reference to the embodiments described below in detail in conjunction with the accompanying drawings. However, the present invention is not limited to the embodiments published below, but may be implemented in various different forms, and only these embodiments complete the disclosure of the present invention, and common knowledge in the technical field to which the present invention pertains It is provided to fully inform those who have the scope of the invention, and the present invention is only defined by the scope of the claims. Like reference numerals refer to like elements throughout.

Unless otherwise defined, all terms (including technical and scientific terms) used herein may be used with the meaning commonly understood by those of ordinary skill in the art to which the present invention belongs. In addition, terms defined in a commonly used dictionary are not to be interpreted ideally or excessively unless clearly defined in particular.

The terminology used herein is for the purpose of describing the embodiments and is not intended to limit the present invention. As used herein, the singular also includes the plural unless specifically stated otherwise in the phrase. As used herein, “comprise” and/or “comprising” does not exclude the presence or addition of one or more other components in addition to the stated components.

In the present specification, the computer includes all of various devices capable of providing a result to a user by performing arithmetic processing. For example, the computer includes a desktop, a laptop, a smart phone, a tablet computer, and the like. Also, the computer may be a server that receives a request from a client and performs information processing.

In the present specification, clothing includes all things that can be worn by a person, and clothing includes tops (eg, shirts, T-shirts, polas, jackets, coats, etc.), bottoms (jeans, shorts, skirts, etc.), hats, socks, shoes, etc. The top or bottom may be a slim fit or a loose fit depending on the type, and includes clothes of various designs.

Hereinafter, with reference to the drawings, a smart three-dimensional avatar generating system and an avatar generating method thereof, a smart three-dimensional clothing texture generating system and a texture generating method thereof, a smart three-dimensional fitting generating system and a texture generating method thereof according to embodiments of the present invention with reference to the drawings and a method of providing a clothing service through the Internet will be described.

1 is a module layout diagram for a clothing fitting generating system according to an embodiment of the present invention.

2 is an overall schematic flowchart of a smart clothing fitting method according to an embodiment of the present invention, FIGS. 3 and 4 are flowcharts of a smart avatar creation method according to an embodiment of the present invention, and FIGS. 5A to 5B is an exemplary diagram related to smart avatar creation.

1 to 5B , a clothing fitting system 10 according to an embodiment of the present invention includes a three-dimensional avatar generating system 100 , a three-dimensional texture generating system 200 , and a three-dimensional fitting generating system 300 . .

The three-dimensional avatar generating system 100 according to an embodiment of the present invention includes a camera, an input module 110 , a numerical calculation module 120 , and an avatar generating module 130 .

A smart clothing fitting method according to an embodiment of the present invention includes: generating a three-dimensional (three-dimensional) avatar from user body information (S100); generating 3D texture data from dimension information required by the user from specific clothes (S200); and fitting the three-dimensional avatar and the three-dimensional texture (S300). Below, each step according to an embodiment of the present invention will be described in order.

The smart three-dimensional avatar generating system 100 according to an embodiment of the present invention is a system for generating a three-dimensional (three-dimensional) avatar photo of the user by receiving a two-dimensional (two-dimensional) photo of the user. a camera that produces a two-dimensional picture; and an avatar generating module that receives the two-dimensional picture and generates a three-dimensional avatar. A 'module' described in this specification and claims is a computer, such as a procedure, a function, or a library, which is loaded from the storage device of the computer and resides in the main memory and executed by the CPU of the computer. It has a broad meaning meaning a program (or a set of programs) that is executed for a specific purpose within the context, and may be used interchangeably with the term unit depending on the situation. The avatar generating system 100 may receive and designate a specific part of the user's body by means of a touch input through the input module 110 . For example, to measure the circumference of the hip (buttocks), you can touch the most protruding points on both sides of the user's buttocks in the front photo. data values can be calculated. Here, the size value is measured by comparing the user's body input by the user with the user's picture input into the computer. Such an avatar generating system 100 may be implemented by the numerical calculation module 120 for calculating the numerical value of a predetermined body part on the user's two-dimensional picture. Meanwhile, the user may input a numerical value measured by himself/herself, which may be implemented through the input module of the three-dimensional avatar generating system 100 . Since the user's body values may be temporally variable, when the user gains or loses weight, the values of a specific body part of the user may also change, so the user may input and reflect new values that can update the corresponding values. Such a variable body of the user can also be calculated using a two-dimensional picture. In addition, the avatar generating module 130 of the avatar generating system may generate a skeleton (skeleton) using two-dimensional pictures (front, side, and back) of the user and the numerical values or updated numerical values directly input by the user. The skeleton (skeleton) may typically start from the user's upright posture, but is not limited thereto, and various postures derived from the movement of the user's skeleton may be implemented. It is also possible to create a surface texture that forms the user's appearance around the user's skeleton. The surface texture may also reflect all the various movements of the user, so that the surface texture may be changed in response to the movement of the skeleton. In particular, when the contact between the two textures and pressure beyond that contact is applied, the shape of the surface texture can be changed. For example, this can be implemented in three dimensions by reflecting the shape in which the user's arms can be folded or both legs are bent so that the rear parts of the legs overlap each other and the flesh can be crushed or contracted.

