KR20250095386A - Clothing and shoe matching system and method in online shopping mall - Google Patents

Abstract

The present invention relates to a system and method for matching clothes and shoes in an online shopping mall. According to the present invention, the system comprises: a 3D model registration unit for registering a 3D model acquired by photographing a user's body from a user terminal connected to an online shopping mall; a product selection unit for selecting a product desired by the user from a list of products sold in the online shopping mall and selecting a size recommendation function for the selected product; a 3D model comparison unit for comparing the 3D model of the user with the 3D models of existing purchasers who have purchased the product when the size recommendation function is selected, and searching for at least one purchaser whose 3D model has a high similarity with the user; a size estimation unit for analyzing reviews of the purchasers regarding the sizes of the products already registered, and estimating the size of the product matching the user's body type; and a size recommendation unit for recommending and providing the estimated size to the user.
According to the present invention, it is possible to recommend a product size that fits the user's body type by considering both the purchase size and registered reviews by buyers who have 3D models with high similarity to the user.

Description

{Clothing and shoe matching system and method in online shopping mall}

The present invention relates to a clothing and shoe matching system and method in an online shopping mall, and more particularly, to a clothing and shoe matching system and method in an online shopping mall for recommending a product of a size that fits a user's body type in a shopping mall selling clothing or shoes.

One of the most difficult things about buying clothes, shoes, etc. at a shopping mall is choosing the size. Usually, when purchasing a product, the user chooses the size that he or she usually wears or wears, but since the sizes of each product are different even for the same size, choosing the size is very difficult, and there are frequent cases where the size needs to be exchanged.

Currently, most people guess the size of a product by reading reviews of the product. Since users usually judge whether a product is larger or smaller than its size by reading reviews, the accuracy of size judgment is low.

Another way to determine the size of a product is to provide the user with an image of the product virtually fitting on the 3D model of the user's body using a 3D model of the clothing or shoe in question and a 3D model of the user's body. This allows the user to try on various sizes in a virtual environment and purchase the size they want.

However, the above-described method has the difficulty that all products must have 3D models.

The technology underlying the present invention is disclosed in Korean Patent Publication No. 2022-0076637 (published on June 8, 2022).

The purpose of this invention is to provide a clothing and shoe matching system and method in an online shopping mall that can recommend a product size that fits the user's body type by combining a user's 3D model, a buyer's 3D model, and a product review registered by the buyer in the shopping mall selling the product.

The present invention comprises: a 3D model registration unit for receiving a 3D model acquired by photographing a user's body from a user terminal connected to an online shopping mall; a product selection unit for receiving a user's selection of a desired product from a list of products sold in an online shopping mall and selecting a size recommendation function for the corresponding product; a 3D model comparison unit for comparing the user's 3D model with a 3D model already registered of a previous purchaser who purchased the product when the size recommendation function is selected to search for at least one purchaser whose 3D model is highly similar to the user; a size estimation unit for analyzing reviews of the previous purchasers regarding the size of the product to estimate a size of a product matching the user's body type; and a size recommendation unit for recommending and providing the estimated size to the user.

In addition, the 3D model registration unit can create a 3D mesh model implemented as a mesh structure based on an image of the user's body or foot taken through the user terminal and register it in the user's account.

In addition, the model comparison unit determines a body part for judging the similarity between the 3D models of the user and the purchaser according to the type of the product, and compares the similarity between the corresponding body parts on the 3D models of the user and the purchaser, and the type of the product may include at least one of upper garments, lower garments, hats, shoes, and gloves.

Additionally, the review of the size of the product may include at least one of a descriptive review in which the review is described in text form and a questionnaire review in which a choice is provided regarding the degree of size, large or small, and is registered in the survey.

In addition, the size recommendation unit can provide data summarizing the estimated size and each buyer's registered review to the user terminal.

In addition, the above size recommendation section can provide summary data of statistical data on actual purchase sizes and size evaluation results extracted from the registered reviews of each purchaser.

In addition, the clothing and shoe matching system may further include a review registration section that receives reviews of the sizes of products from user terminals after the user purchases the products.

