KR101418372B1 - System and method for recommending merchandise using virtual fitting - Google Patents

Abstract

The present invention relates to a product recommendation system and a recommendation method using virtual fittings. More particularly, the present invention relates to a product recommendation system and a recommendation method using virtual fittings, The present invention relates to a product recommendation system and a recommendation method using a virtual fitting for generating a virtual fitting image combining an avatar image of a customer and transmitting the created virtual fitting image to a user terminal.
According to the present invention, a virtual fitting image, which combines an avatar image similar to a user's actual appearance and a product image suitable for a user's purchase taste, is automatically provided to a user, It is possible not only to help the user to select the product himself, but also to promote the sales of the product and increase the satisfaction of the purchase of the product.

Description

{System and method for recommending merchandise using virtual fitting}

The present invention relates to a product recommendation system and a recommendation method using virtual fittings. More particularly, the present invention relates to a product recommendation system and a recommendation method using virtual fittings, The present invention relates to a product recommendation system and a recommendation method using a virtual fitting for generating a virtual fitting image combining an avatar image of a customer and transmitting the created virtual fitting image to a user terminal.

Recently, with the development of internet and e-commerce, online shopping mall selling products through online became popular. Because online shopping malls ease the trouble of visiting shops or department stores to purchase goods, busy modern people are increasingly being used.

However, since it is not about buying things directly, there are many cases in which the purchased items are different from what you think. The return rate is very high if you buy something through an online shopping mall because it is not suitable for you or because it differs from your own idea.

Accordingly, there is a growing demand for an online shopping mall that allows a user to directly confirm whether a product to be purchased matches his / her style such as his / her sex, age, height, weight, In addition to providing detailed virtual fittings for facial photographs, hairstyles, makeup styles, skin tones, and eyeglasses styles as well as gender, age, height, weight, body shape, There is a growing demand for an online shopping mall that will help you to do that.

SUMMARY OF THE INVENTION The present invention provides a product recommendation system and method using a virtual fitting that enables each user to conveniently select a product suitable for him by automatically providing a virtual fitting image suitable for his or her style to each user .

According to an aspect of the present invention, there is provided a merchandise recommendation system using virtual fittings, the merchandise recommendation system including merchandise information including information on type, color, size, material, and product image for each merchandise, A commodity database for adding commodity information of the new commodities; A customer database storing appearance information including a customer's facial image and physical condition, and preference information related to a customer's purchase preference; A product extracting unit for extracting a recommended product matching with customer's preference information among the received new products; A virtual fitting unit for generating an avatar image of a customer from appearance information of a customer and combining a product image of a recommended product extracted in the generated avatar image to generate a virtual fitting image; And a transmitting unit for transmitting the created virtual fitting image to a user's terminal in an e-mail or message form.

Preferably, the customer database further stores purchase history information on a purchase item purchased by a user and information on goods held by a user, and the product recommendation system stores purchase history information and product information Further comprising a merchandise matching unit for analyzing and extracting a list of merchandise held by the user and for extracting a merchandise matching the recommended merchandise from the merchandise list, wherein the virtual fitting unit further combines the merchandise images of the extracted merchandise A virtual fitting image can be generated.

Preferably, the commodity recommendation system includes: a matching information database storing information on a matching rate that is appropriate for matching among various goods; And a matching rate calculation unit for calculating a matching rate between the recommended product and the extracted owned product on the basis of the matching information database, wherein the owned product matching unit can extract the owned product having the highest matching rate with the recommended product.

Preferably, the merchandise recommendation system may further include an other merchandise matching unit for searching for a merchandise database based on the user's preference information and extracting other merchandise matching the recommended merchandise, A virtual fitting image can be generated by further combining images.

Preferably, the recommendation product extracting unit extracts a recommendation product matching the user's preference information and weather information from among the new products received based on the provided weather information, and the virtual fitting unit extracts a recommended product corresponding to weather information and weather information The period can be displayed together with the virtual fitting image.

