JP7473274B1 - Information processing system, program, and information processing method - Google Patents

Abstract

[Problem] To provide an information processing system etc. that can easily grasp information about a group of images used in machine learning. [Solution] According to one aspect of the present invention, an information processing system including a processor is provided. In this information processing system, the processor acquires relevant information related to the group of images used in machine learning in an information acquisition step. In an output step, the processor outputs metadata indicating the relevant information acquired about the group of images in association with an image generated by an artificial intelligence module that performed machine learning using the group of images. [Selected Figure] Figure 1

Description

The present invention relates to an information processing system, a program, and an information processing method.

Patent Document 1 discloses a technology for generating new images of a specific subject based on parameters of an image generator obtained by performing a learning process using a combination of each image in a sub-training image group, which is a group of multiple images of a subject, and posture information of the specific subject in each image in the sub-training image group as training data.

JP 2020-86869 A

In order for the artificial intelligence module to generate images, machine learning is carried out using existing images as training data. In such cases, understanding the images used in the machine learning process can be important in using the generated images.

In view of the above circumstances, the present invention aims to provide an information processing system etc. that can easily grasp information about a group of images used in machine learning.

According to one aspect of the present invention, an information processing system including a processor is provided. In this information processing system, the processor acquires related information related to a group of images used for machine learning in an information acquisition step. In an output step, the processor outputs metadata indicating the related information acquired about the group of images in association with an image generated by an artificial intelligence module that has performed machine learning using the group of images.

This aspect makes it easy to understand information about the group of images used in machine learning.

1 is a diagram showing an example of the overall configuration of an image generating system 1. FIG. FIG. 2 is a diagram illustrating an example of a hardware configuration of a server device 10. FIG. 2 is a diagram illustrating an example of a hardware configuration of an operator terminal 20. FIG. 11 is an activity diagram showing an example of a data set creation process. FIG. 13 is a diagram showing an example of a displayed data set creation screen. FIG. 2 is a diagram showing an example of an image database DB1. FIG. 13 is a diagram showing an example of a displayed candidate image. FIG. 11 is a diagram illustrating an example of generated data set information. FIG. 11 is an activity diagram illustrating an example of a learning process. FIG. 13 is a diagram showing an example of a displayed machine learning screen. FIG. 13 is a diagram showing an example of stored learning model information. FIG. 11 is an activity diagram showing an example of an image generation process. FIG. 13 is a diagram showing an example of a displayed image generation screen. A figure showing an example of a displayed AI image screen. FIG. 13 is a diagram showing an example of a displayed DS image screen. FIG. 13 is a diagram showing an example of a displayed contract information screen. FIG. 11 is a diagram showing an example of displayed metadata. FIG. 11 is a diagram showing an example of displayed metadata. FIG. 13 is a diagram showing an example of a selection screen for displayed rights information. FIG. 13 is a diagram showing an example of a screen for changing a displayed DS image. FIG. 13 is a diagram showing another example of stored learning model information. FIG. 11 is a diagram showing another example of displayed metadata. FIG. 13 is a diagram showing an example of a displayed screen for adding a user image. FIG. 13 is a diagram showing another example of stored learning model information. A figure showing an example of learning model information managed in node format.

The following describes embodiments of the present invention with reference to the drawings. The various features shown in the following embodiments can be combined with each other.

The program for implementing the software used in this embodiment may be provided as a non-transitory computer-readable recording medium, or may be provided so that it can be downloaded from an external server, or may be provided so that the program is started on an external computer and its functions are implemented on a client terminal (so-called cloud computing).

In this embodiment, a "unit" may also include, for example, a combination of hardware resources implemented by a circuit in the broad sense and software information processing that can be specifically realized by these hardware resources. In addition, this embodiment handles various types of information, which may be represented, for example, by physical values of signal values representing voltage and current, high and low signal values as a binary bit group consisting of 0 or 1, or quantum superposition (so-called quantum bits), and communication and calculations may be performed on a circuit in the broad sense.

In the broad sense, a circuit is a circuit realized by at least appropriately combining a circuit, circuitry, a processor, and memory. In other words, it includes application specific integrated circuits (ASICs), programmable logic devices (e.g., simple programmable logic devices (SPLDs), complex programmable logic devices (CPLDs), and field programmable gate arrays (FPGAs)), etc.

<Embodiment 1>
1. System Configuration Hereinafter, the system configuration according to the first embodiment will be described.
FIG. 1 is a diagram showing an example of the overall configuration of an image generation system 1. FIG. 1 shows an overview of each device included in the image generation system 1 and a user who uses the devices. Each overview will be explained from time to time with reference to other figures. The image generation system 1 is an information processing system that generates an image using an artificial intelligence module having a function of generating an image. The image generation system 1 executes a learning process that causes an artificial intelligence module to perform machine learning, and an image generation process that causes a trained artificial intelligence module to generate an image.

The image generation system 1 includes a communication line 2, a server device 10, an operator terminal 20, and a user terminal 30. The communication line 2 is not particularly limited, but may be configured, for example, by the Internet network. The communication line 2 may also include a local area network, a mobile communication network, and a VPN (Virtual Private Network), etc. The communication line 2 mediates data exchange between devices connected to the line. In the example of FIG. 1, the server device 10 is connected to the communication line 2 by a wire, and the operator terminal 20 and the user terminal 30 are connected wirelessly. The connection of each device to the communication line 2 may be wired or wireless.

The server device 10 is an information processing device that executes learning processes, image generation processes, and the like. The server device 10 stores an image database DB1, a dataset database DB2, and a learning model database DB3. The image database DB1 stores images and the like used in machine learning. The dataset database DB2 stores information (hereinafter referred to as "dataset information") related to a dataset that includes a group of images selected from the images stored in the image database DB1.

The dataset referred to here is a collection of data used as training data for machine learning. The dataset information is, for example, information that can identify images (hereinafter referred to as "DS images") included in the dataset. The learning model database DB3 stores information (hereinafter referred to as "learning model information") related to the learning model generated by machine learning performed using the dataset. The dataset information and learning model information will be explained in detail later.

The server device 10 is equipped with an image generation AI module 100. The image generation AI module 100 is an artificial intelligence module that is adjusted (tuned) to generate images using AI (Artificial Intelligence) technology. The image generation AI module 100 performs machine learning using a dataset that includes a DS image identified by the dataset information stored in the dataset database DB2, and generates an image that has similar characteristics to the DS image. Hereinafter, an image generated by the image generation AI module 100 is referred to as an "AI image."

The operator terminal 20 is a terminal whose user is an operator who operates the image generation system 1, such as a laptop computer. The user terminal 30 is a terminal whose user is an image user who uses AI images on web pages, etc., such as a laptop computer. The server device 10 executes a display process for displaying images on the operator terminal 20 and the user terminal 30, and an authentication process for authenticating the users of the operator terminal 20 and the user terminal 30.

The server device 10 performs processes such as generating and transmitting HTML (Hyper Text Markup Language) files as display processes, and displays web pages showing system screens on user terminals such as the operator terminal 20 and the user terminal 30. Note that the user terminal may install an application program for using the image generation system 1, and the server device 10 may perform processes such as generating and transmitting display data in that application as display processes. The server device 10 controls the display of the user terminal by performing these display processes.

The server device 10 stores authentication information (user ID, password, etc.) for authenticating users such as the operator and image users, and authenticates users who input the authentication information. By authenticating users, the server device 10 can restrict access to data and make data entered by users identifiable.

2. Hardware Configuration The hardware configuration according to the first embodiment will be described below.
2 is a diagram showing an example of a hardware configuration of the server device 10. The server device 10 includes a control unit 11, a storage unit 12, a communication unit 13, and a bus 14. The bus 14 electrically connects the various units included in the server device 10.

(Control unit 11)
The control unit 11 has at least one processor. The at least one processor may be configured, for example, by a central processing unit (CPU) (not shown), a micro processing unit (MPU), a graphics processing unit (GPU), one or more integrated circuits, one or more discrete circuits, or a combination of these.

The control unit 11 is a computer that realizes various functions related to the image generation system 1 by reading out a specific program stored in the memory unit 12. In other words, information processing by the software stored in the memory unit 12 is specifically realized by the control unit 11, which is an example of hardware, and can be executed as each functional unit included in the control unit 11. Note that the control unit 11 is not limited to being single, and may be implemented with multiple control units 11 for each function. A combination of these may also be used.

(Memory unit 12)
The storage unit 12 stores various information defined by the above description. This can be implemented, for example, as a storage device such as a solid state drive (SSD) or a hard disk drive (HDD) that stores various programs and the like related to the image generation system 1 executed by the control unit 11, or as a memory such as a random access memory (RAM) that stores temporarily required information (arguments, arrays, etc.) related to the program calculations. The storage unit 12 stores various programs, variables, etc. related to the image generation system 1 executed by the control unit 11.

(Communication unit 13)
The communication unit 13 is configured by a communication module. The communication module may be a wireless communication module conforming to standards such as IEEE802.11a/b/g/n/ac/ax, LTE, 5G, and 6G, or may be a wired communication module conforming to standards such as IEEE802.3. The communication unit 13 is configured to be capable of transmitting various electrical signals from the server device 10 to external components. The communication unit 13 is also configured to be capable of receiving various electrical signals from the external components to the server device 10. More preferably, the communication unit 13 has a network communication function, and may be capable of communicating various information between the server device 10 and an external device via the communication line 2.

