Generation system and generation program

Abstract

To provide a generation system and generation program that allows users to confirm the state in which styling has been applied to an object. [Solution] A server device 2 that generates a styling display image showing a user who is the target of styling, with clothing, which is the object of styling, applied to it, comprising: an acquisition means for acquiring body-related information indicating the body of the target of styling; a first generation means for generating a target of styling display image showing the appearance of the target of styling based on the body-related information acquired by the acquisition means; and a second generation means for generating a styling display image based on the target of styling display image generated by the first generation means and styling object-related information indicating the clothing, which is the object of styling, wherein the first generation means generates a target of styling display image based on skeletal point information, which is reference point information identified by the identification means, and a reference target of styling display image.

Description

Technical Field

[0001] The present invention relates to a generation system and a generation program.

Background Art

[0002] Conventionally, a technique for performing various processes on an image of a user trying on clothes has been known (for example, Patent Document 1).

Prior Art Documents

Patent Documents

[0003]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0004] By the way, for example, when purchasing a styling object such as clothes via the Internet, etc., there has been a demand for a technique to allow a user to confirm the state in which the styling object is applied.

[0005] The present invention has been made in view of the above, and an object thereof is to provide a generation system and a generation program that enable a user to confirm the state in which a styling object is applied.

Means for Solving the Problems

[0006] To solve the above-mentioned problems and achieve the objective, the generation system described in claim 1 is a generation system for generating a styling display image that displays a person to whom a styling object has been applied, comprising: acquisition means for acquiring body-related information indicating the body of the person to whom the styling object has been applied; first generation means for generating a person display image that displays the appearance of the person to whom the styling object has been applied based on the body-related information acquired by the acquisition means; and second generation means for generating the styling display image based on the person display image generated by the first generation means and styling object-related information indicating the styling object.

[0007] The generation system according to claim 2 is the generation system according to claim 1, further comprising: identification means, which, based on the body-related information acquired by the acquisition means, identifies reference point information indicating the positions of a plurality of reference points that form the physical appearance of the person subject to the criteria, with reference to a reference person display image that displays the physical appearance of the person subject to the criteria; and the first generation means generates the reference person display image based on the reference point information identified by the identification means and the reference person display image.

[0008] The generation system according to claim 3 is the generation system according to claim 2, wherein the identification means performs a first identification process to identify the size of a predetermined part of the body of the target person based on the body-related information acquired by the acquisition means, and a second identification process to identify the reference point information based on the body-related information acquired by the acquisition means and the size of the predetermined part identified in the first identification process, and the second generation means generates the styling display image based on the size of the predetermined part of the body of the target person identified in the first identification process, the target person display image generated by the first generation means, and the styling object-related information.

[0009] The generation system according to claim 4 is the generation system according to claim 2, wherein the identification means further identifies the degree of obesity of the target person based on the body-related information acquired by the acquisition means, and the first generation means generates a first target person display image that displays the appearance of the target person, which is generated based on the entire body of the target person, based on the reference point information identified by the identification means, the degree of obesity information identified by the identification means, and the standard target person display image, and the first generation means side first processing that generates a first target person display image that displays the appearance of the target person, which is generated based on the entire body of the standard target person, and face information that shows the face of the target person. The first generation means performs a second process on the first generation means side, which, based on the face information input to the generation system, edits only the face of the target person shown in the first target person display image generated by the first generation means side, thereby generating a second target person display image that displays the appearance of the target person including the face corresponding to the face information. The second generation means generates the styling display image based on the second target person display image generated by the first generation means and the styling object related information.

[0010] The generation system according to claim 5 is the generation system according to claim 1, wherein the acquisition means further acquires styling object-related information indicating the styling object selected by the user of the generation system, the styling object-related information includes at least a styling object display image that displays the styling object and styling object size information indicating the size of the styling object, and the second generation means generates the styling display image based on the target user display image generated by the first generation means, the styling object display image acquired by the acquisition means, and the styling object size information acquired by the acquisition means.

[0011] The generation program according to claim 6 is a generation program for generating a styling display image that shows a person to whom a styling object has been applied, wherein the computer functions as an acquisition means for acquiring body-related information indicating the body of the person to whom the styling object has been applied; a first generation means for generating a person display image that shows the appearance of the person to whom the styling object has been applied based on the body-related information acquired by the acquisition means; and a second generation means for generating the styling display image based on the person display image generated by the first generation means and styling object-related information indicating the styling object. [Effects of the Invention]

[0012] According to the generation system described in claim 1 and the generation program described in claim 6, by generating a styling display image, it becomes possible to confirm, for example, the state in which the styling target object has been applied.

[0013] According to the generation system described in claim 2, by using reference point information that indicates the positions of multiple reference points that serve as criteria for shaping the physical appearance of the person to whom the styling is applied, it is possible to appropriately generate, for example, a display image of the person to whom the styling is applied, and to appropriately confirm the state in which the styling object has been applied.

[0014] According to the generation system described in claim 3, by generating a styling display image based on the size of a predetermined part of the body of the person to whom the styling is applied, for example, it is possible to appropriately generate a person display image taking into consideration the size of a predetermined part of the body of the person to whom the styling is applied, and to appropriately confirm the state in which the styling object is applied.

[0015] According to the generation system described in claim 4, by performing a second processing on the first generation means side to generate a second target person display image based on a first target person display image generated in the first processing on the first generation means side, it is possible to appropriately generate a second target person display image, for example, and to appropriately confirm the state in which the styling target object has been applied.

[0016] According to the generation system described in claim 5, by generating a styling display image based on the styling target object display image and the styling target object size information, for example, it becomes possible to appropriately confirm the state in which the styling target object is applied.

Brief Description of the Drawings

[0017] [Figure 1] It is a block diagram conceptually showing the information system according to this embodiment. [Figure 2] It is a diagram illustrating a base model image. [Figure 3] It is an explanatory diagram of each learned model. [Figure 4] It is a flowchart of image display processing. [Figure 5] It is a diagram illustrating a base model image. [Figure 6] It is a diagram illustrating a base model image. [Figure 7] It is an explanatory diagram of reference skeleton points and skeleton points. [Figure 8] It is an explanatory diagram of an operation example of the 5th skeleton point and the 13th skeleton point. [Figure 9] It is an explanatory diagram of an operation example of the 5th skeleton point and the 13th skeleton point. [Figure 10] It is a diagram illustrating a first user display image. [Figure 11] It is a diagram illustrating a second user display image. [Figure 12] It is a diagram illustrating a first user display image. [Figure 13] It is a diagram illustrating a second user display image. [Figure 14] It is a diagram illustrating a styling display image. [Figure 15] It is a display example of a styling display image.

Embodiments for Carrying Out the Invention

[0018] Hereinafter, embodiments of the generation system and generation program according to the present invention will be described in detail with reference to the drawings. However, the present invention is not limited by these embodiments. Here, the basic concepts will be explained first, followed by a description of specific embodiments.

[0019] (Basic concept) First, let's explain the basic concept. The generation system according to the present invention is a system that generates a styling display image that shows a person to whom a styling object has been applied. This concept includes, for example, a dedicated system for generating styling display images, or a system that is implemented by installing a generation program on a general-purpose system (for example, a general-purpose computer, a server computer, or multiple computers distributed on a network (i.e., a so-called cloud computer)) and implementing a function for generating styling display images.

[0020] "Styling target items" refers to items applied to the target person, and is a concept that includes, for example, clothing, shoes, bags, various accessories (necklaces, rings, etc.), and cosmetics.

[0021] "Target users" refers to those to whom the styling is applied, and is a concept that includes, for example, humans and animals (pets such as dogs and cats).

[0022] A "styling display image" is an image that shows the person to whom the styling is applied.

[0023] In the embodiments described below, for example, an example is given where the "generation system" is applied to a system for realizing a so-called e-commerce site (shopping site, etc.) that sells clothing, etc. via the internet, and the "styling target" is clothing and the "target user" is a person.