According to another aspect of the present invention, there is provided a product recommendation method using virtual fittings, the method comprising the steps of: (a) storing product information including information on types, colors, sizes, materials, Adding product information of new goods received; (b) cross-searching a customer database and a product database storing preference information related to the customer's purchase taste of the customer and the appearance information including the customer's facial image and physical condition, Extracting a matching recommended product; (c) generating an avatar image of the user based on the appearance information, and creating a virtual fitting image by combining the product images of the recommended product extracted in the generated avatar image; And (d) transmitting the virtual fitting image in the form of an e-mail or a message to a user's terminal of the customer.

According to another aspect of the present invention, there is provided a computer-readable recording medium on which a program for realizing a product recommendation method using virtual fitting is recorded.

According to the present invention, a virtual fitting image, which combines an avatar image similar to a user's actual appearance and a product image suitable for a user's purchase taste, is automatically provided to a user, It is possible not only to help the user to select the product himself, but also to promote the sales of the product and increase the satisfaction of the purchase of the product.

In addition, according to the present invention, it is possible to promote sales of a product to a potential customer by providing a virtual fitting image to a potential customer extracted and extracted from the product.

BRIEF DESCRIPTION OF THE DRAWINGS FIG. 1 is a diagram illustrating a product recommendation system and a recommendation method using a virtual fitting according to the present invention; FIG.
2 is a block diagram illustrating a virtual fitting server in a product recommendation system using a virtual fitting according to an embodiment of the present invention.
3 is a block diagram illustrating a user terminal in a product recommendation system using a virtual fitting according to the present invention.
4 is a flowchart illustrating a product recommendation method using a virtual fitting according to the present invention.
5 is a flowchart showing step S30 in FIG.

The present invention is capable of various modifications and various forms, and specific embodiments are illustrated in the drawings and described in detail in the text. It should be understood, however, that the invention is not intended to be limited to the particular forms disclosed, but includes all modifications, equivalents, and alternatives falling within the spirit and scope of the invention.

The terms first, second, etc. may be used to describe various components, but the components should not be limited by the terms. The terms are used only for the purpose of distinguishing one component from another. For example, without departing from the scope of the present invention, the first component may be referred to as a second component, and similarly, the second component may also be referred to as a first component.

The terminology used herein is for the purpose of describing particular embodiments only and is not intended to be limiting of the invention. The singular expressions include plural expressions unless the context clearly dictates otherwise. In the present application, the terms "comprising" or "having ", and the like, are intended to specify the presence of stated features, integers, steps, operations, elements, parts, or combinations thereof, But do not preclude the presence or addition of one or more other features, integers, steps, operations, elements, parts, or combinations thereof.

Unless defined otherwise, all terms used herein, including technical or scientific terms, have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs.

Terms such as those defined in commonly used dictionaries are to be interpreted as having a meaning consistent with the meaning in the context of the relevant art and are to be interpreted as ideal or overly formal in meaning unless explicitly defined in the present application Do not.

Hereinafter, preferred embodiments of the present invention will be described in detail with reference to the accompanying drawings. Like reference symbols in the drawings denote like elements.

1 is a diagram illustrating a product recommendation system and a recommendation method using a virtual fitting according to the present invention.

The product recommendation system and the recommendation method using the virtual fitting according to the present invention include at least one user terminal 200_1 to 200_n connected to the virtual fitting server 100 and the virtual fitting server 100 through a wired / And may further include a merchandising server 300 for providing merchandise information.

The virtual fitting server 100 generates a virtual fitting image by combining the product image matching the user and the avatar image of the user with each of the user terminals 200_1 to 200_n and transmits the generated virtual fitting image to the user terminals 200_1 to 200_n It is a server that performs the role of transmitting.

The user terminals 200_1 to 200_n are terminals owned by respective users and refer to a computer terminal or a portable terminal connected to a wired / wireless communication network or the Internet. A portable terminal includes at least one of a mobile phone, a smartphone, and a tablet computer using local and / or wireless communication.

The virtual fitting image generated in the virtual fitting server 100 is transmitted to the user terminals 200_1 to 200_n using at least one of the format of the e-mail, the multimedia message service (MMS) of the cellular phone, and the push message used in the Internet .

The virtual fitting server 100 includes a database for storing appearance information of each user, preference information related to each user's purchase preference, and product information about a new commodity as well as existing inventory commodities.

