KR20230033041A - Method, server and program that providing user-customized interior information based on artificial intelligence - Google Patents

Abstract

The present invention relates to a user-customized interior information provision method based on artificial intelligence, in which interior product recommendation information which matches the type of space recognized for an input image, the attribute information and emotion information determined for a space element is derived, and a user-customized interior space image is created and provided by applying analysis result data for the input image to the user space.

Description

Method, server and program that providing user-customized interior information based on artificial intelligence {Method, server and program that providing user-customized interior information based on artificial intelligence}

The present invention relates to a method for providing user-customized interior information based on artificial intelligence.

Recently, cases in which individuals directly select and purchase interior products and decorate their homes are rapidly increasing.

In particular, the global interior market is expected to grow at an average annual rate of 4.9%, growing to a market size of about 792.6 billion dollars by 2025.

However, it is more difficult than expected for ordinary people to decorate the interior according to their own taste.

This is because there are more interior elements than expected in one space, and it is too difficult for the general public to consider the personal taste while composing them harmoniously.

Because of these problems, people who do self-interior use interior recommendation technology.

However, since the interior recommendation technologies disclosed so far recommend interior products using a statistical approach that many people purchase, a service that reflects individual tastes in detail is not provided at all.

In order to provide such a service, tagging and analysis of attributes of various interior elements in a photographed image of a specific space are required, but such a technology has not been disclosed at present.

Republic of Korea Patent Publication No. 10-2021-0037203, (2021.04.06)

The present invention for solving the above-mentioned problems recognizes the type of space included in the input image, determines the space element included in the recognized space, selects the attribute information of each space element, and then determines the information about the space based on this. We want to generate emotional information.

In addition, the present invention intends to derive and provide interior product recommendation information matching the type of space recognized with respect to an input image, attribute information and emotion information determined with respect to space elements, and provided to the user.

In addition, the present invention intends to generate and provide a user-customized interior space image by applying analysis result data for the input image to the user space when the user inputs an image of a space to be decorated.

In addition, according to the present invention, when a user inputs a purchase request for a specific interior product, it is stored as the user's behavior pattern data, and when it leads to an actual purchase, it is stored as the user's purchase pattern data, so that the user using the customized interior information providing service We want to collect big data on interior taste of people.

The problems to be solved by the present invention are not limited to the problems mentioned above, and other problems not mentioned will be clearly understood by those skilled in the art from the description below.

An artificial intelligence-based method for providing user-customized interior information according to an embodiment of the present invention for solving the above problems is an artificial intelligence-based method for providing user-customized interior information performed by a computer. Recognizing a space and determining the type of the recognized space; recognizing at least one space element included in the recognized space and determining the type of each recognized space element; determining attribute information of each of the determined spatial elements; generating emotion information about the recognized space based on the type of the recognized space and attributes selected for each of the determined space elements; and generating analysis result data including at least one piece of interior product recommendation information matching the type of the recognized space, the determined attribute information, and the generated emotion information.

In addition, the analysis result data may further include a construction method and a construction tool for constructing the interior product recommendation information.

In addition, receiving user space information including a user's spatial image from the user's terminal; generating the user space analysis data based on the user space information; and generating a user-customized interior space image by applying the analysis result data to the user space based on the user space analysis data.

In addition, the generating of the user space analysis data may include analyzing the user space information, determining at least one spatial element included in the user space, and calculating coordinates and dimensions of the determined at least one spatial element. In the generating of the space image, the user-customized interior space image may be generated by applying the analysis result data to match the calculated coordinates and dimensions.

In addition, the computer analyzes the user space information, determines at least one space element included in the user space, and applies the analysis result data to the space element to be applied to the interior input to the user's terminal. A customized interior space image may be generated, and non-target space elements may maintain the state of the user space information or may be excluded from the user customized interior space image.

In addition, the attribute information is selected from among individual emotional characteristics calculated according to individual external characteristics that mean characteristics of at least one of interior materials, colors, external shapes, and external patterns, and individual external characteristics determined for each space element. and at least one, wherein the computer determines individual external appearance characteristics of each of the recognized spatial elements, and determines individual emotional characteristics of each of the recognized spatial elements based on emotional characteristic scores set for each of the determined individual external appearance characteristics. , and the emotional information may be generated by integrating individual emotional characteristics calculated for spatial elements existing in the recognized space.

In addition, the attribute information is selected from among individual emotional characteristics calculated according to individual external characteristics that mean characteristics of at least one of interior materials, colors, external shapes, and external patterns, and individual external characteristics determined for each space element. and, when a change request signal for a specific space element in the user-customized interior space image is received from the user's terminal, the computer matches the change request attribute information included in the change request signal. The image of the user-customized interior space may be modified by selecting and reflecting it to the specific space element.

Also, the change request signal may include at least one of a change request for attribute information of the specific spatial element and a request for changing a price range of an interior product of the specific spatial element.

