Information processing system, information processing method, and program

Abstract

[Problem] To further improve the convenience of image generation in which an image generation model is used. [Solution] An information processing system comprises: an acquisition unit that acquires input data; and a generation unit that generates depth information on the basis of the input data, and generates an image in which the depth information is reflected in accordance with the generated depth information and selection information pertaining to an image generation model corresponding to a specific object.

Description

Information Processing System, Information Processing Method, and Program

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

[0002] In recent years, development of technologies related to image generation AI (Artificial Intelligence) that automatically generates images from input text has been advancing. An AI model trained on a large number of images is used for the image generation AI.

[0003] For example, the following Patent Document 1 discloses a technology for generating training data used for machine learning using a CG (Computer Graphics) model.

[0004] International Publication No. 2021 / 177324

[0005] When generating an image using image generation AI, a command sentence called a prompt is input as text, but it is difficult to manually input a prompt to generate an intended image, and image generation takes time and effort.

[0006] Accordingly, the present disclosure proposes an information processing system, an information processing method, and a program capable of further enhancing the convenience of image generation using an image generation model.

[0007] According to the present disclosure, there is provided an information processing system comprising: an acquisition unit that acquires input data; and a generation unit that generates depth information based on the input data, and generates an image reflecting the depth information in accordance with the generated depth information and selection information of an image generation model corresponding to a specific target.

[0008] Further, according to the present disclosure, there is provided an information processing method including, by a processor, acquiring input data, generating depth information based on the input data, and generating an image reflecting the depth information in accordance with the generated depth information and selection information of an image generation model corresponding to a specific target.

[0009] Furthermore, this disclosure provides a program that causes a computer to function as an acquisition unit that acquires input data, and a generation unit that generates depth information based on the input data, and generates an image in which the depth information is reflected according to the generated depth information and the selection information of an image generation model corresponding to a specific object.

[0010] This figure shows an example of the configuration of an information processing device 20 according to one embodiment of the present disclosure. This figure is for explaining the generation of an AI image based on a subject image 300 according to this embodiment. This flowchart shows an example of the flow of the AI ​​image generation process using a subject image according to this embodiment. This flowchart shows an example of the flow of the AI ​​image generation process using a 3D avatar image according to this embodiment. This figure shows an example of the input screen for parameters used to generate a 3D avatar according to this embodiment. This flowchart shows an example of the flow of the AI ​​image generation process using a 3D avatar corresponding to a user according to this embodiment. This flowchart shows an example of the flow of the AI ​​image generation process based on a 3D avatar generated from a subject image according to this embodiment. This flowchart shows an example of the flow of the AI ​​image generation process based on a line drawing according to this embodiment. This flowchart shows an example of the flow of the AI ​​image generation process (worn image) based on a clothing image according to this embodiment. This figure shows an example of the display screen in the generation of a worn block diagram shows an example of the configuration of an information processing device 20A according to a modified version of this embodiment. This flowchart shows an example of the flow of the AI ​​image judgment process according to this modified version. This figure shows an example of the display of the person similarity judgment score and hallucination judgment score according to this modified version. This figure shows an example of the display of the evaluation result for a worn image according to this modified version. This figure shows an example of the configuration of an advertising effectiveness utilization system according to a modification of this embodiment. This flowchart shows an example of the process for reflecting the advertising effectiveness measurement results on the worn image according to this modification. This block diagram shows an example of the hardware configuration of the information processing device 900 according to one embodiment of this disclosure.

[0011] Preferred embodiments of this disclosure will be described in detail below with reference to the attached drawings. In this specification and the drawings, components having substantially the same functional configuration are denoted by the same reference numerals, and redundant descriptions will be omitted.

[0012] Furthermore, the explanation will be given in the following order: 1. Configuration 2. Examples of AI image generation 2-1. Image generation based on subject images 2-2. Image generation based on 3D avatar images 2-3. Image generation based on line drawings 2-4. Generation of images of clothing being worn 3. Variations 3-1. Judgment function 3-2. Reflection of advertising effectiveness measurement results 4. Hardware configuration examples 5. Supplementary information

[0013] <1. Configuration> An image generation system (an example of an information processing system) according to one embodiment of this disclosure will be described. The image generation system can be implemented by one or more information processing devices.

[0014] Figure 1 shows an example of the configuration of an information processing device 20 according to one embodiment of the present disclosure. The information processing device 20 can be implemented using a smartphone, tablet terminal, PC (personal computer), HMD (head-mounted display), or server, etc.

[0015] As shown in Figure 1, the information processing device 20 includes a communication unit 210, a control unit 220, an operation display unit 230, and a storage unit 240.

[0016] (Communication Unit 210) The communication unit 210 has a transmitting unit that transmits data to an external device and a receiving unit that receives data from an external device. The communication unit 210 may communicate with an external device or the Internet using, for example, a wired or wireless LAN (Local Area Network), Wi-Fi (registered trademark), Bluetooth (registered trademark), a mobile communication network (LTE (Long Term Evolution), 4G (fourth-generation mobile communication system), 5G (fifth-generation mobile communication system)), etc. The information processing device 20 may acquire input data from an external source via the communication unit 210.

[0017] (Control Unit 220) The control unit 220 functions as an arithmetic processing unit and control unit, and controls the overall operation of the information processing unit 20 according to various programs. The control unit 220 is implemented by electronic circuits such as a CPU (Central Processing Unit), a GPU (Graphics Processing Unit), or a microprocessor. The control unit 220 may also include a ROM (Read Only Memory) for storing programs and calculation parameters to be used, and a RAM (Random Access Memory) for temporarily storing parameters that change as needed.

[0018] The control unit 220 can also function as a display control unit 221, an input data acquisition unit 222, and an image generation processing unit 223.

[0019] The display control unit 221 controls the display of various screens on the operation display unit 230. The display control unit 221 may also control the display of various screens on an external device via the communication unit 210.

[0020] The input data acquisition unit 222 has the function of acquiring input data for image generation using an image generation model via the communication unit 210 or the operation display unit 230, etc. Details of the input data will be described later, but examples include subject images (images of subjects), depth information, 3D avatar parameters, line drawings (illustrations), etc. The method of acquiring the input data may be imaging using a camera (not shown).

[0021] The image generation processing unit 223 has the function of generating an image using an image generation model based on the input data acquired by the input data acquisition unit 222. The image generation model is implemented by an AI model. The internal structure of the AI ​​model can be any structure as long as it can output the desired information in response to the input. For example, the AI ​​model may include a convolutional neural network (CNN) or a generative adversarial network (GAN). The AI ​​model may also include any structure such as a Transformer-based architecture.

