Information processing device, method, program and system

Abstract

To provide a program for supporting preparation of visual content suitable as a record of a character's activities.SOLUTION: A program causes a computer to function as means of: providing a first model input based on profile information that is set for a specific character, to a first model having learned relations between profile information of a plurality of reference characters and features of visual contents disclosed to an outside respectively by the plurality of reference characters to acquire feature information representing a feature of a visual content predicted to be suitable for the specific character; providing a second model input based on the feature information to a second model having learned to generate an image with an appearance of the specific character used as a subject, so as to acquire a candidate content being a visual content with the appearance of the specific character used as the subject and having a feature corresponding to the second model input; and outputting the candidate contents.SELECTED DRAWING: Figure 5

Description

[Technical field]

[0001] The present disclosure relates to an information processing device, a method, a program, and a system. [Background technology]

[0002] Conventionally, there is known an agent dialogue system that dialogues with a user by text or voice. Patent Document 1 discloses a technical idea that a character with a personality automatically replies with a text according to the personality. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] Patent Publication No. 2021-056668 Summary of the Invention [Problem to be solved by the invention]

[0004] The technical idea of ​​Patent Document 1 has the potential to increase the sense of familiarity with the character and the reality of its existence by automatically responding with text that corresponds to the character's personality.

[0005] On the other hand, for example, celebrities and influencers in the real world post photos and videos of their activities during their off-time or on-time on SNS (Social Networking Service) to attract the interest of recipients. Similarly, for fictional characters, visual content such as photos and videos can be used as a tool to increase the affinity that recipients have with the character or the reality of the character's existence. However, it is difficult to say that visual content as a record of a character's activities is sufficient if the character is simply used as a subject, and ideally the background (location of the photo), activity content, clothing, etc. should be appropriate from the perspective of the individual's personality and trends at the time of posting. For this reason, it is not easy to frequently and timely prepare visual content suitable as a record of a character's activities. Even if the technical idea of ​​Patent Document 1 is taken into consideration, visual content suitable as a record of a character's activities cannot be automatically generated.

[0006] The objective of the present disclosure is to assist in the preparation of visual content suitable as a record of a character's activities. [Means for solving the problem]

[0007] A program of one embodiment of the present disclosure causes a computer to function as: a means for acquiring feature information representing the features of visual content predicted to be suitable for a specific character by providing a first model input based on profile information set for a specific character to a first model that has learned the relationship between profile information of multiple reference characters and the features of visual content that each of the multiple reference characters has made public; a means for acquiring candidate content which is visual content that has the appearance of a specific character as its subject and has features corresponding to the second model input by providing a second model input based on the feature information to a second model that has learned to generate images having the appearance of a specific character as its subject; and a means for outputting the candidate content. [Brief description of the drawings]

[0008] [Figure 1] 1 is a block diagram showing a configuration of an information processing system according to an embodiment of the present invention; [Diagram 2] FIG. 2 is a block diagram showing a configuration of a client device according to the present embodiment. [Diagram 3] FIG. 2 is a block diagram showing a configuration of a server according to the present embodiment. [Figure 4] FIG. 2 is an explanatory diagram of an overview of learning of a first model according to the present embodiment. [Diagram 5] FIG. 1 is an explanatory diagram of one aspect of the present embodiment. [Figure 6] 4 is a diagram showing a data structure of a profile table according to the present embodiment. FIG. [Figure 7] 4 is a diagram illustrating a data structure of a post log table according to the embodiment. FIG. [Figure 8] 11 is a flowchart of a posting log collection process according to the present embodiment. [Figure 9] 11 is a flowchart of a visual content acquisition process according to the present embodiment. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0009] Hereinafter, an embodiment of the present invention will be described in detail with reference to the drawings. In the drawings for explaining the embodiment, the same components are generally designated by the same reference numerals, and the repeated description will be omitted.

[0010] (1) Information Processing System Configuration The configuration of the information processing system will be described below. Fig. 1 is a block diagram showing the configuration of the information processing system according to the present embodiment.

[0011] As shown in FIG. 1, the information processing system 1 includes a client device 10 and a server 30. The client device 10 and the server 30 are connected via a network (eg, the Internet or an intranet) NW. The server 30 and the external system 50 are connected via a network NW.