[0024] (composition) First, the configuration of the information system according to this embodiment will be described. Figure 1 is a block diagram that functionally illustrates the information system according to this embodiment.

[0025] The information system 100 is a system that includes a generation system, and is, for example, a system for realizing a so-called e-commerce site for selling clothing and the like via the internet, and one example includes a terminal device 1 and a server device 2.

[0026] (Configuration - Terminal device) The terminal device 1 in Figure 1 is a device used by the user (target person), and is, for example, a tablet terminal or a smartphone, and one example includes a communication unit 11, a touchpad 12, a display 13, a recording unit 14, and a control unit 15.

[0027] Note that other terminal devices, such as a personal computer, can also be used as terminal device 1. Furthermore, the number of terminal devices 1 is arbitrary, but in this embodiment, the configuration shown in Figure 1 will be used for explanation.

[0028] (Configuration - Terminal Equipment - Communication Unit) The communication unit 11 is a communication means for communicating with external devices (for example, server device 2). The specific type and configuration of this communication unit 11 are arbitrary, but it can be configured using, for example, a known communication circuit.

[0029] (Configuration - Terminal device - Touchpad) The touchpad 12 is an operating means that receives various operation inputs from the user when pressed by the user's finger or the like. The specific configuration of the touchpad 12 is arbitrary, but for example, a known one equipped with an operation position detection means such as a resistive type or a capacitive type can be used.

[0030] (Configuration - Terminal device - Display) The display 13 is a display means that displays various images based on the control of the control unit 15. The specific configuration of the display 13 is arbitrary, but for example, a known liquid crystal display or an organic EL display, or other flat panel display, can be used. The touch pad 12 and the display 13 may also be superimposed on each other to form an integrated touch panel.

[0031] (Configuration - Terminal device - Recording unit) The recording unit 14 is a recording means for recording programs and various data necessary for the operation of the terminal device 1, and can be configured using, for example, flash memory (the same applies to the recording units of other devices).

[0032] (Configuration - Terminal device - Control unit) The control unit 15 is a control means for controlling the terminal device 1, and specifically, it is a computer comprising a CPU, various programs interpreted and executed on the CPU (including basic control programs such as an OS, and application programs launched on the OS to realize specific functions), and internal memory such as RAM for storing programs and various data (the control units of other devices are similar). In particular, the program according to this embodiment is installed on the terminal device 1 via any recording medium or network, thereby substantially constituting each part of the control unit 15 (the control units of other devices are similar).

[0033] (Configuration - Server Device) The server device 2 is a generation system and includes, for example, a communication unit 21, a recording unit 22, and a control unit 23.

[0034] Furthermore, all or part of the various functions necessary to realize the e-commerce site may be implemented in this server device 2, or they may be implemented in another server device that can communicate with server device 2.

[0035] Furthermore, since any configuration, including publicly known configurations, can be applied to implement the various functions of the e-commerce site, a description of such configurations will generally be omitted.

[0036] (Configuration - Server device - Communication unit) The communication unit 21 in Figure 1 is a communication means for communicating with an external device (for example, terminal device 1). The specific type and configuration of this communication unit 21 are arbitrary, but it can be configured using, for example, a known communication circuit.

[0037] (Configuration - Server device - Recording unit) The recording unit 22 in Figure 1 is a recording means (storage means) for recording programs and various data necessary for the operation of the server device 2, and is configured, for example, using a hard disk or flash memory (not shown) as an external recording device. However, any other recording medium, including magnetic recording media such as magnetic disks, or optical recording media such as DVDs or Blu-ray discs, can be used instead of or in conjunction with the hard disk or flash memory.

[0038] The recording unit 22 stores, for example, a base model image, obesity degree identification criteria information, a first trained model for generation, a second trained model for generation, and a trained model for styling, as shown in Figure 1.

[0039] (Configuration - Server device - Recording unit - Base model image) Figure 2 is an example of a base model image.

[0040] A "base model image" is an image used to display the appearance of a person to whom the criteria apply, and as shown in Figure 2, for example, it is an image that displays base model B1.

[0041] The term "target subject" refers to a hypothetical target subject that serves as the basis for processing. For example, it refers to a hypothetical subject whose body size and build are predetermined, and in this embodiment, it is also called the "base model." Details of the base model image in Figure 2 will be described later.

[0042] (Configuration - Server device - Recording unit - Obesity degree identification criteria information) Figure 1, "Obesity Degree Identification Criteria Information," shows the criteria for identifying the degree of obesity of the target user.

[0043] "Obesity level" is a concept that indicates the physical condition of a user. For example, it is a concept defined according to BMI (Body Mass Index). One example is to use seven levels, ranging from underweight to overweight (i.e., from small BMI values ​​to large BMI values): underweight, underweight, normal weight, overweight, obese (grade 1), obese (grade 2), and obese (grade 3).

[0044] Note that the definition of obesity and the number of stages presented here are examples only, and other definitions and number of stages may be applied as long as they accurately represent the user's physical condition.

[0045] Furthermore, the criteria information for determining the degree of obesity shown in Figure 1 may include, for example, table information that correlates the BMI range with each state of body size (for example, BMI less than 16.0 is "underweight", BMI 16.0 or more but less than 18.0 is "underweight", etc.), or various programs or calculation formulas that output the degree of obesity when height and weight (or the BMI value itself) are input.

[0046] (Configuration - Server device - Recording unit - First trained model on the generation side) Figure 3 is an explanatory diagram of each trained model. In Figure 3, the input information ("Input" column) and output information ("Output" column) of each trained model are shown as examples.

[0047] The "Generating First Trained Model" in Figure 1 is a trained model generated by machine learning performed in advance. As illustrated by the correspondence with "Number" = "1" in Figure 3, it is a model configured to generate and output a first user display image when a base model image, skeletal point information, and obesity level information are input.

[0048] "Obesity level information" refers to the information indicating the degree of obesity mentioned above. Skeletal point information and the first user display image will be described later.

[0049] The specific machine learning method used for this first pre-trained generator model is arbitrary; for example, so-called supervised learning may be performed according to a predetermined intention (the same applies to the second pre-trained generator model and the styling model described later).

[0050] (Configuration - Server device - Recording unit - Second trained model on the generation side) The "Generating Second Trained Model" in Figure 1 is a trained model generated by machine learning performed in advance. As illustrated by the correspondence with "Number" = "2" in Figure 3, it is a model configured to generate and output a second user-displayed image when the first user-displayed image, first region identification information, and face information are input.

[0051] The first user-displayed image, first region identification information, face information, and second user-displayed image will be described later.

[0052] (Configuration - Server device - Recording unit - Styling side trained model) The "Styling-side trained model" in Figure 1 is a trained model generated by machine learning performed in advance. As illustrated by the correspondence with "Number" = "3" in Figure 3, it is a model configured to generate and output a styling display image when a second user display image, second region identification information, product-related information, and body size information are input.

[0053] As mentioned above, the "styling display image" is an image that displays the person to whom the styling object is applied, and in this embodiment, it is an image that displays the user wearing the clothes selected by the user. The second user display image, second area identification information, product-related information, and body size information will be described later.

[0054] (Configuration - Server device - Control unit) The control unit 23 in Figure 1 is a control means for controlling the server device 2.

[0055] Functionally, this control unit 23 functions, for example, as an acquisition means, a identification means, a first generation means, and a second generation means.

[0056] ===Acquisition method=== The "acquisition means" refers to means of acquiring body-related information that indicates the body of the person to whom the system is applied. The "acquisition means" further acquires, for example, styling object-related information that indicates the styling object selected by the user of the generation system.

[0057] "Physical information" refers to information that describes the physical characteristics of the person to whom the information is applied, such as height, weight, arm length, and shoulder width.