[0022] Furthermore, the AI ​​model may be generated (learned) by any learning process. The AI ​​model may also be a machine learning model that has been learned using any machine learning technique. For example, the AI ​​model may be a model generated by fine-tuning a so-called Foundation Model so that the Foundation Model is applicable to a specific task. The Foundation Model may include, for example, a Convolutional Neural Network (CNN) or a Generative Adversarial Network (GAN). The Foundation Model may also include any structure such as a Transformer-based architecture.

[0023] The image generation processing unit 223 may use an image generation AI specialized in generating images of specific subjects, such as celebrities, fashion models (also referred to as human models), or users, as an example of an AI model, to generate advertising images, fitting simulation images, or promotional images. Such an image generation AI may be constructed, for example, by a method of additionally learning the facial features of a specific person (so-called fine tuning) based on an existing AI model that generates images based on input text (a so-called Text-to-Image model). The training data may be a dataset of facial features obtained by analyzing multiple captured images of a specific person's face. LoRA (Low Rank Adaptation) may be used for additional learning. In this specification, a proper noun model that has learned the features of a specific person using LoRA is also referred to as Model LoRA and is used as an example of an image generation model.

[0024] The method and types of image generation models according to this embodiment are not limited to the examples described above.

[0025] Furthermore, the image generation processing unit 223 takes depth information of the desired composition (pose, screen composition, etc.) and prompts as input data, and uses a model LoRA selected from multiple model LoRAs (hereinafter referred to as the specific model LoRA) to generate an image that reflects the desired composition and the facial features of a specific person. Depth information is obtained by the input data acquisition unit 222 from, for example, a subject image, a 3D avatar image, or a line drawing, and input to the image generation processing unit 223. The image generated by the image generation processing unit 223 is hereinafter referred to as the AI ​​image.

[0026] Each of the functions of the control unit 220 described above may also be implemented by an image generation application installed on the information processing device 20.

[0027] (Operation Display Unit 230) The operation display unit 230 includes the function of an operation unit that receives operation input to the information processing device 20 and the function of a display unit that displays various screens. The operation display unit 230 may have a structure in which the operation unit function and the display unit function are integrated, such as a touch panel display. Alternatively, the operation display unit 230 may be configured with the operation unit function and the display unit function as separate units. The operation unit or the display unit may be implemented by an external device that communicates via the communication unit 210.

[0028] (Storage Unit 240) The storage unit 240 is implemented by a ROM that stores programs and calculation parameters used in the processing of the control unit 220, and a RAM that temporarily stores parameters that change as needed. For example, the storage unit 240 stores the above-mentioned input data, selection information for a specific model LoRA, and the generated AI image. The image generation model itself, various parameters and algorithms in the image generation process can also be stored in the storage unit 240.

[0029] Although the specific configuration of the information processing device 20 according to this embodiment has been described above, the configuration of the information processing device 20 according to this disclosure is not limited to the example shown in Figure 1. The information processing device 20 does not necessarily have to have all the configurations shown in Figure 1. Furthermore, the information processing device 20 may be implemented by multiple devices.

[0030] <2. AI Image Generation Example> Next, an example of AI image generation according to this embodiment will be specifically described. In this embodiment, an image generation AI can be used to generate images for advertising, fitting simulation, or sales promotion. In this case, the image generation processing unit 223 of the information processing device 20 can efficiently generate an AI image that reflects the composition (pose, screen composition, etc.) desired by the user by using depth information of the composition, thereby further enhancing the convenience of image generation.

[0031] <<2-1. Image Generation Based on Subject Images>> The information processing device 20 can generate an AI image that reflects the composition of a subject image captured of a subject. By inputting any subject image, the user can obtain an AI image that reflects the composition of the subject image.

[0032] Figure 2 is a diagram illustrating the generation of an AI image based on a subject image 300 according to this embodiment. The subject image 300 may be a pre-prepared captured image or an image captured by the user.

[0033] As shown in Figure 2, when a subject image 300 is input, the information processing device 20 generates a depth image from the subject image 300. The input (selection) of the subject image 300 can be performed by the user. The depth image contains information (depth information) indicating the distance of each pixel in the image from the camera, and is used in subsequent image generation to reflect the image composition, three-dimensionality, the arrangement of subjects, the poses of subjects, etc.

[0034] Furthermore, the information processing device 20 generates prompts from the subject image 300. These prompts are text that describes the image, such as descriptions of people, landscapes, and objects included in the image, as well as the image's atmosphere, context, actions, and style. The prompts are text instructions given to the image generation AI in subsequent image generation, defining the direction and content of the image generation.

[0035] Furthermore, the information processing device 20 acquires selection information of a model number (an example of identification information for a specific person) based on user operation. The model number corresponds to the Model LoRA, which allows the information processing device 20 to identify the Model LoRA to be used for image generation. The user may select the person model they wish to generate from electronic data or composite materials on paper and manually input the corresponding model number (which may be written in the composite materials). Alternatively, the user may click on the person model they wish to generate from a list of person models (such as a list of facial photographs) displayed on the selection screen, allowing the information processing device 20 to acquire the corresponding model number. Here, person models capable of generating AI images (i.e., for which an image generation AI is available) are listed as selection candidates. The image generation AI may be prepared based on information provided by pre-registered model agencies or individuals. Note that the person model is an example of a "specific person," and various fashion models such as magazine models, show models, and e-commerce models are assumed. The specific person is not limited to such person models; it may also be a talent, celebrity, athlete, etc.

[0036] The information processing device 20 then takes the depth image, the generated prompt, and the selected model LoRA as inputs to perform image generation processing and acquire the AI ​​image 350. Specifically, the information processing device 20 inputs the depth image and the prompt to the selected model LoRA and acquires the AI ​​image 350 from the model LoRA. In this embodiment, by using the depth image and prompt obtained from the subject image 300 and the model LoRA of a specific person, it is possible to generate an AI image 350 of a specific person that reflects the composition of the subject image 300. By using the subject image, the desired AI image can be acquired more easily, reliably, and efficiently without the hassle of manually inputting prompts that describe the composition.

[0037] Note that the information processing device 20 may perform a process of changing the facial expression of the person in the image on the generated AI image 350, and generate an AI image 355 with the changed facial expression. For example, the AI image 355 is an image in which the facial expression of the woman in the AI image 350 is changed to a more smiling expression. The specification of the facial expression may be performed by the user through input of a prompt or the like. This allows the user to flexibly change the facial expression of the person in the image.

[0038] FIG. 3 is a flowchart showing an example of the flow of AI image generation processing using a subject image according to the present embodiment.

[0039] As shown in FIG. 3, first, the information processing device 20 acquires a subject image (S103).