In addition, when the user's terminal receives a purchase request signal for a specific interior product from the user-customized interior space image, the computer provides a purchase page for the specific interior product to the user's terminal, and the specific interior product Characterized in that the product attribute information is stored as the user's behavior pattern data, and when the actual purchase of the specific interior product proceeds, the attribute information of the specific interior product is stored as the user's purchase pattern data.

In addition, the computer requests the user's terminal to select an analysis target spatial element to generate analysis result data from among the determined spatial elements, and generates analysis result data for the analysis target spatial element selected from the user's terminal. characterized by

In addition, the input image is characterized in that at least one of the sample images stored in the memory of the computer is selected according to the user's interior taste, or is an image uploaded from the user's terminal.

In addition, the computer includes an artificial intelligence model learned through an interior image including at least one space element, and the interior image is tagged with a space type for at least one space included in the interior image, , The type of at least one spatial element included in the interior image is tagged, attribute information for each of the at least one spatial element is tagged, and at least one emotion information matching the interior image is tagged. .

Further, the computer may further include requesting designation of a target space when there are a plurality of spaces included in the input image, following the step of recognizing the type of space.

In addition, an artificial intelligence-based user-customized interior information providing device according to an embodiment of the present invention for solving the above problems includes a communication unit; a memory in which an artificial intelligence model learned through an interior image including at least one spatial element is stored; and recognizing a space included in the input image by using the artificial intelligence model, recognizing a type of the recognized space, determining at least one spatial element included in the recognized space, and determining each of the determined spaces. Select attribute information of an element, generate emotion information for the recognized space based on the type of the recognized space and the attribute selected for each of the determined spatial elements, and determine the type of the recognized space, the determined and a processor generating analysis result data including attribute information and at least one interior product recommendation information matching the generated emotion information.

In addition to this, another method for implementing the present invention, another system, and a computer readable recording medium recording a computer program for executing the method may be further provided.

According to the present invention as described above, after recognizing the type of space included in the input image, determining the space elements included in the recognized space, and selecting the attribute information of each space element, emotional information about the space is generated based on this. has the effect of

In addition, the present invention exerts an effect of deriving and providing interior product recommendation information matching the type of space recognized with respect to an input image, attribute information and emotion information determined with respect to space elements, and providing the information to the user.

In addition, according to the present invention, when a user inputs an image of a space to be decorated, the analysis result data for the input image is applied to the user space to generate and provide a user-customized interior space image, so that the user can create a customized interior product for his or her own space. applied can be visually confirmed.

In addition, according to the present invention, when a user inputs a purchase request for a specific interior product, it is stored as the user's behavior pattern data, and when it leads to an actual purchase, it is stored as the user's purchase pattern data, so that the user using the customized interior information providing service It is possible to collect big data on people's interior taste.

The effects of the present invention are not limited to the effects mentioned above, and other effects not mentioned will be clearly understood by those skilled in the art from the description below.

1 is a block diagram of an artificial intelligence-based user-customized interior information providing device according to an embodiment of the present invention.
2 is a diagram illustrating learning of an artificial intelligence model using learning data and various models stored in a memory.
3 and 4 are flowcharts of a method for providing user-customized interior information based on artificial intelligence according to an embodiment of the present invention.
5 is a diagram illustrating generating analysis result data from an input image.
6 is a diagram illustrating the generation of a user-customized interior space image using user space information and analysis result data.
7 is a diagram illustrating types of spaces according to an embodiment of the present invention.
8 is a diagram illustrating types of space elements according to an embodiment of the present invention.
9 is a diagram illustrating that a spatial element determination model determines the type of each spatial element in an input image.
10 is a diagram illustrating types of emotion information that can be used in an embodiment of the present invention.
11 is a diagram illustrating construction of big data for user behavior pattern data and purchase pattern data for a user-customized interior space image.

Advantages and features of the present invention, and methods of achieving them, will become clear with reference to the detailed description of the following embodiments taken in conjunction with the accompanying drawings. However, the present invention is not limited to the embodiments disclosed below, but may be implemented in various different forms, only these embodiments are intended to complete the disclosure of the present invention, and are common in the art to which the present invention belongs. It is provided to fully inform the person skilled in the art of the scope of the invention, and the invention is only defined by the scope of the claims.

Terminology used herein is for describing the embodiments and is not intended to limit the present invention. In this specification, singular forms also include plural forms unless specifically stated otherwise in a phrase. As used herein, "comprises" and/or "comprising" does not exclude the presence or addition of one or more other elements other than the recited elements. Like reference numerals throughout the specification refer to like elements, and “and/or” includes each and every combination of one or more of the recited elements. Although "first", "second", etc. are used to describe various components, these components are not limited by these terms, of course. These terms are only used to distinguish one component from another. Accordingly, it goes without saying that the first element mentioned below may also be the second element within the technical spirit of the present invention.

Unless otherwise defined, all terms (including technical and scientific terms) used in this specification may be used with meanings commonly understood by those skilled in the art to which the present invention belongs. In addition, terms defined in commonly used dictionaries are not interpreted ideally or excessively unless explicitly specifically defined.