[0058] "Styling object-related information" refers to information that indicates a styling object selected by the user of the generation system, and for example, includes at least a styling object display image that shows the styling object, and styling object size information that indicates the size of the styling object.

[0059] ===Specific means=== "Identification means" refers to means that, based on the physical information acquired by the acquisition means, identify reference point information that indicates the positions of multiple reference points that form the physical appearance of the person subject to the criteria, using a reference image of the person subject to the criteria that displays the physical appearance of the person subject to the criteria as a reference. Note that "using the reference image of the person subject to the criteria as a reference" may be interpreted as meaning, for example, "based on the reference image of the person subject to the criteria."

[0060] The "identification means" includes, for example, a first identification process that identifies the size of a predetermined part of the body of the target person based on the body-related information acquired by the acquisition means, and a second identification process that identifies reference point information based on the body-related information acquired by the acquisition means and the size of the predetermined part identified in the first identification process.

[0061] "Identification means" for example, identify the degree of obesity of the target person based on the physical information acquired by the acquisition means.

[0062] ===First generation means=== The "first generation means" is a means for generating a person display image that shows the appearance of the person to be applied to, based on the body-related information acquired by the acquisition means.

[0063] The "first generation means" generates an image of the person to whom the criteria apply based, for example, on the reference point information identified by the identification means and the image of the person to whom the criteria apply.

[0064] The "first generation means" performs, for example, a first processing on the first generation means side which generates a first target person display image that displays the appearance of the target person, which is generated based on the entire body of the target person, based on reference point information identified by the specific means, obesity degree information identified by the specific means, and a target person display image of the standard target person, and a second processing on the first generation means side which generates a second target person display image that displays the appearance of the target person, which includes the face corresponding to the face information, by editing only the face of the target person shown in the first target person display image generated in the first processing on the first generation means side in the first target person display image generated in the first target person display image.

[0065] ===Second generation means=== The "second generation means" is a means for generating a styling display image based on the target person display image generated by the first generation means and styling object-related information indicating the styling object.

[0066] The "second generation means" generates a styling display image based, for example, the size of a predetermined part of the body of the target person identified in the first identification process, the target person display image generated by the first generation means, and information related to the styling object.

[0067] The "second generation means" generates a styling display image based, for example, on the second target person display image generated by the first generation means and the styling object related information.

[0068] The "second generation means" generates a styling display image based, for example, on the target person display image generated by the first generation means, the styling object display image acquired by the acquisition means, and the styling object size information acquired by the acquisition means.

[0069] The processing performed by each part of the control unit 23 will be described later.

[0070] (process) Next, we will describe, for example, the image display process as an example of the processing performed by the information system 100 configured in this way.

[0071] (Processing - Image display processing) Figure 4 is a flowchart of the image display process (each step will be referred to as "S" below). The image display process is performed by server device 2 and, in general terms, involves generating and displaying a styling display image.

[0072] The timing of executing this image display process is arbitrary, but for example, it will be activated when a user (hereinafter also simply referred to as "user"), who is the target of this process, performs a predetermined operation via their terminal device 1, and the explanation will begin from the point where the image display process is activated.

[0073] Here, we will illustrate and explain a case where, for example, application software for making purchases on an e-commerce site is installed on the user's terminal device 1, and the user inputs various information through the terminal device 1.

[0074] As a variation, the system may be configured so that the installation of the application software is unnecessary, and users can access the server device 2, etc., using web browsing software such as a browser to make purchases on the e-commerce site.

[0075] ===SA1=== In SA1 shown in Figure 4, the control unit 23 of the server device 2 acquires body-related information (information indicating height, weight, arm length, and shoulder width), facial information, and product-related information.

[0076] ==Physical Information== While the specific processing of body-related information is optional, for example, if a user inputs information indicating their height, weight, arm length, and shoulder width via their terminal device 1, this information is transmitted from terminal device 1 to server device 2, which then receives and acquires this transmitted information.

[0077] Here, for example, height and weight are configured to be entered as numerical information, arm length is configured to be entered by selecting one option from size-related choices such as "short," "standard," or "long," and shoulder width is configured to be entered by selecting one option from size-related choices such as "narrow," "standard," or "wide."

[0078] Then, if the user enters information indicating their height, weight, arm length, and shoulder width, such as "145cm", "55kg", "standard", and "wide", this information is sent from terminal device 1 to server device 2, and server device 2 receives and acquires this transmitted information.

[0079] ==Facial Information== "Facial information" refers to information that describes the user's facial area (the part from the neck up, including the face and head), and is a concept that includes, for example, the user's facial photograph and information that describes the individual parts of the user's facial area (also referred to as "parts information").

[0080] The content of the "parts information" is arbitrary. For example, images and various other information generated by any image software such as CAD other than photographs may be used. For instance, information indicating parts related to hair (parts that express length, color, bangs type, hair texture, etc.) or information indicating skin color may be used.

[0081] While the specific processing of facial information is optional, when a user performs a first operation (an operation to input a facial photograph) via their terminal device 1 and then takes a photograph of their face using a camera (not shown) on the terminal device 1, facial information indicating the user's facial photograph is transmitted from the terminal device 1 to the server device 2. Furthermore, when a user performs a second operation (an operation to input part information) via their terminal device 1 and then selects part information indicating each part corresponding to themselves, facial information indicating the selected part information is transmitted from the terminal device 1 to the server device 2. Meanwhile, the server device 2 receives this transmitted facial information and acquires the received facial information.

[0082] ==Product-related information== "Product-related information" refers to information related to the item being styled, and specifically, it is information indicating the item selected by the user (in this embodiment, clothing). This concept includes, for example, a product image that displays the product (in this embodiment, an image of clothing), and product size information that indicates the size of the product (in this embodiment, the actual size of the clothing, including the overall size and dimensions of each part (for a skirt, this would include overall sizes such as S, M, L, etc., as well as dimensions such as waist, hip, and skirt length)). It may also be interpreted that the product image corresponds to the "image displaying the item being styled," and the product size information corresponds to the "size information of the item being styled."

[0083] While the specific processing of product-related information is optional, for example, let's assume that product-related information (product image and product size information) for each product sold on the e-commerce site is recorded in the recording unit 22 of the server device 2. When a user selects a product sold on the e-commerce site via the terminal device 1 to consider purchasing it, information indicating the selected product is transmitted from the terminal device 1 to the server device 2. The server device 2 receives this transmitted information and, based on the received information, retrieves the product-related information (product image and product size information) corresponding to the product selected by the user from the product-related information of each product recorded in the recording unit 22.

[0084] Here, for example, if a user selects a specific skirt from among the products sold on the e-commerce site, information indicating the selected skirt is transmitted from terminal device 1 to server device 2. Meanwhile, server device 2 receives this transmitted information and, based on the received information, retrieves product-related information (product image and product size information) corresponding to the specific skirt selected by the user from the product-related information of each product recorded in recording unit 22.

[0085] ===SA2=== In SA2 in Figure 4, the control unit 23 of the server device 2 identifies skeletal point information corresponding to the user based on the body-related information acquired in SA1 and the base model image in the recording unit 22.

[0086] ==Base Model Image== Figures 5 and 6 are illustrative diagrams of the base model image, and Figure 7 is an explanatory diagram of the reference skeleton points and skeleton points. In Figure 5, each reference skeleton point is shown in the base model image of Figure 2, and in Figure 6, in addition to the information in Figure 5, the coordinate system set in the base model image is illustrated.

[0087] As mentioned above, the "base model image" in Figure 2 is an image representing the target user of the criteria, and is the image that displays base model B1. As mentioned above, "base model" B1 represents a hypothetical user who is a target user of the criteria, and whose body size and build are predetermined. For example, it may be set to be 165cm tall, with an arm length of 70cm, shoulder width of 40cm, face width of 14.5cm, waist width of 35cm, etc.