In this way, the user can create a final three-dimensional avatar by manufacturing his/her overall shape and then applying a head texture using the head and/or face part from the user's photo or the like.

The smart three-dimensional avatar creation method (S100) according to an embodiment of the present invention takes a picture image of the user's front view, left and right side views (hereinafter referred to as 'side view') and back view of the user more specifically, The user's body (key) is input (S110). The main parts of the user's front view, side view, and back view are marked (S120). In order to mark the main part of the user, it is determined by marking the outer part of each part (S130). A user's main part is typically a part for measuring data related to clothing and fitting, and it is necessary to calculate the length of the neck, shoulder, arm, chest, waist, pelvis, thigh, and leg. In order to calculate the circumference as well as the length of the user's body part, each part of the side is calculated in addition to the numerical values of the front and the back (S140). The numerical value of the part may be directly input by the user, or may be calculated by a computer based on a body image obtained based on the user's photo and body information input by the user. For example, the user may input a numerical value measured directly by the user's chest circumference, or may input the corresponding data by measuring the waist circumference. The computer can receive and use full-body images taken from the front/rear/side in order to obtain the overall size information of the user, but obtain body size information for a specific body part (eg, upper body parts such as neck, chest, waist, etc.) For this purpose, an image including the corresponding part may be received and used.

On the other hand, the computer may receive the user's front, back, and side photos and convert them into data. Through this, the neck circumference, shoulder circumference, chest circumference, waist circumference, pelvis circumference, thigh circumference, leg length, arm length, etc. are calculated and converted into data. The length may be calculated by calculating the proportion of the body size through the photograph and the actual height of the user. The computer calculates the size from the front, side, and back photos taken with a smartphone or the like and the body input by the user (S141). For example, the computer may be used to create a three-dimensional avatar shape by combining the front, side, and back photos of the user as a whole. To explain in more detail, the computer can calculate the shoulder width of the user by using the values calculated from the front, back, and side photos of the user.

A virtual three-dimensional skeleton (skeleton) is generated through a computer using the calculated size (S142). Meanwhile, the computer combines the three-dimensional skin texture according to the size data (S143). In this way, the computer generates the user's body shape with the combination of the user's three-dimensional avatar and the skeleton (skeleton) and skin texture already created by the three-dimensional avatar generating program, and then reflects the user's face to generate the three-dimensional avatar. can

6 and 7 are flowcharts of a method for generating a 3D texture image according to an embodiment of the present invention, and FIGS. 8A and 8B are exemplary diagrams related to the generation of a 3D texture image.