Hereinafter, embodiments of the present invention will be described in detail with reference to the accompanying drawings.

1 is a block diagram of an artificial intelligence-based user-customized interior information providing apparatus 100 according to an embodiment of the present invention.

2 is a diagram illustrating learning of an artificial intelligence model using training data and various models stored in the memory 130 .

3 and 4 are flowcharts of a method for providing user-customized interior information based on artificial intelligence according to an embodiment of the present invention.

5 to 9 are various exemplary views for explaining an artificial intelligence-based user-customized interior information providing apparatus 100 and method according to an embodiment of the present invention.

Referring to FIG. 1 , an artificial intelligence-based user-customized interior information providing apparatus 100 according to an embodiment of the present invention includes a processor 110, a communication unit 120, and a memory 130.

The artificial intelligence-based user-customized interior information providing device 100 according to an embodiment of the present invention is implemented by means such as information processing means, computers, and computing devices, and communicates with the user terminal 50 to provide the user with artificial intelligence-based interior information. It may be implemented in the form of a server to provide a user-customized interior information providing service.

When implemented as an artificial intelligence-based user-customized interior information providing server, the server may provide services through the web or may provide services to users using a service application developed by itself.

In this case, the user may use the service by accessing the web using the terminal 50 or may use the service by installing a service application in the terminal 50 .

However, in some embodiments, the server may include fewer or more components than those shown in FIG. 1 .

In some embodiments, the artificial intelligence-based user-customized interior information providing device 100 may further include an input/output unit 140 that receives various control signals and outputs results.

The communication unit 120 may communicate with the user terminal 50 connected to the server.

Specifically, the user-customized interior information providing server and device 100 may receive various control signals, images, etc. from the user terminal 50 through the communication unit 120, and may transmit analysis result data, space images, etc. to the user terminal ( 50) can be transmitted.

The memory 130 stores an artificial intelligence model learned through an interior image including at least one spatial element.

In addition, the memory 130 stores a plurality of space types, space elements, attribute information, and emotion information encoded and stored.

In one embodiment, the processor 110 may input several/plural space types, space elements, attribute information, and emotion information to the coding module 150, encode each of them, and store the generated code together with the respective information. there is.

Referring to FIG. 2 , data for learning includes a plurality of interior images.

At this time, in the interior image, a space type of at least one space included in the interior image is tagged, and a type of at least one space element included in the interior image is tagged.

In addition, the interior image is tagged with attribute information for each of at least one space element, and at least one emotion information matched with each interior image is tagged.

Scores are set according to a plurality of different emotion types for each attribute information of each spatial element, and by learning this, the artificial intelligence model calculates a score according to the emotion type for the spatial elements existing in the input image, Sentiment information on the input image may be generated by synthesizing the information.

An artificial intelligence model is learned by inputting such a plurality of interior images as learning data, and the artificial intelligence model may include various models as shown in FIG. 2 .

In detail, the artificial intelligence model may include a spatial recognition model, a space type determination model, a spatial element determination model, an attribute information determination model, an emotion information recognition model, and an analysis result generation model.

In addition, the memory 130 stores various commands, algorithms, and the like for a method of providing user-customized interior information based on artificial intelligence.

The processor 110 is in charge of controlling all components in the artificial intelligence-based user-customized interior information providing device 100, and can perform each process by executing commands and algorithms stored in the memory 130.

The apparatus 100 for providing user-customized interior information based on artificial intelligence according to an embodiment of the present invention may further include a coding module 150 .

The coding module 150 may code various data and information generated by the operation of the processor 110 and store them in the memory 130 .

Referring to FIGS. 3 and 4 below, the operation of the processor 110 will be described in detail.

5 is a diagram illustrating generating analysis result data from an input image, and FIG. 6 is a diagram illustrating generating a user-customized interior space image using user space information and analysis result data.

First, the device 100 receives an input image from the user terminal 50 through the communication unit 120 . (S110)

In an embodiment of the present invention, the input image may be at least one selected from among sample images stored in the memory 130 of the device 100 as a user's interior preference, or an image uploaded from the user's terminal 50 .

In detail, when an AI-based user-customized interior information providing service is requested from a user, the processor 110 outputs a plurality of sample images from among a plurality of sample images stored in the memory 130 to the user terminal 50, A request may be made to select a sample image corresponding to the user's interior taste as an input image.

Alternatively, when a user requests a service for providing user-customized interior information based on artificial intelligence, the processor 110 may request the user terminal 50 to transmit a specific image captured or possessed by the user as an input image.

In this way, since various cases can be applied to receiving or receiving an input image from the user terminal 50, the practitioner of the present invention can easily select it.

The processor 110 recognizes a space included in the input image and determines the type of the recognized space. (S120)

7 is a diagram illustrating types of spaces according to an embodiment of the present invention.

More specifically, the processor 110 recognizes a space included in an input image using a space recognition model built through learning, and determines the type of the recognized space.

