KR20210036666A - System and method for recommending eyewear based on sales data by face type and size - Google Patents

Abstract

The present invention is a technology for recommending glasses based on a face mesh generated by scanning a customer's face and sales data by face type of a customer. A glasses recommendation system according to an embodiment comprises: a face mesh generation unit for generating a face mesh of a customer using a face scan result; a feature data extraction unit for extracting feature data from the generated face mesh of the customer; a face type determination unit for determining a face type of the customer based on the extracted feature data; a sales data extraction unit for extracting at least one or more sales data corresponding to the determined face type of the customer with reference to a database in which sales data by face type is recorded; a recommended product determination unit for determining at least one recommended product by reflecting customer's preference information in the extracted sales data; and a recommendation processing unit for processing the determined at least one or more recommended products to be provided to the customer.

Description

System and method for recommending eyewear based on sales data by face type and size}

The present invention is a technology for recommending glasses based on a face mesh generated by scanning a customer's face and sales data for each customer's face type, and sales data according to the size and shape of the face through cluster analysis and Bayesian analysis of the data. It connects with and recommends the optimal eyewear design with only face recognition.

The iPhone released by Apple can generate at least 30,000 infrared dots on a user's face, read the coordinates of the irradiated infrared dots using an infrared sensing camera, and use them to create a face mesh.

The created face mesh can be used for various purposes, such as being used for security purposes to check whether a registered user is correct based on the user's face data.

In particular, the created face mesh can be used for various purposes because it has absolute information on the size and shape of the face. Among these, it can be used for a technology that recommends suitable glasses for each individual by measuring the distance, angle, and the like between a specific coordinate of the face and the coordinates.

Korean Patent Publication No. 2015-0180244 "Method of recommending glasses for online glasses sales system" US Patent No. 9529213 "Method and system to create custom, user-specific eyewear"

An object of the present invention is to recommend an optimal eyeglass design only by recognizing the face by linking with sales data according to the size and shape of a face through cluster analysis and Bayesian analysis of data.

The present invention analyzes the customer's face through the Face Ruler function based on the result of the face scan in the iOS app, and based on this, the size and design of the glasses are recommended according to the size and shape of the face, and big data analysis is performed in the future. It aims to improve accuracy through.

An eyeglass recommendation system according to an embodiment includes a face mesh generator that generates a face mesh of a customer using a face scan result, a feature data extractor that extracts feature data from the generated face mesh of the customer, and the extracted feature data A face type determination unit that determines the customer's face type based on a sales data extraction unit that extracts at least one sales data corresponding to the determined customer's face type by referring to a database in which sales data for each face type are recorded. , A recommended product determination unit configured to determine at least one recommended product by reflecting the customer's preference information in the extracted sales data, and a recommendation processing unit configured to provide the determined at least one or more recommended products to the customer. have.

The face mesh generator according to an embodiment may use a result of scanning the face of the customer using a true depth camera supported through an iOS app as the face scan result.

The face type determiner according to an embodiment may determine the face type in consideration of a ratio, distance, and depth between the extracted feature data.

The face type determination unit according to an embodiment processes size analysis, shape analysis, shape ratio analysis, and nose analysis in consideration of the ratio, distance, and depth between the extracted feature data, and uses the results of the processed analyzes. Thus, the face type can be determined.

The face type determination unit according to an embodiment may calculate at least one of a face width for the customer, a width between pupils, a width between corners, a width between corners of the eyes, and a width of a nose pillar for the customer through the size analysis.

The face type determination unit according to an embodiment may determine at least one of a nose tip width, a cheekbone width, a mandible width, a vertical length, a forehead length, a forehead-nose tip length, and a nose tip-chin tip length for the customer through the shape analysis. Can be calculated.

The face type determination unit according to an embodiment may calculate at least one of a vertical-horizontal ratio, a mandibular-face width ratio, and a face width-total bone ratio for the customer through the shape ratio analysis.

The face type determination unit according to an embodiment may calculate at least one of a height of a nose, a height of a nose, a length of a nose, and an angle of the nose for the customer through the nose analysis.