[0012] The client device 10 is an example of an information processing device that transmits a request to the server 30. The client device 10 is, for example, a smartphone, a tablet terminal, or a personal computer. The user of the client device 10 is, for example, a person who has the authority to decide whether or not to permit the publication of visual content related to a specific character (hereinafter, referred to as an "administrative user"). The administrative user may be, for example, a person who holds a non-fungible token (NFT) that certifies the administrative authority of a specific character, and in this case, the administrative user changes with the transfer of the NFT.

[0013] The server 30 is an example of an information processing device that provides the client device 10 with a response in response to a request transmitted from the client device 10. The server 30 is, for example, a server computer.

[0014] (1-1) Client device configuration The configuration of the client device will now be described with reference to Fig. 2, which is a block diagram showing the configuration of the client device of this embodiment.

[0015] 2, the client device 10 includes a storage device 11, a processor 12, an input / output interface 13, and a communication interface 14. The client device 10 is connected to a display 21.

[0016] The storage device 11 is configured to store programs and data. The storage device 11 is, for example, a combination of a Read Only Memory (ROM), a Random Access Memory (RAM), and a storage (for example, a flash memory or a hard disk).

[0017] The programs include, for example, the following programs: ·OS (Operating System) programs · Programs for applications that process information (e.g. web browsers)

[0018] The data includes, for example, the following data: Databases referenced in information processing Data obtained by performing information processing (i.e., the results of performing information processing)

[0019] The processor 12 is a computer that realizes the functions of the client device 10 by running a program stored in the storage device 11. The processor 12 is, for example, at least one of the following: ·CPU(Central Processing Unit) ·GPU(Graphic Processing Unit) ·ASIC(Application Specific Integrated Circuit) ·FPGA(Field Programmable Gate Array)

[0020] The input / output interface 13 is configured to obtain information (eg, a user's instruction) from an input device connected to the client device 10, and to output information (eg, an image) to an output device connected to the client device 10.

[0021] The input device is, for example, a keyboard, a pointing device, a touch panel, or a combination thereof. The output device is, for example, a display 21, a speaker, or a combination thereof.

[0022] The communication interface 14 is configured to control communications between the client device 10 and an external device (eg, a server 30).

[0023] The display 21 is configured to display an image (a still image or a moving image). The display 21 is, for example, a liquid crystal display or an organic EL display.

[0024] (1-2) Server configuration The configuration of the server will now be described with reference to Fig. 3, which is a block diagram showing the configuration of the server according to this embodiment.

[0025] As shown in FIG. 3, the server 30 includes a storage device 31, a processor 32, an input / output interface 33, and a communication interface .

[0026] The storage device 31 is configured to store programs and data, and is, for example, a combination of a ROM, a RAM, and a storage (for example, a flash memory or a hard disk).

[0027] The programs include, for example, the following programs: -OS programs Application programs that perform information processing

[0028] The data includes, for example, the following data: Databases referenced in information processing Results of information processing

[0029] The processor 32 is a computer that realizes the functions of the server 30 by starting a program stored in the storage device 31. The processor 32 is, for example, at least one of the following. ·CPU GPU ·ASIC FPGA

[0030] The input / output interface 33 is configured to obtain information (eg, a user's instruction) from an input device connected to the server 30, and to output information (eg, an image) to an output device connected to the server 30.

[0031] The input device is, for example, a keyboard, a pointing device, a touch panel, or a combination thereof. The output device is, for example, a display.

[0032] The communication interface 34 is configured to control communications between the server 30 and an external device (eg, the client device 10, or an external system 50).

[0033] (2) One aspect of the embodiment An embodiment of the present invention will be described below. Fig. 4 is an explanatory diagram of an overview of learning of the first model of the present invention. Fig. 5 is an explanatory diagram of an embodiment of the present invention.

[0034] As shown in FIG. 4, a poster posts visual content such as a photo or video with the appearance of a reference character P1 as a subject to an external system 50 (hereinafter referred to as "visual content of the reference character P1"). The reference character P1 may be the same person as the poster, or may be a fictional character played by the poster. Typically, the reference character P1 is a real-life influencer, celebrity, talent, model, idol, actor, singer, dancer, athlete, etc., but may also be a fictional character such as an avatar, a character in a story, a local character, etc. In addition, the number of reference characters P1 may be any number of 2 or more, but it is preferable to ensure a scale sufficient for learning a first model LM31 described later.