As an embodiment, the processor 110 may determine the type of the recognized space by recognizing a code of the recognized space and comparing the recognized code with a code list in the memory 130 .

A detailed analysis can be performed by including one space in the input image, but is not limited thereto, so two or more spaces may coexist in the input image.

As an embodiment, the processor 110 may request the user terminal 50 to designate a space to be analyzed when there are a plurality of spaces included in the input image in S120.

As an embodiment, if there are a plurality of spaces included in the input image in S120, the processor 110 may perform analysis on each of the plurality of spaces.

In one embodiment, if there are a plurality of spaces included in the input image in S120, the processor 110 analyzes the space when the proportion of the specific space in the input image is greater than or equal to a preset reference proportion, and in other parts can be omitted.

For example, if the portion of the kitchen is 80% and the portion of the living room is 20% in the input image, the processor 110 determines that the living room was captured while filming the kitchen, and excludes the living room portion for the kitchen. analysis can proceed.

As described above, since various embodiments can be applied when a plurality of spaces exist in the input image, the embodiment suitable for the situation is applied so that the user can easily select the space so that the user feels convenient.

The processor 110 recognizes space elements included in the space recognized in S120 and determines the type of each recognized space element. (S130)

In detail, the processor 110 recognizes all spatial elements included in the recognized space in the input image by using the spatial element determination model built through learning, and determines the type of each recognized spatial element.

As an embodiment, the processor 110 may determine the type of each spatial element by recognizing the code of each recognized spatial element and matching it with a code list in memory.

8 is a diagram illustrating types of space elements according to an embodiment of the present invention.

The spatial element determination model recognizes all spatial elements included in the recognized space in the input image and determines the type of each spatial element. The type of such spatial element is illustrated in FIG. 8 and is coded and stored. has been

The spatial element determination model analyzes the image of each spatial element recognized in the input image to determine the spatial element, and basically determines the type of the recognized spatial element based on the type of the corresponding space.

For example, when the type of the corresponding space is a kitchen, the space element judgment model mainly determines that the space element corresponds to a kitchen.

In detail, the spatial element determination model analyzes the image of each recognized spatial element in the input image to determine the spatial element, and basically recognizes the code of the recognized spatial element based on the type of the corresponding space, and the processor ( 110) may determine the type of the recognized spatial element by matching the recognized code with a code in the memory.

Therefore, since it is possible to eliminate errors in recognizing spatial elements that cannot come to the kitchen, the judgment accuracy of the spatial element judgment model is dramatically increased.

In addition, in the embodiment of the present invention, the learning data emphasizes the boundary and boundary of each spatial element included in a plurality of interior images, and the spatial element determination model learns the boundary between different spatial elements included in the input image. can be accurately identified.

In some embodiments, the interior information providing device learns a boundary determination model to clearly distinguish and separate the boundaries between different spaces and between different spatial elements in the input image, and stores the information in the memory 130. can be saved

The boundary determination model may extract a code for a pixel of a predetermined region from an input image, and may determine the region as a boundary when a change in a pixel value is greater than or equal to a predetermined level.

However, it is not limited thereto, and the boundary determination model learns features such as boundary characteristics according to the type of space and boundary characteristics according to the type of spatial elements that can be arranged in the corresponding space according to the type of space, and converts the input image to can be analyzed in detail.

The spatial element determination model has an effect of accurately determining the boundary of spatial elements in the user space image included in the user space information by applying the same boundary surface determination function between spatial elements when analyzing user space information.

9 is a diagram illustrating that a spatial element determination model determines the type of each spatial element in an input image.

The spatial element determination model recognizes spatial elements existing in the input image as shown in FIG. 9 and determines the type of each spatial element, and the tagged numbers in FIG. 9 are the same as the spatial element table in FIG. 8 .

In detail, the spatial element determination model determines the type of each spatial element by recognizing a code of a spatial element existing in an input image and matching the recognized code to a code list in the memory 130 .

As an embodiment, the processor 110 may request the user terminal 50 to select a spatial element to be analyzed from among spatial elements determined in S130 .

For example, the user may want to obtain analysis result data by performing analysis on all spatial elements included in the input image, but he likes a specific spatial element included in the input image and proceeds with the analysis to obtain the analysis result. You may want to acquire data.

For this case, if the processor 110 proceeds to step S130 and determines the spatial elements existing in the input image, it requests an input to the user terminal 50 regarding which spatial elements among the determined spatial elements to be analyzed. can

Through the configuration described above, the user may analyze the input image as a whole to generate analysis result data, and use this to check the user-customized interior space image.

In addition, through the above-described configuration, the user selects only the spatial elements that suit his or her interior taste as analysis targets in the input image, generates analysis result data, and uses this to confirm the user-customized interior space image. Analysis result data may be acquired by overall analysis, or analysis result data may be obtained by selecting a spatial element that the user wants to design among spatial elements included in the input image.