The appearance information is information including a user's facial image and physical condition, and the physical condition includes information on at least one of the user's sex, height, weight, and body shape.

The preference information is information indicating a user's purchase preference, and is information including at least one of a product type, a product style, a size, a color, and a brand. On the other hand, the preference information may further include ranking information in which the user's taste of goods purchased is sorted by the order. As will be described later, the ranking information is included in the purchase preference of the user, so that the matching rate of the recommended product and the user provided by the recommended product offering system according to the present invention can be calculated.

Appearance information and preference information are stored in the database when the user registers himself / herself with his / her face image, physical condition, and his / her purchase preference when registering the user as a user in the virtual fitting server 100.

The virtual fitting server 100 extracts a recommended product suitable for each user's taste based on the user's preference information, and transmits the virtual fitting image obtained by combining the extracted product image of the recommended product and the user's avatar image to the user .

In the present invention, the following two methods can be used as a method for recommending a product suitable for a user's taste and providing it to a user.

In the first method, when a new product is received, a product matching the user's preference information is extracted from the received new product using the user's preference information, and the product image of the extracted product and the avatar image of the user are combined It is a method to provide to users.

In the second method, when a new product is received, a target customer having preference information matching each new product is extracted for each new product, and the extracted product image of the new product and the avatar image of the target customer are combined To the customer.

2 is a block diagram illustrating a virtual fitting server in a product recommendation system using a virtual fitting according to an embodiment of the present invention.

Referring to FIG. 2, the virtual fitting server 100 includes a product registration unit 110, a recommended product extraction unit 120, an item addition unit 130, a virtual fitting unit 140, a transmission unit 150, 160, a customer database 170, a matching information database 180, a weather information providing unit 190, and an order processing unit 195.

The product registration unit 110 registers product information of the new product in the product database 160 by adding the product information of the new products received in the new product to the product database 160.

The merchandise database 160 stores and updates merchandise information for each merchandise, and the merchandise information includes information on the type, color, size, material, and merchandise image of the merchandise.

The customer database 170 is a database for storing preference information 171 of the user's purchase preference and user's appearance information 172 for each user.

The user's appearance information 172 includes a user's facial image and physical condition, and is used to generate an avatar image similar to a user's actual appearance.

The user's preference information 171 includes at least one of information about the type, style, size, color, material, brand, and fashion trend of the user (e.g., office look, sight look, etc.). For example, in the case of women's clothes, there are various kinds of clothes such as dresses, suits, skirts, pants, shirts, blouses, t-shirts and underwears, and there are also various styles according to colors, materials and brands. The type, style, size, material, and color of the user's favorite clothes of various kinds, materials, and colors are stored in the product database 160 as preference information, and the preference information 171 is stored in the product database 160 And is used for extracting a recommended product suitable for the user's usual purchase taste from the product information of the stored various products.

In addition, the customer database 170 may store purchase history information for purchase items purchased by the user for each user, and stores the purchase history information about the items (clothes, sundries, accessories, etc.) Information may also be stored for each user. The purchase history information of the purchased product of the user can be automatically acquired through the history purchased from the site, and the stored product information of the product currently held by the user can be directly input from the customer in the user terminal 200 of the customer And can be obtained by receiving identification information of goods to be received or held.

The recommended product extracting unit 120 extracts a recommended product matching the user's preference information from the received new products when a new product is received. To this end, the recommendation product extracting unit 120 cross-compares the preference information of the user stored in the customer database 170 with the product information of the new products stored in the product database 160 to obtain a recommendation product matching the preference information .

On the other hand, the recommended product extracting unit 120 may extract a recommendation item matching the user's preference information and the weather among the new products based on the weather information provided from the weather information providing unit 190. For example, when the rainy season is started or the wind is heavy, the recommended product extraction unit 120 may extract the recommended product in consideration of the climate change, thereby providing the user with useful coordination information on the product suitable for the climate.

The item adding unit 130 adds a commodity item that can be matched to the recommended commodity, thereby matching various commodities harmoniously. When a specific pants is extracted as a recommended item, various item items such as a shirt, a blouse, a jacket, a shoe, a handbag and the like coordinated together with item items matching the specific pants extracted as a recommended item, for example, It is desirable to show.