[0035] The external system 50 provides users with services including sharing of visual content. The services provided by the external system are, for example, but not limited to, SNS such as Instagram (registered trademark). Furthermore, sharing of visual content may be an ancillary service rather than a main service of the external system. The external system 50 accumulates the posted visual content of the reference character P1 and makes it available to users.

[0036] The server 30 acquires profile information of the reference character P1. The profile information of the reference character P1 may be acquired from an external system 50, or may be created based on information available about the reference character P1. The profile information of the reference character P1 may be created by a human (e.g., an administrator of the information processing system 1), or may be created by, for example, summarizing and shaping information available about the reference character P1 into a trained model.

[0037] The server 30 also retrieves the visual content of the reference character P1 (which may include metadata of the visual content) from an external system 50.

[0038] The server 30 uses the acquired information as learning data to create a first model LM31 that learns the relationship between the profile information of the reference character P1 and the features of the visual content of the reference character P1. As a result of the learning, the first model LM31 can infer, for given profile information, the features of the visual content that are appropriate for that profile information.

[0039] 5, the server 30 acquires profile information of a specific character C2. The profile information of the specific character C2 may be created by a human (e.g., a management user M3 of the specific character C2), or may be created by having a learned model summarize and format information designated by a human (e.g., a management user M3 of the specific character C2, a fan of the character C2, or a combination thereof) about the specific character C2.

[0040] The server 30 provides a first model input based on profile information of a particular character C2 to a first model LM31 to obtain feature information representing the characteristics of visual content predicted to be suitable for the character C2 (e.g., suitable for the character C2 to publish).

[0041] Furthermore, the server 30 acquires visual content using the second model LM32. Here, the second model LM32 has been trained in advance to generate an image having the appearance of a specific character C2 as a subject. The server 30 provides the second model input based on the acquired feature information to the second model LM32 to acquire visual content (hereinafter referred to as "candidate content") having the appearance of the specific character C2 as a subject and having features according to the feature information. Note that the second model LM32 may generate multiple candidate contents at the same time, not limited to one.

[0042] The server 30 outputs (transmits) the acquired candidate content to the client device 10 operated by an administrator user M3 of the specific character C2. The administrator user M3 checks whether the candidate content is suitable as visual content for the specific character C2, and instructs the client device 10 to approve or reject the publication. The client device 10 transmits the approval or rejection of the publication of the candidate content to the server 30. The server 30 publishes the candidate content that has been approved for publication to the outside.

[0043] The disclosure of visual content to the outside refers to making the visual content available to a third party (a person different from the administrative user M3, typically a fan of a specific character C2) to view the whole or part of the visual content, or a processed version of either, with or without conditions. The disclosure of visual content to the outside may be any of the following: Posting visual content to SNS (Social Networking Service) or content sharing services, etc. Submitting visual content to a service that sells (which may include an auction) visual content (which may include NFTs of visual content);

[0044] (3) Database The tables constituting the database of this embodiment will be described below. The following tables are stored in the storage device 31.

[0045] (3-1) Profile Table The profile table of this embodiment will now be described with reference to Fig. 6, which is a diagram showing the data structure of the profile table of this embodiment.

[0046] The profile table stores profile information, which is information related to the profile of a reference character.

[0047] 6, the profile table includes a "character ID" field, a "gender" field, an "occupation" field, an "age" field, and a "number of followers" field. Each field is associated with the others.

[0048] The "character ID" field stores a character ID. The character ID is information for identifying a reference character corresponding to a corresponding record. As an example, the character ID may be an account name of the reference character in a service provided by the external system 50.

[0049] The "gender" field stores gender information. The gender information is information about the gender of the reference character corresponding to the record.

[0050] The "occupation" field stores occupation information. The occupation information is information about the occupation of the reference character corresponding to the record.

[0051] The "age" field stores age information. The age information is information about the age of the reference character corresponding to the record.

[0052] The "Number of Followers" field stores information about the number of followers. The information about the number of followers is, for example, information about the number of followers of the account of the reference character corresponding to the record in a service provided by an external system.