The at least one or more sales data corresponding to the customer's face type according to an embodiment may include design information and size information for glasses.

The preference information according to an embodiment may include information input from the customer when the customer registers as a member.

The recommended product determination unit according to an embodiment performs at least one of a data cluster analysis, a user-based collaborative filtering, an item-based collaborative filtering, and a naive Bayesian algorithm using the determined face type and the extracted sales data. By applying the machine learning, it is possible to determine the recommended product using the result of the machine learning.

The recommended product determination unit according to an embodiment determines at least one recommended product by reflecting the customer's preference information in the extracted sales data, but when the customer does not select a recommended product, the recommended product is determined again. And, the recommendation processing unit may process to provide the re-determined recommended product to the customer.

The recommended product determination unit according to an embodiment determines at least one recommended product by reflecting the customer's preference information in the extracted sales data, but when the customer does not select a recommended product, the customer's face type The recommended products to which priority is given may be determined based on, and the recommendation processing unit may process to provide the determined recommended products to the customer.

The recommended product determination unit according to an embodiment determines at least one recommended product by reflecting the customer's preference information in the extracted sales data, but when the customer does not select a recommended product, a user-based collaboration The number of products selected through filtering is determined as recommended products, and from the next recommendation, recommended products related to the result of item-based collaborative filtering are determined, and the recommendation processing unit provides the determined recommended products to the customer. Can be handled to do so.

The operating method of the eyeglass recommendation system according to an embodiment includes generating a face mesh of a customer using a face scan result, extracting feature data from the generated face mesh of the customer, and based on the extracted feature data. Determining the customer's face type, extracting at least one sales data corresponding to the determined customer's face type, referring to a database in which sales data for each face type are recorded, the customer from the extracted sales data It may include determining at least one recommended product by reflecting the preference information of, and processing to provide the determined at least one or more recommended products to the customer.

Generating the face mesh according to an embodiment may include using a result of scanning the face of the customer as the face scan result using a true depth camera supported through an iOS app.

The determining of the face type of the customer according to an embodiment may include determining the face type in consideration of a ratio, distance, and depth between the extracted feature data.

In the determining of the face type of the customer according to an embodiment, size analysis, shape analysis, shape ratio analysis, and nose analysis are processed in consideration of the ratio, distance, and depth between the extracted feature data, and the processed The results of the analyzes can be used to determine the face type.

Determining the face type of the customer according to an embodiment, through the size analysis, calculate at least one of the face width for the customer, the width between the pupils, the width between the corners of the eyes, the width between the corners of the eyes, and the width of the nose pillar. It may include the step of.

Determining the face type of the customer according to an embodiment, through the shape analysis, the nose tip width, cheekbones width, mandibular width, vertical length, forehead length, forehead-nose tip length, and nose tip-chin tip for the customer At least one of the lengths can be calculated.

Determining the face type of the customer according to an embodiment, through the shape ratio analysis, calculating at least one of the vertical-horizontal ratio, the mandible-face width ratio, and the face width-radius ratio for the customer. It may include steps.

Determining the face type of the customer according to an embodiment may include calculating at least one of a height of a nose, a height of a nose, a length of a nose, and an angle of the nose for the customer through the nose analysis. .

The determining of the at least one recommended product according to an embodiment may include analyzing a data cluster, user-based collaborative filtering, item-based collaborative filtering, naive bay using the determined face type and the extracted sales data. It may include performing machine learning by applying at least one of the cyan algorithms, and determining the recommended product by using the result of the machine learning.

According to an embodiment, through cluster analysis and Bayesian analysis of data, it is possible to recommend an optimal eyeglass design only by recognizing the face by connecting with sales data according to the size and shape of the face.

According to an embodiment, the face of the customer is analyzed through the Face Ruler function based on the result of the face scan in the iOS app, and based on this, the size and design of the glasses are recommended according to the size and shape of the face. Accuracy can be improved through data analysis.