The item adding unit 130 may include a merchandise matching unit 131, another merchandise matching unit 132, and a matching rate calculator 133.

The merchandise matching unit 131 analyzes the purchase history information stored in the customer database 170 to extract a list of owned merchandise held by the user, and matches the recommended merchandise with the recommended merchandise list extracted based on the user's preference information And extracts the corresponding merchandise. The images of the merchandise corresponding to the recommended merchandise are provided to the user in combination with the video of the recommended merchandise and the avatar image of the user, and the user can be assisted in ordering the recommended merchandise matching the merchandise owned by the user .

The present invention combines the recommended products extracted from the avatar images of the user and provides them to the user when the recommended products corresponding to the user's preferences are extracted. If the blouse is extracted as a recommended product, if the blouse image alone is combined with the avatar image of the user, there may occur a problem that the avatar image without the bottom is provided to the user. Also, when purchasing clothes or accessories, users are worried that "recommended products, clothing and accessories, do they fit well with other items I have".

In order to solve such problems and troubles, it is desirable to match the items possessed by the user with the recommended products and display them to the user. As described above, by providing the user with a matching image of the goods possessed by the user in the recommended product, the user can efficiently purchase the clothes suitable for him or herself, and when the image is recommended, the avatar image is provided without the bottom It is not so good, and it gives good effect even if it is beauty appreciation.

To help understand the explanation, it is assumed that the user has jeans, jeans, seven pants, jeans, hooded T-shirts, and the user has a favorite blouse.

The recommended product extracting unit 120 extracts a blouse that matches the user's preference information among the received new products.

Then, the merchandise matching unit 131 extracts, from the customer database 170, the lower information held by the user (jeans, jeans, pants, trousers, shorts, etc.)

Then, the virtual fitting unit 140 generates a virtual fitting image by combining the avatar image of the customer with the blouse, the jeans, the blouse and the shade, the blouse and the pants, the pants, . As described above, according to the present invention, the user has the advantage of being able to efficiently confirm how well the costume and the recommended costume he / she has with himself / herself.

At this time, the virtual fitting unit 140 may provide a virtual fitting image that is a combination of a blouse as a recommended product and all the bottoms possessed by the user, selects a bottom that best matches the blouse, And may provide an image.

The other goods matching unit 132 searches the goods database 160 based on the user's preference information and extracts other goods matching the recommended goods. The image of the other product matching the recommended product is provided to the user in combination with the image of the recommended product and the avatar image of the user.

The matching information database 180 is a database that stores information about matching rates that are appropriate for various products. The matching information database 180 may include, for example, the matching information of a particular style, the degree of suitability of the bottom of the particular style, the degree to which clothes of a particular color meet each other, the suitability of a particular style of apparel accessories such as bags, scarves and sunglasses, Information on the matching rate is stored in a database.

Match rate will be explained in more detail.

Basically, the top is matched with the bottom, and in detail, the shirt or the blouse in the top is matched with the jacket or the cardigan, and the jacket can be matched with the court. In addition, a miscellaneous goods such as a bag, a shoe, and a scarf are matched in a state in which both the top and bottom are aligned, and it is preferable that the tops and bottoms such as sunglasses, necklaces and earrings are matched with each other.

In the case of goods for layered look, the bottom skirt and leggings can be matched together, but long jeans or long cotton pants can not be matched together. At the same time, it is rare to wear jeans and cotton pants together.

On the other hand, even if the top and bottom can be basically matched, the matching degree may vary depending on the type, color, and material of the top and bottom. For example, jeans are not well suited to Hanbok jeogori, and clothing of certain colors may be avoided. In addition, the accessory and miscellaneous goods are different depending on the color and the material of the clothes of various colors.

As such, the matching rate information means information obtained by statistically matching the degrees of matching among the respective products. The matching jeans and cotton pants can not be worn together and can not be worn together. In conclusion, the higher the matching degree among the items that are matched together, the higher the matching rate value, and the lower the matching rate value means that the matched products do not match each other.