Figure 3 is a diagram showing an example of the hardware configuration of the operator terminal 20. The operator terminal 20 includes a control unit 21, a memory unit 22, a communication unit 23, an input unit 24, an output unit 25, and a bus 26. The bus 26 electrically connects the various units included in the operator terminal 20. The control unit 21, the memory unit 22, and the communication unit 23 are similar hardware to the control unit 11, the memory unit 12, and the communication unit 13 shown in Figure 2, although the specifications, model, etc. may be different.

(Input unit 24)
The input unit 24 has keys, buttons, a touch screen, a mouse, etc., and receives input from the user. The input unit 24 also has a microphone, and receives voice input from the user.

(Output unit 25)
The output unit 25 has a display, a speaker, and the like, and displays visual information generated in a manner that is visible to the user, such as screens, images, icons, text, and the like, on the display surface of the display, and outputs sounds including voice.

The user terminal 30 shown in FIG. 3 has the same hardware configuration as the operator terminal 20. In the following description of the user terminal 30, only the control unit 31 is given a different reference number from the control unit 21 of the operator terminal 20.

3. Information Processing Hereinafter, information processing according to the embodiment will be described. In the following description, the server device 10, the operator terminal 20, and the user terminal 30 are described as the subjects of each information processing, but the information processing is executed by a processor possessed by the control unit of each device. The information processing executed by each processor includes the learning processing and image generation processing described above, as well as a dataset creation processing for creating a dataset used in machine learning.

Figure 4 is an activity diagram showing an example of a dataset creation process. The dataset creation process shown in Figure 4 is initiated when a user of the operator terminal 20 performs an operation to display a dataset creation screen. First, the server device 10 generates screen data showing the dataset creation screen, and transmits the generated screen data to the operator terminal 20 (A11). The operator terminal 20 displays the dataset creation screen shown in the transmitted screen data (A12).

On the dataset creation screen, the operator selects DS images (images to be included in the dataset). Methods for selecting DS images include individually inputting the file name and path name of the image to be selected as the DS image, or specifying a folder in which the image to be used as the DS image is stored. In the example of Figure 5, a method is described in which images with common characteristics are narrowed down from a large number of images as candidate images for the DS image (hereinafter referred to as "candidate images"), and then the image to be used as the DS image is selected.

Figure 5 is a diagram showing an example of a displayed dataset creation screen. The dataset creation screen G1 shown in Figure 5 displays the character string "Please select images to include in the dataset," an input field E11 for the image user, an input field E12 for the dataset name, an input field E13 for image attribute conditions, an input field E14 for image rights conditions, an image display field E15, a select all button B11, a deselect button B12, and a confirm button B13.

As described above, the person who will use the AI image is input as the image user in input field E11. The name of the dataset to be created is input in input field E12. Filtering conditions for narrowing down the candidate images are input in input fields E13 and E14. These filtering conditions are conditions that can be determined based on information associated with each image in the image database DB1 described below.

In input field E13, conditions for the attributes of the candidate images are input as narrowing-down conditions. Examples of image attribute conditions include image size, resolution, and subject. When the image is a photograph, the subject is the main object that appears in the photograph, and includes people and landscapes. Note that an image with a "person" as the subject is an image in which the person is identifiable and portrait rights apply to that person. The attribute conditions listed here are merely examples, and any attribute (for example, the date and time of shooting, the location of shooting, or the name of the person who is the subject) may be used as long as it is a condition for narrowing down the candidate images.

For example, when "large", "medium", and "small" are entered as the image size in input field E13, only one size may be entered, multiple sizes may be entered, or a condition that does not narrow down the sizes, such as "all sizes", may be entered. In this way, it is sufficient for one or more conditions to be entered in the image attribute conditions. Note that if no conditions are entered, the conditions may not be narrowed down, just as in the case where "all sizes" is entered.

In input field E14, conditions related to the rights held by the image user with respect to the candidate image are entered as conditions for narrowing down the candidate images. The rights held by the image user include, for example, the copyright given to the person who created the candidate image, the copyright held by the creator (author) of the work depicted in the candidate image, portrait rights that can be claimed by the person depicted in the candidate image, or the right to use the candidate image granted by the owner of those rights (hereinafter simply referred to as "permission").

In input field E14, information such as whether or not the image can be used, the purpose of use, and the expiration date of use is entered. Whether or not the image can be used indicates whether or not the image user is currently able to use the image, in other words, whether or not the image user currently has any rights that allow the image to be used. In what follows, among the rights that allow the image to be used, copyright or permission to use it is referred to as "copyright-related rights," and portrait rights or permission to use it is referred to as "portrait-related rights."

In the following, if the subject is a person, the image can be used if the image user has both copyright-related rights and portrait-related rights, and is marked as "○", if the image user has only one of copyright-related rights or portrait-related rights, and is marked as "△", and if the image user does not have either copyright-related rights or portrait-related rights, and is marked as "×". In addition, if the subject is not a person, the image can be used if the image user has copyright-related rights, and is marked as "○", and if the image user does not have copyright-related rights, and is marked as "×". Note that even if an image is marked as "△" or "×", it is possible that the image user will obtain copyright-related rights or portrait-related rights and use the image if they want to use it for machine learning, so it is possible to use it as a filtering condition.

If the image user holds the copyright and portrait rights, there are no particular restrictions on the purpose of use of the image, but if the image user is a licensee who has been granted permission to use the image by the rights holder who holds those rights, the purpose of use may be restricted by contract with the licensor. Such restricted purpose of use is entered in the input field for purpose of use. Examples of purposes of use include use on a web page, use in advertising, use within the company, and use outside the company.

If the image user holds the copyright and portrait rights, the image can be used indefinitely as long as the image user holds those rights, but if the image user is a licensor, the end of the usage period, i.e., the usage deadline, may be determined by contract with the licensor. The usage deadline input field is used to input the period during which the image user can use the image. For example, if the deadline "until the end of 2025" is entered, the images will be narrowed down to those that can be used at least until the end of 2025.

In input field E14, for example, only "○" may be entered as information indicating whether the image can be used, or multiple conditions such as "○" and "△" may be entered, or a condition such as "all" may be entered that does not narrow down the images based on whether the image can be used. In this way, as with image attribute conditions, it is sufficient for one or more conditions to be entered into the image rights conditions. Note that if no conditions are entered, the conditions may not be narrowed down, just as in the case where "all" is entered.

The operator terminal 20 accepts input of narrowing down conditions for candidate images by the user (A13). The operator terminal 20 transmits the input narrowing down conditions to the server device 10. The server device 10 reads out images that satisfy the transmitted narrowing down conditions from the images stored in the image database DB1 as candidate images (A14).

Figure 6 is a diagram showing an example of image database DB1. In image database DB1, the image file name of each image, image attribute information, and contract information are stored in association with each other. The image file name may include the path to the folder in which the image data is stored in addition to the file name of the image data. The image attribute information stores the image size, resolution, subject, etc. Examples of image subjects include people and landscapes. In addition to these, the subjects may also include buildings, vehicles, animals, etc.

The image contract information stores the image copyright holder, copyright licensee, copyright license conditions, portrait right owner, portrait right licensee, and portrait right license conditions. Note that portrait rights do not arise if the subject is not a person, so if the subject is a landscape, the portrait rights are marked "NA" (not applicable). The license conditions include the purpose of use of the image and the expiration date of the image, as described in Figure 5.

In addition to these conditions, the permission conditions may include conditions that clearly define the scope of the copyrighted work or portrait in the image, restrictions on altering the copyrighted work or portrait in the image, usage fees (including remuneration or royalties to the copyright and portrait right holder), usage period (including not only the usage period but also the start time of usage), an obligation to indicate the copyright holder or portrait right holder, an obligation to protect the privacy and reputation of the portrait right holder, or a usage area. The contract information may also include information such as the contract holder, the contract date, or the contract number, and image data of the contract. The server device 10 reads out images whose attribute information and contract information satisfy the narrowing conditions and transmits them to the operator terminal 20. The operator terminal 20 displays the transmitted images as candidate images (A15).

Figure 7 is a diagram showing an example of a displayed candidate image. Figure 7 shows the dataset creation screen G1 shown in Figure 5. In the example of Figure 7, "Company C1" has been entered as the image user, and "DSv1.1 for Company C1" has been entered as the dataset name. The operator terminal 20 displays the transmitted candidate images in the image display field E15 shown in Figure 5, and displays check boxes E16 in association with these candidate images. The candidate images displayed in the display field E15 are only a portion of the images, and other candidate images can be displayed by operations such as scrolling.

The administrator checks the check boxes E16 of the candidate images that they want to include in the data set. When the Select All button B11 is operated, all check boxes E16 are checked, and when the Deselect button B12 is operated, all check boxes E16 are unchecked. Every time the administrator changes the conditions entered in the image attribute condition input field E13 and the image rights condition input field E14, operations A13 to A15 are performed to update the displayed candidate images.

The operator selects the candidate images to be included in the data set by, for example, roughly narrowing down the candidate images using the filtering conditions, checking all the check boxes E16, and then unchecking the check boxes E16 of unnecessary candidate images. Note that there may be a function to display candidate images that do not satisfy the filtering conditions in a separate display area so that candidate images that do not satisfy the filtering conditions can be added. When the operator has decided on the candidate images to be included in the data set, he or she operates the decision button B13.

The operator terminal 20 accepts the input operation of the narrowing down conditions, the check operation, and the operation of the decision button B13 as the selection operation and decision operation of the candidate image (A16), and transmits selection result data indicating the selection result to the server device 10. The server device 10 confirms the candidate image indicated by the transmitted selection result data as a DS image, and generates information regarding the dataset including the confirmed DS image as dataset information (A17). The server device 10 then stores and preserves the generated dataset information in the dataset database DB2 shown in FIG. 1 (A18).