And, the present invention relates to a clothing and shoe matching method performed by a clothing and shoe matching system in an online shopping mall, comprising: a step of registering a 3D model acquired by photographing a user's body from a user terminal connected to the online shopping mall; a step of having the user select a desired product from a list of products sold in the online shopping mall and selecting a size recommendation function for the corresponding product; a step of comparing the 3D model of the user with a 3D model already registered of an existing purchaser who purchased the product when the size recommendation function is selected, and searching for at least one purchaser whose 3D model has a high similarity between the user and the user; a step of analyzing a review of the product already registered by the purchaser to estimate a size of a product matching the user's body type; and a step of recommending and providing the estimated size to the user.

In addition, the clothing and shoe matching method may further include a step of receiving a review of the size of the product from the user terminal after the user purchases the product.

According to the present invention, a shopping mall selling clothing or shoes can recommend a product size that fits the user's body type by combining 3D models of users and buyers and registered reviews of buyers.

In this way, the present invention allows users to provide a 3D model of their body shape to a shopping mall, and receive recommendations for clothing or shoes of a size most similar to their 3D model.

In addition, the present invention can recommend a size by considering both the purchase size and the registered reviews by buyers who have 3D models with high similarity to the user.

FIG. 1 is a drawing showing a clothing and shoe matching system in an online shopping mall according to an embodiment of the present invention.
Figure 2 is a drawing showing in detail the configuration of the clothing and shoe matching system illustrated in Figure 1.
Figure 3 is a drawing explaining a clothing and shoe matching method using the system of Figure 2.
FIG. 4 is a drawing showing the result of generating a 3D mesh model for a user's body or foot part according to an embodiment of the present invention.
FIG. 5 is a diagram showing two examples of a method for comparing similarity between 3D mesh models in an embodiment of the present invention.

Hereinafter, with reference to the attached drawings, embodiments of the present invention will be described in detail so that those with ordinary skill in the art can easily implement the present invention. However, the present invention may be implemented in various different forms and is not limited to the embodiments described herein. In addition, in order to clearly describe the present invention in the drawings, parts that are not related to the description are omitted, and similar parts are assigned similar drawing reference numerals throughout the specification.

Throughout the specification, when a part is said to be "connected" to another part, this includes not only the case where it is "directly connected" but also the case where it is "electrically connected" with another element in between. Also, when a part is said to "include" a component, this does not mean that it excludes other components, but rather that it may include other components, unless otherwise stated.

FIG. 1 is a drawing showing a clothing and shoe matching system in an online shopping mall according to an embodiment of the present invention.

As shown in Fig. 1, the clothing and shoe matching system (100) in an online shopping mall according to an embodiment of the present invention can be connected to a user terminal (200) through a wired, wireless, or wired/wireless combination network to transmit and receive mutual information. The wireless network can include at least one of RF, WLAN, Wi-Fi, and Bluetooth methods, and can use various known wireless network methods.

The clothing and shoe matching system (100) may correspond to a server that provides clothing and shoe matching services in an online shopping mall. The clothing and shoe matching system (100) may store and manage data collected through a user terminal (200) and provide various data related to the shopping mall to the user terminal (200). Of course, the clothing and shoe matching system (100) may be included in the operation service platform of an online shopping mall site.

The clothing and shoe matching system (100) can provide a product matching service platform implemented as an application or web to a network-connected user terminal (200). The service platform can be an application program running in a mobile app or web environment.

The user terminal (200) may include a device such as a PC, a desktop, a smart phone, a tablet, a notebook, etc., which can connect to a wired or wireless network and exchange information. For convenience of explanation, the drawing illustrates one user terminal, but in reality, a plurality of user terminals may be connected through the network.

The clothing and shoe matching system (100) according to an embodiment of the present invention can recommend a product size that fits the user's body type by considering both the 3D model of the user who wishes to purchase the product and the 3D models and registered reviews of existing purchasers of the product.

Figure 2 is a drawing showing in detail the configuration of the clothing and shoe matching system illustrated in Figure 1.

As shown in FIG. 2, the clothing and shoe matching system (100) according to the embodiment of the present invention includes a 3D model registration unit (110), a product selection unit (120), a 3D model comparison unit (130), a size estimation unit (140), a size recommendation unit (150), and a review registration unit (160). Here, the operation of each unit (110 to 160) and the data flow between each unit can be controlled by a control unit (not shown).