[0088] ==Base Model Image - Size== Furthermore, the size of the base model image is predetermined (XX pixels high x YY pixels wide). The display of base model B1 within the base model image is set to an appropriate size relative to the overall size of the base model image (for example, as shown in Figure 2, a size that leaves predetermined margins on all sides of base model B1, ensuring good visibility when displayed).

[0089] ==Base Model Image - Reference Skeleton Points== Furthermore, as shown in Figure 5, the base model image has 18 reference skeletal points defined, for example, from the first reference skeletal point 901 to the 18th reference skeletal point 918 (see the "Reference Skeletal Points" column in Figure 7).

[0090] Furthermore, each reference skeletal point is collectively referred to simply as a "reference skeletal point." The reference skeletal points exemplified here are for the sake of explanation; for example, many reference skeletal points may be defined in the hand part of base model B1, or reference skeletal points may be defined in other parts, or some reference skeletal points may be omitted.

[0091] "Reference skeletal points" are a set of reference points that serve as the basis for shaping the external form of the body of base model B1. This concept includes, for example, positions corresponding to predetermined body parts such as joints. In this embodiment, the first reference skeletal points 901 to the 18th reference skeletal points 918 are defined as reference skeletal points, corresponding to the positions of the "parts" in Figure 7. That is, for example, the first reference skeletal point 901 corresponds to the "nose" of base model B1, the second reference skeletal point 902 corresponds to the "right eye," and so on.

[0092] ==Base Model Image - Coordinate System== Furthermore, as shown in Figure 6, the base model image has a defined coordinate system consisting of, for example, the X-axis (horizontal axis) and the Y-axis (vertical axis), and each position in the base model image (i.e., the position of the reference skeletal point, etc.) can be identified by coordinates based on this coordinate system.

[0093] In detail, the coordinate system in Figure 6 is defined such that, for example, the X-axis has coordinates where the values ​​increase from left to right, and the Y-axis has coordinates where the values ​​increase from top to bottom. The values ​​of each coordinate may be set based on, for example, the pixels of the base model image. Other coordinate systems may also be used as variations.

[0094] ==Skeletal Point Information== "Skeleton point information" (not shown in Figures 5 and 6) is reference point information, specifically information indicating the position of skeletal points (reference points).

[0095] "Skeletal points" refer to multiple reference points that serve as the basis for shaping the physical appearance of the subject, and specifically refer to multiple reference points that serve as the basis for shaping the physical appearance of the user model representing the user (in detail, the first user model U1 and the second user model U2 described later), and are a concept that includes, for example, positions corresponding to predetermined parts of the body such as joints.

[0096] ==Processing== The specific processing of SA2 in Figure 4 is arbitrary, but for example, the following first process (first specific process) and second process (second specific process) are performed.

[0097] =First Process= The first process, in general terms, involves identifying the size of a predetermined part of the user's body based on the body-related information acquired by SA1. The predetermined part whose size is identified is the part necessary to identify the skeletal point information that indicates the position of the skeletal point corresponding to the reference skeletal point part in Figure 7 (there are 18 parts in Figure 7).

[0098] Here, we will specifically illustrate and explain the case where, for example, the size of shoulder width and waist width are specified as the size of certain parts of the user's body.

[0099] First, regarding shoulder width, for example, in the body-related information acquired by SA1, as mentioned above, there are three levels of information: "narrow," "standard," and "wide." Based on height and whether it is "narrow," "standard," or "wide," a calculation formula is predetermined to determine the specific size of the shoulder width (e.g., XX cm). Based on this calculation formula, for example, 38 cm is determined as the specific size.

[0100] Furthermore, regarding waist width, for example, since waist width is not included in the body-related information obtained with SA1, the waist width size is determined as XX cm using an arbitrary identification method (for example, a method in which table information or calculation formulas showing the relationship between height, shoulder width and waist width are predetermined, and this predetermined table information or calculation formula is used for identification).

[0101] In addition to shoulder width and waist width, the sizes of other body parts (for example, arm length, face width, and other body parts) necessary to identify the skeletal point information that indicates the position of the skeletal point corresponding to the reference skeletal point parts in Figure 7 (18 parts in Figure 7) are identified using any method (for example, a method that uses a predetermined calculation formula to identify the size of each body part from the body-related information acquired by SA1).

[0102] =Second Process= The second process, in general terms, is the process of identifying skeletal point information based on the body-related information acquired in SA1 and the size of the predetermined body part identified in the first process described above.

[0103] In this second process, for example, the skeletal point information is identified using the positions (coordinates) of the reference skeletal point information in Figures 6 and 7, so that the user model representing the user (specifically, the first user model U1 and the second user model U2 described later) is displayed at an appropriate size in an image the same size as the base model image.

[0104] For details, follow steps 1 and 2 below.

[0105] <<<Step 1>>> In the first step, the coordinates (i.e., positions) of the reference skeleton points defined in the base model image are identified. Here, for example, the coordinates of the first reference skeleton point 901 to the 18th reference skeleton point 918 are identified using the coordinate system set in the base model image in Figure 6 as a reference.

[0106] <<<Step 2>>> In the second step, based on the processing results from the first step, the coordinates (i.e., positions) of each skeletal point corresponding to each part of the reference skeletal point are identified as skeletal point information. The skeletal point corresponding to the "nose," which is part of the first reference skeletal point 901, will be referred to as the "first skeletal point," and similarly, the skeletal points corresponding to each of the parts of the second reference skeletal point 902 to the 18th reference skeletal point 918 will be referred to as the "second skeletal point" to the "18th skeletal point."

[0107] In general terms, the coordinates of the skeletal points corresponding to each reference skeletal point are determined based on the coordinates of each reference skeletal point. That is, for example, the coordinates of the first skeletal point (the skeletal point corresponding to the user's "nose") corresponding to the first reference skeletal point 901 are determined based on the coordinates of the first reference skeletal point 901.

[0108] <<Fixed>> While the details are optional, for items marked "Fixed" in the "X-coordinate" and "Y-coordinate" columns of Figure 7, the same coordinates as the reference skeletal point are used to identify the skeletal point's coordinates.

[0109] Here, for example, for the first skeletal point of the "part" = "nose" in Figure 7, both the X and Y coordinates associated with the "first reference skeletal point 901" (in the "reference skeletal point" column in Figure 7) corresponding to that first skeletal point are "fixed." Therefore, the same X and Y coordinates as those of the first reference skeletal point 901 identified in the first step are identified as the coordinates of the first skeletal point. In other words, the coordinates that indicate the same position as the first reference skeletal point 901 in Figure 6, etc., become the position of the first skeletal point corresponding to the user's "nose."

[0110] Furthermore, for example, in Figure 7, for the fifth skeletal point of the "part" = "right shoulder," only the Y coordinate is "fixed." Therefore, the same Y coordinate as the fifth reference skeletal point 905 identified in the first step is identified as the Y coordinate of this fifth skeletal point. The method for identifying the X coordinate of this fifth skeletal point will be described later.

[0111] Then, the coordinates of the skeletal points in other parts of Figure 7 that are "fixed" are identified in the same way.

[0112] <<Other than fixed>> Furthermore, for all locations in Figure 7 except those marked "Fixed" in the "X-coordinate" and "Y-coordinate" columns (blank or "corresponding to the right"), the coordinates of the skeletal points are determined by performing a predetermined calculation and reflecting the calculation result in the coordinates of the reference skeletal points.

[0113] The details of the predetermined calculations may be defined for each body part, but we will explain in detail the calculation examples for the 5th and 13th skeletal points of the "body parts" = "right shoulder" and "left shoulder" in Figure 7, and only an overview will be given for the other skeletal points.

[0114] <Non-fixed - X coordinates of the right and left shoulders> Figures 8 and 9 are explanatory diagrams illustrating calculation examples for the 5th and 13th skeletal points. Figure 8 corresponds to a partially enlarged view of Figure 6.