Figure 8 is a diagram showing an example of generated dataset information. In the dataset database DB2 shown in Figure 8, the filtering conditions, image information, and rights information are stored as dataset information in association with the dataset name. The filtering conditions store the conditions entered in the image attribute condition input field E13 and the image rights condition input field E14 on the dataset creation screen G1. Not all candidate images narrowed down by these filtering conditions are included in the dataset, but the filtering conditions are stored as reference information that indicates the characteristics of the images included in the dataset.

The image information includes information such as the image file name and the subject. The image file name indicates the file names (which may also include path names) of all DS images included in the dataset. The server device 10 can read out the DS images included in the dataset by referencing these image file names.

The subject indicates the percentage of the subject that appears in all DS images included in the dataset. For example, in the case of a dataset called "DSv1.1 for Company C1," the subject percentage is shown as 80% people and 10% landscapes. The server device 10 calculates this percentage by counting the number of subjects in all DS images stored in the image database DB1. Note that the image information may also store information other than the subject, such as the number of images, image size, and resolution. In short, it is sufficient if the image information contains information that represents the characteristics of those DS images, making it easy to get an idea of what kind of DS images are included in the dataset.

The rights information includes the image user (referred to as "user" in the diagram), permission information, copyright ownership, copyright permission, permission conditions, portrait right ownership, portrait right permission, and permission conditions. The image user is the user of the DS image indicated by the rights information.

The permission information is information indicating whether all DS images included in the dataset can be used. As described above, the permission of a DS image is indicated as "○" if the image user has both copyright-related rights and portrait-related rights (only copyright-related rights if the subject does not include a person), "△" if they only have one of them, and "×" if they do not have either of them (only copyright-related rights if the subject does not include a person). The permission information is indicated as "○" if all DS images are "○", "△" if there is even one "△", and "×" if there is even one "×".

"Copyright Owned" indicates the percentage of all DS images for which the image user owns the copyright. "Copyright Licensed" indicates the percentage of all DS images for which the image user has been granted copyright. "License Conditions" for copyright indicates the percentage of all DS images for which the license conditions are common. In the example of Figure 8, the percentage of images for which the license conditions include the same period of use is shown. In the case of a dataset called "DSv1.1 for Company C1," 50% are "unlimited," 25% are "up to 20xx," and 10% are "up to 20yy."

"Portrait rights owned" indicates the percentage of all DS images with a person subject for which the image user owns the portrait rights. "Portrait rights licensed" indicates the percentage of all DS images with a person subject for which the image user has been licensed for the portrait rights. "License conditions" for portrait rights indicates the percentage of all DS images with a person subject that have common license conditions. In the example of Figure 8, the percentage of images with common license conditions for the period of use is shown. In the case of a dataset called "DSv1.1 for Company C1", "unlimited" is shown at 40%, "up to 20xx" at 30%, and "up to 20yy" at 15%.

The server device 10 calculates the above-mentioned rights information by referring to the contract information of each DS image stored in the image database DB1, and stores the information in the dataset database DB2. Note that the rights information shown in FIG. 8 is an example, and is not limited to this information. The rights information may include, for example, license conditions other than the usage period, or may include only the license conditions that have the highest proportion of common images. In short, it is sufficient that the rights information includes information that describes the characteristics of the rights of those DS images, so that it is easy to estimate the rights status of the DS images included in the dataset.

Next, a learning process will be described in which machine learning is performed using the dataset created by the dataset creation process described above.
Fig. 9 is an activity diagram showing an example of a learning process. The learning process shown in Fig. 9 is started when a user of the operator terminal 20 performs an operation to display a machine learning screen. First, the server device 10 generates screen data showing the machine learning screen and transmits the generated screen data to the operator terminal 20 (A21). The operator terminal 20 displays the machine learning screen shown by the transmitted screen data (A22).

Figure 10 is a diagram showing an example of a displayed machine learning screen. The machine learning screen G2 shown in Figure 10 displays the character string "Please select the dataset to be used for machine learning.", an input field E21 for the image user, an input field E22 for the learning model name, a display field E23 for a list of datasets, and an execute machine learning button B21. In the input field E21, a person who uses the AI image is entered as the image user. In the example of Figure 10, "Company C1" is entered as the image user.

When the operator terminal 20 accepts input of the image user (A23), it transmits user data indicating the input image user to the server device 10. When the user data is transmitted, the server device 10 refers to the dataset database DB2 and reads out the dataset information associated with the image user indicated by the user data (A24). The server device 10 transmits the read dataset information to the operator terminal 20. The operator terminal 20 displays the transmitted dataset information in the display field E23 of the dataset list (A25).

In the example of FIG. 10, the operator terminal 20 displays the dataset name, whether or not it can be used ('Yes' in the figure), the subject, copyright permission conditions, and portrait right permission conditions as dataset information. In the example of FIG. 10, the dataset information of the datasets 'DSv1.1 for company C1', 'DSv1.2 for company C1', 'DSv1.3 for company C1', 'DSv2.1 for company C1', and 'DSv2.2 for company C1' that were created with company C1 as the image user is displayed.

The dataset information displayed in the display field E23 is not limited to that shown in FIG. 10. For example, the number of images, resolution, the ratio of copyright and its permission, or the ratio of portrait rights and its permission, etc. may be displayed in the display field E23. In short, it is sufficient that information expressing the characteristics of the DS images and the characteristics of the rights contained in the dataset is displayed as dataset information as a basis for deciding when selecting a dataset to be used for machine learning.

The operator refers to the displayed dataset information and selects a dataset to be used for machine learning. In the example of FIG. 10, the operator selects "DS v1.3 for company C1." Note that, although only one dataset is selected in the example of FIG. 10, two or more datasets may be selected. The operator inputs the name of the learning model to be generated by machine learning. In the example of FIG. 10, the operator inputs "Model v1.3 for company C1." The operator operates the execute machine learning button B21 to cause the image generation AI module 100 to execute machine learning on the selected dataset.

The operator terminal 20 accepts the above operation as an operation to select a dataset (A26) and an operation to instruct machine learning (A27), and transmits instruction data indicating the selected dataset and the name of the input learning model to the server device 10. The server device 10 references the file name of the DS image included in the dataset indicated by the transmitted instruction data in the dataset database DB2, and reads out the image with the referenced file name as a dataset (A31).

Next, the server device 10 instructs the image generation AI module 100 to perform machine learning using the read dataset (A32). The image generation AI module 100 performs machine learning using the instructed dataset (A33). As a result of performing the machine learning, the image generation AI module 100 generates a learning model capable of outputting an image similar to the DS image included in the dataset. The server device 10 stores and saves the generated learning model in the learning model database DB3 in association with learning model information related to the learning model (A34).

Figure 11 is a diagram showing an example of stored learning model information. In the learning model database DB3 shown in Figure 11, learning model information is stored in association with the learning model. The learning model information includes a dataset name, image information, and rights information. The dataset name indicates the name of the dataset used in machine learning when generating the learning model. For example, "Model v1.1 for company C1" is associated with "DSv1.1 for company C1". Furthermore, "Model v1.2 for company C1" is associated with two datasets, "DSv1.1 for company C1" and "DSv2.1 for company C1".

The image information and rights information stored in the image information and rights information database DB2 in association with the corresponding dataset is read and stored. For example, "Model v1.1 for Company C1" stores image information including the proportion of the subject, such as "People: 80%", in association with the image user, "Company C1", the availability of use, such as "○", copyright permission conditions, such as "Unlimited: 50%", and portrait right permission conditions, such as "Unlimited: 40%".

The server device 10 refers to the image information and rights information stored in the dataset database DB2, reads out information associated with the datasets used in machine learning, and stores it in the learning model database DB3. Note that for a learning model that has two or more datasets used in machine learning, such as "Model v1.2 for Company C1," the server device 10 refers to the image information and contract information associated with the images included in those datasets in the image database DB1, recalculates the proportion of subjects and the proportion of permission conditions, etc., and stores them as image information and rights information.

The learning model information shown in FIG. 11 is an example and is not limited to this information. The learning model information may include image information such as the number of images or resolution, and may include only license conditions other than the usage period or license conditions with the highest proportion of common images as rights information. In short, it is sufficient that the learning model information includes information that represents the characteristics of the DS images and the rights characteristics so that it is easy to estimate the characteristics of the DS images and the rights status of those DS images included in the dataset used in the machine learning of the learning model.

Next, we will explain the image generation process that generates an AI image using the learning model generated by the learning process described above.
Fig. 12 is an activity diagram showing an example of image generation processing. The image generation processing shown in Fig. 12 is started when a user of the operator terminal 20 performs an operation to display an image generation screen. First, the server device 10 generates screen data showing the image generation screen and transmits the generated screen data to the operator terminal 20 (A41). The operator terminal 20 displays the image generation screen shown by the transmitted screen data (A42).

Figure 13 is a diagram showing an example of a displayed image generation screen. The image generation screen G3 shown in Figure 13 displays the character string "Please select the learning model to be used for image generation.", an input field E31 for the image user, a display field E32 for a list of learning models, an input field E33 for a prompt, and an execute image generation button B31. In the input field E31, a person who will use the AI image is entered as the image user. In the example of Figure 13, "Company C1" is entered as the image user.

When the operator terminal 20 accepts input of the image user (A43), it transmits user data indicating the input image user to the server device 10. When the server device 10 receives the user data, it reads out the learning model information associated with the image user indicated by the user data (A44). The server device 10 transmits the read learning model information to the operator terminal 20. The operator terminal 20 displays the transmitted learning model information in the display field E32 of the list of learning models (A45).