1 is a block diagram illustrating a system for recommending glasses according to an exemplary embodiment.
2A is a diagram illustrating a face mesh generated through a face scan according to an embodiment.
2B is a diagram illustrating an embodiment of extracting feature data from a face mesh and using the extracted feature data to determine a customer's face type.
FIG. 2C is a diagram illustrating sales data including design of glasses sold for each face type of a customer and sales amount.
2D is a diagram illustrating an algorithm for recommending glasses using a customer's face type and sales data.
3 is a block diagram illustrating a method of operating a system for recommending glasses according to an exemplary embodiment.

Specific structural or functional descriptions of embodiments according to the concept of the present invention disclosed in the present specification are exemplified only for the purpose of describing embodiments according to the concept of the present invention, and embodiments according to the concept of the present invention They may be implemented in various forms and are not limited to the embodiments described herein.

Since the embodiments according to the concept of the present invention can apply various changes and have various forms, the embodiments will be illustrated in the drawings and described in detail in the present specification. However, this is not intended to limit the embodiments according to the concept of the present invention to specific disclosed forms, and includes changes, equivalents, or substitutes included in the spirit and scope of the present invention.

Terms such as first or second may be used to describe various elements, but the elements should not be limited by the terms. The terms are only for the purpose of distinguishing one component from other components, for example, without departing from the scope of the rights according to the concept of the present invention, the first component may be referred to as the second component, Similarly, the second component may also be referred to as a first component.

When a component is referred to as being "connected" or "connected" to another component, it is understood that it may be directly connected or connected to the other component, but other components may exist in the middle. It should be. On the other hand, when a component is referred to as being "directly connected" or "directly connected" to another component, it should be understood that there is no other component in the middle. Expressions describing the relationship between components, for example, "between" and "just between" or "directly adjacent to" should be interpreted as well.

The terms used in the present specification are only used to describe specific embodiments, and are not intended to limit the present invention. Singular expressions include plural expressions unless the context clearly indicates otherwise. In the present specification, terms such as "comprise" or "have" are intended to designate that the specified features, numbers, steps, actions, components, parts, or combinations thereof exist, but one or more other features or numbers, It is to be understood that the presence or addition of steps, actions, components, parts, or combinations thereof does not preclude the possibility of preliminary exclusion.

Unless otherwise defined, 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 the present invention belongs. Terms as defined in a commonly used dictionary should be construed as having a meaning consistent with the meaning of the related technology, and should not be interpreted as an ideal or excessively formal meaning unless explicitly defined in the present specification. Does not.

Hereinafter, embodiments will be described in detail with reference to the accompanying drawings. However, the scope of the patent application is not limited or limited by these embodiments. The same reference numerals shown in each drawing indicate the same members.

1 is a block diagram illustrating a system 100 for recommending glasses according to an exemplary embodiment.

According to an embodiment, through cluster analysis and Bayesian analysis of data, it is possible to recommend an optimal eyeglass design only by recognizing the face by connecting with sales data according to the size and shape of the face. In addition, the face of the customer is analyzed through the Face Ruler function based on the result of the face scan in the iOS app, and based on this, the size and design of the glasses are recommended according to the size and shape of the face, and through big data analysis in the future. Accuracy can be improved.

To this end, the glasses recommendation system 100 includes a face mesh generation unit 110, a feature data extraction unit 120, a face type determination unit 130, a sales data extraction unit 140, a recommended product determination unit 150, And a recommendation processing unit 160.

First, the face mesh generation unit 110 may generate a customer's face mesh using the face scan result. In particular, the face mesh generation unit 110 may use a result of scanning a face for a customer using a true depth camera supported through an iOS app as the face scan result.

In addition, the feature data extraction unit 120 may extract feature data from the generated customer's face mesh.

The customer's face mesh can be created using a true depth camera,

Feature data can be interpreted as information calculated using distance, angle, and depth between specific coordinates in the face mesh. For example, the feature data may be feature data that can be obtained from a face mesh, such as a distance between the pupil and the pupil.