The matching rate calculation unit 133 can calculate the matching rate between the recommended product and the owned product, and between the recommended product and the other product based on the matching rate information stored in the matching information database 180. [

The calculated matching rate is provided to the virtual fitting unit 140, and may be included in the virtual fitting image and provided to the user. The user can be provided with the matching rate of the recommended product, the owned product, and the recommended product and the other product together with the virtual fitting image, so that it is possible to confirm which products are well compatible with each other and which product is suitable for me .

As described above, the user can receive images matching the recommended product with all the possessed products of the user. According to the present invention, since the calculated matching result of each virtual fitting image can be displayed together with the virtual fitting image combined with the recommended product and the corresponding article, the user can easily recognize which recommended product the user owns, It is possible to know objectively which product is compatible with which product of the commodity.

The calculated matching rate is provided to the corresponding merchandise matching unit 131 and the other merchandise matching unit 132. The owned merchandise matching unit 131 and the other merchandise matching unit 132 compare the calculated matching rate with the recommended merchandise And the other products having the highest matching rate with other products can be extracted.

On the other hand, the merchandise matching unit 131 and the other merchandise matching unit 132 do not extract the merchandise and other merchandise whose matching rate with the recommended merchandise is 0 on the basis of the matching rate calculation result. As described above, since the matching rate is 0 for a commodity item that can not be worn together, it is possible to prevent the same kind of commodity item that can not be worn together with a recommended commodity from being extracted as a commodity or another commodity.

For example, when a jeans is extracted as a recommended product, the matching product ratio between the base product and the pants as the other product matching the recommended product is zero, thereby preventing the jeans and the pants from being matched at the same time.

On the other hand, the item adding unit 130 may extract the recommended goods and the other goods, which best match the recommended goods and weather information, based on the current or future weather information provided from the weather information providing unit 190 . The weather information is provided to the virtual fitting unit 140 and transmitted to the user terminal 200 together with the virtual fitting image.

For example, in the case of a rainy season a week later, the recommended product extracting unit 120 extracts the raincoat as a recommended product, and the item adding unit 130 stores the raincoat, rain boots, or waterproof pants And the like can be extracted. Then, the virtual fitting unit 140 creates a virtual fitting image by combining the extracted recommended product, the owned product, or the recommended product and the image of the other product with the user's avatar image, and transmits the generated virtual fitting image to the user terminal 200 ). At this time, it is preferable that the information about the period corresponding to the weather information and the weather information is displayed together with the virtual fitting image. According to the present invention, since weather information of a specific period is provided to a user together with a virtual fitting image, a user can know the weather of a specific period in advance, and clothes corresponding thereto can be recommended in advance.

The virtual fitting unit 140 extracts the product image of the recommended product from the product database 160 and extracts the appearance information of the user from the customer database 170 to generate the avatar image. Then, the virtual fitting unit 140 generates a virtual fitting image by combining the product image of the recommended product and the avatar image of the user.

The transmitting unit 150 transmits the generated virtual fitting image to the user terminal 200 of the corresponding user using at least one of e-mail, multimedia message service (MMS) of the cellular phone, and push message used in the Internet.

Since the avatar image of the user is generated based on the appearance information including the user's facial image and physical condition, it is similar to the actual appearance characteristic of the user. The present invention automatically provides a user with a virtual fitting image in which an image of a recommended product is combined with an avatar image similar to a real appearance of a user so that the user can receive information on a recommended product that he / If you want to order more easily, you can easily order the recommended product by returning the received e-mail or message.

Meanwhile, the virtual fitting unit 140 may generate a virtual fitting image by further combining the product image of the recommended product and the avatar image of the user with the product image of the product that can be matched to the recommended product, It is also possible to display the matching rate of the recommended product and the corresponding product.

The virtual fitting unit 140 may further generate a virtual fitting image by further combining the product image of the recommended product and the product image of the other product that can be matched to the recommended product to the user's avatar image, The matching rate between the recommended product and the other product can also be displayed.

Of course, the virtual fitting unit 140 combines all of the product image of the recommended product, the avatar image of the user, the product image of the owned product that can be matched to the recommended product, and the product image of the other product that can be matched to the recommended product, A fitting image can be generated.