In the example of FIG. 13, the operator terminal 20 displays the learning model name, whether or not it can be used ("Yes" in the figure), the subject, copyright permission conditions, and portrait right permission conditions as learning model information. In the example of FIG. 13, learning model information is displayed for learning models generated by machine learning performed with company C1 as the image user, namely "Model v1.1 for company C1", "Model v1.2 for company C1", "Model v1.3 for company C1", "Model v2.1 for company C1", and "Model v2.2 for company C1".

The learning model information displayed in the display field E32 is not limited to that shown in FIG. 13. For example, the number of images, resolution, the ratio of copyright and its permission, or the ratio of portrait rights and its permission, etc. may be displayed in the display field E32. In short, it is sufficient that information representing the characteristics of the DS image and the characteristics of the rights contained in the dataset used in the machine learning of the learning model is displayed as learning model information as a basis for deciding when selecting a learning model to be used for AI image generation.

The operator refers to the displayed learning model information and selects the learning model to be used to generate the AI image. In the example of FIG. 13, the operator selects "Model v1.3 for company C1." The operator inputs a sentence (a so-called prompt) that indicates the AI image to be generated in the prompt input field E33. Then, the operator operates the image generation execution button B31 to cause the image generation AI module 100 to generate an AI image using the selected learning model using the input prompt.

The operator terminal 20 accepts the above operations as an operation to select a learning model (A46), an operation to input a prompt (A47), and an operation to instruct the generation of an AI image (A48), and transmits instruction data indicating the selected learning model and the input prompt to the server device 10. The server device 10 executes a process to generate an AI image based on the transmitted instruction data (A51). Specifically, the server device 10 executes a process to instruct the image generation AI module 100 to generate an AI image using the learning model and prompt indicated by the instruction data. The image generation AI module 100 generates an AI image based on this instruction.

Next, the server device 10 reads out learning model information associated in the learning model database DB3 with the learning model used to generate the AI image (A52). The server device 10 then generates metadata for the generated AI image based on the read learning model information (A53). The server device 10 then generates screen data showing an AI image screen that displays the generated AI image and metadata, and transmits the generated screen data to the operator terminal 20 (A54). The operator terminal 20 displays the AI image screen indicated by the transmitted screen data (A55).

Figure 14 is a diagram showing an example of a displayed AI image screen. On the AI image screen G4 shown in Figure 14, the character string "AI image is completed," an AI image display field E41, a metadata display field E42, a back button B41, and an image regenerate button B42 are displayed. The generated AI image F41 is displayed in display field E41. The generated metadata M41 is displayed in display field E41.

In the example of FIG. 14, the server device 10 generates metadata M41 including an "image ID", a "learning model name", "model safety", a "dataset name", a link L41 to a DS image, and a link L42 to contract information. "Image ID" indicates the identification information of the AI image issued by the image generation AI module 100 when generating the AI image F41. "Learning model name" indicates the name of the learning model used to generate the AI image F41. "Dataset name" indicates the name of the dataset used for machine learning of the learning model used to generate the AI image F41.

"Model safety" indicates the security of the rights of the learning model used to generate the AI image F41. The server device 10 indicates "availability" included in the learning model information of the learning model used to generate the AI image F41 as "model safety." In other words, the server device 10 indicates "○" if both copyright-related rights and portrait-related rights are available for all DS images included in the dataset used in the machine learning of the learning model used to generate the AI image F41, "△" if at least one image lacks either of these rights, and "×" if at least one image lacks both of these rights.

When the back button B41 is operated, the server device 10 returns to the previous screen, the image generation screen G3 shown in FIG. 13, and displays the screen displayed immediately before the image generation execution button B31 was operated. The operator can select a learning model again on the displayed image generation screen G3. When the image regeneration button B42 is operated, the server device 10 executes the process from A51 onwards again without changing the learning model, and displays the regenerated AI image and metadata on the operator terminal 20.

The link L41 to the DS image indicates a link to an image included in the dataset indicated in "Dataset Name". When the link L41 is operated, the operator terminal 20 accepts the operation and transmits to the server device 10 the name of the dataset linked by the operated link L41 and instruction data instructing the display of the DS image (A61). The server device 10 reads out from the image database DB1 the DS images included in the dataset whose name is indicated in the transmitted instruction data (A62), and generates a DS image screen showing a list of the read DS images (A63). The server device 10 generates screen data showing the generated DS image screen, and transmits the generated screen data to the operator terminal 20. The operator terminal 20 displays the DS image screen indicated by the transmitted screen data (A64).

Figure 15 is a diagram showing an example of a displayed DS image screen. The DS image screen G5 shown in Figure 15 displays the character string "This image is included in the dataset used for machine learning," a display field E51 for the image user, a display field E52 for the dataset name, a display field E53 for the image attribute conditions, a display field E54 for the image rights conditions, and a display field E55 for the DS image. The server device 10 controls the display of these display fields by referring to the image database DB1, the dataset database DB2, and the learning model database DB3.

The server device 10 displays in display field E51 the name of the image user who uses the generated AI image, and in display field E52 the name of the dataset used for machine learning. The server device 10 also displays in display fields E53 and E54 the filtering conditions (attribute conditions and rights conditions) satisfied by the DS images used for machine learning, and in display field E55 displays the DS images included in the dataset. The DS images displayed in display field E55 are only a portion of the images, and other DS images can be displayed by operations such as scrolling.

Figure 15 shows an example of the display of DS images when candidate images are selected after narrowing down using the narrowing conditions as described in Figure 5 etc. In the example of Figure 15, by displaying the narrowing conditions, it is possible to understand what common characteristics the DS images have. Note that if the DS images are selected by specifying the file name, path name or folder name, the file name etc. may be displayed instead of the narrowing conditions.

The link L42 to the contract information indicates a link to the contract information of the image included in the dataset indicated in "Dataset Name". When the link L42 is operated, the operator terminal 20 accepts the operation and transmits to the server device 10 the name of the dataset linked by the operated link L42 and instruction data instructing the display of the contract information (A71). The server device 10 reads out the contract information of the DS image included in the dataset whose name is indicated by the transmitted instruction data from the image database DB1 (A72), and generates a contract information screen showing a list of the read contract information (A73). The server device 10 generates screen data showing the generated contract information screen, and transmits the generated screen data to the operator terminal 20. The operator terminal 20 displays the contract information screen indicated by the transmitted screen data (A74).

Figure 16 is a diagram showing an example of a displayed contract information screen. The contract information screen G6 shown in Figure 16 displays the character string "This is the contract information for the images included in the dataset used for machine learning," a display field E61 for the image user, a display field E62 for the dataset name, and a display field E63 for a list of contract information for DS images. The server device 10 controls the display of these display fields by referring to the image database DB1, the dataset database DB2, and the learning model database DB3.

The server device 10 displays in display field E61 the name of the image user who uses the generated AI image, and in display field E62 the name of the dataset used in the machine learning. In addition, the server device 10 displays in display field E63 the file names of the DS images used in the machine learning in association with the contract information for those DS images. The contract information includes the image copyright holder, copyright licensee, copyright license conditions, portrait right owner, portrait right licensee, and portrait right license conditions. For DS images in which the subject is not a person, the portrait-related rights are shown as "NA" (not applicable).

As described above, in the image generation system 1, the server device 10 functions as an example of an information acquisition unit that acquires related information related to a group of images used in machine learning. In other words, the group of images used in machine learning is a group of images included in a dataset, i.e., multiple DS images. The server device 10 acquires, for example, learning model information as related information in A52 shown in FIG. 12. As shown in FIG. 11, the learning model information is information indicating image information and rights information for multiple DS images. In this way, the related information (learning model information) acquired by the server device 10 includes rights information for the group of images used in machine learning.

The rights information is information indicating the right to use images (DS images) included in the image group used for machine learning. In the example of FIG. 11, the rights information includes the image user who holds the right to use the DS image, whether the DS image can be used, the copyright permission conditions, and the portrait right permission conditions. The server device 10 functions as an example of an output unit that outputs metadata indicating related information obtained about the image group in association with AI images generated by an artificial intelligence module (one example of which is the image generation AI module 100) that has performed machine learning using the image group.

In the example of FIG. 14, the server device 10 outputs metadata M41 to the operator terminal 20 in association with the AI image F41 and displays it. The metadata M41 indicates rights information that the "safety of the model" (the same information as the availability of the DS image) is "○" for the DS image used for machine learning by the image generation AI module 100 that generated the AI image F41, i.e., the DS image included in the dataset called "DSv1.3 for company C1".

According to this aspect, compared to the case where metadata such as that shown in FIG. 14 is not output, it is easier to grasp information about the group of images used in machine learning, i.e., the multiple DS images used in machine learning by the artificial intelligence module that generated the AI image. In the example of FIG. 14, it is easier to grasp rights information about the multiple DS images used in machine learning by the artificial intelligence module to generate the AI image F41.

If machine learning were to be performed using an image group that includes DS images that the AI image user does not have the right to use, i.e., an image group with incomplete rights, it could be determined that there was an intention to infringe on the rights of others, or that care was not taken to avoid infringing on the rights of others. Conversely, if machine learning were to be performed using an image group that includes only DS images that the AI image user has the right to use, i.e., an image group with no incomplete rights, it would be more likely to be found that there was no intention to infringe on the rights of others, or that care was taken to avoid infringing on the rights of others.