6 to 8B , the 3D texture generation system 200 according to an embodiment of the present invention includes a camera, an input module 210 , a numerical calculation module 220 , and a texture generation module 230 . The camera generates a two-dimensional photo by inputting or receiving a photo of the front/side (sometimes can be ignored)/rear from the user, and the texture generation module 230 generates a three-dimensional texture photo from the photo of the garment by receiving or receiving the input or receiving the photo from the user. can do. In addition, the user may include an input module 210 capable of inputting a predetermined portion on a specific garment by the user with a touch, and receiving the upper and lower lengths of the garment and a unique elastic modulus of the garment. For example, the elastic modulus may be different depending on the fabric material of the garment, and the elastic modulus value increases as the elastic modulus is made of a material with good elasticity. Even if clothes of the same shape have different elastic modulus, a difference occurs in the appearance of a user who wears the clothes. For example, when the user wears leggings, the shape may appear according to the user's body type as it is, depending on the degree of elasticity that the clothing of the leggings can withstand. On the other hand, in the case of a material with low elasticity, such as wool, the corresponding clothing may appear as loose or tight fitting to the user's body depending on the user's figure. In other words, changes in physical shape of various postures can be reflected by applying characteristics according to the material of the fabric, and when clothes are worn on a three-dimensional avatar, changes in the shape of clothes can be appropriately simulated according to the material and elastic modulus. . The computer may receive the corresponding information from the clothing seller, receive physical properties such as material and elastic modulus for each part, and reflect the information to generate simulated clothing. In addition to the elastic modulus, it may include a degree of change of shape according to frictional force and gravity.

In addition, the computer may receive the characteristic information of the fabric and material of each garment from the terminal or the server. In addition, the computer may receive each clothing brand, product name, clothing type, size information, and size name. In addition, it is possible to receive and reflect information on a small number of original clothes, not limited to the already made ready-to-wear.

On the other hand, the texture generation module 230 calculates numerical values of predetermined parts on the photograph of the clothing selected by the user as well as independently, or calculates the calculated values and reflects the calculated values and the elastic modulus of the clothing to reflect the calculated values and the elastic modulus of the clothing. It may additionally include a numerical calculation module 220 for calculating extension values corresponding to the expandable size. In addition, the input module 210 includes being able to receive new numerical values that can update numerical values automatically calculated by the numerical calculation module 220 or directly measured and input numerical values and extended numerical values. do. In this way, the three-dimensional texture generating system 200 generates a three-dimensional texture photo of the garment using the two-dimensional photo of the garment designated by the user through the texture generating module 230 and the values measured from the photo, and the elasticity of the garment. As a result of additionally reflecting the coefficients, it is possible to generate an extended three-dimensional texture picture of the clothing.

On the other hand, the method (S200) of generating three-dimensional texture data according to an embodiment of the present invention is more specifically, the front/side (sometimes negligible)/rear photo image of the clothes specified by the user and the vertical length of the clothes. input (S210). The main parts of the front and back views of the clothes designated by the user are marked (S220). In order to mark the main parts of the garment, the outer part of each part is marked and judged. The main parts of the garment are usually the collar, shoulder, chest, waist, upper arm circumference, arm length, pelvis, thigh, knee, calf, and leg. It is the main part of clothing corresponding to the part of the back. In order to finally measure the length of the part, the length of the upper and lower parts of the corresponding clothes may be input, and the length of the remaining part is calculated corresponding to the input length between the upper and lower parts (S230). After automatically calculating and dataizing the size of the main part of the garment, the expanded size of the main part of the garment can also be automatically calculated and dataized in consideration of the elasticity of the garment itself. The method may further include the step of reviewing and correcting the automatically calculated size by the user.

On the other hand, the computer can receive the front, back, and side photos of clothes and convert them into data. Through this, the neck circumference, shoulder circumference, chest circumference, waist circumference, pelvis circumference, thigh circumference, leg length, arm length, etc. are calculated and converted into data (S231). The length may be calculated by calculating a degree of proportionality between the photo and the upper and lower lengths of the clothes input by the user. The components of the texture for each part are extracted using the calculated size (S232). After extracting these components, three-dimensional texture data is extracted (S233). After extracting the three-dimensional texture data, the extended three-dimensional texture data is extracted by reflecting the elastic modulus and the like (S234). Through this process, an image of an extended three-dimensional texture can be generated.