The processor 110 determines the attribute information of each spatial element recognized in S120 using the attribute information determination model. (S140)

The processor 110 generates emotion information about the recognized space using the emotion information recognition model. (S150)

As an embodiment, the attribute information determination model determines the attribute information of each spatial element by recognizing the code for the attribute of the spatial element existing in the input image and matching the recognized code to the code list in the memory 130. can

10 is a diagram illustrating types of emotion information that can be used in an embodiment of the present invention.

At this time, the attribute information includes at least one of individual external characteristics, which means characteristics of interior materials, colors, external patterns, etc., and individual emotional characteristics calculated according to the individual external characteristics determined for each space element.

Therefore, the individual appearance characteristics refer to specific information expressing what kind of interior materials were used for space elements, and what colors, external shapes, and external patterns were used.

In addition, the individual emotional characteristics refer to the emotional characteristics of the space element expected according to the individual external appearance characteristics.

Spatial elements used in interiors may have specific emotional characteristics according to material, color, external shape, and externally exposed pattern, and emotional characteristics as illustrated in FIG. 10 may be applied.

As an embodiment, when spatial elements are recognized in the input image and the type of each spatial element is determined, the processor 110 determines the individual appearance characteristics of each spatial element using an attribute information determination model, and determines the determined individual appearance. Based on the characteristics, individual emotional characteristics for the corresponding spatial element can be calculated.

To this end, scores for a plurality of different emotional characteristics are coded and stored for each individual appearance characteristic in the memory 130 of the device 100 .

In addition, the attribute information determination model may calculate individual emotional characteristics for a corresponding spatial element by integrating emotional characteristic scores of individual external appearance characteristics determined for a specific spatial element.

Based on this, let's give a more specific example.

It is assumed that a room in the input image is determined as a type of space, and there is a closet among spatial elements in the room.

The processor 110 recognizes codes for the material, color, external shape, and external pattern of the closet using the attribute information determination model, and determines individual external characteristics based on this.

Next, the processor 110 loads emotional characteristic scores for each material, color, external shape, and external pattern of the closet in the memory 130 .

Then, the processor 110 calculates the individual emotional characteristics of the closet by integrating the loaded emotional characteristic scores.

For example, when the emotional characteristic scores for each individual emotional characteristic are summed up, if Minimal is 82 points, Classic is 40 points, and Casual is 25 points, the processor 110 determines the wardrobe It is judged to have the individual emotional characteristics of minimalism.

Finally, the processor 110 synthesizes the individual emotional characteristics calculated for each spatial element existing in the recognized space and generates emotional information about the space recognized in the input image.

The processor 110 generates analysis result data including at least one interior product recommendation information matched with the recognized type of space, attribute information determined for each space element, and generated emotion information. (S160)

In more detail, the processor 110 uses an analysis result generation model to analyze individual appearance characteristics determined for each space element and analysis result data including at least one interior product recommendation information matched with the generated emotion information. generate

As an embodiment, the analysis result data may consist of only at least one piece of interior product recommendation information.

As an embodiment, the analysis result data may further include construction methods and construction tools for constructing interior product recommendation information.

As such, since information on construction methods and construction tools is further provided, the user can purchase interior products based on the analysis result data and proceed with self-interior.

As an embodiment, the analysis result data may include a user-customized interior space image generated by applying interior product recommendation information, which will be described in detail below.

The processor 110 receives user space information including a user space image from the user terminal 50 . (S170)

The processor 110 generates user space analysis data based on user space information. (S180)

The processor 110 generates a user-customized interior space image by applying the analysis result data to the user space based on the user space analysis data. (S190)

The artificial intelligence-based user-customized interior information providing device 100 may provide a user-customized interior space image by applying the analysis result data generated through S110 to S160 to the user's space.

The processor 110 requests the user terminal 50 to photograph the user space where the user intends to decorate the interior, and the user photographs his/her own space through the terminal 50 and uploads the image to the device 100 .

In this case, the user space information may further include the size of the user space, the type of the user space, the types of spatial elements included in the user space, and the like, in addition to the user space image.

As such, when the user directly inputs various information into the user space information, the artificial intelligence model can generate a more accurate analysis result for the user space.

When analyzing the user space (S180), the processor 110 analyzes the user space information, determines at least one spatial element included in the user space, and calculates coordinates and dimensions of the determined at least one spatial element. .

In detail, the processor 110 analyzes the user space information, recognizes a code of at least one spatial element included in the user space, and matches the recognized code with a code list in the memory 130 to determine each spatial element's code. When generating the space image (S190), the processor 110 generates a user-customized interior space image by applying the analysis result data of S160 to match the coordinates and dimensions calculated in S180. .

As an embodiment, the processor 110 may generate a user-customized interior space image by selecting an interior product having the highest matching degree with the input image from at least one piece of interior product recommendation information in the analysis result data.

In one embodiment, the processor 110 excludes an interior product that is determined to be inapplicable to the user space due to the coordinates and dimensions calculated for the user space in S180 from the interior recommendation information and selects another interior product to provide a user-customized interior. You can create spatial images.