The order processing unit 195 receives an order application from the user terminal 200 and processes the settlement procedure based on the received order application.

1, a target customer having preference information matching with a new product is extracted for each new product, and the extracted product image of the new product and the avatar image of the target customer are combined to be transmitted to the target customer Or the like.

In this embodiment, the virtual fitting server 100 compares the preference information 171 of each user stored in the customer database 170, and compares the preference information 171 with a target customer extracting unit Time). The target customer list matched to each new product is provided to the item adding unit 130 and the virtual fitting unit 140. Each component other than the target customer extracting unit is the same as each of the constituent elements shown in FIG. 1, so that a duplicate description will be omitted.

3 is a block diagram illustrating a user terminal in a product recommendation system using a virtual fitting according to the present invention.

Referring to FIG. 3, the user terminal 200 includes an input / output unit 210, an additional fitting unit 220, and an ordering unit 230.

The input / output unit 210 outputs the virtual fitting image transmitted from the virtual fitting server 100 to the screen, and receives various input information including an order request from the user.

Since the virtual fitting image is an image obtained by combining the avatar image of the user generated from the user's appearance information stored in advance and the product image of the recommended product in accordance with the size, the virtual fitting image may be slightly different from the current actual user appearance over time.

Therefore, the input / output unit 210 further includes an additional feature input unit 215 to provide an effect that the user feels fitting. The additional feature input unit 215 may receive an additional feature including at least one of a face image, a hair style, a makeup style, a skin tone, a glasses style, and the like from a user.

The additional fitting unit 220 changes the virtual fitting image according to the received additional property, and outputs the modified virtual fitting image to the screen.

Also, the user can rotate the virtual fitting image in a desired direction. For this, the input / output unit 210 may receive input information for rotating the virtual fitting image from the user in a desired direction.

 When the input information for rotation of the virtual fitting image is inputted, the additional fitting unit 220 rotates the virtual fitting image according to the input information of the user or changes the virtual fitting image according to the input additional property.

The ordering unit 230 is a component that performs a role of receiving an ordering request from a user. The order request includes payment information that the user needs to complete the payment process.

Hereinafter, a product recommendation method using the virtual fitting according to the present invention will be described. The product recommendation method using the virtual fitting according to the present invention is essentially the same as the product recommendation system using the virtual fitting, and thus a detailed description and a duplicate description will be omitted.

4 is a flowchart illustrating a product recommendation method using a virtual fitting according to the present invention.

The preferred merchandise registration unit 110 adds merchandise information of the received new merchandise to the merchandise database 160 and registers the merchandise information of the new merchandise in the merchandise database 160 (S10). The merchandise database 160 stores and updates merchandise information for each merchandise, and the merchandise information includes information on the type, color, size, material, and merchandise image of the merchandise.

After the new product is received, the recommended product extraction unit 120 extracts a recommended product matching the user's preference information from the received new products (S20). For this, the recommendation product extractor 120 searches for the preference information 171 of the user stored in the customer database 170 and the product information of the new products stored in the product database 160, Extract recommended product.

The virtual fitting unit 140 combines the product image of the recommended product with the avatar image of the user to generate a virtual fitting image (S30). The avatar image of the user is generated on the basis of the appearance information of the user extracted from the customer database 170, and the product image of the recommended product is extracted from the product database 160.

The transmitting unit 150 transmits the generated virtual fitting image to the user terminal 200 using at least one of e-mail, a multimedia message service (MMS) of the cellular phone, and a push message used in the Internet at step S40.

The user may check the virtual fitting image through the user terminal 200 and then input the order request to the user terminal 200 when the user wants to purchase the recommended product included in the virtual pitching image. When an order request is input to the user terminal 200, the inputted order request is transmitted to the virtual fitting server 100.

Upon receiving the order request, the virtual fitting server 100 performs a payment process for the order request (S50).

FIG. 5 is a detailed flowchart of step S30 in FIG. 4, and step S30 may include at least one of the following steps.

The merchandise matching unit 131 analyzes the purchase history information stored in the customer database 170 to extract a list of owned merchandise held by the user, and matches the recommended merchandise with the recommended merchandise list extracted based on the user's preference information (S31).