[0040] Next, the information processing device 20 generates a depth image from the acquired subject image (S106).

[0041] Next, the information processing device 20 generates a prompt from the subject image (S109).

[0042] Subsequently, the information processing device 20 selects (identifies) the LoRA model based on a user operation (selection operation) (S112). Such a LoRA model is an image generation AI capable of generating an AI image of the face of the model selected by the user.

[0043] Next, the information processing device 20 performs image generation using the depth image, the prompt, and the selected LoRA model (S115).

[0044] Next, the information processing device 20 performs a process of changing the facial expression of the person in the generated AI image as necessary (S118). Specifically, a case is assumed where the user specifies a change in facial expression.

[0045] Further, the information processing device 20 performs a process of changing the clothing worn by the person in the image as necessary (S121). Specifically, a case is assumed where the user specifies a change in clothing. Details of the clothing change processing will be described later.

[0046] As described above, based on an input subject image, an AI image reflecting the composition of the subject image can be generated with the face of the model selected by the user. The generated AI image is used, for example, for advertisement distribution. By preparing the base subject image, the user can efficiently obtain an AI image with a desired composition, and the burden of image generation work is reduced.

[0047] For example, when an actual model is engaged to perform shooting for an advertisement, a huge amount of time and cost is required to generate a single advertisement image, including planning, scheduling, negotiation with the model, securing a shooting location, preparing shooting equipment, preparing shooting staff, and the like. In the present embodiment, by using a model LoRA obtained by converting a real human model into an AI model, a large number of advertisement images of the human model for which rights processing has been completed can be generated.

[0048] As users of the image generation system according to the present embodiment, for example, business companies, individual stores, advertising agencies, and the like are assumed. Further, when using the image generation system, it is also assumed that a monthly usage fee is charged to the user. In addition, the provider side of the image generation system may perform pay-per-use charging according to the number of generated images, or obtain an advertisement adoption fee when a generated image is adopted for an advertisement.

[0049] <<2-2. Image Generation Based on 3D Avatar Image>> The acquisition of depth information is not limited to acquisition from the subject image described above, and can also be obtained from other input data. For example, the information processing apparatus 20 may generate a depth image from a 3D avatar image and use it for generating an AI image. Thereby, an AI image reflecting the composition of the 3D avatar image can be generated. Further, the information processing apparatus 20 may generate the 3D avatar image based on a user operation. The user can arbitrarily adjust the pose, arrangement, camera angle, and the like of the 3D avatar, thereby flexibly generating an input image having a composition desired by the user.

[0050] FIG. 4 is a flowchart showing an example of the flow of AI image generation processing using a 3D avatar image according to the present embodiment.

[0051] As shown in Figure 4, first the information processing device 20 generates a 3D avatar (S133).

[0052] Next, the information processing device 20 sets the physical characteristics of the 3D avatar, such as height and body shape, to be the same as those of a specific model (human model) (S136). The physical characteristic data to be set may be manually entered by the user. Figure 5 shows an example of an input screen for parameters used to generate a 3D avatar according to this embodiment. As shown in Figure 5, the input screen 400 displays numerical values ​​for physical characteristics such as height, weight, chest circumference, waist circumference, hips, and inseam in the form of sliders. By adjusting these parameters, the user can construct a 3D avatar that faithfully reproduces the physical characteristics of a specific person (model). Note that the input of physical characteristic data (an example of parameters) is not limited to manual input by the user. For example, physical characteristic data for each model may be stored in the storage unit 240, and the information processing device 20 may obtain the corresponding physical characteristic data from the storage unit 240 according to the model selected by the user.

[0053] Next, the information processing device 20 generates a 3D avatar image (S139). The user can adjust each element related to the composition of the 3D avatar, such as its pose, placement, and camera position. This allows the user to easily obtain an input image with the desired composition, even if a subject image has not been prepared in advance or if there is no subject image with the desired composition. Furthermore, by making the physical characteristics of the 3D avatar identical to those of a specific model, the AI ​​image generated by the image generation AI can reflect the characteristics of a specific model not only in the face but also in the body of the person.

[0054] Next, the information processing device 20 generates a depth image from the 3D avatar image (S142).

[0055] Next, the information processing device 20 selects (identifies) the model LoRA based on the user operation (S145).

[0056] Next, the information processing device 20 performs image generation using the depth image and the selected model LoRA (S148). The AI ​​image generated here reflects the composition and features of the 3D avatar. In this way, by inputting the physical characteristics of a specific model as parameters for 3D avatar generation and generating an AI image based on the 3D avatar image, it becomes possible to generate images with higher accuracy and greater consistency with the specific model. Furthermore, since the pose of the 3D avatar and the camera position can be arbitrarily set when generating the 3D avatar image, flexible and expressive image generation can be achieved.

[0057] Furthermore, the information processing device 20 can, if necessary, perform processing to change the facial expression of a person in the generated AI image (S151), or to change the clothing worn by a person in the generated AI image (S154).

[0058] Based on the above, it is possible to generate an AI image using the face of a model selected by the user, reflecting the composition of a 3D avatar image based on the input parameters (physical characteristic data of a specific model, pose specifications, camera position, etc.). The generated AI image can be used, for example, for advertising distribution.

[0059] (Setting the user's physical characteristics in the 3D avatar) In the embodiment described above, the physical characteristic data of a real model was set in the 3D avatar, but the system is not limited to this, and by setting the user's physical characteristic data, the image generation system according to this embodiment can also be used for try-on simulations.

[0060] Figure 6 is a flowchart showing an example of the AI ​​image generation process using a 3D avatar corresponding to a user according to this embodiment.

[0061] As shown in Figure 6, first, the information processing device 20 generates a 3D avatar (S163).

[0062] Next, the information processing device 20 sets the physical characteristics of the 3D avatar, such as height and body shape, to be the same as those of the user (S166). In this way, the information processing device 20 obtains a 3D avatar that resembles the user.

[0063] Next, the information processing device 20 generates a 3D avatar image (S169). The user can adjust various elements related to the composition of the 3D avatar, such as its pose, placement, and camera position.

[0064] Next, the information processing device 20 generates a depth image from the 3D avatar image (S172).

[0065] Next, the information processing device 20 selects a user LoRA, which is an image generation model corresponding to the user, based on the user's operation (S175). Such a user LoRA is a specific image generation model constructed by learning the user's face image, etc.

[0066] Next, the information processing device 20 performs image generation using the depth image and the selected user LoRA (S178).