Through this configuration, since the processor 110 generates a user-customized interior space image excluding interior products that are determined not to be applicable to the user space, an error in which interior products that are not applicable to the user are recommended can be prevented in advance. There is an effect that can be done.

In one embodiment, the processor 110 analyzes the user space information and determines at least one spatial element included in the user space.

Specifically, the processor 110 recognizes a space included in the user space information using a space recognition model, determines the type of the recognized space using a space type determination model, and uses a space element determination model to determine the user space. It is possible to determine the spatial elements included in .

In addition, the processor 110 requests the user terminal 50 to select a spatial element to which interior is to be applied from among the spatial elements recognized for the user space to which spatial element to apply the analysis result data to generate a user-customized interior space image. can

Accordingly, the processor 110 generates a user-customized interior space image by applying the analysis result data to the interior application target space element input through the user terminal 50 .

In this case, the processor 110 may maintain the user space information state of the non-target space element or may exclude it from the user-customized interior space image.

For example, when a specific non-target spatial element is requested from the user terminal 50 to maintain the state of user space information, the processor 110 causes the designated non-target spatial element to maintain the state of user space information, A user-customized interior space image may be created.

In addition, when a specific non-target spatial element is requested to be excluded from the user-customized interior space image from the user terminal 50, the processor 110 may perform image processing to exclude the corresponding spatial element from the user-customized interior space image. .

Accordingly, the user can create a user-customized interior space image by reflecting the analysis result data on other spatial elements while maintaining the spatial elements included in the current user space to maintain the current state.

Alternatively, the user may generate a user-customized interior space image by excluding spatial elements included in the current user space and reflecting analysis result data on other spatial elements while excluding spatial elements to be excluded.

Through this configuration, the user can apply a customized interior product to all spatial elements in the current user space, apply it only to specific spatial elements, or exclude specific spatial elements from the user-customized interior space image. Therefore, it is possible to obtain various customized information.

As an embodiment, when a change request signal for a specific spatial element within a user-customized interior space image is received from the user terminal 50, the processor 110 selects an interior product that matches the change request attribute information included in the change request signal. By selecting and reflecting on the corresponding space element, the user-customized interior space image can be modified.

For example, although the processor 110 generates a user-customized interior space image and provides it to the user terminal 50, the user may have a spatial element desired to change after viewing the user-customized interior space image.

In this case, the user may request modification of the user-customized interior space image by inputting a change request signal including attribute information on a specific spatial element to be changed through the terminal 50 .

At this time, as mentioned above, the change request for the attribute information that the user can request for change is applicable to the change request for the interior material, color, external shape and external pattern, and the change request for the individual emotional characteristics of the space element.

As an embodiment, the processor 110 may code the user's change request and modify the user-customized interior space image based on the coded change request.

As an embodiment, the processor 110 may provide price information of interior products included in the user-customized interior space image to the user terminal 50 .

In this case, in addition to the unit price of the interior product, the area of the user's space may also affect the price.

Accordingly, the processor 110 may calculate price information by considering the unit price of the interior product included in the image of the user-customized interior space, the user's space, and the area of the space element.

Further, the change request signal may include a change request for attribute information of a specific spatial element as well as a request for changing a price range of an interior product of a specific spatial element.

Accordingly, when a request to change the price range of a specific space element is received from the user terminal 50, the processor 110 may modify the user-customized interior space image by deriving other interior products from the analysis result data based on the requested price range. there is.

For example, although a user-customized interior space image is generated to suit the user's interior taste, at least one interior product included in the user-customized interior space image may be too expensive than the user's desired price range.

In this case, the user may request a change in the price range of the interior product, and the processor 110 may select an appropriate interior product based on the request for change and analysis result data to modify the image of the user-customized interior space.

Through this configuration, the artificial intelligence-based user-customized interior information providing apparatus 100 according to an embodiment of the present invention has an effect of modifying not only individual appearance characteristics and individual emotional characteristics, but also the price range of interior products. do.

In addition, the method and apparatus 100 for providing customized interior information according to an embodiment of the present invention can be applied to each space type because the user selects the interior for each space, and the space elements within each space (eg: floors, ceilings, walls, doors, etc.) are recognized separately, so there is an effect of being able to analyze each space element and apply it to the user space.

In these configurations, the processor 110 codes each through the coding module 150 and stores them in the memory 130, recognizes the input image and codes each, thereby enabling accurate recognition and user becomes customizable.

For example, if you like other space elements in your space but don't like the floor, you can change the floor to another attribute.

According to the above-described embodiment, the user can search for and apply actual interior products for each spatial element while designing and interiorizing his/her own house.

In addition, the method and apparatus 100 for providing user-customized interior information according to an embodiment of the present invention analyzes user space information to determine at least one space element included in the user space, as well as to determine the area of each space element, Since even the dimensions are accurately calculated, the analysis result data can be accurately applied and provided to the user.

11 is a diagram illustrating construction of big data for user behavior pattern data and purchase pattern data for a user-customized interior space image.