[0115] As stated in the "<Premise>" column of Figure 9, the number of pixels corresponding to the height of base model B1 in the base model image of Figure 6 (that is, the number of pixels in the Y-axis direction corresponding to the length L in Figure 6, also referred to as the "base model height pixel count") is 890 pixels (890 pixels), and the base model's body type is defined as follows: height 165cm, arm length 70cm, shoulder width 40cm, face width 14.5cm, waist width 35cm, etc. Furthermore, the user's body type is identified as: height 145cm, arm length 65cm, shoulder width 38cm, face width XXcm, waist width XXcm, etc., based on the body-related information acquired by SA1 and the processing results in "First Processing".

[0116] Then, the following calculations are performed using these values.

[0117] Specifically, the first conversion ratio is calculated by performing the calculation shown in the formula illustrated in the upper part of "<Conversion Ratio>" in Figure 9. The "first conversion ratio" is a concept that represents the number of pixels per centimeter of height for the base model B1. Here, for example, the calculation 890 (pixels) ÷ 165 cm = 5.4 is performed.

[0118] Furthermore, the second conversion ratio is calculated by performing the calculation shown in the formula illustrated in the lower part of "<Conversion Ratio>" in Figure 9. The "second conversion ratio" is a concept that represents the number of pixels per centimeter of the user's height. Here, for example, the calculation 890 (pixels) ÷ 145 cm = 6.14 is performed.

[0119] Next, the X-coordinate of the fifth skeletal point is calculated by performing the calculation shown in the formula illustrated in the upper part of "<Coordinate Calculation>" in Figure 9. Note that "Shoulder Width (Base)" and "Shoulder Width (User)" refer to the shoulder width of the base model B1 and the user's shoulder width, respectively. For example, as shown in Figure 8, if the coordinates of the fifth reference skeletal point 905 are (200, Ys5), the X-coordinate of the fifth skeletal point is calculated and identified as 191 by performing the calculation shown in Figure 9.

[0120] In this case, as shown in Figure 7, the Y coordinate of the fifth skeletal point is "fixed" and is the same as the Y coordinate of the fifth reference skeletal point. Therefore, as shown in Figure 8, the coordinates of the fifth skeletal point are determined by calculating (191, Ys5).

[0121] Furthermore, the X-coordinate of the 13th skeletal point is calculated by performing the calculation shown in the formula illustrated in the lower part of "<Coordinate Calculation>" in Figure 9.

[0122] Here, as indicated in the "X-coordinate" column for the 13th reference skeletal point 913 in Figure 7, which states "corresponds to the right side," the calculation is basically the same as for the coordinate of the 5th skeletal point, which is the right shoulder corresponding to the left shoulder. However, from the perspective of reflecting the left-right inversion with respect to the vertical axis centerline in the base model B1 in Figure 6, the operator (the "+" part in "Xs13 (X-coordinate of the 13th reference skeletal point 913) + (shoulder width (base)...)" in Figure 9) will be "+" instead of "-" (see the calculation formula for the 5th skeletal point in the upper part of "<Coordinate Calculation>" in Figure 9).

[0123] Similarly, the coordinates of the skeletal points in other parts of Figure 7 that correspond to the "right side" are identified in the same manner.

[0124] For example, as shown in Figure 8, if the coordinates of the 13th reference skeleton point 913 are (300, Ys13), the Y coordinate of the 13th skeleton point is calculated as 309 by performing the calculation shown in Figure 9.

[0125] In this case, as shown in Figure 7, the Y coordinate of the 13th skeletal point is "fixed" and is the same as the Y coordinate of the 13th reference skeletal point 913. Therefore, as shown in Figure 8, the coordinate of the 13th skeletal point is determined by calculating (309, Ys13).

[0126] <X coordinates of the right and left hips, excluding fixed positions> The X coordinates of the 8th and 16th skeletal points corresponding to the right and left hips are determined in the same manner as described above. In this case, the hip width may be used for calculation and determination instead of the shoulder width of the base model B1 and the user.

[0127] <Other than fixed - X coordinates of the right and left eyes> The X coordinates of the second and eleventh skeletal points corresponding to the right and left eyes are determined in the same manner as described above. In this case, the distance between the eyes may be used for calculation instead of the shoulder width of the base model B1 and the user. The distance between the eyes may be determined by performing a predetermined calculation based on the face width (for example, half the length of the face width).

[0128] <Other than fixed - X coordinates of the right and left ears> The X coordinates of the third and twelfth skeletal points corresponding to the right and left ears are determined in the same manner as described above. In this case, the face width may be used for calculation and determination instead of the shoulder width of the base model B1 and the user.

[0129] <X coordinates of the right and left elbows, excluding fixed positions> The X-coordinates of the 6th and 14th skeletal points corresponding to the right and left elbows are determined in the same manner as described above. In this case, the distance between the elbows may be used for calculation instead of the shoulder width of the base model B1 and the user. The distance between the elbows may be determined by performing a predetermined calculation based on the shoulder width (for example, 1.2 times the shoulder width).

[0130] <X coordinates of non-fixed components - right wrist, left wrist, right knee, left knee, right ankle, left ankle> The X-coordinates of the 7th and 15th skeletal points corresponding to the right and left wrists, the 9th and 17th skeletal points corresponding to the right and left knees, and the 10th and 18th skeletal points corresponding to the right and left ankles are determined in the same manner as the X-coordinates of the 6th and 14th skeletal points corresponding to the right and left elbows.

[0131] <Y coordinates of non-fixed points - right elbow, left elbow, right wrist, left wrist> The Y coordinates of the 6th and 14th skeletal points corresponding to the right and left elbows, and the 7th and 15th skeletal points corresponding to the right and left wrists, may be determined by performing a calculation on the Y axis similar to the calculation in "<Coordinate Calculation>" in Figure 9. In this case, the calculation may be performed using half the length of the arm or the entire length, instead of the shoulder width. Although not shown in Figure 7, these Y coordinates may also be "fixed" as with the others.

[0132] The methods for identifying each skeletal point shown here are illustrative examples, and the coordinates designated as "fixed" in Figure 7 are also illustrative examples. You may use other methods to identify the points, or you may arbitrarily change the reference points designated as "fixed."

[0133] ===SA3=== In SA3 of Figure 4, the control unit 23 of the server device 2 generates a first user display image based on the information acquired in SA1 and the skeletal point information (i.e., the coordinates of each skeletal point) identified in SA2.

[0134] ==First User Display Image== Figure 10 is an example of a first user display image. The "first user display image" is the first target user display image (target user display image), and specifically, it is an image that displays the appearance of the first user model U1 representing the user (i.e., the user's appearance). The first user display image can also be interpreted as an image that displays the user's appearance generated based on the entire body of the base model B1.

[0135] Furthermore, the size of the first user display image may be set arbitrarily, but in this embodiment, for example, it will be the same as the size of the base model image in Figure 2 (i.e., a predetermined size of XX pixels vertically x YY pixels horizontally). Also, for example, the size of the second user display image and the styling display image described later will be the same as the size of the base model image.

[0136] ==Processing== The specific processing of SA3 in Figure 4 is arbitrary, but for example, the following first and second processes are performed.

[0137] =First Process= The first process is, in general terms, the process of determining the user's degree of obesity. Specifically, it determines the user's height and weight based on the physical information acquired by SA1, calculates the BMI based on the determined height and weight, and determines the degree of obesity corresponding to the calculated BMI by referring to the degree of obesity determination criteria information in the recording unit 22 in Figure 1 (i.e., determining the degree of obesity information).

[0138] Here, for example, if the user's height and weight are identified as "145cm" and "55kg," the corresponding BMI is calculated, and then the user is identified as "overweight." Note that the numbers and processing results here are for illustrative purposes only (the same applies to other results and numbers).

[0139] =Second Process= The second process, in general terms, is the process of generating the first user display image using the processing results of the first process, etc.