[0067] Furthermore, the information processing device 20 can, if necessary, perform processing to change the facial expression of a person in the generated AI image (S181) or to change the clothing worn by the person in the generated AI image (S184). The clothing change processing function can be used, for example, in a try-on simulation when a user is considering purchasing clothes on an e-commerce site. Based on user operation, the information processing device 20 processes the clothing in the AI ​​image, which is generated using user LoRA and reflects the user's characteristics, to change the clothing to the target clothing, and presents the modified AI image to the user. This allows the user to understand what it would look like to wear the target clothing.

[0068] Thus, in this embodiment, by utilizing depth information obtained from a 3D avatar image using a 3D avatar that reflects the user's physical characteristics, and a user-specific image generation model, it becomes possible to generate AI images that faithfully reproduce the user. Furthermore, since elements such as pose, facial expression, and clothing can be flexibly adjusted, highly accurate image generation that meets the individual expression needs of each user can be achieved.

[0069] (Generating a 3D avatar from a subject image) A 3D avatar may be generated from an input subject image. This will be explained below with reference to Figure 7.

[0070] Figure 7 is a flowchart showing an example of the process flow for generating an AI image based on a 3D avatar generated from a subject image according to this embodiment.

[0071] As shown in Figure 7, first, the information processing device 20 acquires an image of the subject (S203).

[0072] Next, the information processing device 20 generates a 3D avatar image from the subject image (S206). At this time, the user can arbitrarily adjust the pose and camera position of the 3D avatar. By converting the subject image into a 3D avatar image, the user can flexibly set the composition.

[0073] Next, the information processing device 20 generates a depth image from the 3D avatar image (S209).

[0074] Next, the information processing device 20 selects a LoRA model to be used for image generation based on user input (S212).

[0075] Next, the information processing device 20 performs image generation processing using the depth image and the selected model LoRA (S215). As a result, the user obtains an AI image of a specific model that reflects the composition of the 3D avatar image based on the subject image.

[0076] Furthermore, the information processing device 20 can, if necessary, perform processing to change the facial expression of a person in the generated AI image (S218), or to change the clothing worn by a person in the generated AI image (S221).

[0077] <<2-3. Image Generation Based on Line Drawings>> The information processing device 20 can generate depth images from input line drawings and use them for AI image generation. The following explanation will be given with reference to Figure 8.

[0078] Figure 8 is a flowchart showing an example of the process flow for generating an AI image based on a line drawing according to this embodiment.

[0079] As shown in Figure 8, first, the information processing device 20 acquires a line drawing (S233). The line drawing may be a simple sketch of a person drawn by the user.

[0080] Next, the information processing device 20 generates a depth image from the line drawing (S236). Specifically, the information processing device 20 may perform a process to convert the line drawing into a photographic image and generate a depth image from the converted photographic image.

[0081] Next, the information processing device 20 generates a prompt from the line drawing (S239).

[0082] Next, the information processing device 20 selects a model LoRA, which is an AI image generation model of a specific person to be used for image generation, based on user input (S242).

[0083] Next, the information processing device 20 performs image generation using the depth image, prompt, and selected model LoRA (S245).

[0084] Furthermore, the information processing device 20 can, if necessary, perform processing to change the facial expression of a person in the generated AI image (S248), or to change the clothing worn by a person in the generated AI image (S251).

[0085] Thus, in this embodiment, even from simple input data such as a line drawing, a highly accurate AI image that reflects the composition, such as the pose and placement of the person drawn in the line drawing, can be generated using the face of a specific person model, through the generation of a depth image. The image generation system according to this embodiment can also flexibly respond to the need to intuitively generate images from hand-drawn images and the like.

[0086] <<2-4. Generation of Wearing Images>> Next, the process of changing the clothing in the generated AI image will be explained in detail. The information processing device 20 generates an image in which the clothing of the person in the AI ​​image has been changed, specifically an image that reflects the characteristics of the clothing shown in the clothing image (wearing image), from the generated AI image and the input clothing image. This image generation system according to this embodiment, which can generate wearing images in response to the input clothing image, is expected to be used, for example, by e-commerce apparel companies and contracting companies that are commissioned to create information for product pages on e-commerce apparel sites.

[0087] Figure 9 is a flowchart showing an example of the process flow for generating an AI image (worn image) based on a clothing image according to this embodiment. Figures 10 to 12 show examples of display screens in the worn image generation process according to this embodiment. The worn image generation process will be described below with reference to the display screen examples shown in Figures 10 to 12 as appropriate.

[0088] As shown in Figure 9, first, the information processing device 20 generates a prompt from the input clothing image (S263). The user inputs a clothing image of the clothing they want to dress. The generated prompt is text information that reflects the shape, color, design, etc. of the clothing, and is used in the subsequent image generation. As an example of the prompt generation screen for clothing images, refer to the display screen 420 in Figure 10.

[0089] As shown in Figure 10, the display screen 420 is provided with an input image display area 422, a prompt display area 423, and a prompt generation button 424. The input image display area 422 displays a flat lay image of clothing, that is, an image taken with clothing placed on the floor or a table (an example of a clothing image). Note that the clothing image is not limited to a flat lay image, but may also be an image of clothing on a mannequin or hanging on a hanger. In the example shown in Figure 10, a long-sleeved striped shirt is displayed. A prompt generation button 424 is provided at the bottom of the display screen 420, and when the user operates the prompt generation button 424, the information processing device 20 generates a clothing prompt from the clothing image.

[0090] The information processing device 20 displays a prompt for clothing generated from the clothing image in the prompt display area 423. This prompt functions as an instruction used in subsequent image generation to generate an image in which a person in the AI ​​image is wearing the clothing shown in the clothing image.

[0091] Next, the information processing device 20 selects a model LoRA corresponding to a specific person model based on the user operation (S266).

[0092] Next, the information processing device 20 uses the clothing prompt generated in the process of S263 and the selected model LoRA to generate an AI image of a specific person model (S269). At this time, the information processing device 20 can generate an AI image that reflects the composition desired by the user by further using the depth image and the like mentioned above as input data to the model LoRA. The AI ​​image generated at this stage is expected to be a base image in which the general features such as shape and pattern are reflected based on the input clothing prompt, but it does not faithfully reproduce the target clothing. As an example of a screen for generating such a base image, refer to the display screen 430 in Figure 11.

[0093] As shown in Figure 11, the display screen 430 includes a prompt input field 432, a LoRA setting field 433, a number of generated images setting field 434, an image generation button 435, and a generated image display area 436. The prompt input field 432 displays the prompt generated from the clothing image in Figure 10. The user can also edit this prompt as needed. In addition, prompts obtained from the subject image, as explained with reference to Figure 2, may be added to the prompt input field 432.