[0053] The profile information is not limited to the above examples. The profile information may not include all or part of the above examples, and may include the following information. -Regional information about the residential area of ​​the reference character corresponding to the record Appearance information about the characteristics of the reference character corresponding to the record (e.g., race, body type, etc.) -Preference information regarding the clothing or makeup preferences of the reference character corresponding to the record Personality information about the personality of the reference character corresponding to the record Life rhythm information about the reference character's life rhythm (e.g., the time of morning preparation, bathing, eating, or working) corresponding to the record

[0054] In addition to the reference characters, profile information for characters for which candidate content may be created (ie, potential subjects of candidate content) may also be managed using the same or a similar table.

[0055] (3-2) Post log table The post log table of this embodiment will be described below with reference to Fig. 7, which is a diagram showing the data structure of the post log table of this embodiment.

[0056] The post log table stores post log information. The post log information is information related to a log of visual content of a reference character that has been posted to a service provided by the external system 50.

[0057] 7, the post log table includes a "posting date and time" field, a "character ID" field, and a "visual feature" field. Each field is associated with the others.

[0058] The "posted date and time" field stores posted date and time information. The posted date and time information is information about the posted date and time of the visual content corresponding to the record.

[0059] The "Character ID" field stores a character ID. The character ID is information that identifies the reference character that is the subject of the visual content corresponding to the record. The character ID is the same as or has a one-to-one relationship with the character ID in the profile table (Figure 6).

[0060] The "visual feature" field stores visual feature information. The visual feature information is information that represents the visual features of the visual content corresponding to the record. The visual feature information is typically text information. The server 30 can acquire such visual feature information by using a trained model that outputs text that represents the visual features of the visual content in response to a model input based on the visual content. However, the visual feature information may be the visual content itself or information indicating its location (e.g., a URL).

[0061] (4) Information processing The information processing of this embodiment will be described.

[0062] (4-1) Post log collection process The post log collection process of this embodiment will be described with reference to the flowchart of FIG 8.

[0063] The post log collection process can be started, for example, when the following start conditions are met. An instruction to start the post log collection process is given to the server 30 by, for example, the administrator of the information processing system 1. - A specified period of time has passed since the last post log collection process was performed. The appointed time has arrived. The profile table (Figure 6) was updated (which could include inserting or deleting records). -Tables that manage the profile information of characters for which candidate content can be created have been updated (which may include inserting or deleting records).

[0064] As shown in FIG. 8, the server 30 acquires the posted content (S130). Specifically, the server 30 acquires visual content posted to the external system 50. Here, the acquired visual content may be limited to that posted during a predetermined period of time in the past. The predetermined period of time may be, for example, the period from the present time to the time when the previous posting log collection process was performed.

[0065] As an example, the server 30 may obtain the visual content using an API (Application Programming Interface) provided by the external system 50 or by scraping.

[0066] Server 30 may further obtain metadata for each visual content. The metadata may include at least one of the following: The time at which visual content was posted (which may include the posting date and time) The account name of the person posting the visual content · Location of visual content The response to the publication of visual content (e.g. number of views or reactions from viewers) Characters tagged in visual content

[0067] After step S130, the server 30 executes the identification of the reference character (S131). Specifically, server 30 identifies a reference character corresponding to each visual content acquired in step S130 (i.e., a reference character that is a subject of each visual content). As an example, server 30 identifies a reference character corresponding to the visual content based on the account name of a poster of the visual content. This associates each visual content acquired in step S130 with profile information of the corresponding reference character.

[0068] After step S131, the server 30 updates the posting log table (S132). Specifically, the server 30 saves the posting log information corresponding to each visual content acquired in step S130 in a posting log table (FIG. 7). As an example, the server 30 acquires text expressing visual features of the visual content by having an existing image generation AI (Artificial Intelligence) service read the visual content. This text corresponds to a prompt (command sentence) for causing the image generation AI to generate visual content identical or similar to the visual content that is the source of the text. As the image generation AI service, for example, Midjourney or Stable Diffusion can be used. The server 30 may further convert the acquired prompt into a format suitable for the second model. For example, a large-scale language model such as ChatGPT can be used for the conversion.

[0069] The server 30 saves in the posting log table (FIG. 7) a record that associates the posting date and time information acquired in step S130, the visual feature information of the visual content acquired in step S130 (or the visual content itself), and the character ID of the reference character identified in step S131.