This clothing and shoe matching system (100) may be implemented as a computer device physically configured to include a processor, memory, a user interface input/output device and a storage device, a network input/output unit, etc., or may be implemented as software including an application program running on a computer device or a user terminal.

The 3D model registration unit (110) registers a 3D model obtained by taking a picture of the user's body from a user terminal (200) connected to an online shopping mall. This 3D model registration unit (110) can register, store, and manage 3D models for multiple users.

The product selection section (120) allows the user to select a desired product from a list of products sold in an online shopping mall, and selects a size recommendation function for the product. To this end, a size recommendation button for each product sold in the shopping mall can be provided.

When the size recommendation function is selected in this way, the 3D model comparison unit (130) compares the user's 3D model with the registered 3D models of existing buyers who purchased the product to search for at least one buyer with a high degree of similarity between the user and the 3D model. At this time, the one buyer with the highest degree of similarity or the top N buyers with the highest degree of similarity (e.g., 3 or 5) may be searched for.

The size estimation unit (140) analyzes reviews of the searched purchasers regarding the sizes of the products registered above to estimate the size of the product that matches the user's body type.

The size recommendation unit (150) provides a recommendation for the estimated size through the user terminal (200). At this time, summary data on the registered reviews of each purchaser may be provided together with the estimated size.

The review registration section (160) receives reviews from users about the size of a product after the user purchases the product. The review data registered in this way can be used as data for recommending the optimal size for the product to other users with similar body types.

The following describes in more detail a method for matching clothing and shoes in an online shopping mall according to an embodiment of the present invention.

Figure 3 is a drawing explaining a clothing and shoe matching method using the system of Figure 2.

First, the 3D model registration unit (110) registers a 3D model obtained by photographing the user's body from a user terminal (200) connected to an online shopping mall (S310). Here, the 3D model may correspond to a 3D mesh model implemented in a mesh structure.

Specifically, the 3D model registration unit (110) can create a 3D mesh model, which is a 3D model of a mesh structure, based on images of the user's body or feet taken from various angles through the user terminal (200), and can register the created 3D mesh model to the user's account.

A technique for 3D modeling a person or object based on an image taken from multiple angles can use various techniques known in the art. According to an embodiment of the present invention, a 3D mesh model can be created based on a single image taken of a user's body (full body) or foot area or images taken from multiple angles.

FIG. 4 is a drawing showing the result of generating a 3D mesh model of a user's body or foot part according to an embodiment of the present invention. This FIG. 4 shows a procedure for generating a 3D model of a user's body and uploading it to the user's shopping mall account.

In order to create a 3D model, a body or foot part must be photographed, and a 3D model can be created using a single photographed image or images from various angles. This can be done through deep learning inference on a user terminal (smartphone) or a server, and the created 3D mesh model can be stored on the user terminal or the server.

The 3D model created in this way is the result of modeling the body as a 3D mesh and corresponds to a model that includes the actual body size.

The product selection section (120) selects a product to be purchased from a list of products sold in an online shopping mall and selects a size recommendation for the product (S320). For example, a user can search for a product to be purchased on an online shopping mall site and click the "size recommendation" button for the product.

Next, when the size recommendation button is selected, the 3D model comparison unit (130) compares the user's 3D model with the registered 3D models of existing buyers who have purchased the product, and searches for at least one buyer whose 3D model has a high degree of similarity with the user (S330).

Here, the 3D model of the existing purchaser may correspond to information previously stored in the 3D model registration unit (110). Other users can also register 3D models of their bodies through their user terminals.

In this way, when the “size recommendation” button for a product is selected, the system (100) of the present invention executes a preset recommendation algorithm to compare the user’s 3D model with the 3D models of previous purchasers of the product, and search for at least one purchaser among the purchasers of the product who has a 3D model that is most similar or has a high degree of similarity to the user’s 3D model.

In the embodiment of the present invention, the similarity between 3D mesh models can be measured using various methods known in the art. Two examples are briefly described below, but the method for comparing similarity between 3D models applied to the embodiment of the present invention can be applied in various ways, and deep learning-based similarity analysis technology can also be utilized.