That is, the feature data extractor 120 according to an embodiment may extract feature data using specific coordinates measured from the face mesh.

For example, feature data may be extracted using at least one of a distance, an angle, and a depth between specific coordinates measured from the face mesh. More specifically, the feature data extracting unit 120 includes the distance between the pupil and the pupil measured from the face mesh, the distance between the inner end of the eye, the distance between the outer end of the eye, the height of the nose, the height of the glabellar and the pupil, the width of the face, and the face. Feature data can be extracted using length, inner folds of each ear, temporal bone, sphenoid bone, or cheekbone arch.

The feature data extraction unit 120 may directly generate a face mesh. For example, the feature data extractor 120 may collect at least 30,000 or more infrared dots by irradiating at least 30,000 or more infrared rays onto the customer's face and then controlling the true depth camera in response to the irradiated infrared rays. Among the collected 30,000 infrared dots, about 1,200 significant coordinates can be used to extract feature data.

In addition, the feature data extraction unit 120 may generate a face mesh of a customer based on the coordinates of the collected infrared dots.

The face type determination unit 130 may determine the face type of the customer based on the extracted feature data. In particular, the face type determination unit 130 may determine the face type in consideration of the ratio, distance, and depth between the extracted feature data. For example, the face type determination unit 130 may process size analysis, shape analysis, shape ratio analysis, and nose analysis in consideration of the ratio, distance, and depth between the extracted feature data. In addition, the face type can be determined using the results of the processed analyzes.

In addition, the face type determination unit 130 may calculate at least one of a face width for a customer, a width between pupils, a width between corners, a width between corners of the eyes, and a width of the nose pillar through size analysis.

In addition, the face type determination unit 130 may calculate at least one of the nose tip width, the cheekbone width, the mandible width, the vertical length, the forehead length, the forehead-nose tip length, and the nose tip-chin length for the customer through shape analysis. I can.

In addition, the face type determination unit 130 may calculate at least one of a vertical-horizontal ratio, a mandibular-face width ratio, and a face width-total bone ratio for the customer through shape ratio analysis.

In particular, the face type determination unit 130 may calculate at least one of the height of the nose, the height of the nose, the length of the nose, and the angle of the nose for the customer through nose analysis.

The sales data extraction unit 140 may extract at least one sales data corresponding to the determined customer's face type by referring to a database in which sales data for each face type is recorded.

To this end, the glasses recommendation system may further include a database in which sales data for each face type is recorded, or may be linked with the database.

For example, at least one or more sales data corresponding to the customer's face type may include design information and size information for glasses.

On the other hand, the preference information may include information input from the customer when the customer registers as a member. In addition, preference information may be extracted from information newly inputted from the customer after membership registration, or may be inferred based on the customer's behavior pattern or profiling.

The recommended product determination unit 150 may determine at least one recommended product by reflecting customer preference information in the extracted sales data.

Specifically, the recommended product determination unit 150 uses the determined face type and the extracted sales data to analyze a data cluster, a user based collaborative filtering, and a Naive Bayesian algorithm. Baysian Algorithm) can be applied to perform machine learning. In addition, the results of machine learning can be used to determine recommended products.

The recommendation processing unit 160 according to an embodiment may process to provide the determined at least one or more recommended products to a customer.

For example, a function of controlling a display may be performed to determine a priority for sorting the determined at least one or more recommended products, or to display recommended products arranged according to the determined priority.

In addition, in the case of making recommendations based only on the face, in most cases, the consumer may not immediately adopt the glasses for which the first recommendation was made.

In this situation, additional recommendations are essential.

To this end, the recommended product determination unit 150 determines at least one recommended product by reflecting the customer's preference information in the extracted sales data, but when the customer does not select the recommended product, the recommended product may be determined again. .

In addition, the recommendation processing unit 160 may process to provide the determined recommended product to the customer.

For a specific example, when the customer does not select the recommended product, the recommended product determination unit 150 may determine recommended products to which priority is given based on the customer's face type. At this time, the recommendation processing unit 160 may process the determined recommended products to be provided to the customer.