The other goods matching unit 132 searches the goods database 160 based on the user's preference information and extracts other goods matching the recommended goods (S32).

The matching rate calculation unit 133 calculates a matching rate between the recommended product and the owned product, and between the recommended product and the other product based on the matching rate information stored in the matching information database 180 (S33). Here, the matching information database 180 is a database that stores information on matching rates that are appropriate for various products.

The calculated matching rate is provided to the merchandise matching unit 131 and the other merchandise matching unit 132. The merchandise matching unit 131 and the other merchandise matching unit 132 compare the matching rate with the recommended merchandise according to the calculated matching rate. It is possible to extract the goods and other goods having the highest matching rate.

On the other hand, the merchandise matching unit 131 and the other merchandise matching unit 132 extract the recommended merchandise and the other merchandise that best fits the current weather, based on the weather information received from the weather information provider 190 .

The merchandise and other merchandise matching the extracted recommendation merchandise are provided to the virtual fitting unit 140. The virtual fitting unit 140 stores the merchandise image of the recommended merchandise, the avatar image of the user, A product image of a product, and a product image of another product that can be matched to a recommended product can be combined to create a virtual fitting image.

The method of the present invention can also be embodied as computer readable code on a computer readable recording medium. A computer-readable recording medium includes all kinds of recording apparatuses in which data that can be read by a computer system is stored. Examples of the computer-readable recording medium include a ROM, a RAM, a CD-ROM, a magnetic tape, a floppy disk, an optical data storage device, and the like, and may be implemented in the form of a carrier wave (for example, transmission via the Internet) . The computer-readable recording medium may also be distributed over a networked computer system so that computer readable code can be stored and executed in a distributed manner.

While the present invention has been particularly shown and described with reference to exemplary embodiments thereof, it is to be understood that the invention is not limited to the disclosed embodiments, but, on the contrary, is intended to cover various modifications and equivalent arrangements included within the spirit and scope of the invention. Accordingly, the scope of protection of the present invention should be construed in accordance with the following claims, and all technical ideas within the scope of equivalents and equivalents thereof should be construed as being covered by the scope of the present invention.

100, a virtual fitting server 200, a user terminal
110, a product registration unit 120, a recommended product extraction unit
130, an item adding unit 140, a virtual fitting unit
150, a transmission unit 160, a goods database
170, a customer database 180, a matching information database
190, weather information supply

Claims (17)