[0094] In the LoRA settings section 433, an image generation AI (model LoRA) corresponding to a specific person model is selected. This determines the image generation AI to be used to generate the AI ​​image. Although not shown in Figure 11, the display screen 430 may also include a check box for using depth images. Specifically, for example, this assumes the use of depth images obtained from subject images as explained with reference to Figure 2.

[0095] The number of images to generate setting field 434 is an interface for specifying the number of person images to generate, and the number can be adjusted by operating a slider. After the setting is complete, when the user operates the image generation button 435, the information processing device 20 executes the AI ​​image generation process based on the prompt and the selected model LoRA.

[0096] As a result of the image generation process, the generated image display area 436 displays an AI image of a specific person model wearing clothing that reflects the information from the clothing image. In the example shown in Figure 11, the person in the generated image is depicted wearing a long-sleeved striped shirt naturally. However, at this stage, while the clothing image has a full striped pattern, the AI ​​image displayed in the generated image display area 436 only has stripes on parts such as the sleeves, and the design of the stripes also differs slightly from the clothing image, resulting in an image that is somewhat ambiguous in terms of clothing reproduction. In this embodiment, the AI ​​image at this stage is used as the base image to generate the worn image next.

[0097] Returning to Figure 9, let's continue the explanation. The information processing device 20 generates a worn image using the generated AI image and the clothing image of the clothes to be worn (S272). The clothing image is the same as the clothing image used in S263 above. By combining the clothing image with the AI ​​image, the information processing device 20 generates a worn image and can change the clothing of the person in the AI ​​image. As an example of such a worn image generation screen, refer to the display screen 440 in Figure 12.

[0098] As shown in Figure 12, the display screen 440 includes a model display area 442, an input image display area 443, a worn image generation button 444, and a worn image (composite image) display area 445. The model display area 442 displays the base AI image generated in Figure 11 (specifically, a person image generated by the LoRA model).

[0099] The input image display area 443 displays a flat lay image of the clothing to be worn by the model (an example of a clothing image) and a mask image of the clothing area (in this case, the top) of the AI ​​image displayed in the model display area 442. The mask image is used to identify the outline of the clothing and to composite the clothing image only into a predetermined area of ​​the model's AI image.

[0100] When the user operates the worn image generation button 444, the information processing device 20 performs a process to synthesize clothing onto the AI ​​image of the model using the input clothing image and mask image. After the synthesis process, the worn image display area 445 on the right side of the display screen 440 displays a new AI image, i.e., a worn image, in which the model's clothing has been changed.

[0101] As shown in Figure 11, the configuration allows for the efficient generation of realistic images of clothing being worn by combining arbitrary clothing images with an AI image of a human model. Users can easily obtain a composite image of a specific human model wearing the clothing simply by preparing the clothing images. This embodiment provides an image generation function that can be used for a variety of purposes, such as virtual try-on, product suggestions, and advertising material creation. Furthermore, this embodiment provides an extremely useful function in the fashion industry as a means of generating content for try-on simulations and sales promotion.

[0102] <3. Modified Forms> Next, modified forms of this embodiment will be described with reference to the drawings.

[0103] <<3-1. Judgment Function>> Figure 13 is a block diagram showing an example of the configuration of the information processing device 20A according to a modification of this embodiment. The information processing device 20A is equipped with a judgment function for evaluating the generated AI image.

[0104] This section will mainly describe the configuration that differs from the information processing device 20 shown in Figure 1. As shown in Figure 13, the control unit 220A of the information processing device 20A also functions as a display control unit 221, an input data acquisition unit 222, an image generation processing unit 223, and a determination unit 224.

[0105] The determination unit 224 is responsible for performing various evaluations on the generated AI image. For example, the determination unit 224 may perform a person-likeness determination (first determination function) that evaluates the similarity between the face of a person in the generated AI image and a specific person (selected model), a hallucination determination (second determination function) that evaluates the presence or absence of unnatural elements (hallucination) in the AI ​​image, or a clothing determination (third determination function) that evaluates the accuracy of the reproduction of clothing in the worn image. Each determination function by the determination unit 224 may be implemented by a determination AI. The determination AI can employ any internal structure as long as it can output the desired information in response to the input. Furthermore, there are no particular limitations on the method or type of construction of the determination AI. The determination results from the determination unit 224 are output by the display control unit 221 and presented to the user.

[0106] The memory unit 240 stores input data, generated AI images, various generation models, prompts, depth images, and various judgment algorithms and their evaluation results.

[0107] Thus, the information processing device 20A according to this modified version has the function of not only generating AI images but also performing quality evaluation of the generated images, contributing to improved reliability of image generation processing and increased efficiency of image selection by the user.

[0108] (Processing) Figure 14 is a flowchart showing an example of the AI ​​image determination process using this modified example.

[0109] As shown in Figure 14, first, the information processing device 20A generates an image using the LoRA model (S303). This image generation process corresponds to the generation process of each AI image described above.

[0110] Next, the information processing device 20A calculates a facial similarity score for the generated AI image using the person similarity determination AI (S306). This score indicates the degree to which the faces of the people in the generated AI image match the faces of a specific person.

[0111] Next, the information processing device 20A uses a hallucination judgment AI to evaluate the generated AI image to determine if it contains any unnatural elements or defects, and calculates a judgment score (S309). This evaluation allows for confirmation of the naturalness and reliability of the image.

[0112] Next, the information processing device 20A modifies the clothing in the generated AI image and generates a worn image (S312).

[0113] The information processing device 20A then uses the clothing image recognition AI to evaluate the reproducibility of the clothing worn by the person in the generated clothing image and calculates a clothing recognition score (S315). This score indicates whether the wearing state is appropriately reproduced based on the clothing image and can be used as a quality indicator in advertising and product proposals.

[0114] Thus, this modified version incorporates functions to evaluate image quality and reliability from multiple perspectives, in addition to image generation, thereby enabling the construction of a practical system that supports high-precision image generation processing and appropriate image selection.

[0115] (Example of display of judgment results) Figure 15 shows an example of the display of the personal resemblance judgment score and hallucination judgment score using this modified example.

[0116] As shown in Figure 15, the display screen 500 includes a matrix area 502 on which the personal resemblance judgment score is plotted on the vertical axis and the hallucination judgment score is plotted on the horizontal axis.

[0117] The vertical axis displays a person-similarity score ranging from "good (similar)" to "bad (not similar)," evaluating how closely the generated image of a person matches a specific individual. The horizontal axis, on the other hand, displays a hallucination score ranging from "good (low hallucination)" to "bad (high hallucination)," evaluating the naturalness and consistency of the image.

[0118] Within the matrix region 502, multiple generated images to be evaluated are positioned according to their respective scores. This makes it easier for the user to visually identify similar and natural images (upper left region) and to eliminate unnatural images with high hallucination (lower right region).