On the other hand, according to the present invention, glasses in the order of 1-5 are recommended through user-based collaborative filtering at first, and then the first recommendation is made through item-based collaborative filtering from the next, but there is an additional recommendation request. You can recommend related glasses (items) in conjunction with.

To this end, the recommended product determination unit 150 determines at least one recommended product by reflecting the customer's preference information in the extracted sales data, but when the customer does not select the recommended product, user-based collaborative filtering The number of products selected through may be determined as recommended products. Also, from the next recommendation, the recommended product determination unit 150 may determine recommended products as recommended products related to a result of item-based collaborative filtering.

Accordingly, the recommendation processing unit 160 may provide the determined recommended products to the customer.

Hereinafter, FIGS. 2A to 2D will be described in more detail an embodiment in which recommended products are provided after a face scan.

2A is a diagram illustrating a face mesh generated through a face scan according to an embodiment.

The system for recommending glasses according to an embodiment may scan a face by irradiating infrared rays onto the face of a customer (211).

In this process, the glasses recommendation system may collect at least 30,000 or more infrared dots by irradiating at least 30,000 or more infrared rays onto the customer's face and then controlling the true depth camera in response to the irradiated infrared rays. In addition, a face mesh of the customer may be generated based on the coordinates (x, y, z) of the collected infrared dots (212). Among more than 30,000 infrared dots, about 1,200 significant infrared dots can be used as feature data. Feature data can be used to calculate values such as ratio, distance, size, length, etc. that are calculated to classify the type of face.

2B is a diagram illustrating an embodiment 220 in which feature data is extracted from a face mesh and the extracted feature data is used to determine a customer's face type.

Size analysis (221), shape analysis (222), shape ratio analysis (223), and nose analysis (224) are processed in consideration of the ratio, distance, and depth between feature data, and the face type using the results of the processed analyzes. Can be determined.

The eyeglass recommendation system according to an embodiment may calculate at least one of a face width for a customer, a width between pupils, a width between corners, a width between corners of the eyes, and a width of the nose pillar through the size analysis 221.

The glasses recommendation system according to an embodiment calculates at least one of the nose tip width, the cheekbone width, the mandible width, the vertical length, the forehead length, the forehead-nose tip length, and the nose tip-chin tip length for the customer through the shape analysis 222 can do.

The glasses recommendation system may calculate at least one of a vertical-horizontal ratio, a mandibular-face width ratio, and a face width-optical bone ratio for the customer through the shape ratio analysis 223.

The glasses recommendation system may calculate at least one of the height of the nose, the height of the nose, the length of the nose, and the angle of the nose for the customer through the nose analysis 224.

Reference numeral 225 denotes a form in which infrared dots extracted as feature data are distinguished among all infrared dots in the face mesh viewed from the front of the customer's face.

Meanwhile, reference numeral 226 denotes a form in which infrared dots extracted as feature data are distinguished among all infrared dots in the face mesh viewed from the customer's face side. Analysis results 221, 222, 223, and 224 may be calculated based on the feature data displayed at 225 and 226.

FIG. 2C is a diagram for explaining sales data 230 including design of glasses sold for each face type of a customer and sales amount.

In the sales data 230 according to an embodiment, a face type may be recorded through the'customer's face type' field, and information about the eyeglass design purchased in association with the face type may be recorded through the'design' field. have. In addition, the sales volume for each design may be recorded in the'Sales Volume' field.

As shown in the sales data 230, in the case of customers whose face type is'B type', most of them have purchased'ROUND N'design glasses, and it can be seen that the sales volume for this is '89'.

On the other hand, in the case of customers whose face type is'A type', they purchased'ROUND N'and'ROUND C'design glasses, and it can be seen that the sales for them are '23' and '12', respectively.

In other words, it can be seen that 65% of customers who are'A type' chose the design of'ROUND N', and the design of'ROUND C'was selected for the rest.

2D is a diagram illustrating an algorithm for recommending glasses using a customer's face type and sales data.