A commodity database for storing commodity information including information on type, color, size, material and image of each commodity for each commodity, and adding new commodities of the new commodities when new commodities are received;
Stores appearance information including the customer's facial image and physical condition, preference information related to the purchase preference of the customer, purchasing history information of the purchased product purchased by the user, and information on the product possessed by the user Customer database;
A product extracting unit for extracting a recommended product matching the customer's preference information from the received new products;
A merchandise matching unit for analyzing the purchase history information and the merchandise information to extract a list of merchandise held by the user and extracting a merchandise matching the recommended merchandise from the merchandise list;
A virtual fitting unit for generating an avatar image of a customer from the customer's appearance information, combining the product image of the extracted recommended product and the product image of the extracted product with the generated avatar image, and creating a virtual fitting image; And
And a transmitting unit for transmitting the generated virtual fitting image to a user's terminal in the form of an e-mail or a message. The method according to claim 1,
An additional feature input unit receiving an additional feature including at least one of a face image, a hairstyle, a makeup style, a skin tone, a glasses style, and the like from a user; And
And an additional fitting unit for changing the virtual fitting image according to an input additional property and outputting the changed virtual fitting image to a screen. The product recommendation system according to claim 1,
A matching information database storing information on a matching rate that is suitable for matching among various goods; And
And a matching rate calculator for calculating a matching rate between the recommended product and the extracted article based on the matching information database,
Wherein the merchandise matching unit extracts a merchandise item having a highest matching rate with the recommended merchandise item. The product recommendation system according to claim 1,
A matching information database storing information on a matching rate that is suitable for matching among various goods; And
And a matching rate calculator for calculating a matching rate between the recommended product and the extracted article based on the matching information database,
Wherein the merchandise matching unit extracts all of the merchandise whose matching rate with the recommended merchandise is 0, excluding the merchandise whose matching rate with the recommended merchandise is 0,
Wherein the virtual fitting unit generates the virtual fitting image by combining the extracted product images of all the owned products with the product image and the avatar image of the recommended product, respectively. The product recommendation system according to claim 1, further comprising an other goods matching unit that searches the goods database based on the user's preference information and extracts other goods matched to the recommended goods,
Wherein the virtual fitting unit further generates the virtual fitting image by further combining product images of the extracted other products. 6. The product recommendation system according to claim 5,
A matching information database storing information on a matching rate that is suitable for matching among various goods; And
And a matching rate calculator for calculating a matching rate between the recommended product and the extracted other product based on the matching information database,
And the other goods matching unit extracts the other goods having the highest matching rate with the recommended goods. The apparatus of claim 1, wherein the product extracting unit extracts a user's preference information and a recommended product matching the weather information from the received new products based on the received weather information,
Wherein the virtual fitting unit displays a period corresponding to the weather information and the weather information together with the virtual fitting image. The product recommendation system according to claim 1,
Further comprising an order processing unit for processing the received order request when an order request is received from the user terminal. (a) storing product information including a type, a color, a size, a material, and information on a product image of the product database, and adding product information of the new goods when the new product is in stock;
(b) The product extracting unit extracts information on the appearance of the customer, including appearance information including the customer's facial image and physical condition, preference information related to the customer's product purchase preference, purchase history information on the purchased product purchased by the user, A step of cross-searching a customer database storing stored product information and the product database to extract a recommended product matching the user's preference information from the received new products;
(c) analyzing the purchase history information and the stored goods information to extract a list of owned goods owned by the user, and extracting a corresponding merchandise matching the recommended goods from the owned goods list;
(d) The virtual fitting unit generates an avatar image of the user based on the appearance information, combines the product image of the extracted recommended product and the product image of the extracted owned product in the generated avatar image, ; And
(e) transmitting a virtual fitting image to the user's terminal in the form of an e-mail or a message by the transmitting unit. 10. The product recommendation method according to claim 9,
The user terminal of the customer receiving an additional feature including at least one of a facial image, a hairstyle, a makeup style, a skin tone, a glasses style, etc. from the user;
Further comprising changing the virtual fitting image according to the input additional property and outputting the modified virtual fitting image to the screen. 10. The method of claim 9, wherein step (c)
Calculating a matching rate between the recommendation product and the extracted owned goods by searching a matching information database storing information on a matching rate that matches each other among various goods; And
The method of claim 12, further comprising: extracting a product having a highest matching rate with the recommended product. 10. The method of claim 9, wherein step (c)
Calculating a matching rate between the recommendation product and the extracted owned goods by searching a matching information database storing information on a matching rate that matches each other among various goods; And
The method according to any one of claims 1 to 3, further comprising the steps of: extracting all the merchandise whose matching rate is not 0, excluding the merchandise whose matching rate with the recommended merchandise is 0,
Wherein the virtual fitting unit generates the virtual fitting image by combining the extracted product images of all the owned products with the product images and the avatar images of the recommended products, respectively. 10. The product recommendation method according to claim 9,
The other goods matching unit may further include searching the goods database based on the user's preference information and extracting other goods matching the recommended goods,
Wherein the virtual fitting image is generated by further combining images of the extracted other products. 14. The product recommendation method according to claim 13,
Calculating a matching rate between the recommended product and the extracted other product by searching a matching information database storing information on a matching rate that is appropriate for matching among various products; And
Further comprising the step of extracting the other goods having the highest matching rate with the recommended goods, by the other goods matching unit. 10. The method of claim 9, wherein step (b)
Wherein the product extracting unit extracts a user's preference information and a recommended product matching the weather information from the received new products based on the received weather information,
Wherein the virtual fitting image is displayed together with the weather information and the period corresponding to the weather information. 10. The method of claim 9,
Further comprising processing the received order request when the order processing unit has received the order request from the user terminal. A computer-readable recording medium storing a program for realizing the method according to any one of claims 9 to 16.

Images