[0119] Each image may be labeled with its ranking as a candidate for adoption. The information processing device 20 may highlight images with a high ranking as candidates for adoption on the display screen 500.

[0120] In this way, by presenting a matrix display like the display screen 500 shown in Figure 15, users can intuitively and efficiently select high-quality images while comparing multiple generated images. By evaluating both the facial reproduction of a specific person and the naturalness of the image in a balanced manner, the accuracy and work efficiency of image selection are dramatically improved.

[0121] Figure 16 shows an example of displaying the evaluation results for a worn image using this modified example. As shown in Figure 16, the display screen 520 includes a matrix area 522 on which the horizontal axis represents the clothing judgment score and the vertical axis represents the person similarity judgment score (or hallucination judgment score).

[0122] The horizontal axis, representing the clothing classification score, indicates the degree to which the clothing is reproduced in the generated worn image, i.e., how accurately the input clothing image is reproduced. The scores are arranged from left to right, in order from "good (high clothing reproduction accuracy)" to "poor (low clothing reproduction accuracy)."

[0123] The vertical axis displays the personal resemblance score (or hallucination score), with evaluations ranging from "good (similar / natural)" to "bad (not similar / unnatural)" from top to bottom. This allows for intuitive identification of model images appropriately wearing highly reproducible clothing. Alternatively, the vertical axis could display both the personal resemblance score and the hallucination score, with evaluations ranging from "good (similar and natural)" to "bad (not similar and unnatural)" from top to bottom.

[0124] Within the matrix region 522, each evaluated generated image is placed at its corresponding score position. Images in the upper left are those of particularly high quality (e.g., top 1-3 candidates for adoption). Images in the lower right are those of low quality, making it easy to exclude unnecessary images.

[0125] As shown in Figure 16, the matrix display allows for a comprehensive evaluation of generated images from two different perspectives: the reproducibility of clothing and the reproducibility of the person, enabling the efficient selection of the optimal image. This is particularly useful for the practical application of quality assurance and automated selection processing of worn images in the fields of fashion e-commerce and advertising.

[0126] <<3-2. Reflection of Advertising Effectiveness Measurement Results>> This section explains the mechanism for using the generated AI images for advertising purposes and regenerating the AI ​​images according to their advertising effectiveness.

[0127] Figure 17 shows an example of the configuration of an advertising effectiveness utilization system based on a modified version of this embodiment. As shown in Figure 17, this system is composed of an information processing device 20B, an apparel company terminal 30, and an advertising distribution system 35.

[0128] The information processing device 20B is a device equipped with the image generation function described above. The information processing device 20B may be the information processing device 20 in Figure 1 or the information processing device 20A in Figure 13.

[0129] The information processing device 20B generates an image of a specific model wearing the clothing based on an AI image and clothing image that reflect the face and other features of the specific model. The image is then transmitted to the apparel company terminal 30.

[0130] The apparel company terminal 30 manages the images of clothing being worn that it receives from the information processing device 20B, and provides these images to the advertising distribution system 35, which then distributes them as advertisements.

[0131] The advertising distribution system 35 distributes advertisements, including images of people wearing the products, through online advertising media and social networking services (SNS), and measures the effectiveness of the advertisements, such as user responses, click-through rates, and conversion rates.

[0132] These measurement results (advertising effectiveness measurement results) are sent back from the advertising distribution system 35 to the apparel company terminal 30, and then transmitted from the apparel company terminal 30 to the information processing device 20B. Based on the received advertising effectiveness measurement results, the information processing device 20B can partially change the image generation conditions and regenerate new images of the garment being worn.

[0133] (Processing) Figure 18 is a flowchart showing an example of the process for reflecting the advertising effectiveness measurement results to the worn images using this modified example.

[0134] As shown in Figure 18, first, the information processing device 20B generates an AI image of a specific person using the LoRA model (S333). This image generation process corresponds to the AI ​​image generation process described above.

[0135] Next, the information processing device 20B modifies the clothing in the generated AI image and generates a worn image (S336). By combining the clothing images, the information processing device 20B can generate a composite image in which a specific model is dressed in arbitrary clothing.

[0136] Next, the generated images of the clothing being worn are transmitted to the apparel company terminal 30 (S339), and provided to the advertising distribution system 35 via the apparel company terminal 30. After the advertising distribution is performed, the information processing device 20B obtains the advertising effectiveness measurement results from the apparel company terminal 30 (S342).

[0137] The information processing device 20B modifies some of the generation conditions for the worn image (or the base AI image) based on the advertising effectiveness measurement results and generates a new worn image (S345). Examples of the generation conditions to be modified include the person model, pose, facial expression, background, and clothing style. The newly generated worn image is then transmitted again to the apparel company terminal 30 and used for advertising distribution (S339).

[0138] The information processing device 20B repeats steps S339 to S345 until the processing condition is met (S348 / No), and if the processing condition is met (S348 / Yes), it terminates the processing.

[0139] Thus, in this modified version, an automatic feedback loop is realized that optimizes image generation conditions based on advertising effectiveness, thereby improving the quality of advertising materials and maximizing marketing results.

[0140] <5. Hardware Configuration Example> An embodiment of the present disclosure has been described above. Next, with reference to Figure 19, an example of a hardware configuration used in the information processing device 20, information processing device 20A, or information processing device 20B according to an embodiment of the present disclosure will be described. Hereinafter, the information processing device 20, information processing device 20A, and information processing device 20B will be collectively referred to as the information processing device 20.

[0141] Figure 19 is a block diagram showing an example of the hardware configuration of an information processing device 900 according to one embodiment of the present disclosure. The information processing device 900 is an example of the hardware configuration applied to the information processing device 20 according to this embodiment. Note that the information processing device 900 does not necessarily have all of the hardware configurations shown in Figure 19.

[0142] As shown in Figure 19, the information processing device 900 includes a processing circuit 901, a ROM (Read Only Memory) 902, and a RAM (Random Access Memory) 903. The information processing device 900 may also include a host bus 907, a bridge 909, an external bus 911, an interface 913, an input device 915, an output device 917, a storage device 919, a drive 921, a connection port 923, and a communication device 925.

[0143] The processing circuit 901 functions as an arithmetic processing unit and control unit, and controls the overall operation or a part of the operation within the information processing unit 900 according to various programs recorded in the ROM 902, RAM 903, storage device 919, or removable recording medium 927. The ROM 902 stores programs and calculation parameters used by the processing circuit 901. The RAM 903 temporarily stores programs used in the execution of the processing circuit 901 and parameters that change as appropriate during its execution. The processing circuit 901, ROM 902, and RAM 903 are interconnected by a host bus 907, which is composed of an internal bus. Furthermore, the host bus 907 is connected to an external bus 911, such as a PCI (Peripheral Component Interconnect / Interface) bus, via a bridge 909.