FIG. 5 is a diagram showing two examples of a method for comparing similarity between 3D mesh models in an embodiment of the present invention.

First, the first method of Fig. 5 (a) is a method of comparing the differences between corresponding vertices, edges, and polygons. For example, if there are two models, the difference in the Euclidean distance of the edges (egde) at the same location is measured. By measuring all edges of the entire model and averaging them, the similarity (inversely proportional to the distance difference) between the two models can be measured.

Next, the second method of Fig. 5 (b) is a method of comparing the differences between corresponding curvature distributions. In this case, the corresponding value varies depending on the shape of the curvature, and the similarity between the two models can be measured by utilizing the fact that the difference value increases as the shapes of the curvatures differ.

Here, in the similarity analysis step between these models, the 3D model comparison unit (130) determines body parts for judging the similarity between the 3D models of the user and the buyer according to the type of product, and can compare the similarity between the corresponding body parts on the 3D models of the user and the buyer.

At this time, the type of product may include at least one of top, bottom, hat, shoes, and gloves. If the selected product is bottom (pants), the lower body part can be selected on the 3D model and the similarity of the 3D model for this can be compared. In addition, if the selected product is shoes, the volume of the foot part can be selected on the 3D model and the similarity of the 3D model for this can be compared.

In addition, the 3D model comparison unit (130) may set priorities when comparing the user's 3D model with the buyer's 3D model. In the case of pants, the priorities for similarity comparison may be set in the order of waist circumference, length, width, etc. This may be preset as a default value in the system or may be changed by the user through the user terminal (200).

Next, when a buyer with a high similarity between the user and the 3D model is searched, the size estimation unit (140) analyzes reviews of sizes already registered by the searched buyer for the product to estimate the size of the product that matches the user's body type (S340).

Here, the review on size may include at least one of a descriptive review in which the review is described in text form, and a survey-type review that provides choices for the degree of size (e.g., very small, small, right, large, very large) and is registered through a survey. In the case of a survey-type review, since numerical conversion is possible, the size recommendation algorithm can be operated more easily. In the case of a descriptive review, the review content related to size can be analyzed through context analysis using text recognition.

In an embodiment of the present invention, the size estimation unit (140) can recommend a size by synthesizing the size information that the purchaser actually purchased from the review of the size of the product and the purchase review of the purchaser's size, i.e., size reviews such as "very small, small, right, big, very large."

For example, among shoe product purchasers, the buyer's registered review with the highest similarity to the 3D model of the user's foot can be combined with the buyer's actual purchased size and the size review results to recommend the optimal shoe size that fits the user's body type. If the actual purchased size is 240 and the review result for the size is "small," size 240 can be recommended, and if it is "very small," size 250 can be recommended.

As another example, the optimal shoe size for the user's body type can be recommended by combining the actual sizes purchased and size review results of three buyers from the registered reviews of the top N buyers with high similarity using a technique such as an average.

In this way, in the case of an embodiment of the present invention, among previous purchasers of the product, purchasers who have a 3D model similar to the user are first filtered and searched, and then the optimal size suitable for the user can be recommended based on the size reviews of the searched purchasers.

Next, the size recommendation unit (150) recommends the estimated size to the user (S350).

At this time, the size recommendation unit (150) can provide the estimated size and data summarizing the registered reviews of each buyer to the user terminal (200). Specifically, the size recommendation unit (150) can provide statistical data on the actual purchase size and size evaluation results extracted from the registered reviews of each buyer as summary data.

For example, statistical data on the actual purchase size and size review evaluation results for each buyer are processed and provided in the form of tables, graphs, images, etc., so that users can intuitively check the results. Of course, at the S350 stage, reviews from each buyer can be collected and displayed.

Afterwards, the user goes through the process of purchasing the product, wearing it to check the size, and leaving a review.

That is, the review registration unit (160) receives a review about the size of a product from a user terminal (200) after the user purchases the product (S360).