By indicating rights information in the metadata, users of the AI image can see whether the artificial intelligence module that generated the AI image corresponding to that metadata used machine learning from a group of images with inappropriate rights or a group of images with no inappropriate rights, allowing them to understand the copyright safety of the AI image.

The server device 10 also functions as an example of a selection receiving unit that receives the selection of images (DS images) to be included in the image group used for machine learning. In the examples of Figures 5 and 7, the server device 10 receives the selection of DS images by displaying a dataset creation screen G1. As shown in Figure 6, the selected DS images are associated with contract information that indicates the contents of a contract that specifies the right to use those DS images.

The server device 10 functions as an example of a generating unit that generates rights information for a group of images based on contract information associated with the DS images included in the group of images through the above selection. For example, in the example of FIG. 8, the server device 10 generates rights information indicating whether the DS images can be used and the percentage of DS images that have common license conditions based on contract information indicating copyright-related rights and portrait-related rights associated with the DS images. According to this aspect, it is possible to reduce the effort required to prepare rights information compared to a case in which a user manually prepares similar rights information.

The server device 10 also functions as an example of an output unit that outputs metadata so that it can be displayed on a user terminal. The operator terminal 20 is an example of a user terminal, and in the example of FIG. 14, it displays metadata M41. The server device 10 also functions as an example of an image display unit that displays, when a display operation of the image group used in machine learning is performed on the user terminal on which the metadata is displayed, images included in the image group whose related information is indicated by the metadata. According to this aspect, it is possible to make it easier to verify the actual image generated by machine learning, compared to a case in which the display control of the DS image is not performed by the above-mentioned display operation.

The server device 10 also functions as an example of a contract display unit that displays, on a user terminal on which metadata is displayed, contract information that is the source of the rights information indicated by the metadata, when a display operation for details of the rights is performed on the user terminal on which the metadata is displayed. In the example of FIG. 14, when an operation on the link L42 to the contract information included in the metadata M41 is performed as a display operation for details of the rights to use the image group used in machine learning, the server device 10 displays, on the operator terminal 20, contract information of the DS image whose rights information is indicated by the metadata M41, as shown in FIG. 16. According to this aspect, it is possible to easily verify the contract contents indicating the rights of the image, compared to a case in which the display control of the contract information by the above-mentioned display operation is not performed.

<Variation: Rights Information>
The rights information included in the metadata is not limited to the information shown in FIG. 14 (safety of the model = whether it can be used or not).
Fig. 17 is a diagram showing an example of displayed metadata. In the example of Fig. 17, the server device 10 displays metadata M42 in the metadata display field E42 shown in Fig. 14. The metadata M42 includes "copyright-related rights" and "portrait-related rights" in addition to the rights information (safety of the model) shown in Fig. 14.

Specifically, copyright-related rights include the "right to use" of "copyright = 4%" and "copyright permission = 96%", the "expiry of use" of "unlimited: 45%, up to 20xx: 35%, up to 20yy: 5%, ...", and the "purpose of use" of "web pages, advertisements, internal documents, external documents, ...". Portrait-related rights include the "right to use" of "portrait rights = 2%" and "portrait rights permission = 98%", the "expiry of use" of "up to 20xx: 30%, up to 20yy: 25%, unlimited: 20%, ...", and the "purpose of use" of "web pages, advertisements, internal documents, external documents, ...".

In this way, the rights information may include information indicating the conditions under which the right to use a DS image becomes valid. The conditions under which the right to use a DS image becomes valid include the usage period and purpose of use shown in FIG. 17, as well as the region in which the DS image can be used, the license renewal conditions, the license fee, and other conditions. By including these conditions in the metadata, it is possible to alert users of AI images so as not to invalidate their right to use the DS image.

For example, if a user of an AI image is alerted that the image contains a DS image whose usage period ends in 20xx, the user can take measures such as limiting the use of the generated AI image to 20xx or extending the usage period of the relevant DS image. In addition, the user of an AI image can be careful not to use the generated AI image for purposes other than those permitted.

The rights information may also include information indicating the amount of DS images that have the same conditions for validating the right to use the DS images. In the example of FIG. 17, the rights information includes the percentage of DS images that have the same usage period, such as "unlimited: 45%, up to 20xx: 35%, up to 20yy: 5%, ...". Note that the information indicating the amount of DS images is not limited to the percentage, and may be expressed by the number of images or data size, etc. According to this aspect, it is possible to grasp the degree of impact when it is assumed that the right to use some DS images has been invalidated.

Furthermore, when a condition for the right to use a DS image to be valid is likely to be unfulfilled, the server device 10 may display the rights information indicating that condition in a manner different from other rights information. For example, when there is a DS image with less than one year until the expiration date, the server device 10 may display the expiration date in bold red, or may add a text string such as "Less than one year left until expiration date!" to draw special attention. According to such a manner, it is possible to make it less likely that the right to use a DS image will be invalidated, compared to when only a uniform display is performed.

The rights information may also include information about the type of rights to use the DS image. The types of rights to use the DS image are, for example, "copyright-related rights" and "portrait-related rights" as shown in FIG. 17. Copyright-related rights are an example of rights related to the person who created the DS image, and portrait-related rights are an example of rights related to the person depicted in the DS image. It is sufficient that at least one of these rights is included.

The rights relating to the person who created the DS image are not limited to copyright, and may include, for example, design rights or trademark rights. Furthermore, the rights relating to the person depicted in the DS image may include publicity rights or privacy rights. When any of these rights are included in the rights information, the security of the rights can be evaluated in more detail than, for example, when the rights information only includes whether the DS image can be used.

The rights information may also include information indicating the amount of DS images that share a common type of right to use the DS images. In the example of FIG. 17, the rights information includes the percentage of DS images that share a common type of right, such as "Copyright = 4%" and "Copyright permission = 96%." Note that the information indicating the amount of DS images is not limited to a percentage, and may be expressed by the number of images or data size, etc. According to this aspect, it is possible to grasp the balance of the types of rights and, if necessary, take measures to adjust the balance (for example, measures such as lowering the percentage of permission and increasing the copyright ownership rate).

In addition, when the subject of a DS image is a person, the areas protected by the granted portrait rights may be limited. For example, if the DS image is a close-up of a face, it is highly likely that the AI image showing the close-up of the face will be protected by the granted portrait rights for the DS image, but it is highly likely that the AI image showing the whole body will not be protected by those rights. Conversely, when the DS image is a full-body photograph, it is highly likely that the AI image showing the whole body will be protected by the granted portrait rights for the DS image, but it is highly likely that the AI image showing the close-up of the face will not be protected by those rights.

The rights information may therefore include information indicating the extent to which the rights apply to the objects depicted in the images included in the image group. The "rights" referred to here are the rights to use the DS image, such as rights granted by the owner of portrait rights. In this case, information indicating the extent to which the rights apply (hereinafter referred to as "scope information") is stored in image database DB1 in association with each image. If the subject depicted in the DS image is a person, "face", "whole body", "upper body" or "profile view" etc. are stored as scope information.

When DS images to be included in the data set are selected, the server device 10 refers to the image database DB1, reads out range information associated with those DS images, and generates rights information based on the read out range information. For example, the server device 10 generates information indicating the amount of DS images that have a common range covered by the rights as rights information. Then, when the server device 10 generates an AI image, it outputs metadata including rights information based on the range information.

Figure 18 is a diagram showing an example of displayed metadata. In the example of Figure 18, the server device 10 displays metadata M43 in the metadata display field E42 shown in Figure 14. In addition to the rights information (safety of the model) shown in Figure 14, metadata M43 includes "extent of rights: face = 65%, upper body = 30%, ..." as "portrait-related rights."

In this case, the user of the AI image can determine the composition of the AI image based on the rights information indicated by the metadata. For example, in the example of FIG. 18, the range indicated by the rights information, i.e., the range of the rights granted for the DS image, is mostly limited to the part above the upper body, so when generating an AI image depicting a person, a prompt can be issued to generate an AI image with a composition that largely depicts the part above the upper body. According to this aspect, it is possible to generate an AI image with a composition that poses fewer rights risks. Furthermore, by including rights information indicating the amount of DS images with a common range of rights in the metadata, it is possible to intuitively understand compositions with fewer rights risks.

In addition to the above, the rights information may also include information on items such as the holder of the rights to use the DS image, the owner of copyright-related rights and portrait-related rights, or the usage fee. It is desirable for the metadata to include, as rights information, information on items that can be used as reference when a user of an AI image understands the rights status of the DS image that is the source of the AI image, judges rights-related risks, and takes measures to avoid those risks.

The rights information may also include multiple items, as in the examples of Figures 17 and 18. The multiple items preferably include two or more of the following: the presence or absence of rights (one example is "safety of model"), the conditions under which the rights become valid (one example is "expiration date of use"), the type of rights (one example is "copyright-related rights" and "portrait-related rights"), the scope of the rights, the type of title of the rights (one example is "title of use"), or the name of the rights. According to this aspect, the rights-related safety of the generated AI image can be evaluated from multiple perspectives.

<Variation: Selection of rights information>
The rights information included in the metadata may be selectable. In this case, the server device 10 displays a selection screen for selecting the rights information. Since the selection of the rights information item may be performed by either the manager or the user of the AI image, the selection screen may be displayed on either the manager terminal 20 or the user terminal 30.

Figure 19 is a diagram showing an example of a displayed rights information selection screen. The rights information selection screen G7 shown in Figure 19 displays a check box E71 for the rights information item to be selected, a metadata image E72, and a decision button B71. The check box E71 contains rights information items such as "whether or not there is a right", "type of right", "name of right", "expiration date of use", "area of use", "usage fee", and "renewal conditions", as well as their check boxes (where a "○" indicates that the item is checked).