[0144] The input device 915 is a device operated by the user, such as a button. The input device 915 may also include a mouse, keyboard, touch panel, switch, and lever. The input device 915 may also include a microphone that detects the user's voice. The input device 915 may be, for example, a remote control device that uses infrared or other radio waves, or an external connection device 929 such as a mobile phone that is compatible with the operation of the information processing device 900. The input device 915 includes an input control circuit that generates an input signal based on information input by the user and outputs it to the processing circuit 901. By operating this input device 915, the user inputs various data to the information processing device 900 or instructs it to perform processing operations.

[0145] The input device 915 may also include an imaging device and sensors. The imaging device is a device that captures real space and generates an image using various components such as an image sensor, such as a CCD (Charge Coupled Device) or CMOS (Complementary Metal Oxide Semiconductor), and a lens for controlling the imaging of a subject onto the image sensor. The imaging device may capture still images or motion images. The sensors are various types of sensors, such as distance sensors, acceleration sensors, gyro sensors, geomagnetic sensors, vibration sensors, light sensors, and sound sensors. The sensors acquire information about the state of the information processing device 900 itself, such as the orientation of the housing of the information processing device 900, and information about the surrounding environment of the information processing device 900, such as the brightness and noise around the information processing device 900. The sensors may also include a GPS sensor that receives GPS (Global Positioning System) signals and measures the latitude, longitude, and altitude of the device.

[0146] The output device 917 is comprised of a device capable of visually or audibly notifying the user of the acquired information. The output device 917 may be, for example, a display device such as an LCD (Liquid Crystal Display) or an organic EL (Electro-Luminescence) display, or an audio output device such as a speaker or headphones. The output device 917 may also include a PDP (Plasma Display Panel), a projector, a hologram, a printer, etc. The output device 917 outputs the results obtained from the processing of the information processing device 900 as images such as text or pictures, or as sound such as voice or sound. The output device 917 may also include a lighting device that brightens the surroundings.

[0147] The storage device 919 is a data storage device configured as an example of the storage unit of the information processing device 900. The storage device 919 is composed of, for example, a magnetic storage device such as an HDD (Hard Disk Drive), a semiconductor storage device, an optical storage device, or a magneto-optical storage device. This storage device 919 stores programs and various data executed by the processing circuit 901, as well as various data acquired from external sources.

[0148] The drive 921 is a reader / writer for removable recording media 927 such as magnetic disks, optical disks, magneto-optical disks, or semiconductor memory, and is either built into or external to the information processing device 900. The drive 921 reads information recorded on the installed removable recording media 927 and outputs it to the RAM 905. The drive 921 also writes data to the installed removable recording media 927.

[0149] The connection port 923 is a port for directly connecting equipment to the information processing device 900. The connection port 923 may be, for example, a USB (Universal Serial Bus) port, an IEEE 1394 port, or a SCSI (Small Computer System Interface) port. Alternatively, the connection port 923 may be an RS-232C port, an optical audio terminal, or an HDMI (High-Definition Multimedia Interface) port. By connecting an external device 929 to the connection port 923, various types of data can be exchanged between the information processing device 900 and the external device 929.

[0150] The communication device 925 is a communication interface, for example, consisting of a communication device for connecting to an external network 50. The communication device 925 may be, for example, a communication card for wired or wireless LAN (Local Area Network), Bluetooth®, Wi-Fi®, or WUSB (Wireless USB). Alternatively, the communication device 925 may be a router for optical communication, a router for ADSL (Asymmetric Digital Subscriber Line), or a modem for various types of communication. The communication device 925 transmits and receives signals, for example, to the Internet or other communication devices using a predetermined protocol such as TCP / IP. The external network 50 connected to the communication device 925 is a network connected by wire or wireless, for example, the Internet, a home LAN, infrared communication, radio wave communication, or satellite communication.

[0151] For example, when the information processing device 900 functions as the information processing device 20 according to the embodiment of this disclosure, the processing circuit 901 of the information processing device 900 functions as a control unit 220 (or control unit 220A) by executing a program loaded on the RAM 903. The storage device 919 stores the information processing program according to this disclosure and various data stored in the storage unit 240. The processing circuit 901 reads and executes the program data from the storage device 919, but as another example, these programs may be obtained from other devices via an external network 50. In other words, the storage device 919 is not limited to being inside the information processing device 900, but may be located outside the information processing device 900. The processing circuit 901 is an example of an integrated circuit, and CPU (Central Processing Unit), MPU (Micro Processing Unit), GPU (Graphics Processing Unit), APU (Accelerated Processing Unit), ASIC (Application Specific Integrated Circuit), and FPGA (Field Programmable Gate Array) can all be considered integrated circuits.

[0152] Furthermore, when the information processing device 900 functions as the information processing device 20 according to the embodiment of this disclosure, the communication device 925 of the information processing device 900 corresponds to the communication unit 210. Also, the input device 915 and output device 917 of the information processing device 900 correspond to the operation display unit 230.

[0153] <6. Supplementary Information> Although preferred embodiments of the present disclosure have been described in detail above with reference to the attached drawings, the present technology is not limited to such examples. It is clear that a person with ordinary skill in the art of the present disclosure may conceive of various modifications or alterations within the scope of the technical idea described in the claims, and these will naturally be understood to fall within the technical scope of the present disclosure.

[0154] Furthermore, one or more computer programs can be created to enable the information processing device 20 to perform its functions, using the CPU, ROM, RAM, and other hardware components built into the information processing device 20. A computer-readable storage medium containing these one or more computer programs is also provided.

[0155] Furthermore, the effects described herein are merely descriptive or illustrative and not limiting. In other words, the technology relating to this disclosure may produce other effects that will be apparent to those skilled in the art from the description herein, in addition to or in lieu of the effects described herein.