[0070] After step S132, the server 30 executes learning of the first model (S133). Specifically, the server 30 performs machine learning of the first model using learning data based on the update contents of the posting log table in step S132 (i.e., added posting log information) and profile information of the reference character corresponding to each posting log information. For example, the learning data includes explanatory variables based on the profile information of the reference character and answer data based on features of the corresponding visual content (e.g., posting date and time information, visual feature information, or a combination thereof). As an example, the server 30 may create or update the first model by performing additional learning (e.g., fine tuning or transfer learning) of an existing trained model such as ChatGPT.

[0071] The learning of the first model (S133) may be skipped if it is not necessary. That is, the learning of the first model (S133) may be repeatedly executed at intervals longer than the posting log collection process. Alternatively, the learning of the first model (S133) may be executed at a timing independent of the posting log collection process (for example, periodically or in response to a request from the administrator of the information processing system 1).

[0072] (4-2) Visual content acquisition processing The visual content acquisition process of this embodiment will be described with reference to the flowchart of FIG 9.

[0073] The visual content acquisition process may be started, for example, when the following start conditions are met. An instruction to start the visual content acquisition process is given to the server 30 by, for example, the administrator of the information processing system 1 or an administrative user of a particular character. A certain period of time has passed since the last visual content acquisition process was performed. The appointed time has arrived. The profile table (Figure 6) was updated (which could include inserting or deleting records). -Tables that manage the profile information of characters for which candidate content can be created have been updated (which may include inserting or deleting records). New post log information has been stored in the post log table (Figure 7).

[0074] As shown in FIG. 9, the server 30 executes acquisition of profile information (S230). Specifically, the server 30 acquires profile information of characters that are to be the targets of acquiring visual content (hereinafter, referred to as "target characters"). The target characters may be all or a part of the characters that can be the targets of creating candidate content.

[0075] After step S230, the server 30 performs obtaining (S231) the characteristics of the desired visual content. Specifically, the server 30 obtains feature information representing features of visual content predicted to be suitable for the target character (i.e., desired visual content) by providing the first model with a first model input based on the profile information of each target character obtained in step S230. The feature information may be data in text format. As an example, the server 30 may list the profile information of the target character and use as the first model input a prompt requesting output of features of visual content suitable for the target character (likely to be posted). The feature information may include, for example, a prompt for causing the image generation AI to generate an image and information on a time or date suitable for publishing the generated image.

[0076] After step S231, the server 30 executes generation of model inputs (S232). Specifically, the server 30 generates a second model input for each target character based on the feature information of the target character acquired in step S231. As an example, the server 30 generates the second model input by removing information on features that the server 30 does not want to be reflected in the candidate content from the feature information acquired in step S231 (e.g., deleting it from the prompt). The features that the server 30 does not want to be reflected in the candidate content may include, for example, at least one of the following: -Static features of appearance - Quasi-static characteristics of appearance (when the first condition is met)

[0077] Here, static features of appearance refer to features that do not change over a long period of time (e.g., one month or more), such as the size, shape, or color of a part or the whole of the body (e.g., breast size, eye, skin or hair color, hair length, height). Semi-static features of appearance refer to features that may change in a short cycle (e.g., about one day), such as makeup, clothing, or hairstyle, but do not usually change within the same cycle. On the other hand, dynamic features of appearance include, for example, facial expressions, posture, and gestures.

[0078] The first condition may include, as a necessary condition, that the scheduled publication date and time of the candidate content belongs to a time period from after the target character gets ready in the morning to before taking a bath (hereinafter referred to as an "active time period"). Here, the scheduled publication date and time is, for example, a time or date that matches a time or date included in the characteristic information. The target character's morning preparation time and bath time may be included in the profile information. That is, for example, when the scheduled publication date and time is before the target character's morning preparation time (and after the wake-up time) or after the bath time (and before bedtime), the server 30 may not remove quasi-static features such as "nightwear" and "no makeup" from the characteristic information. The first condition may further include, as a necessary condition, that the candidate content is a visual content that is published second or later in the active time period of the day to which the scheduled publication date and time belongs. That is, when the candidate content is a visual content that is published first in the active time period of the day to which the scheduled publication date and time belongs, the server 30 may not remove quasi-static features from the characteristic information. This makes it possible to change quasi-static features from day to day, while suppressing fluctuations in quasi-static features such as clothing and makeup during active periods on the same day.