Image E72 shows an image of metadata when the items checked in the check box E71 are included. In the example of FIG. 19, "Rights", "Type of rights", "Expiration date of use" and "Usage fee" are checked, and in image E72, "Rights" is represented as "Safety of model", and "Type of rights" is represented as "Copyright-related rights" and "Portrait-related rights". In addition, "Expiration date of use" and "Usage fee" are represented by the same names as the items.

The user (the operator or the user of the AI image) checks the image E72 while checking the items in the check box E71, and when the items to be included in the metadata are decided, the user operates the decision button B71. In this way, the server device 10 functions as an example of a selection display unit that displays on the user terminal a selection image that accepts a selection operation of items to be included in the metadata from among the multiple items included in the rights information. The operation of checking the check box E71 and the operation of the decision button B71 are examples of selection operations, and the selection screen G7 is an example of a selection image. Then, the server device 10 (an example of an output unit) outputs data indicating the items selected by the selection operation as metadata. According to this aspect, the user can include desired items in the metadata.

<Modification: Changing the DS image>
The DS image included in the dataset may be changed after the dataset is created. In this case, the server device 10 displays a change screen for changing the DS image instead of the dataset creation screen in A11 and A12 of the dataset creation process shown in Fig. 4. Since the DS image may be changed by either the operator or the user of the AI image, the change screen may be displayed on either the operator terminal 20 or the user terminal 30.

Figure 20 is a diagram showing an example of a displayed screen for changing a DS image. The screen G8 for changing a DS image shown in Figure 20 displays the character string "Please reselect the images to be included in the dataset.", an input field E81 for the image user, an input field E82 for the dataset name, a display field E83 for the DS image, a check box E84, an input field E85 for the image attribute conditions, an input field E86 for the image rights conditions, a display field E87 for candidate images, a check box E88, an input field E89 for the changed dataset name, and a confirm button B81.

When a user inputs an image user in input field E81, the name of the dataset associated with that image user can be input in input field E82, and when a dataset name is input in input field E82, the DS images included in that dataset are displayed in display field E83. The displayed DS images have the corresponding check boxes E84 checked, but can be removed from the dataset by unchecking them.

In the image attribute condition input field E85 and the image rights condition input field E86, conditions for narrowing down the candidate images to be added to the dataset are input. When the narrowing down conditions are input, operations A13 to A15 shown in FIG. 4 are performed, and candidate images that satisfy the narrowing down conditions input by the user are displayed in the display field E87. Of the displayed candidate images, those with the corresponding check box E88 checked are added to the dataset.

The dataset name after the DS image has been changed is entered in input field E89. When the confirm button B81 is operated in this state, operations A16 to A18 shown in FIG. 4 are performed, and new dataset information including the dataset name entered in input field E89 is generated and saved. At this time, rights information for the newly generated dataset is stored in dataset database DB2 shown in FIG. 8.

In this way, the server device 10 functions as an example of a change acceptance unit that accepts changes to the DS images to be included in the image group of the dataset, for example, based on a user operation on the change screen G8. Then, the server device 10 (an example of a generation unit) generates rights information for the image group based on the contract information associated with the DS images included in the image group of the dataset after the change has been accepted. According to this aspect, even after the dataset is changed, the effort required for preparing rights information can be reduced compared to when similar rights information is manually prepared by the user.

The server device 10 may automatically generate a changed dataset name and display it in the input field E89. For example, the server device 10 refers to the input field E82, and if the input dataset name contains a portion "v(N1).(N2)" (N1 and N2 are natural numbers), the server device 10 generates a dataset name "v(N1).(N2+1)" and displays it in the input field E89. For example, in the example of FIG. 20, the server device 10 generates a changed dataset name "DSv3.2 for company C1" based on the dataset name "DSv3.1 for company C1" and displays it in the input field E89.

Furthermore, the server device 10 may generate a dataset name (v(N1+1), (N2)) with a major version upgrade if the changes to the DS images satisfy a condition indicating a major change, and may generate a dataset name (v(N1), (N2+1)) with a minor version upgrade if the condition is not satisfied. The condition indicating a major change is, for example, a condition that is met when a predetermined percentage (e.g., 30% or 50%) or more of the DS images included in the dataset have changed. This aspect can eliminate the need to input the changed dataset name.

<Modification: Component>
An artificial intelligence module (e.g., the image generation AI module 100) may have multiple components. A component is a part that constitutes a part of an artificial intelligence module. The artificial intelligence module generates one AI image by each of the multiple components operating. Each of these multiple components performs machine learning using the same or different image groups.

When the artificial intelligence module has the above-mentioned multiple components, the server device 10 stores learning model information different from the example of FIG. 11 in the learning model database DB3.
Fig. 21 is a diagram showing another example of stored learning model information. Learning models such as "model ML1", "model ML2", and "model ML3" are stored in the learning model database DB3 shown in Fig. 21.

Each learning model is associated with three components, "CMP1", "CMP2", and "CMP3", and the names of the datasets used in the machine learning of each component. In the example of Figure 21, datasets DS1, DS2, and DS3 are used for machine learning. For example, in the case of "Model ML1", DS1 is associated with CMP1, DS1 and DS2 are associated with CMP2, and DS3 is associated with CMP3.

In the case of "Model ML2", CMP1 is associated with DS2, CMP2 is associated with DS3, and CMP3 is associated with DS2. In the case of "Model ML3", CMP1 and CMP2 are associated with DS1 and DS2, and CMP3 is associated with DS2 and DS3. Each component is also associated with image information and rights information according to the dataset used for machine learning. This image information and rights information is information that has been read out from the dataset database DB2 or recalculated, as in the example of Figure 11.

When an instruction to generate an AI image is given, the server device 10 (an example of an output unit) outputs data indicating related information acquired for the group of images used in the machine learning by each of the multiple components as metadata, in association with the AI image generated by the artificial intelligence module. For example, similar to the example of FIG. 14, the server device 10 outputs metadata indicating rights information for the group of images included in the dataset used in the machine learning as related information.

Figure 22 is a diagram showing another example of displayed metadata. In the example of Figure 22, the server device 10 displays metadata M44 in the metadata display field E42 shown in Figure 14. Metadata M44 includes a learning model name of "Model ML1", component names of "CMP1", "CMP2", and "CMP3", dataset names of "DS1", "DS1+DS2", and "DS3", and model safety ratings of "○", all of which are associated with each other.

In the example of Figure 22, the user of the image can easily understand the learning content (dataset) of each component compared to when metadata is not output, and can easily understand that each component is performing machine learning using DS images that are secure in terms of rights.

<Variation: User-provided image>
The images included in the data set are images prepared in advance for the image generation system 1, but are not limited thereto, and an image provided by a user may be added. In this case, the server device 10 displays an addition screen for adding an image provided by a user (hereinafter referred to as a "user image"). Since the addition of a user image may be performed by either the operator or the user of the AI image, the addition screen may be displayed on either the operator terminal 20 or the user terminal 30.

Figure 23 is a diagram showing an example of a displayed screen for adding a user image. The screen G9 for adding a user image shown in Figure 23 displays the text "Please enter the user image to add and its provision conditions.", an input field E91 for the file name of the user image to be added, an input field E92 for the user of the AI image, an input field E93 for the provision conditions related to copyright, an input field E94 for the provision conditions related to portrait rights, and a confirm button B91.

In input field E91, a path name may be input in addition to the file name of the user image, and the file names of two or more user images may also be input. In input field E92, a user who will use the AI image generated when the added user image is treated as a DS image is input. The provision conditions input in input fields E93 and E94 are the provision conditions for this user.

In input field E93, the conditions of provision such as "Copyright holder," "Copyright transfer," "License for use," "Expiration date of use," "Purpose of use," and "Usage fee" are entered. In "Copyright holder," the name of the copyright holder of the user image is entered. In "Copyright transfer," "○" is entered if the copyright of the user image is transferred to the image user. In "License for use," "○" is entered if use is permitted without transferring the copyright. In "Expiration date of use," "Purpose of use," and "Usage fee," the conditions for permitting use are entered.

In the input field E94, in the case where a portrait right exists for the user image, the provision conditions are entered in the same manner as in the input field E93. When each input field is filled in and the decision button B91 is operated, the server device 10 inquires of the image user whether or not they agree to the entered provision conditions. If the image user responds that they agree to the provision conditions, the server device 10 stores the user image in the image database DB1. At that time, the server device 10 stores the provision conditions entered in the addition screen G9 in the image database DB1 as contract information.

For example, the server device 10 stores the image user as the "copyright holder" in the image database DB1 if the copyright has been transferred, and stores the input copyright holder if the copyright has not been transferred. The stored information on the image user, etc. is the name of a corporation or an individual. The server device 10 stores "NA" (not applicable) as the "copyright licensee" if the copyright has been transferred, and stores the image user if the copyright has been licensed.

The server device 10 stores provision conditions such as "usage period," "purpose of use," and "usage fee" as "license conditions" for copyright. In addition, if there are provision conditions for portrait rights, the server device 10 also stores the "portrait right owner," "portrait right licensee," and "license conditions" for the portrait rights. The provision conditions stored in the image database DB1 in this way are used as contract information for the dataset creation process shown in Figure 4. As a result, the rights information stored in the dataset database DB2 during the dataset creation process will include these provision conditions.