The glasses recommendation system according to an embodiment may provide the recommendation information 240 using a face mesh corresponding to the reference numeral 220, feature data extracted therefrom, and sales data 230.

The recommendation information 240 may display a design, a size, and a recommendation ranking for recommended glasses.

For example, a recommendation ranking may be determined in consideration of sales volume.

In order to determine a recommended product, the glasses recommendation system according to an embodiment includes data cluster analysis, user-based collaborative filtering, item-based collaborative filtering, and naive Bayesian algorithm using the determined face type and extracted sales data. Machine learning can be performed by applying at least one.

That is, using the present invention, it is possible to recommend an optimal eyeglass design only by recognizing faces by connecting with sales data according to the size and shape of a face through cluster analysis and Bayesian analysis of data through the Face Ruler.

3 is a block diagram illustrating a method of operating a system for recommending glasses according to an exemplary embodiment.

In the operating method of the eyeglass recommendation system according to an embodiment, a face mesh of a customer may be generated by using the face scan result (step 310).

As an example, the operating method of the glasses recommendation system to generate a face mesh may use a result of scanning the face of the customer using a true depth camera supported through an iOS app as a face scan result.

In addition, the operating method of the glasses recommendation system may extract feature data from the generated face mesh of the customer (step 320), and determine the face type of the customer based on the extracted feature data (step 330).

For example, in order to determine the face type of the customer, the operating method of the eyeglass recommendation system may include determining the face type in consideration of a ratio, distance, and depth between the extracted feature data.

In particular, in order to determine the face type of the customer, the size analysis, shape analysis, shape ratio analysis, and nose analysis are processed in consideration of the ratio, distance, and depth between the extracted feature data, and the face type using the results of the processed analyzes. Can be determined.

In addition, through size analysis to determine the customer's face type, at least one of the customer's face width, the width between the pupils, the width between the corners of the eyes, the width between the corners of the eyes, and the width of the nose pillar may be calculated.

The operation method of the eyeglass recommendation system is to determine the customer's face type through shape analysis, among the nose tip width, cheekbone width, mandible width, vertical length, forehead length, forehead-nose length, and nose tip-chin length for the customer. At least one can be calculated.

In addition, the operating method of the glasses recommendation system calculates at least one of the vertical-horizontal ratio, the mandibular-face width ratio, and the face width-optical bone ratio for the customer through shape ratio analysis to determine the customer's face type. can do.

The operating method of the eyeglass recommendation system may calculate at least one of the height of the nose, the height of the nose, the length of the nose, and the angle of the nose for the customer through nose analysis to determine the customer's face type.

The operating method of the glasses recommendation system may extract at least one or more sales data corresponding to the determined customer's face type by referring to a database in which sales data for each face type is recorded (step 340).

In addition, the operating method of the glasses recommendation system may determine at least one recommended product by reflecting the customer's preference information in the extracted sales data (step 350).

As an example, the operating method of the glasses recommendation system includes data cluster analysis, user-based collaborative filtering, item-based collaborative filtering, using the determined face type and the extracted sales data to determine at least one recommended product, Machine learning may be performed by applying at least one of the Naive Bayesian algorithms. In addition, the results of machine learning can be used to determine recommended products.

In addition, the operating method of the glasses recommendation system may process to provide the determined at least one or more recommended products to the customer (step 360).

Consequently, using the present invention, it is possible to recommend an optimal eyeglass design only by recognizing the face by connecting with sales data according to the size and shape of the face through cluster analysis and Bayesian analysis of the data. In addition, the face of the customer is analyzed through the Face Ruler function based on the result of the face scan in the iOS app, and based on this, the size and design of the glasses are recommended according to the size and shape of the face, and big data analysis is performed in the future. Through this, accuracy can be improved.