[0156] Furthermore, this technology can also take the following configurations: (1) An information processing system comprising: an acquisition unit that acquires input data; and a generation unit that generates depth information based on the input data and generates an image that reflects the depth information according to the generated depth information and the selection information of an image generation model corresponding to a specific object. (2) The information processing system according to (1), wherein the image generation model is an image generation model that generates an image of a specific person, and the generation unit uses the selected image generation model to generate an image of the specific person that reflects the composition shown in the input data. (3) The information processing system according to (2), wherein the composition includes the pose and arrangement of the person. (4) The information processing system according to any one of (1) to (3), wherein the input data includes an image captured of a subject. (5) The information processing system according to (4), wherein the generation unit generates a depth image and a prompt from the image captured, and uses the generated depth image and the prompt to generate an image using the selected image generation model. (6) The information processing system according to any one of (1) to (5) above, wherein the input data includes parameters relating to the generation of a 3D avatar image. (7) The information processing system according to (6) above, wherein the parameters include at least one of the following: numerical values ​​of the physical characteristics of a specific person used to generate a 3D avatar, pose information of the 3D avatar, and camera angle information when generating the 3D avatar image. (8) The information processing system according to (7) above, wherein the generation unit generates a 3D avatar image based on the parameters, generates a depth image from the generated 3D avatar image, and generates an image using the generated depth image with the selected image generation model.(9) The information processing system according to any one of (1) to (8), wherein the input data includes an image of a subject, the generation unit generates a 3D avatar of the subject from the image, generates 3D avatar images of an arbitrary pose and camera angle, generates a depth image from the generated 3D avatar images, and generates the image using the generated depth image with the selected image generation model. (10) The information processing system according to any one of (1) to (9), wherein the input data includes a line drawing, the generation unit generates a depth image and prompts from the line drawing, and generates the image using the generated depth image and prompts with the selected image generation model. (11) The information processing system according to any one of (1) to (10), wherein the generation unit generates an image with a changed expression in response to the input of a prompt regarding the expression for the generated image. (12) The information processing system according to any one of (1) to (11), wherein the generation unit generates a prompt from a clothing image, generates a person image according to the generated prompt and selection information of an image generation model for a specific person, and generates a worn image by compositing the clothing image onto the clothing region of the person image based on the generated person image and the clothing image. (13) The information processing system according to (12), wherein the generation unit further generates the person image using the depth information. (14) The information processing system according to any one of (2) to (13), further comprising a first determination unit that determines the similarity between a face in the generated image and the face of a specific person, and controlling the display of the determination result by the first determination unit. (15) The information processing system according to any one of (2) to (14), further comprising a second determination unit that determines the hallucination of the generated image, and controlling the display of the determination result by the second determination unit.(16) The information processing system according to (12) or (13), further comprising a third determination unit for determining the degree of reproduction of clothing in the generated worn image, and controlling the display of the result of the determination by the third determination unit. (17) The information processing system according to any one of (1) to (16), wherein the generation unit partially changes the image generation conditions according to the measurement result of the advertising effect of the image and generates a new image. (18) An information processing method comprising: a processor acquiring input data; generating depth information based on the input data; and generating an image in which the depth information is reflected according to the generated depth information and selection information of an image generation model corresponding to a specific target. (19) A program that causes a computer to function as: an acquisition unit for acquiring input data; and a generation unit for generating depth information based on the input data; and generating an image in which the depth information is reflected according to the generated depth information and selection information of an image generation model corresponding to a specific target. (20) An information processing system comprising: an acquisition unit that acquires input data including an image of clothing and selection information for an image generation model corresponding to a specific person; and a generation unit that generates an image of the specific person wearing the clothing from the clothing image based on the input data.

[0157] 20 Information Processing Unit 210 Communication Unit 220 Control Unit 221 Display Control Unit 222 Input Data Acquisition Unit 223 Image Generation Processing Unit 224 Determination Unit 230 Operation Display Unit 240 Storage Unit

Claims

1. An information processing system comprising: an acquisition unit that acquires input data; and a generation unit that generates depth information based on the input data and generates an image that reflects the depth information according to the generated depth information and selection information of an image generation model corresponding to a specific object.

2. The information processing system according to claim 1, wherein the image generation model is an image generation model that generates an image of a specific person, and the generation unit uses the selected image generation model to generate an image of the specific person that reflects the composition shown in the input data.

3. The information processing system according to claim 2, wherein the composition includes the poses and arrangement of figures.

4. The information processing system according to claim 1, wherein the input data includes an image of a subject captured by imaging.

5. The information processing system according to claim 4, wherein the generation unit generates a depth image and a prompt from the captured image, and generates an image using the generated depth image and the prompt by the selected image generation model.

6. The information processing system according to claim 1, wherein the input data includes parameters relating to the generation of a 3D avatar image.

7. The information processing system according to claim 6, wherein the parameters include at least one of the following: numerical values ​​of the physical characteristics of a specific person used to generate a 3D avatar, pose information of the 3D avatar, and camera angle information when generating the 3D avatar image.

8. The information processing system according to claim 7, wherein the generation unit generates a 3D avatar image based on the parameters, generates a depth image from the generated 3D avatar image, and generates an image using the generated depth image according to the selected image generation model.

9. The information processing system according to claim 1, wherein the input data includes an image of a subject, the generation unit generates a 3D avatar of the subject from the image, generates 3D avatar images of arbitrary poses and camera angles, generates depth images from the generated 3D avatar images, and generates the image using the generated depth images according to the selected image generation model.

10. The information processing system according to claim 1, wherein the input data includes a line drawing, the generation unit generates a depth image and a prompt from the line drawing, and uses the generated depth image and prompt to generate the image using the selected image generation model.

11. The information processing system according to claim 1, wherein the generation unit generates an image with a modified facial expression in response to a prompt input regarding the facial expression for the generated image.

12. The information processing system according to claim 1, wherein the generation unit generates a prompt from a clothing image, generates a person image according to the generated prompt and selection information of an image generation model for a specific person, and generates a worn image by compositing the clothing image onto the clothing region of the person image based on the generated person image and the clothing image.

13. The information processing system according to claim 12, wherein the generation unit further uses the depth information to generate the person image.

14. The information processing system according to claim 2, further comprising a first determination unit that determines the degree of similarity between a face in the generated image and the face of a specific person, and controlling the display of the result of the determination by the first determination unit.

15. The information processing system according to claim 2, further comprising a second determination unit for determining the hallucination of the generated image, and controlling the display of the result of the determination by the second determination unit.

16. The information processing system according to claim 12, further comprising a third determination unit for determining the degree of reproduction of clothing in the generated wearing image, and controlling the display of the determination result by the third determination unit.

17. The information processing system according to claim 1, wherein the generation unit modifies some of the image generation conditions and generates a new image according to the measurement results of the advertising effect of the image.

18. An information processing method comprising: a processor acquiring input data; generating depth information based on the input data; and generating a target image that reflects the depth information according to the generated depth information and selection information of an image generation model corresponding to a specific target.

19. A program that causes a computer to function as: an acquisition unit that acquires input data; a generation unit that generates depth information based on the input data and generates a target image that reflects the depth information according to the generated depth information and selection information of an image generation model corresponding to a specific target.

Images