As an embodiment, the processor 110 provides a purchase page for the corresponding interior product to the user terminal 50 when a purchase request signal for a specific interior product is received from the user terminal 50 in the interior space image customized for the user. .

Further, the processor 110 may build big data on behavior patterns of users who use the artificial intelligence-based user-customized interior information providing service by storing attribute information of the corresponding interior product as user behavior pattern data.

In addition, when actual purchase of a specific interior product is made, the processor 110 stores attribute information of the corresponding interior product as the user's purchase pattern data, thereby determining the purchase pattern of users using the artificial intelligence-based user-customized interior information providing service. It is possible to build big data for

In addition, the processor 110 stores a change request for attribute information of a specific spatial element received from the user terminal 50 to determine what users using the service mainly request for change, and which spatial element to request change You can collect big data about what it is.

In addition, the processor 110 stores a price range change request for a specific interior recommended product received from the user terminal 50, and by using the service, the user can change the preferred price range or non-preferred price range for each interior product of each space element. Big data can be collected on

In addition, the processor 110 stores information on spatial elements to be applied to interior and spatial elements to which interior is not applied, which are received from the user terminal 50, so that the user can generate a user-customized interior space image. It is possible to collect big data on spatial elements that you do not want to reflect.

By collecting and analyzing such big data, the processor 110 has an effect of being able to determine the preference order and reason for preference/non-preference for recommending interior products by using the user-customized interior information providing service.

Referring again to the above-described learning of the artificial intelligence model, the device 100 includes an artificial intelligence model learned through an interior image including at least one spatial element.

At this time, the interior image used as learning data is tagged with a space type for at least one space included in the interior image.

In addition, the interior image used as learning data is tagged with at least one type of spatial element included in the interior image.

In addition, the interior image used as learning data is tagged with attribute information for each of at least one spatial element.

In addition, the interior image used as learning data is tagged with at least one emotional information matching the interior image.

More specifically, each space element included in the interior image used as learning data is tagged with attribute information, and the attribute information includes individual appearance characteristics and each It includes at least one of the individual emotional characteristics calculated according to the individual appearance characteristics determined with respect to the spatial element.

In addition, in the memory 130 of the device 100, scores are set according to a plurality of different emotion types for each attribute information of a spatial element, and since the artificial intelligence model learns this, it exists in the input image when the input image is received. Scores are calculated according to the type of emotion for spatial elements that are played, and emotion information for the input image can be generated by synthesizing the scores.

In an embodiment, the apparatus 100 separately learns a space recognition model, a space type determination model, a space element determination model, an attribute information determination model, an emotion information recognition model, and an analysis result generation model to generate independent models. can be built

The method according to an embodiment of the present invention described above may be implemented as a program (or application) to be executed in combination with a server, which is hardware, and stored in a medium.

The aforementioned program is C, C++, JAVA, machine language, etc. It may include a code coded in a computer language of. These codes may include functional codes related to functions defining necessary functions for executing the methods, and include control codes related to execution procedures necessary for the processor of the computer to execute the functions according to a predetermined procedure. can do. In addition, these codes may further include memory reference related code for determining where (address address) of the computer's internal or external memory the additional information or media required for the computer's processor to execute the functions should be referenced. there is. In addition, when the processor of the computer needs to communicate with any other remote computer or server in order to execute the functions, the code uses the communication module of the computer to determine how to communicate with any other remote computer or server. It may further include communication-related codes for whether to communicate, what kind of information or media to transmit/receive during communication, and the like.

The storage medium is not a medium that stores data for a short moment, such as a register, cache, or memory, but a medium that stores data semi-permanently and is readable by a device. Specifically, examples of the storage medium include ROM, RAM, CD-ROM, magnetic tape, floppy disk, optical data storage device, etc., but are not limited thereto. That is, the program may be stored in various recording media on various servers accessible by the computer or various recording media on the user's computer. In addition, the medium may be distributed to computer systems connected by a network, and computer readable codes may be stored in a distributed manner.

Steps of a method or algorithm described in connection with an embodiment of the present invention may be implemented directly in hardware, implemented in a software module executed by hardware, or implemented by a combination thereof. A software module may include random access memory (RAM), read only memory (ROM), erasable programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), flash memory, hard disk, removable disk, CD-ROM, or It may reside in any form of computer readable recording medium well known in the art to which the present invention pertains.

Although the embodiments of the present invention have been described with reference to the accompanying drawings, those skilled in the art to which the present invention pertains can be implemented in other specific forms without changing the technical spirit or essential features of the present invention. you will be able to understand Therefore, it should be understood that the embodiments described above are illustrative in all respects and not restrictive.