[0079] After step S232, the server 30 executes acquisition of candidate contents (S233). Specifically, the server 30 obtains candidate content for each target character by providing the second model input for each target character generated in step S232 to the second model. The second model is created in advance by additional learning for an image generation AI (e.g., Stable Diffusion) so that the image is generated with the target character's appearance as the subject. For example, the following machine learning methods can be used for the additional learning, but are not limited to these. Dreambooth Textual Inversion LoRA LoCon ·LoHA HyperNetworks DreamArtist Aesthetic Gradients

[0080] By providing a second model input, the second model generates and outputs one or more visual contents having the appearance of the target character as a subject and features according to the second model input. The server 30 acquires the output visual contents as candidate contents.

[0081] After step S233, the server 30 executes output of the candidate contents (S234). Specifically, the server 30 outputs (transmits) the candidate content for each target character generated in step S233 to the client device 10 operated by the administrator user of the target character. The client device 10 displays the received candidate content on the display 21 and asks for approval from the administrator user. The administrator user checks whether the candidate content is of sufficient quality and whether the candidate content is appropriate from the perspective of the target character's branding and fan base, and instructs the client device 10 to approve if it is determined that it can be made public, or instructs the client device 10 to reject if it is determined that it cannot be made public. The client device 10 transmits the result of approval or rejection of each candidate content to the server 30.

[0082] If the visual content output in step S234 is approved by the administrative user, the server 30 executes the publication of the candidate content (S235). Specifically, the server 30 publishes the candidate content to the outside as visual content of the target character. As an example, the server 30 puts the candidate content up for sale on a service that sells visual content or its NFT. As another example, the server 30 shares the candidate content on a membership-based or fee-based visual content sharing service. As yet another example, the server 30 posts the candidate content on a social networking site.

[0083] On the other hand, if the visual content output in step S234 is rejected by the administrator user, the server 30 ends the candidate content acquisition process. Alternatively, the server 30 may re-execute the candidate content acquisition process for the target character for which all candidate contents have been rejected. This allows the server 30 to repeatedly provide candidate contents until the administrator user of the target character is satisfied.

[0084] (5) Summary As described above, the server 30 of the present embodiment obtains feature information representing the features of visual content predicted to be suitable for a specific character by providing a first model input based on the profile information set for a specific character to a first model that has learned the relationship between the profile information of a plurality of reference characters and the features of visual content that each of the plurality of reference characters has made public. The server 30 obtains candidate content, which is visual content having the appearance of a specific character as a subject and having features according to the second model input, by providing a second model input based on the obtained feature information to a second model that has learned to generate an image having the appearance of a specific character as a subject. The server 30 outputs the candidate content. This makes it possible to obtain candidate content having the appearance of a specific character as a subject and having features suitable for the profile setting of the character, that is, candidate content that has the reality of an activity record of the character.

[0085] The features of the visual content disclosed by each of the plurality of reference characters may include the visual content itself or text representing the visual features of the visual content, thereby enabling the first model to be appropriately trained and inferred.

[0086] The features of the visual content published by each of the multiple reference characters may further include information regarding at least one of the publication time of the visual content and the reaction to the publication of the visual content, which makes it possible to perform learning or inference using the publication time of the visual content as a feature, or to perform learning or inference by placing emphasis on visual content that has received a large reaction to its publication.

[0087] The feature information may include text representing visual features of visual content predicted to be suitable for a particular character. The server 30 may generate the second model input based on the text representing the visual features excluding text representing static features of appearance. This can prevent static features of the character's appearance that do not change over a long period of time (e.g., one month or more) from being changed unnaturally frequently.

[0088] The static characteristics of appearance may include the size, shape, or color of a part or all of the body, which can prevent the size, shape, or color of a part or all of the body of a character from being changed unnaturally frequently in the candidate content.

[0089] The feature information may include text representing visual features of the visual content predicted to be suitable for the particular character. When the first condition is satisfied, the server 30 may generate the second model input based on text that is obtained by excluding text representing quasi-static features of appearance from text representing the visual features. This makes it possible to prevent quasi-static features, which are features of the character's appearance that may vary in a short cycle (e.g., about one day) but do not usually vary within the same cycle, from being changed with unnatural frequency.