[0140] Specifically, the control unit 23 acquires the base model image of the recording unit 22 in Figure 1 (Figure 6, etc.), the skeletal point information identified by SA2 (i.e., the coordinates of each skeletal point, including the coordinates of the 5th skeletal point "(191, Ys5)" and the 13th skeletal point "309, Ys13" in Figure 8), and the obesity degree information indicating "overweight" which was identified in the first processing. This acquired information is then input into the first trained model on the generation side (numbered "1" in Figure 3). In this case, the first user display image in Figure 10 is output from the first trained model on the generation side, meaning that the control unit 23 generates the first user display image.

[0141] As mentioned above, the first user model U1 represents a user, so the first user display image in Figure 10 can be said to be an image that displays the user's appearance.

[0142] Furthermore, as mentioned above, the first pre-trained model on the generating side has been trained using so-called supervised learning in accordance with the intent, and with the base model image, skeletal point information, and obesity level information as input, as shown in Figure 10, a first user display image is generated that displays the first user model U1 which appropriately reflects the user's body, based on the entire body of the base model B1 in Figure 6 (an image in which each skeletal point is reflected and the state of being slightly overweight).

[0143] Furthermore, these settings ensure that the first user model U1 is set to an appropriate size relative to the overall size of the first user display image (for example, as shown in Figure 10, a size that leaves predetermined margins on all sides of the first user model U1, ensuring good visibility when displayed) (the same applies to the second user display image and the styling display image described later).

[0144] Furthermore, this SA3 processing may be interpreted as corresponding to the "first processing on the first generation means side."

[0145] ===SA4=== In SA4 of Figure 4, the control unit 23 of the server device 2 generates a second user display image based on the information acquired in SA1 and the first user display image generated in SA3.

[0146] ==Second User Display Image== Figure 11 is an example of a second user display image. The "second user display image" is a second target user display image (target user display image), and specifically, it is an image that displays the appearance of the second user model U2 representing the user (i.e., the user's appearance). Unlike the first user display image, for example, it is an image that reflects the face corresponding to the user.

[0147] ==Processing== The specific processing of SA4 in Figure 4 is arbitrary, but for example, the following first and second processes are performed.

[0148] =First Process= Figure 12 is an example of the first user-displayed image. The first process is, in general terms, the process of identifying the first region M1 (Figure 12) in the first user-displayed image.

[0149] The "first region" M1 is the region corresponding to the face in the first user model U1 in the first user display image. Specifically, as shown in Figure 12, it is the region surrounding the part of the first user model U1 from the neck up (including the face and head).

[0150] This first region M1 is the area surrounding the unmasked portion of the first user-displayed image (the portion intended to be edited (i.e., modified) in the second process below). In other words, the portion of the first user-displayed image other than the first region M1 (the hatched portion in Figure 12, also referred to as the "masked portion") is the portion not intended to be edited (i.e., modified) in the second process below (i.e., the portion intended to be excluded from modification).

[0151] In detail, the process involves using an arbitrary method (such as using a known image processing program for recognizing different parts of a person, or a method that identifies parts based on their characteristics) to identify the face of the first user model U1 in the first user display image generated by SA3, and then identifying the region surrounding the face as the first region M1.

[0152] Here, for example, in the first user-displayed image of Figure 10 generated by SA3, the first region M1 in Figure 12 is identified.

[0153] =Second Process= The second process, in general terms, is the process of generating a second user display image using the processing results of the first process, etc.

[0154] Specifically, the control unit 23 acquires the first user display image generated by SA3, the first region identification information indicating the first region M1 identified in the first processing (for example, a group of coordinates indicating the first region M1), and the face information acquired by SA1 (i.e., the face information input to the server device 2), and inputs this acquired information to the generation-side second trained model (number = "2" in Figure 3). In this case, the generation-side second trained model outputs, for example, the second user display image shown in Figure 11, meaning that the control unit 23 generates the second user display image.

[0155] As mentioned above, the second user model U2 (Figure 11) represents a user, so the second user display image in Figure 11 can be said to be an image that displays the user's appearance.

[0156] As mentioned above, the second pre-trained model on the generating side has been trained using so-called supervised learning in accordance with the intent, and since the inputs are the first user-displayed image, the first region identification information, and the face information, the second user-displayed image is generated that reflects the intent regarding the first region M1 in Figure 12 (that is, the intent to edit only the part surrounded by the first region M1 and to display the face corresponding to the input face information), as shown in Figure 11.

[0157] In other words, by editing only the face portion of the user shown in the first user display image generated by SA3 (specifically, the first user model U1), a second target user display image is generated that displays the appearance of the user (specifically, the second user model U2) that includes the face portion corresponding to the face portion information.

[0158] Furthermore, this SA4 processing may be interpreted as corresponding to the "second processing on the first generation means side."

[0159] ===SA5=== In SA5 of Figure 4, the control unit 23 of the server device 2 generates a styling display image based on the information acquired in SA1 and the second user display image generated in SA4.

[0160] Specifically, although optional, one might perform the following first and second processes.

[0161] =First Process= Figure 13 is an example of a second user-displayed image. The first process is, in general terms, the process of identifying the second region M2 (Figure 13) in the second user-displayed image.

[0162] The "second area" M2 is the area in the second user-displayed image that corresponds to the part of the garment selected by the user (in the case of Figure 13, the skirt worn on the lower half of the garment) that is worn. Specifically, if the garment is a skirt, as shown in Figure 13, it is a relatively wide area that comfortably surrounds the part where all types of skirts, including miniskirts and long skirts, are worn (the same applies to other types of garments such as tops, jackets, and pants, depending on the type of garment).

[0163] This second region M2 is the area surrounding the unmasked portion of the second user-displayed image (the portion intended to be edited (i.e., modified) in the second process below). In other words, the portion of the second user-displayed image other than the second region M2 (the hatched portion in Figure 13, also referred to as the "masked portion") is the portion not intended to be edited (i.e., modified) in the second process below (i.e., the portion intended to be excluded from modification).

[0164] In detail regarding the processing, depending on the type of product, the part to be worn (for example, the lower body part for a skirt, the upper body part for a jacket, etc.) is predetermined. In the second user display image generated by SA4, the part to be worn corresponding to the product selected by the user in the second user model U2 is identified using an arbitrary method (such as a method using a known image processing program for recognizing each part of a person, or a method that identifies based on the characteristics of each part), and the area that comfortably surrounds the said part to be worn is identified as the second region M2.

[0165] Here, for example, if SA1 obtains product-related information corresponding to a specific skirt selected by the user, SA4 identifies the second region M2 in Figure 13 in the second user display image in Figure 11 generated by SA4.

[0166] =Second Process= Figure 14 is an example of a styling display image. The second process is, in general terms, a process that generates a styling display image using the processing results of the first process, etc.

[0167] Specifically, the control unit 23 acquires the second user display image generated by SA4, second region identification information indicating the second region M2 identified in the first processing (for example, a group of coordinates indicating the second region M2), product-related information acquired by SA1 (in detail, product image and product size information), and body size information.

[0168] "Body size information" refers to information indicating the size of the user's body. For example, it is information indicating the size of the body identified based on body-related information acquired in SA1. This concept includes, for example, height and the same information as the size of a specific part of the user's body identified in the first processing of SA2.

[0169] In particular, by using the same information as that used to identify skeletal point information in SA2 (such as information indicating the size of a specific part of the user's body identified in the first processing of SA2, or information indicating the user's height) as "body size information," it is possible to improve the processing accuracy here.

[0170] Here, we will illustrate and explain the case where, for example, information indicating the user's height (height identified from body-related information obtained in SA1), arm length (arm length identified in the first processing of SA2), shoulder width (shoulder width identified in the first processing of SA2), and waist width (waist width identified in the first processing of SA2) is used as "body size information."

[0171] Then, the previously acquired information (i.e., the second user display image, the second region identification information, the product-related information, and the body size information) is input to the pre-trained styling model (numbered "3" in Figure 3). In this case, the pre-trained styling model outputs, for example, the styling display image shown in Figure 14 (an image displaying the second user model U2 wearing the skirt selected by the user), meaning that the control unit 23 generates the styling display image.