The apparatus described above may be implemented as a hardware component, a software component, and/or a combination of a hardware component and a software component. For example, the devices and components described in the embodiments include, for example, a processor, a controller, an arithmetic logic unit (ALU), a digital signal processor, a microcomputer, a field programmable array (FPA), It may be implemented using one or more general purpose or special purpose computers, such as a programmable logic unit (PLU), a microprocessor, or any other device capable of executing and responding to instructions. The processing device may execute an operating system (OS) and one or more software applications executed on the operating system. Further, the processing device may access, store, manipulate, process, and generate data in response to the execution of software. For the convenience of understanding, although it is sometimes described that one processing device is used, one of ordinary skill in the art, the processing device is a plurality of processing elements and/or a plurality of types of processing elements. It can be seen that it may include. For example, the processing device may include a plurality of processors or one processor and one controller. In addition, other processing configurations are possible, such as a parallel processor.

The software may include a computer program, code, instructions, or a combination of one or more of these, configuring the processing unit to behave as desired or processed independently or collectively. You can command the device. Software and/or data may be interpreted by a processing device or, to provide instructions or data to a processing device, of any type of machine, component, physical device, virtual equipment, computer storage medium or device. , Or may be permanently or temporarily embodyed in a transmitted signal wave.

The software may be distributed over networked computer systems and stored or executed in a distributed manner. Software and data may be stored on one or more computer-readable recording media.

The method according to the embodiment may be implemented in the form of program instructions that can be executed through various computer means and recorded in a computer-readable medium. The computer-readable medium may include program instructions, data files, data structures, and the like alone or in combination. The program instructions recorded on the medium may be specially designed and configured for the embodiment, or may be known and usable to those skilled in computer software. Examples of computer-readable recording media include magnetic media such as hard disks, floppy disks, and magnetic tapes, optical media such as CD-ROMs and DVDs, and magnetic media such as floptical disks. -A hardware device specially configured to store and execute program instructions such as magneto-optical media, and ROM, RAM, flash memory, and the like. Examples of program instructions include not only machine language codes such as those produced by a compiler, but also high-level language codes that can be executed by a computer using an interpreter or the like. The hardware device described above may be configured to operate as one or more software modules to perform the operation of the embodiment, and vice versa.

Claims (23)