[0090] The first condition may include, as a necessary condition, that the scheduled release date and time of the candidate content belongs to a time period from after the specific character gets ready in the morning until before taking a bath. This makes it possible to suppress fluctuations in the quasi-static characteristics of the character's appearance from the time the character gets ready in the morning until taking a bath.

[0091] The first condition may include, as a necessary condition, that the scheduled release date and time of the candidate content belongs to a time slot after a specific character gets ready in the morning and before taking a bath, and that the candidate content is a visual content that is released second or later within that time slot on the day to which the scheduled release date and time belongs. In this way, while quasi-static characteristics of the character's appearance can be determined based on the feature information in the candidate content that is released first in a day, the quasi-static characteristics can also be maintained in candidate content that may be released later in the same day.

[0092] The profile information set for the specific character and the profile information for the multiple reference characters may each include information on at least one of gender, occupation, age, residential area, appearance characteristics, or clothing or makeup preferences, thereby making it possible to obtain appropriate characteristic information from the first model.

[0093] The server 30 may output the candidate content to an administrative user associated with a specific character, and may release the candidate content to the outside in response to approval of the candidate content from the administrative user. This makes it possible to prevent release of candidate content that is determined by the administrative user to be of insufficient quality or that is determined to be inappropriate from the standpoint of the character's branding, fan base, etc.

[0094] The externally published visual content may include photos or videos posted to a social networking service (SNS) or a content sharing service, which may allow for obtaining relevant feature information from the first model.

[0095] (6) Other modifications The storage device 11 may be connected to the client device 10 via a network NW. The display 21 may be built into the client device 10. The storage device 31 may be connected to the server 30 via the network NW.

[0096] Each step of the above information processing can be executed by either the client device 10 or the server 30.

[0097] In the above description, it is assumed that the visual content of the reference character is posted on an SNS or a content sharing service. However, the visual content of the reference character is not limited to posted content, and may include content that is publicly available in various ways (e.g., content that is exhibited).

[0098] The server 30 may reflect the decision-making contents of an organization consisting of one or more members associated with the target character in the second model or the second model input. The one or more members associated with the target character may be, for example, holders of governance tokens or NFTs related to the target character, or members of the target character's fan club. The organization may be a decentralized autonomous organization (DAO), in which case the organization's decision-making is made by voting by holders of governance tokens (however, each person's voting rights depend on the number of governance tokens held). The decision-making contents are, for example, changes to the target character's appearance characteristics (typically quasi-static characteristics, but may include other characteristics) and actions that the target character should take (food that the user should eat, places that the user should go, etc.). For example, the server 30 may embed a prompt according to the decision-making contents in the second model input, or may perform additional learning of the second model according to the decision-making contents. This allows the target character's appearance to be closer to the preferences of the target character's fans.

[0099] The server 30 may reflect the tendency of the comments or remarks of the members in the community associated with the target character in the second model or the second model input. One or more members associated with the target character may be, for example, holders of governance tokens or NFTs related to the target character, or members of the fan club of the target character. The community may be, for example, a chat group. When the comments or remarks of the members tend to be positive, the server 30 may embed prompts regarding features (e.g., facial expressions, posture, gestures, makeup, clothing, hairstyle, activity) associated with a positive impression in the second model input, or may perform additional learning of the second model according to the features. On the other hand, when the comments or remarks of the members tend to be negative, the server 30 may embed prompts regarding features (e.g., facial expressions, posture, gestures, makeup, clothing, hairstyle, activity) associated with a negative impression in the second model input, or may perform additional learning of the second model according to the features. This allows the generation of candidate content that shows the target character feeling happy or sad in response to the fans' reactions, allowing the fans to feel the target character's presence in a more realistic way.

[0100] Although the embodiment of the present invention has been described in detail above, the scope of the present invention is not limited to the above embodiment. Furthermore, the above embodiment can be improved or modified in various ways without departing from the spirit of the present invention. Furthermore, the above embodiment and the modified examples can be combined. [Explanation of symbols]

[0101] 1: Information processing system 10: Client device 11:Storage device 12: Processor 13: Input / Output Interface 14: Communication interface 21: Display 30: Server 31:Storage device 32 : Processor 33: Input / Output Interface 34: Communication interface 50: External system

Claims

1. Computer, a means for obtaining feature information representing features of visual content predicted to be suitable for a specific character by providing a first model input based on profile information set for the specific character to a first model that has learned a relationship between profile information of a plurality of reference characters and features of visual content that each of the plurality of reference characters has publicly disclosed; a means for acquiring candidate content, which is visual content having the appearance of the specific character as a subject and having characteristics according to the second model input, by providing a second model input based on the feature information to a second model that has been trained to generate an image having the appearance of the specific character as a subject; means for outputting the candidate content; A program that functions as a

2. The characteristics of the visual content disclosed to the outside by each of the plurality of reference characters include the visual content itself or text representing the visual characteristics of the visual content. The program according to claim 1.