50: user terminal
100: AI-based user-customized interior information providing device
110: processor
120: communication department
130: memory
140: input/output unit
150: coding module

Claims (15)

As an artificial intelligence-based user-customized interior information providing method performed by a computer,
Recognizing a space included in an input image and determining a type of the recognized space;
recognizing at least one space element included in the recognized space and determining the type of each recognized space element;
determining attribute information of each of the determined spatial elements;
generating emotion information about the recognized space based on the type of the recognized space and attributes selected for each of the determined space elements; and
Generating analysis result data including at least one interior product recommendation information matching the type of the recognized space, the determined attribute information, and the generated emotion information,
A method for providing user-customized interior information based on artificial intelligence.
According to claim 1,
The analysis result data,
Characterized in that it further comprises a construction method and a construction tool for constructing the interior product recommendation information,
A method for providing user-customized interior information based on artificial intelligence.
According to claim 1,
receiving user space information including a user space image from a user terminal;
generating the user space analysis data based on the user space information; and
Based on the user space analysis data, generating a user-customized interior space image by applying the analysis result data to the user space,
A method for providing user-customized interior information based on artificial intelligence.
According to claim 3,
The step of generating the user space analysis data,
Analyzing the user space information, determining at least one spatial element included in the user space, and calculating coordinates and dimensions of the determined at least one spatial element;
Generating the spatial image,
Characterized in that the user-customized interior space image is generated by applying the analysis result data to match the calculated coordinates and dimensions,
A method for providing user-customized interior information based on artificial intelligence.
According to claim 3,
the computer,
Analyzing the user space information to determine at least one spatial element included in the user space;
Creating a user-customized interior space image by applying the analysis result data to the interior application target space element input to the user's terminal;
Characterized in that the non-target space element maintains the state of the user space information or is excluded from the user-customized interior space image.
A method for providing user-customized interior information based on artificial intelligence.
According to claim 3,
The attribute information is
Including at least one of individual emotional characteristics calculated according to individual external characteristics determined for each space element and individual external characteristics that mean characteristics of at least one of interior materials, colors, external shapes and external patterns,
the computer,
Determining individual external appearance characteristics of each of the recognized spatial elements, calculating individual emotional characteristics of each of the recognized spatial elements based on emotional characteristic scores set for each of the determined individual external external characteristics,
Characterized in that the emotional information is generated by synthesizing individual emotional characteristics calculated for spatial elements existing in the recognized space,
A method for providing user-customized interior information based on artificial intelligence.
According to claim 1,
The attribute information is
Including at least one of individual emotional characteristics calculated according to individual external characteristics determined for each space element and individual external characteristics that mean characteristics of at least one of interior materials, colors, external shapes and external patterns,
the computer,
When a change request signal for a specific spatial element in the user-customized interior space image is received from the user's terminal,
Characterized in that the user-customized interior space image is modified by selecting an interior product matching the change request attribute information included in the change request signal and reflecting it to the specific space element.
A method for providing user-customized interior information based on artificial intelligence.
According to claim 7,
The change request signal,
Including at least one of a change request for attribute information of the specific spatial element or a price change request for an interior product of the specific spatial element,
A method for providing user-customized interior information based on artificial intelligence.
According to claim 3,
the computer,
When a purchase request signal for a specific interior product is received from the user's terminal from the user-customized interior space image, a purchase page for the specific interior product is provided to the user's terminal;
Store the attribute information of the specific interior product as the user's behavior pattern data,
Characterized in that, when the actual purchase of the specific interior product proceeds, attribute information of the specific interior product is stored as purchase pattern data of the user.
A method for providing user-customized interior information based on artificial intelligence.
According to claim 1,
the computer,
Requesting the user's terminal to select a spatial element to be analyzed to generate analysis result data from among the determined spatial elements;
Characterized in that the analysis result data for the spatial element to be analyzed selected from the user's terminal is generated.
A method for providing user-customized interior information based on artificial intelligence.
According to claim 1,
The input image,
At least one of the sample images stored in the memory of the computer is selected as the user's interior taste, or is an image uploaded from the user's terminal.
A method for providing user-customized interior information based on artificial intelligence.
According to claim 1,
the computer,
It includes an artificial intelligence model learned through an interior image containing at least one spatial element,
The interior image,
A space type for at least one space included in the interior image is tagged,
The type of at least one spatial element included in the interior image is tagged,
Attribute information for each of the at least one spatial element is tagged,
At least one emotion information matching the interior image is tagged,
A method for providing user-customized interior information based on artificial intelligence.
According to claim 1,
the computer,
Following the spatial type recognition step,
Further comprising the step of requesting designation of a target space when there are a plurality of spaces included in the input image.
A method for providing user-customized interior information based on artificial intelligence.
A computer-readable recording medium in which a program for executing the method of any one of claims 1 to 13 is stored in combination with a computer, which is hardware.
communications department;
a memory in which an artificial intelligence model learned through an interior image including at least one spatial element is stored; and
Using the artificial intelligence model,
Recognizing a space included in an input image, recognizing a type of the recognized space,
determining at least one spatial element included in the recognized space;
Selecting attribute information of each of the determined spatial elements;
Generating emotion information for the recognized space based on the type of the recognized space and attributes selected for each of the determined space elements;
A processor generating analysis result data including at least one interior product recommendation information matching the type of the recognized space, the determined attribute information, and the generated emotion information,
AI-based user-customized interior information provision device.

Images