9 is an exemplary diagram of a three-dimensional fitting generating system 300 according to an embodiment of the present invention. Referring to FIG. 9 , a three-dimensional fitting generating system 300 according to an embodiment of the present invention provides a three-dimensional avatar of a human body and a three-dimensional texture generating system 200 generated by the three-dimensional avatar generating system 100 described in detail above. ), a receiving module 310 that receives the 3D texture of clothes generated by 320). The three-dimensional fitting module 320 is configured to provide circumference values for designated parts of the three-dimensional avatar, for example, the user's neck, shoulder, chest, and waist circumference values (hereinafter referred to as 'avatar values'), clothing specified by the user, and the like. Physical properties such as elastic modulus in the specified parts of the three-dimensional texture of, for example, the neck, shoulder, chest, and waist circumference values specified by the user (hereinafter referred to as 'texture values') and the three-dimensional texture of the clothing specified by the user If the difference between the avatar value and the texture value is within a predetermined error range compared with the reflected and numerically expressed neck, shoulder, chest, and waist circumference values (hereinafter referred to as the 'extended texture value'), the three-dimensional texture is calculated. It can be applied to the three-dimensional avatar to render and generate a signal that the clothes fit the body. The error range specified herein may be an error derived from the difference in individual characteristics of each person in the process of converting a numerical value derived from a two-dimensional photograph into a three-dimensional numerical value of a circumference, and a numerical value appearing in a linear two-dimensional photograph The error that may occur in the process of being converted into a three-dimensional avatar in a curved form is taken into account. First, in the case of an error appearing from a user's individual human body characteristics, the structure of a specific part can be derived from a number of people through a certain type. For example, in the case of the user's neck, people with short or long necks and those with thick or relatively thin necks can be categorized. An error range within a certain range can be created by recognizing the average difference of the derived body type. In addition, in order to obtain, for example, waist circumference, pelvis circumference, chest circumference, etc., which are difficult to measure with only the frontal image, errors in the characteristics of each individual are reflected in the process of calculating the ratio of the width to the length of each body part for each body type. range can be specified.

If the difference between the avatar value and the texture value is within a predetermined error range, the 3D texture is applied to the 3D avatar to render it, and a signal indicating that the clothes fit the body well is generated.

If the texture value of the clothing specified by the user is larger than a predetermined error range compared to the avatar value, a dummy having a size determined in proportion to the size of the actual error range is added to the 3D avatar to the avatar (in 3D) ), then apply the 3D texture to render and generate a signal that the costume is larger than the body.

If the avatar value is larger than the texture value but less than the extended texture value, the 3D avatar is rendered by applying the extended 3D texture and the clothes are tight fit. create

If the avatar value is greater than the texture value and greater than the predetermined error range and is greater than the extended texture value, the extended texture is applied to the avatar in a differentiated form for rendering, and a signal indicating that the clothes are smaller than the body is generated.

The differentiated form of rendering may be a form in which a button or a zipper is not fastened, or a form in which the length of the user's avatar is shorter than that of the user's avatar.

In addition, when the circumference of the clothes is larger, in addition to the method of putting on the virtual dummy texture and putting clothes on, it is a shape that the clothes are hung in response to the user's body. It is also possible to express it in an existing form. A user may optionally select a loose fit expressed when wearing large clothes as a fashion option.

In addition, when the circumference of the clothes is smaller, the user feels uncomfortable or does not fit the clothes when wearing the clothes, but the elastic modulus of the clothes is activated. With the effect of slim fit or leggings, the appearance of clothes can be stretched by the user's body type, and it can be expressed that the clothes fit tightly. If the elastic modulus of the clothing is exceeded, when the user purchases clothes online, the user will not be able to wear the actual clothes. Therefore, if the clothing extension circumference is smaller, the center may be rendered open. In addition, since each user has a different body structure, the shoulder may be wide, but the waist may be thin. may be within the specified error range. In this case, it is also possible to visualize the relevant part by dividing it, and it is possible to visualize the shoulder with a tight fit and the waist with a loose fit as if worn in reality.

10 and 11 are a method of improving a clothing size calculation formula according to an embodiment of the present invention.