3. The characteristics of the visual content publicly released by each of the plurality of reference characters further include information regarding at least one of the publication time of the visual content or the response to the publication of the visual content. The program according to claim 2.

4. the feature information includes text describing visual features of visual content predicted to be suitable for the particular character; causing the computer to function as a means for generating the second model input based on text that is obtained by excluding text that represents static features of appearance from text that represents the visual features; The program according to claim 1.

5. The static characteristics of the appearance include the size, shape, or color of a part or whole of the body; The program according to claim 4.

6. the feature information includes text describing visual features of visual content predicted to be suitable for the particular character; causing the computer to function as a means for generating the second model input based on text that is obtained by excluding text that represents quasi-static features of appearance from text that represents the visual features when a first condition is satisfied; The program according to claim 1.

7. the first condition includes, as a necessary condition, that the scheduled release date and time of the candidate content belongs to a time period from after the specific character gets ready in the morning until before the specific character takes a bath; The program according to claim 6.

8. the first condition includes, as a necessary condition, that a scheduled release date and time of the candidate content belongs to a time slot from after the specific character gets ready in the morning to before taking a bath, and that the candidate content is a visual content that is released second or later within the time slot on the day to which the scheduled release date and time belongs; The program according to claim 6.

9. The profile information set for the specific character and the profile information for the plurality of reference characters each include information on at least one of gender, occupation, age, residential area, appearance characteristics, or clothing or makeup preferences; The program according to claim 1.

10. The means for outputting the candidate content outputs the candidate content to an administrative user associated with the specific character; causing the computer to further function as a means for disclosing the candidate content to the outside in response to approval of the candidate content from the administrative user; The program according to claim 1.

11. The computer is further configured to function as a means for reflecting decision-making content of an organization consisting of one or more members associated with the specific character in the second model or the second model input. The program according to claim 1.

12. The computer is further caused to function as a means for reflecting a tendency of a comment or a statement of a member in a community of members associated with the specific character in the second model or the second model input. The program according to claim 1.

13. The visual content published to the outside includes photos or videos posted on a social networking service (SNS) or a content sharing service. The program according to claim 1.

14. The computer a step of obtaining feature information representing features of visual content predicted to be suitable for a specific character by providing a first model input based on profile information set for the specific character to a first model that has learned a relationship between profile information of a plurality of reference characters and features of visual content that each of the plurality of reference characters has publicly disclosed; providing a second model input based on the feature information to a second model trained to generate images having the specific character's appearance as a subject, thereby obtaining candidate content that is visual content having the specific character's appearance as a subject and having features according to the second model input; outputting the candidate content; How to do it.

15. a means for obtaining feature information representing features of visual content predicted to be suitable for a specific character by providing a first model input based on profile information set for the specific character to a first model that has learned a relationship between profile information of a plurality of reference characters and features of visual content that each of the plurality of reference characters has publicly disclosed; a means for obtaining candidate content, which is visual content having the specific character's appearance as a subject and having characteristics according to the second model input, by providing a second model input based on the feature information to a second model that has been trained to generate images having the specific character's appearance as a subject; A means for outputting the candidate content; An information processing device comprising:

16. A system including a first information processing device and a second information processing device, The first information processing device is a means for obtaining feature information representing features of visual content predicted to be suitable for a specific character by providing a first model input based on profile information set for the specific character to a first model that has learned a relationship between profile information of a plurality of reference characters and features of visual content that each of the plurality of reference characters has publicly disclosed; a means for obtaining candidate content, which is visual content having the specific character's appearance as a subject and having characteristics according to the second model input, by providing a second model input based on the feature information to a second model that has been trained to generate images having the specific character's appearance as a subject; means for outputting the candidate content to the second information processing device; Equipped with system.

Images