[0172] As mentioned above, the second user model U2 represents the user, so the styling display image in Figure 14 can be said to be an image showing the user wearing the skirt selected by the user.

[0173] Furthermore, as mentioned above, the styling-side trained model has been trained using so-called supervised learning in accordance with the intent, and the inputs are the second user display image, second region identification information, product-related information (specifically, product image and product size information), and body size information. As shown in Figure 14, a styling display image is generated that reflects the intent regarding the second region M2 in Figure 13 (that is, the intent to edit only the part enclosed by the second region M2, and to display the product corresponding to the input product-related information being worn on the body according to the input body size information).

[0174] In particular, since product size information and body size information are input from product-related information, machine learning of the pre-trained styling model can be performed taking these into consideration, making it possible to reflect body size and product size in the styling display image.

[0175] In other words, for example, although not shown in this embodiment, it is possible to reflect the wearing condition when a relatively tall user wears a skirt that is relatively small in size (such as a skirt that is intended to reach around the ankles but ends up reaching around the knees, or a skirt that is intended to be worn loosely but ends up being worn tightly), so it is possible to display a wearing condition that is closer to the actual situation.

[0176] ===SA6=== In SA6 in Figure 4, the control unit 23 of the server device 2 transmits the styling display image generated in SA5 to the user's terminal device 1, thereby displaying it on the display 13 of the terminal device 1.

[0177] Figure 15 shows an example of a styling display image. Here, for example, by transmitting the styling display image in Figure 14 generated by SA5, the styling display image is displayed on the display 13 of the terminal device 1, as shown in Figure 15.

[0178] Then, based on the styling display image in Figure 15, the user can check how the selected skirt looks when worn on the e-commerce site and consider whether or not to purchase the skirt. This concludes the explanation of the image display process.

[0179] (Effects of the embodiment) Thus, according to this embodiment, by generating a styling display image, it becomes possible to, for example, confirm the state in which the styling target object has been applied.

[0180] Furthermore, by using skeletal point information that indicates the positions of multiple skeletal points that serve as the basis for shaping the physical appearance of the user to whom the system is applied, it becomes possible to appropriately generate, for example, a first user display image and a second user display image, and to allow users to properly confirm the state in which the clothing, which is the object of styling, is applied.

[0181] Furthermore, by generating styling display images based on the size of predetermined parts of the user's body, it becomes possible to appropriately generate application display images that take into account the size of predetermined parts of the user's body, allowing users to properly confirm how the clothing, which is the styling target, looks when applied.

[0182] Furthermore, by performing a process to generate a second user display image based on the first user display image, it becomes possible to appropriately generate, for example, a second target user display image, and to properly confirm the state in which the clothing, which is the styling target object, is applied.

[0183] Furthermore, by generating styling display images based on product images and product size information, it becomes possible to properly confirm, for example, how the clothing being styled looks when applied.

[0184] [Modifications of the embodiment] While embodiments of the present invention have been described above, the specific configurations and means of the present invention can be arbitrarily modified and improved within the scope of the technical idea of ​​the present invention as described in the claims. Such modifications will be described below.

[0185] (Regarding the problems to be solved and the effects of the invention) First, the problems that the invention aims to solve and the effects of the invention are not limited to those described above, and may vary depending on the implementation environment and details of the invention's configuration. It may solve only some of the problems described above, or achieve only some of the effects described above. Furthermore, the features of this application may be interpreted as being able to solve problems other than those explicitly stated.

[0186] (Regarding decentralization and integration) Furthermore, the aforementioned electrical components are functional concepts and do not necessarily need to be physically configured as shown in the illustrations. In other words, the specific forms of distribution and integration of each part are not limited to those shown in the illustrations, and all or part of them can be functionally or physically distributed or integrated in any unit according to various loads and usage conditions. Also, the term "device" in this application is not limited to a single device, but includes devices composed of multiple devices.

[0187] (Regarding numerical values ​​and time series) With regard to the components illustrated in the embodiments and drawings, the numerical or chronological relationships can be arbitrarily modified and improved within the scope of the technical concept of the present invention.

[0188] (Regarding product selection) Furthermore, while the above embodiment illustrates the case where only a skirt is selected, if multiple items such as a skirt and a top are selected, the styling display image will show the selected items being worn.

[0189] (Regarding physical information) Furthermore, while the above embodiment describes a case where information indicating height, weight, arm length, and shoulder width is used as body-related information, it is not limited to this. For example, the system may be configured to use only one, two, or three of these types of information as body-related information. Alternatively, it may be configured to use other types of information (for example, waist circumference, inseam length, etc.).

[0190] (Regarding product-related information) Furthermore, in the above embodiment, product feature information (information indicating the characteristics of the product, such as material features like glossiness, or information indicating the attributes of the target audience of the product, such as women in their 20s) may be used as part of the information that constitutes the product-related information.

[0191] (Regarding pre-trained models) Furthermore, each of the trained models in Figure 3 of the above embodiment may be arbitrarily changed. For example, the trained models numbered "1" and "2" may be combined to create a trained model that generates and outputs a second user display image when a base model image, skeletal point information, obesity level information, and facial information are input, and this model may be used to perform the processing corresponding to SA3 and SA4 in Figure 4.

[0192] Furthermore, in each of the trained models shown in Figure 3 of the above embodiment, some of the input information may be omitted. For example, in the trained model number "1", either the obesity level information or the base model image, or both, may be omitted. Also, in the trained model number "3", body size information may be omitted.

[0193] Furthermore, each process may be configured to use a predetermined processing algorithm (a program that does not use a trained model) instead of the trained model.

[0194] (Regarding recommended features) Furthermore, the information system 100 in Figure 1 may be configured to record previously generated styling display images in the recording unit 22 for each of the multiple users utilizing the system. In this case, the server device 2 may be configured to output and display recommended information about wearing clothing, which is a product, to a specific user based on these recorded styling display images.

[0195] For example, the recording unit 22 may record information indicating each user's height and weight, skeletal type (a type classified from skeletal point information), and personal color (the color each user prefers). The control unit 23 of the server device 2 may then identify a user corresponding to a specific user (a user to whom recommendation information is to be output) based on the information about each user recorded in the recording unit 22 (hereinafter referred to as the "corresponding user"), select a styling display image that has been previously generated for the identified corresponding user, and display the selected styling display image on the terminal device 1 of the aforementioned specific user.

[0196] In this case, the method for identifying the corresponding user is arbitrary, but for example, users with the same height and weight may be identified, users with the same skeletal type may be identified, or users with the same personal color may be identified. Alternatively, the system may be configured to identify users who have two or more of the following characteristics: height and weight, skeletal type, or personal color.

[0197] If multiple styling display images have been generated in the past, you may configure the system to display all of them, or you may configure it to select and display only some of the images.

[0198] (Regarding other features) Furthermore, the above embodiment may be configured to use other types of information. For example, the system may be configured to have the user input their age and to identify skeletal point information while taking that age into consideration.

[0199] Furthermore, the SA4 in Figure 4 may be configured to regenerate the second user display image. For example, after processing the SA4, the second user display image may be displayed on the user's terminal device 1, and then, if the user performs an update operation such as pressing the "Update" button via the terminal device 1, the SA4 may be processed again to regenerate the second user display image.

[0200] (Regarding combinations) Furthermore, the features of the above embodiment and the features of the modified example may be combined in any way.

[0201] (Note) The generation system described in Appendix 1 is a generation system for generating a styling display image that shows a person to whom a styling object has been applied, comprising: an acquisition means for acquiring body-related information indicating the body of the person to whom the styling object has been applied; a first generation means for generating a person display image that shows the appearance of the person to whom the styling object has been applied based on the body-related information acquired by the acquisition means; and a second generation means for generating the styling display image based on the person display image generated by the first generation means and styling object-related information indicating the styling object.