Referring to FIGS. 10 and 11 , it is also possible to improve the formula for calculating the clothes size specified by the user. A clothing seller can accumulate data corrected by the screen in the app or test it with the app itself and accumulate it as data. More specifically, the method of improving the calculation formula for the clothes size creates a formula for extracting the size for each part (S410). It is possible to automatically calculate the size of the main part of clothes and convert it into data, and it is possible to automatically calculate the expanded size of the main part of clothes and convert it into data in consideration of the elastic modulus according to the material. An official ID is assigned to the formula for each part of the clothing size and applied (S420). The formula applied to the designated clothes may be a new formula, but data is accumulated by continuously applying it for a certain period of time. While accumulating data, data on measured values and correction values can be accumulated for each official ID, and [Official ID-Measured Value-Correction Value] is converted into data in the form of one set and accumulated (S430). Calculate the error degree for the clothes size that can appear for each formula corresponding to each official ID. This can be continuously measured or measured during a regular period to calculate the degree of error (S440). Among the formulas corresponding to the official ID, the formula with the lowest degree of error is adopted, and it is reflected in the data of the [Official ID-Measured Value-Correction Value] set to improve the error degree for each formula (S450).

A method of providing a clothing service through the Internet according to an embodiment of the present invention includes: receiving a service request from a user who wants to use a virtual fitting service for displayed clothes through a site or an app (S510); receiving the ID and password of the user who requested the service and allowing login (S520); retrieving the information about the user and the information about the clothes from a database (DB) of a server, and then configuring a 3D avatar for the user and a 3D texture or an extended 3D texture for the clothes, respectively (S530); and rendering the configured 3D avatar and the 3D texture or the extended 3D texture and displaying the result (S540). The method may further include: when the clothing seller receives an order request from a user using the app, transmitting the body size data of the user to the server or a third server (S550). In addition, the method of providing a clothing service through the Internet may further include displaying a clothing production completion signal transmitted from the server or the third server on the site or the app (S560).

12 is a learning diagram of fitting fitness using AI according to an embodiment of the present invention. Referring to FIG. 12 , a three-dimensional fitting generating system and a fitting generating method according to an embodiment of the present invention may use artificial intelligence (AI) learning, such as artificial intelligence (AI), to improve a fitting value. More specifically, through the rendering logic model learning in the three-dimensional fitting generation system of the present invention, the user's body size, clothing size, fabric characteristics for each main part of clothing (eg, cotton, hemp, silk, wool, synthetic fiber, etc.), main By accumulating data on wrinkle patterns for each part (e.g., shoulder tip, armpit, chest, waist, elbow, wrist, pelvis, knee, ankle, etc.), a rendering logic model can be created through unsupervised learning. . As in unsupervised learning, if unrefined data is input, clustering can be performed by summarizing the characteristics of the data even if there is no training data, and it is okay to not set a target value. Unsupervised learning according to an embodiment of the present invention can utilize both clustering and pattern recognition. In the case of clustering, distance-based clustering, density-based clustering, etc. can be used, and in the case of pattern recognition, preprocessing/feature extraction method or model A selection/recognition method is available. By using the above learning, the user can further improve and enhance the fitting suitability of his or her 3D avatar and clothing texture.

Clothing fitting generation system: 10 3D avatar generation system: 100
Input module 1: 110 Numerical calculation module 1: 120
Avatar generation module: 130 3D texture generation system: 200
Input module 2: 210 Numerical module 2: 220
Texture Generation Module: 230 3D Fitting Generation System: 300
Receiving module: 310 Three-dimensional fitting module: 320

Claims (16)