As described above, the server device 10 functions as an example of an image acquisition unit that acquires user images provided by the user. As described above, these user images are acquired based on the provision conditions agreed to by the user and the image user. Then, when the artificial intelligence module performs machine learning using an image group including the acquired user images, the server device 10 functions as an example of an information acquisition unit that acquires information including the provision conditions of the user images as rights information for the image group. According to this aspect, the user can include images that he or she likes in the dataset as DS images, so that the tendency of the generated AI images can be tailored to the user's preferences.

<Variation: Multiple Rights Holders>
There may be multiple holders of copyrights, portrait rights, and other rights to a DS image. For example, not only the photographer who took the DS image, but also the copyright holder of a copyrighted work may hold rights to the DS image if the DS image contains such a work. Also, if multiple people are shown in a DS image in a manner that allows them to be identified, each of those people may hold portrait rights.

In such cases, the image database DB1 shown in FIG. 6 stores contract information for multiple rights holders in association with one image. This allows the server device 10 to narrow down candidate images to images for which a specific rights holder has the rights, for example by allowing the rights holder to be input as an image rights condition on the dataset creation screen G1 shown in FIG. 5. The server device 10 may also store dataset information and learning model information indicating the rights holder in the dataset database DB2 and learning model database DB3, respectively.

By doing so, the server device 10 can display information about the rights holder of the DS image (e.g., information indicating that the copyright of the DS image is shared between companies C1 and C2) in, for example, the display field E23 of the list of datasets on the machine learning screen G2 shown in FIG. 10 and the display field E32 of the list of learning models on the image generation screen G3 shown in FIG. 13, so that the information can be used as a reference when selecting a dataset and a learning model.

<Variation: Re-learning>
The image generation AI module 100 may perform re-learning by performing machine learning again using a learning model once generated and a dataset (either a new dataset or a trained dataset). In this case, the server device 10 stores the learning model used for re-learning in the learning model database DB3 as a learning source model.

Figure 24 is a diagram showing another example of stored learning model information. The learning model database DB3a shown in Figure 24 stores learning model information including the original learning model. The original learning model indicates the name of the learning model used when generating the corresponding learning model. In the example of Figure 24, the first generated "Model v1.1 for Company C1" is not a re-learning, so only the dataset "DSv1 for Company C1" is associated with it, and the original learning model is "NA" (not applicable).

The "Company C1 model v1.1" is used as the original model, and the learning model generated by re-learning using the dataset "Company C1 DSv2" is "Company C1 model v1.2." The learning model generated by re-learning using the "Company C1 model v1.2" as the original model and using the dataset "Company C1 DSv3" is "Company C1 model v1.3," and the learning model generated by re-learning using the "Company C1 model v1.2" as the original model and using the dataset "Company C1 DSv4" is "Company C1 model v2.1."

In addition, the learning model generated by re-learning using the data set "Company C1 DSv5" with "Company C1 model v2.1" as the learning source model is "Company C1 model v2.2", and the learning model generated by re-learning using the data set "Company C1 DSv4" with "Company C1 model v1.3" as the learning source model is "Company C1 model v3.1". In this way, according to the example of FIG. 24, it is possible to track the learning models that are the learning sources of the re-learned learning models over many generations. Note that while the management of learning model information including the learning source models is performed in a table format in the example of FIG. 24, it may also be performed in a node format.

Figure 25 is a diagram showing an example of learning model information managed in a node format. In the example of Figure 25, the server device 10 stores nodes Nm1 to Nm6 indicating learning models and nodes Nd1 to Nd5 indicating datasets. Nodes Nm1 to Nm6 indicate "v1.1", "v1.2", "v1.3", "v2.1", "v2.2", and "v3.1" of "Model for Company C1", respectively. Nodes Nd1 to Nd5 indicate "v1", "v2", "v3", "v4", and "v5" of "DS for Company C1", respectively. Rights information corresponding to each dataset is associated with nodes Nd1 to Nd5. Note that learning model information corresponding to each learning model may be associated with nodes Nm1 to Nm6.

The server device 10 stores connection information indicating the connections between each node. In the example of FIG. 25, the connections between each learning model due to re-learning shown in FIG. 24 are shown. For example, the connection information indicates that node Nm2 (model v1.2 for company C1) is a learning model that was generated by re-learning using node Nd2 (DSv2 for company C1) with node Nm1 (model v1.1 for company C1) as the learning source model.

The connection information indicates that node Nm2 branches into node Nm3 (model v1.3 for company C1) which has undergone a minor version upgrade, and node Nm4 (model v2.1 for company C1) which has undergone a major version upgrade, and that only node Nm5 (model v3.1 for company C1) which has undergone a major version upgrade is connected to node Nm3. The server device 10 may be able to display an image visually representing the relationship between the nodes as shown in FIG. 25 based on the information and connection information of each node. This allows the relationship between the learning model, the learning source model, and the dataset to be intuitively understood.

<Modification: Related information>
The related information related to the image group used for machine learning is not limited to rights information (information indicating the right to use the DS image). For example, the related information may be version information of the dataset or image information related to the DS image. The image information of the DS image includes the number of DS images included in the dataset, the resolution of the DS image, or the subject of the DS image. The image information such as the resolution of the DS image and the subject of the DS image may be information indicating the amount of DS images that have common image information.

<Modification: DS image viewing function>
15 has been described as an example of a function for viewing a DS image from metadata, but the following aspect may also be used: For example, an AI image and metadata are viewed by an image user, and the image user may include those who have the right to view, edit, or access the DS image (hereinafter referred to as "viewing right, etc.") and those who do not.

In that case, the server device 10 may output metadata including a link L41 to the DS image shown in FIG. 15 to a user terminal used by a person who has the authority to view the DS image, and output metadata that does not include a link L41 to the DS image to a user terminal used by a person who does not have the authority to view the DS image. According to this aspect, it is possible to prevent unauthorized persons from viewing the DS image.

In addition, the DS image screen G5 shown in FIG. 15 may display a huge number of DS images. In such a case, simply displaying a list of DS images makes it difficult to verify the DS images. Therefore, the server device 10 may calculate, for example, feature amounts used in image recognition (HOG (Histograms of Oriented Gradients) or SIFT (Scaled Invariance Feature Transform), etc.) for the DS images included in the dataset, and preferentially display representative images extracted from similar images that fall within a common range of calculated feature amounts. According to this aspect, DS images with different features can be compared in a short time compared to a case where DS images are displayed regardless of the image features.

<Modification: Contract information viewing function>
In the example of Fig. 16, an example of a function for viewing contract information from metadata has been described, but the following aspect may also be used. As with the DS image described above, the server device 10 may output metadata including a link L42 to the contract information shown in Fig. 16 to a user terminal used by a person who has the authority to view the contract information, and output metadata that does not include a link L42 to the contract information to a user terminal used by a person who does not have the authority to view the contract information. According to this aspect, it is possible to prevent unauthorized persons from viewing the contract information.

<Examples of variations: composition variations>
The configuration (overall configuration, hardware configuration, functional configuration, etc.) shown in FIG. 1 and the like is an example, and other configurations may be used as long as there is no inconvenience in implementation. For example, the server device 10 may be distributed across two or more devices, and may be provided in the form of SaaS (Software as a Service) or a cloud computing system. Furthermore, the information processing performed by the server device 10 may be collectively executed by a user terminal such as the operator terminal 20. In short, as long as the necessary information processing is executed in the entire image generation system 1, the devices that execute the information processing may have any configuration.

The correspondence between information may be performed in various ways. For example, the correspondence may be performed by storing each piece of information in a corresponding area in a database or table, or by creating a table or the like that shows the correspondence between each piece of information. The correspondence may also be performed by adding information that identifies one piece of information to the other piece of information. The correspondence may also be performed in various other well-known ways. The output destination of information or data (hereinafter referred to as "information, etc.") may be another device, a display, a memory unit (including built-in memory units and external memory units), etc. Acquisition of information, etc. includes a mode of acquiring information, etc. transmitted from another device, as well as a mode of acquiring information, etc. generated by one's own device.

The above-described aspects of the embodiment are information processing devices such as the server device 10, the operator terminal 20, and the user terminal 30, and information processing systems such as the image generation system 1 that include these information processing devices, but may also be information processing methods. The information processing method includes the same steps as those executed by the information processing system. The above-described aspects of the embodiment may also be programs. The program causes a computer to execute the same steps as those executed by the information processing system.

<Additional Notes>
Furthermore, it may be provided in the following aspects:

(1) An information processing system including a processor, in which, in an information acquisition step, the processor acquires related information related to a group of images used in machine learning, and in an output step, outputs metadata indicating the related information acquired for the group of images in association with an image generated by an artificial intelligence module that has performed machine learning using the group of images.

This aspect makes it easy to understand information about the group of images used in machine learning.

(2) In the information processing system described in (1) above, the related information includes rights information about the image group, and the rights information is information indicating the right to use the images included in the image group.

According to this embodiment, it is possible to ascertain the copyright safety of the generated image.

(3) In the information processing system described in (2) above, the rights information includes information indicating the conditions under which the rights are valid.

In this manner, it is possible to draw attention to the expiration of rights.

(4) In the information processing system described in (2) or (3) above, the rights information includes information about the type of the rights, and the type of the rights includes at least one of the rights related to the person who created the image and the rights related to the person depicted in the image.

This embodiment allows the security of rights to be evaluated in detail.

(5) An information processing system according to any one of (2) to (4) above, wherein the rights information includes information indicating the scope of the rights among objects depicted in images included in the image group.

This aspect makes it possible to generate images with compositions that pose fewer copyright risks.

(6) An information processing system according to any one of (2) to (5) above, wherein the rights information includes a plurality of items, and the plurality of items include two or more of the following: the existence or nonexistence of the right, the conditions under which the right is valid, the type of the right, the scope of the right, the type of title of the right, or the name of the right.