[0202] The generation system described in Appendix 2 further comprises, in the generation system described in Appendix 1, identification means that, based on the body-related information acquired by the acquisition means, identify reference point information indicating the positions of a plurality of reference points that form the physical appearance of the person subject to the criteria, using a reference image of the person subject to the criteria that displays the physical appearance of the person subject to the criteria as a reference, and the first generation means generates the reference point image of the person subject to the criteria based on the reference point information identified by the identification means and the reference image of the person subject to the criteria.

[0203] The generation system described in Appendix 3 is the generation system described in Appendix 2, wherein the identification means performs a first identification process to identify the size of a predetermined part of the body of the target person based on the body-related information acquired by the acquisition means, and a second identification process to identify the reference point information based on the body-related information acquired by the acquisition means and the size of the predetermined part identified in the first identification process, and the second generation means generates the styling display image based on the size of the predetermined part of the body of the target person identified in the first identification process, the target person display image generated by the first generation means, and the styling object-related information.

[0204] The generation system in Appendix 4 is the generation system described in Appendix 2, wherein the identification means further identifies the degree of obesity of the target person based on the body-related information acquired by the acquisition means, and the first generation means generates a first target person display image that displays the appearance of the target person, which is generated based on the entire body of the target person, based on the reference point information identified by the identification means, the degree of obesity information identified by the identification means, and the standard target person display image, and the first generation means side first processing that generates a first target person display image that displays the appearance of the target person, which is generated based on the entire body of the standard target person, and face information that shows the face of the target person, Based on the face information input to the generation system, the first generation means performs a second process to generate a second target person display image that displays the appearance of the target person including the face corresponding to the face information by editing only the face of the target person shown in the first target person display image generated in the first target person display image generated in the first processing on the first generation means. The second generation means generates the styling display image based on the second target person display image generated by the first generation means and the styling object related information.

[0205] The generation system of Appendix 5, in the generation system described in Appendix 1, further acquires styling object-related information indicating the styling object selected by the user of the generation system, wherein the styling object-related information includes at least a styling object display image that displays the styling object and styling object size information indicating the size of the styling object, and the second generation means generates the styling display image based on the target user display image generated by the first generation means, the styling object display image acquired by the acquisition means, and the styling object size information acquired by the acquisition means.

[0206] The generation program described in Appendix 6 is a generation program that generates a styling display image showing a person to whom a styling object has been applied, and causes a computer to function as an acquisition means for acquiring body-related information indicating the body of the person to whom the styling object has been applied; a first generation means for generating a person display image showing the appearance of the person to whom the styling object has been applied based on the body-related information acquired by the acquisition means; and a second generation means for generating the styling display image based on the person display image generated by the first generation means and styling object-related information indicating the styling object.

[0207] (Effect of the note) According to the generation system described in Appendix 1 and the generation program described in Appendix 6, by generating a styling display image, it becomes possible to, for example, confirm the state in which the styling target object has been applied.

[0208] According to the generation system described in Appendix 2, by using reference point information that indicates the positions of multiple reference points that serve as the basis for shaping the physical appearance of the person to whom the styling is applied, it is possible to appropriately generate, for example, a display image of the person to whom the styling is applied, and to appropriately confirm the state in which the styling object has been applied.

[0209] According to the generation system described in Appendix 3, by generating a styling display image based on the size of a predetermined part of the target person's body, for example, it is possible to appropriately generate a target person display image taking into account the size of a predetermined part of the target person's body, and to appropriately confirm the state in which the styling object has been applied.

[0210] According to the generation system described in Appendix 4, by performing a second processing on the first generation means side to generate a second target person display image based on the first target person display image generated in the first processing on the first generation means side, it is possible to appropriately generate the second target person display image, for example, and to appropriately confirm the state in which the styling target object has been applied.

[0211] According to the generation system described in Appendix 5, by generating a styling display image based on the styling target display image and styling target size information, it becomes possible to appropriately confirm, for example, the state in which the styling target has been applied. [Explanation of Symbols]

[0212] 1. Terminal device 2 Server devices 11 Communications Department 12 Touchpads 13 displays 14 Records Section 15 Control Unit 21 Communications Department 22 Records Section 23 Control Unit 100 Information Systems 901 First Reference Skeleton Point 902 Second Reference Skeleton Point 903 Third Reference Skeleton Point 904 Fourth Reference Skeleton Point 905 Fifth Reference Skeleton Point 906 Sixth Reference Skeleton Point 907 7th Reference Skeleton Point 908 8th Reference Skeleton Point 909 9th Reference Skeleton Point 910 10th Reference Skeleton Point 911 11th Reference Skeleton Point 912 12th Reference Skeleton Point 913 13th Reference Skeleton Point 914 14th Reference Skeleton Point 915 15th Reference Skeleton Point 916 16th Reference Skeleton Point 917 17th Reference Skeleton Point 918 18th Reference Skeleton Point B1 Base Model L Length M1 1st area M2 2nd area U1 First User Model U2 Second User Model

Claims

1. A generation system that generates a styling display image showing a person to whom a styling object has been applied, A means for acquiring body-related information that indicates the physical body of the person to be applied, A first generation means generates an image of the person to be applied, which displays the appearance of the person to be applied, based on the physical information acquired by the acquisition means. A second generation means generates the styling display image based on the target person display image generated by the first generation means and the styling object-related information indicating the styling object, A generation system equipped with the following features.

2. Based on the body-related information acquired by the acquisition means, the system further comprises a means for identifying reference point information that indicates the positions of multiple reference points that serve as the basis for shaping the physical appearance of the person subject to the criteria, using a display image of the person subject to the criteria, which displays the physical appearance of the person subject to the criteria, as a reference. The first generation means generates the target person display image based on the reference point information identified by the identification means and the target person display image. The generation system according to claim 1.

3. The aforementioned specifying means is, Based on the body-related information acquired by the acquisition means, a first identification process is performed to identify the size of a predetermined part of the body of the person to be applied, A second identification process is performed to identify the reference point information based on the body-related information acquired by the acquisition means and the size of the predetermined body part identified in the first identification process. The second generation means generates the styling display image based on the size of a predetermined part of the body of the person to be applied, which was identified in the first identification process, the person to be applied display image generated by the first generation means, and the styling object related information. The generation system according to claim 2.

4. The identification means further identifies the degree of obesity of the subject based on the physical information acquired by the acquisition means, The first generating means is A first processing step on the first generation means side generates a first target person display image that displays the appearance of the target person, which is generated based on the entire body of the target person, based on the reference point information identified by the identification means, the degree of obesity information identified by the identification means, and the target person display image of the target person, The first generation means performs a second process that generates a second target person display image which displays the appearance of the target person including the face corresponding to the face information, by editing only the face of the target person shown in the first target person display image which was generated in the first processing on the first generation means based on the face information input to the generation system. The second generation means generates the styling display image based on the second target person display image generated by the first generation means and the styling object related information. The generation system according to claim 2.

5. The acquisition means further acquires styling object-related information indicating the styling object selected by the user of the generation system, The aforementioned information relating to the object to be styled includes, at a minimum, A styling target display image that shows the styling target, Includes, a size information of the object to be styled, which indicates the size of the object to be styled, The second generating means is, Based on the target person display image generated by the first generation means, the styling target object display image acquired by the acquisition means, and the styling target object size information acquired by the acquisition means, the styling display image is generated. The generation system according to claim 1.

6. A generation program that generates a styling display image showing the person to whom the styling object is applied, Computers, A means for acquiring body-related information that indicates the physical body of the person to be applied, A first generation means generates an image of the person to be applied, which displays the appearance of the person to be applied, based on the physical information acquired by the acquisition means. A second generation means generates the styling display image based on the target person display image generated by the first generation means and the styling object-related information indicating the styling object, A generation program that functions as such.

Images