A computer-implemented system for generating a three-dimensional avatar photo of a user by receiving a two-dimensional photo of a user, the system comprising:
a camera that captures the user's body to generate a two-dimensional picture; and
and an avatar generating module that receives the two-dimensional picture and generates a three-dimensional avatar,
A three-dimensional avatar creation system.
According to claim 1,
The three-dimensional avatar generation system further comprising an input module 1 capable of receiving specific user body parts on the two-dimensional picture to be inputted with a touch to designate it, or to receive a user's key.
According to claim 1,
The avatar creation module comprises:
and a numerical calculation module 1 for calculating the numerical values of the body parts designated in advance on the two-dimensional picture of the user.
4. The method of claim 3,
The input module 1 is,
The three-dimensional avatar generating system that can receive new numerical values for updating the numerical values automatically calculated by the numerical calculation module 1.
5. The method of claim 4,
The avatar creation module comprises:
A skeleton and a surface texture are generated using the two-dimensional photo and the numerical values or the new numerical values, and a final three-dimensional (3D) texture is applied by applying a head texture using the head and/or face part of the user on the two-dimensional photo. A three-dimensional avatar generation system that creates an avatar.
6. The method of claim 5,
The skeleton is to implement various postures including an upright posture, a three-dimensional avatar generation system.
A system for receiving a two-dimensional picture of clothes and generating a three-dimensional texture picture of the clothes,
a camera that takes a picture of the costume and creates a two-dimensional picture; and
and a texture generating module configured to receive the two-dimensional photo and generate a three-dimensional texture photo of the garment.
8. The method of claim 7,
The three-dimensional texture generating system further comprising an input module 2 capable of receiving input of predetermined specific portions on the two-dimensional photo of the garment by touch, and receiving the upper and lower length of the garment and the elastic modulus of the garment.
8. The method of claim 7,
The texture generation module includes:
3D, comprising a numerical calculation module 2 that calculates numerical values of predetermined parts on the two-dimensional photo of the garment, and calculates expansion values corresponding to the expandable size of the garment using the calculated numerical value and the elastic modulus Texture generation system.
10. The method of claim 9,
The input module 2,
The three-dimensional texture generation system, which can receive new values capable of updating the numerical values automatically calculated by the numerical calculation module and the extended numerical values.
11. The method of claim 10,
The texture generation module includes:
A system for generating a three-dimensional texture of the garment by using the two-dimensional photo of the garment and the numerical values, and generating an extended three-dimensional texture photo of the garment by using the two-dimensional photo of the garment and the expanded numerical values.
The three-dimensional avatar of the human body generated by the three-dimensional avatar generating system according to any one of claims 1 to 6 and the clothes generated by the three-dimensional texture generating system according to any one of claims 9 to 11 a receiving module for receiving a three-dimensional texture; and
and a three-dimensional fitting module configured to generate a three-dimensional avatar wearing a three-dimensional costume by performing rendering by applying the input three-dimensional texture to the input three-dimensional avatar.
13. The method of claim 12,
The three-dimensional fitting module,
Perimeter values for designated parts of the three-dimensional avatar (avatar figures), perimeter figures for designated parts of the three-dimensional texture (texture figures), and perimeter figures for designated parts of the extended three-dimensional texture (extended texture) numbers) by comparing
if the difference between the avatar value and the texture value is within a predetermined error range, apply the three-dimensional texture to the three-dimensional avatar to render it, and generate a signal that the clothes fit the body;
If the texture value is larger than the preset error range than the avatar value, the 3D avatar is determined in proportion to the size of the actual error range.
After adding a dummy of size, apply the three-dimensional texture to render and generate a signal that the costume is larger than the body,
if the avatar value is greater than the texture value, but is smaller than the extended texture value, the 3D avatar is rendered by applying the extended 3D texture, and a signal is generated that the clothes fit the body;
If the value of the avatar is greater than the predetermined error range and is greater than the value of the extended texture than the value of the texture, the extended 3D texture is applied to the 3D avatar in a differential form for rendering, and a signal indicating that the costume is smaller than the body is generated A three-dimensional fitting generation system, configured to
In the method of providing a clothing service through the Internet,
receiving a service request from a user who wants to use a virtual fitting service for displayed clothes through a site or an app;
receiving the ID and password of the user who requested the service and allowing login;
retrieving the information about the user and the information about the clothes from a DB of a server, and then configuring a 3D avatar for the user and a 3D texture or an extended 3D texture for the clothes;
Rendering the configured 3D avatar and the 3D texture or the extended 3D texture and displaying a result thereof.
15. The method of claim 14,
When receiving an order request, the step of transmitting the body size data of the user to the server or a third server; clothing service providing method through the Internet further comprising a.
16. The method of claim 15,
Displaying the garment production completion signal transmitted from the server or the third server on the site or the app; further comprising, a method for providing a garment service through the Internet.

Images