A face mesh generator that generates a face mesh of a customer using the face scan result;
A feature data extracting unit for extracting feature data from the generated customer's face mesh;
A face type determination unit that determines a face type of the customer based on the extracted feature data;
A sales data extracting unit for extracting at least one sales data corresponding to the determined customer's face type by referring to a database in which sales data for each face type is recorded;
A recommended product determination unit configured to determine at least one recommended product by reflecting the customer's preference information in the extracted sales data; And
Recommendation processing unit processing to provide the determined at least one or more recommended products to the customer
Glasses recommendation system comprising a. The method of claim 1,
The face mesh generation unit,
Glasses recommendation system using the result of scanning the face of the customer as the face scan result using a true depth camera supported through the iOS app. The method of claim 1,
The face type determination unit,
Glasses recommendation system for determining the face type in consideration of a ratio, distance, and depth between the extracted feature data. The method of claim 3,
The face type determination unit,
A glasses recommendation system that processes size analysis, shape analysis, shape ratio analysis, and nose analysis in consideration of the ratio, distance, and depth between the extracted feature data, and determines the face type using the results of the processed analyzes. The method of claim 4,
The face type determination unit,
Through the size analysis, the eyeglass recommendation system for calculating at least one of a face width, a width between pupils, a width between corners, a width between corners of the eyes, and a width of the nose pillar for the customer. The method of claim 4,
The face type determination unit,
Through the shape analysis, the glasses recommendation system for calculating at least one of a nose tip width, a cheekbone width, a mandible width, a vertical length, a forehead length, a forehead-nose tip length, and a nose tip-chin tip length for the customer. The method of claim 4,
The face type determination unit,
A glasses recommendation system for calculating at least one of a vertical-horizontal ratio, a mandibular-face width ratio, and a face width-optical bone ratio for the customer through the shape ratio analysis. The method of claim 4,
The face type determination unit,
Through the nose analysis, a glasses recommendation system for calculating at least one of a height of a nose, a height of a nose, a length of a nose, and an angle of the nose for the customer. The method of claim 1,
The at least one or more sales data corresponding to the customer's face type,
Glasses recommendation system that includes design information and size information for glasses. The method of claim 1,
The preference information,
Glasses recommendation system including information input from the customer when the customer registers as a member. The method of claim 1,
The recommended product determination unit,
By using at least one of data cluster analysis, users based collaborative filtering, and Naive Baysian Algorithm using the determined face type and the extracted sales data. A glasses recommendation system that performs machine learning and determines the recommended product by using the result of the machine learning. The method of claim 1,
The recommended product determination unit,
At least one recommended product is determined by reflecting the customer's preference information in the extracted sales data,
If the customer does not select the recommended product, re-determine the recommended product,
The recommendation processing unit is a system for recommending glasses that processes the re-determined recommended product to be provided to the customer. The method of claim 1,
The recommended product determination unit,
At least one recommended product is determined by reflecting the customer's preference information in the extracted sales data,
If the customer does not select the recommended product, determine the recommended products that are given priority based on the customer's face type,
The recommendation processing unit is a system for recommending glasses that processes the determined recommended products to be provided to the customer. The method of claim 1,
The recommended product determination unit,
At least one recommended product is determined by reflecting the customer's preference information in the extracted sales data,
When a customer does not select a recommended product, the number of products selected through user-based collaborative filtering is determined as recommended products, and from the next recommendation, recommended products related to the result of item-based collaborative filtering are determined. And
The recommendation processing unit is a system for recommending glasses that processes the determined recommended products to be provided to the customer. Generating a face mesh of a customer using the face scan result;
Extracting feature data from the generated customer's face mesh;
Determining a face type of the customer based on the extracted feature data;
Extracting at least one or more sales data corresponding to the determined customer's face type by referring to a database in which sales data for each face type is recorded;
Determining at least one recommended product by reflecting the customer's preference information in the extracted sales data; And
Processing to provide the determined at least one recommended product to the customer
Operating method of the glasses recommendation system comprising a. The method of claim 15,
The step of generating the face mesh,
Using a face scan result of the customer as the face scan result using a true depth camera supported through an iOS app
Operating method of the glasses recommendation system comprising a. The method of claim 15,
The step of determining the customer's face type,
Determining the face type in consideration of the ratio, distance, and depth between the extracted feature data
Operating method of the glasses recommendation system comprising a. The method of claim 17,
The step of determining the customer's face type,
Of the glasses recommendation system that processes size analysis, shape analysis, shape ratio analysis, and nose analysis in consideration of the ratio, distance, and depth between the extracted feature data, and determines the face type using the results of the processed analysis. How it works. The method of claim 18,
The step of determining the customer's face type,
Calculating at least one of a face width for the customer, a width between pupils, a width between corners, a width between corners, and a width of the nose pillar for the customer through the size analysis.
Operating method of the glasses recommendation system comprising a. The method of claim 18,
The step of determining the customer's face type,
Through the shape analysis, the operating method of the glasses recommendation system for calculating at least one of the nose tip width, the cheekbone width, the mandible width, the vertical length, the forehead length, the forehead-nose tip length, and the nose tip-chin tip length for the customer. The method of claim 18,
The step of determining the customer's face type,
Calculating at least one of a vertical-horizontal ratio, a mandibular-face width ratio, and a face width-optical bone ratio for the customer through the shape ratio analysis
Operating method of the glasses recommendation system comprising a. The method of claim 18,
The step of determining the customer's face type,
Calculating at least one of the height of the nose, the height of the tip of the nose, the length of the nose, and the angle of the nose for the customer through the nose analysis
Operating method of the glasses recommendation system comprising a. The method of claim 15,
The step of determining the at least one recommended product,
Performing machine learning by applying at least one of data cluster analysis, user-based collaborative filtering, item-based collaborative filtering, and naive Bayesian algorithm using the determined face type and the extracted sales data; And
Determining the recommended product using the result of the machine learning
Operating method of the glasses recommendation system comprising a.

Images