This type of approach allows the security of rights to be evaluated from multiple perspectives.

(7) In the information processing system described in any one of (2) to (6) above, the rights information includes a plurality of items, and in the selection and display step, the processor displays on the user terminal a selection image that accepts a selection operation of an item from the plurality of items to be included in the metadata, and in the output step, outputs data indicating the item selected by the selection operation as the metadata.

According to this embodiment, desired items can be included in the metadata.

(8) In the information processing system described in any one of (2) to (7) above, the processor, in a selection receiving step, receives a selection of images to be included in the image group, and the images are associated with contract information representing the contents of a contract that specifies the right to use the images, and in a generation step, generates the rights information for the image group based on the contract information associated with the images included in the image group by the selection.

This embodiment can reduce the effort required to prepare rights information.

(9) In the information processing system described in (8) above, the processor, in the change acceptance step, accepts changes to the images to be included in the image group, and in the generation step, generates the rights information for the image group based on the contract information associated with the images included in the image group after the changes have been accepted.

This aspect reduces the effort required to prepare rights information even after the data set is changed.

(10) In the information processing system described in (8) or (9) above, in the output step, the processor outputs the metadata to a user terminal so that the metadata can be displayed, and in the contract display step, when an operation to display details of the rights is performed on the user terminal on which the metadata is displayed, the contract information on which the rights information indicated by the metadata is based is displayed on the user terminal.

This aspect makes it easier to verify the contract details indicating the rights to the image.

(11) In the information processing system described in any one of (1) to (10) above, in the output step, the processor outputs the metadata to a user terminal so that the metadata can be displayed, and in the image display step, when a display operation of the image group is performed on the user terminal on which the metadata is displayed, the processor causes the user terminal to display an image included in the image group for which the related information is indicated by the metadata.

This aspect makes it easier to verify the actual images that have been learned.

(12) In the information processing system described in any one of (1) to (11) above, the artificial intelligence module has a plurality of components, each of which performs machine learning using the same or different groups of images, the artificial intelligence module generates one image by each of the plurality of components operating, and in the output step, the processor outputs, as the metadata, data indicating the related information obtained for the groups of images used in the machine learning by each of the plurality of components, in association with the image generated by the artificial intelligence module.

This aspect makes it easy to understand the learning content of each component.

(13) In the information processing system described in any one of (2) to (10) above, in the image acquisition step, the processor acquires a user image provided by a user, and the user image is acquired based on the provision conditions agreed to by the user, and in the information acquisition step, when the artificial intelligence module performs machine learning using an image group including the acquired user image, the processor acquires information including the provision conditions of the user image as the rights information for the image group.

This aspect allows the generated image to be tailored to the user's preferences.

(14) A program that causes a computer to execute each step executed by the information processing system described in any one of (1) to (13) above.

This aspect makes it easy to understand information about the group of images used in machine learning.

(15) An information processing method, in which a processor included in an information processing system executes each step executed by the information processing system described in any one of (1) to (13) above.

According to this aspect, it is possible to easily grasp information about the group of images used in machine learning.
Of course, this is not the case.
Furthermore, the above-described embodiments and modifications may be combined in any desired manner.

Finally, although various embodiments of the present invention have been described, these are presented as examples and are not intended to limit the scope of the invention. New embodiments can be implemented in various other forms, and various omissions, substitutions, and modifications can be made without departing from the spirit of the invention. The embodiments and their modifications are within the scope and spirit of the invention, and are also within the scope of the invention and its equivalents as set forth in the claims.

1: Image generation system 2: Communication line 10: Server device 11: Control unit 20: Operator terminal 21: Control unit 30: User terminal 31: Control unit 100: Image generation AI module DB1: Image database DB2: Data set database DB3: Learning model database

Claims (11)

An information processing system including a processor,
The processor,
In the information acquisition step, related information related to an image group used as training data when generating a learning model by machine learning and connection information indicating a connection between multiple versions of the learning model generated by re-learning the learning model are acquired,
In the first output step, metadata indicating the related information acquired for the image group is output in association with an image generated by an artificial intelligence module using a learning model generated as a result of machine learning using the image group as training data ;
The output metadata indicates, as the related information, version information of a learning model on which machine learning was performed using the images indicated by the metadata,
In a second output step, based on the acquired connection information, information indicating an old version of a learning source model that was used as a basis for re-learning the learning model whose version information is indicated in the metadata and a group of images that were used as training data by the learning source model are output.
Information processing system. An information processing system including a processor,
The processor,
In the information acquisition step, related information related to the image group to be used as training data when generating a learning model by machine learning is acquired,
In the output step, metadata indicating the related information acquired for the image group is output in association with an image generated by an artificial intelligence module using a learning model generated as a result of machine learning using the image group as training data;
In the display control step, a screen for inputting a user of the image generated by the artificial intelligence module is displayed;
In the information acquisition step, right information indicating a right of the input user to use the image group is acquired as the related information.
Information processing system. An information processing system including a processor,
The processor,
In the information acquisition step, related information related to the image group to be used as training data when generating a learning model by machine learning is acquired,
In the output step, metadata indicating the related information acquired for the image group is output in association with an image generated by an artificial intelligence module using a learning model generated as a result of machine learning using the image group as training data;
The learning model is associated with learning model information including information related to a group of images used by the learning model for machine learning;
In the information acquisition step, the related information is acquired based on the learning model information associated with the learning model used to generate the image;
In the information generation step, the artificial intelligence module generates learning model information including rights information indicating a right to use the image group used in the machine learning in association with the learning model generated by the machine learning;
In the display control step, a screen for inputting a user of the image generated by the artificial intelligence module is displayed;
In the information generating step, information including right information indicating the right of the input user to use the image group is generated as the learning model information.
Information processing system. An information processing system including a processor,
The processor,
In the information acquisition step, related information related to the image group to be used as training data when generating a learning model by machine learning is acquired,
the related information includes rights information for the group of images;
the rights information is information indicating a right to use an image included in the image group,
In the output step, metadata indicating the related information acquired for the image group is output in association with an image generated by an artificial intelligence module using a learning model generated as a result of machine learning using the image group as training data;
the rights information includes information indicating a portion of a subject appearing in an image included in the image group or information indicating a direction of the subject;
Information processing system. An information processing system including a processor,
The processor,
In the information acquisition step, related information related to the image group to be used as training data when generating a learning model by machine learning is acquired,
In the output step, metadata indicating the related information acquired for the image group is output in association with an image generated by an artificial intelligence module using a learning model generated as a result of machine learning using the image group as training data, and the metadata is further output so as to be displayable on a user terminal;
In the image display step, when a display operation of the image group is performed on the user terminal on which the metadata is displayed, an image included in the image group for which the related information is indicated by the metadata is displayed on the user terminal, and further, an attribute condition satisfied by the image or a condition of a right to use the image is displayed on the user terminal together with the image included in the image group for which the related information is indicated by the metadata.
Information processing system. An information processing system including a processor,
The processor,
In the information acquisition step, related information related to the image group to be used as training data when generating a learning model by machine learning is acquired,
the related information includes rights information for the group of images;
the rights information is information indicating a right to use an image included in the image group,
In the output step, metadata indicating the related information acquired for the image group is output in association with an image generated by an artificial intelligence module using a learning model generated as a result of machine learning using the image group as training data;
In the image acquisition step, a user image provided by a user is acquired, the user being a user of the image generated by the artificial intelligence module;
The user image is acquired based on a provision condition agreed to by the user,
In the information acquisition step, when the artificial intelligence module performs machine learning using an image group including the acquired user image as training data, information including the provision conditions of the user image is acquired as the right information for the image group.
Information processing system. An information processing system including a processor,
The processor,
In the information acquisition step, related information related to the image group to be used as training data when generating a learning model by machine learning is acquired,
The machine learning is performed by an artificial intelligence module having a function of generating images,
The related information is rights information indicating the rights of a user of the images generated by the artificial intelligence module to use the group of images,
In the output step, metadata indicating the related information acquired for the image group is output in association with an image generated by an artificial intelligence module using a learning model generated as a result of machine learning using the image group as training data.
Information processing system. An information processing system including a processor,
The processor,
In the information acquisition step, related information related to the image group to be used as training data when generating a learning model by machine learning is acquired,
Each image included in the image group has an attribute represented in a plurality of patterns;
the association information includes information indicating the amount of images that have a common pattern of the attribute among the images included in the image group;
In the output step, metadata indicating the related information acquired for the image group is output in association with an image generated by an artificial intelligence module using a learning model generated as a result of machine learning using the image group as training data.
Information processing system. An information processing system including a processor,
The processor,
In the information acquisition step, related information related to the image group to be used as training data when generating a learning model by machine learning is acquired,
The learning model includes information related to a group of images used by the learning model for machine learning.
The learning model information is associated with
In the output step, metadata indicating the related information acquired for the image group is output in association with an image generated by an artificial intelligence module using a learning model generated as a result of machine learning using the image group as training data ;
In the information generation step, the artificial intelligence module generates learning model information including rights information indicating a right to use the image group used in the machine learning in association with the learning model generated by the machine learning;
said right being a right of a user of an image generated by said artificial intelligence module,
In the information acquisition step, the related information is acquired based on the learning model information associated with the learning model used to generate the image.
Information processing system. A program causing a computer to execute each step executed by the information processing system according to any one of claims 1 to 9 . 1. An information processing method, comprising:
A processor included in the information processing system
Executing each step executed by the information processing system according to any one of claims 1 to 9 .
Information processing methods.