At this time, the degree of fit of the size can be registered from the user terminal (200) in the form of a text description or a questionnaire using choices. Registration of a review using choices is a numerical review because it allows for numerical conversion of the selected item. In this way, reviews can be descriptive, but it is desirable to include a numerical review as well in order for the size recommendation algorithm to work well. In the case of shoes, it is necessary to induce the user to check in several stages such as large or small, long or short.

In this way, review data registered after a user purchases a product can be used as reference data for size recommendations when other users purchase the product in the future.

According to the present invention as described above, a shopping mall selling clothing or shoes can recommend a product size that fits the user more accurately and conveniently by combining 3D models of users and buyers with registered reviews of buyers.

In this way, the present invention allows users to receive recommendations for clothing or shoes of the size most similar to their 3D model when they provide a 3D model of their body shape to a shopping mall. In addition, the present invention can recommend sizes by considering both the purchase sizes and registered reviews by buyers who have 3D models with a high degree of similarity to the user.

Although the present invention has been described with reference to the embodiments shown in the drawings, these are merely exemplary, and those skilled in the art will understand that various modifications and equivalent other embodiments are possible therefrom. Accordingly, the true technical protection scope of the present invention should be determined by the technical idea of the appended claims.

100: Clothing and footwear matching system 110: 3D model registry
120: Product Selection Section 130: 3D Model Comparison Section
140: Size estimation section 150: Size recommendation section
160: Review register 200: User terminal

Claims (9)

A 3D model registration unit that registers a 3D model obtained by taking a picture of the user's body from a user terminal connected to an online shopping mall;
A product selection section in which the user selects a desired product from a list of products sold in an online shopping mall and a size recommendation function for the product is selected;
A 3D model comparison unit that compares the 3D model of the user with the registered 3D models of existing purchasers who have purchased the product when the above size recommendation function is selected to search for at least one purchaser whose 3D model has a high similarity with the user;
A size estimation unit that analyzes reviews of the above-mentioned purchaser's registered size of the above-mentioned product and estimates the size of the product that matches the user's body type; and
A clothing and shoe matching system in an online shopping mall including a size recommendation section that recommends the above-mentioned estimated sizes to the user. In claim 1,
The above 3D model registration section,
A clothing and shoe matching system in an online shopping mall that creates a 3D mesh model implemented in a mesh structure based on an image of the user's body or feet taken through the user terminal and registers it in the user's account. In claim 1,
The above model comparison section is,
Depending on the type of the above product, the body part for judging the similarity between the 3D models of the above user and the purchaser is determined, and the similarity between the corresponding body part on the 3D models of the above user and the purchaser is compared.
The above product type is a clothing and shoe matching system in an online shopping mall including at least one of tops, bottoms, hats, shoes, and gloves. In claim 1,
Reviews on the size of the above product,
A clothing and shoe matching system in an online shopping mall, comprising at least one descriptive review in which the review is described in text format and one survey-type review that provides options for the degree of size, large or small, and is registered as a survey. In claim 1,
The above size recommendations are:
A clothing and shoe matching system in an online shopping mall that provides the estimated size and data summarizing each buyer's registered review to the user terminal. In claim 1,
The above size recommendations are:
A clothing and shoe matching system in an online shopping mall that provides summary data on statistical data on actual purchase sizes and size evaluation results extracted from the registered reviews of each buyer. In claim 1,
An online shopping mall that further includes a review registration section for receiving reviews of the size of a product from a user terminal after the user purchases the product. In a clothing and shoe matching method performed by a clothing and shoe matching system in an online shopping mall,
A step of registering a 3D model obtained by taking a picture of the user's body from a user terminal connected to an online shopping mall;
A step for having the user select a desired product from a list of products sold in an online shopping mall and selecting a size recommendation function for the product;
When the above size recommendation function is selected, a step of comparing the 3D model of the user with the registered 3D models of existing purchasers who have purchased the product to search for at least one purchaser whose 3D model has a high similarity with the user;
A step of estimating the size of a product that matches the user's body type by analyzing the reviews of the above-mentioned purchaser regarding the above-mentioned product; and
A method for matching clothes and shoes in an online shopping mall, comprising a step of recommending the estimated size to the user. In claim 8,
A method for matching clothing and shoes in an online shopping mall, further comprising the step of receiving a review of the size of the product from the user terminal after the user purchases the product.