Information processing device, information processing method, and program

Abstract

An information processing device according to the present invention comprises a generation unit that generates suggestion information for a clothing selection for a user on the basis of information that indicates the characteristics of clothing selections by the user or a learning model that has learned the characteristics of clothing selections by the user.

Description

Information processing device, information processing method, and program

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

[0002] With the recent development of information processing technology, users are increasingly deciding their own actions based on suggestions from information processing devices. For example, a user may decide what clothes to wear based on suggestions from the information processing device.

[0003] Japanese Patent Application Laid-Open No. 2021-182172

[0004] However, it is difficult for conventional information processing devices to provide highly accurate suggestions that a user truly desires. For example, assume that the information processing device is a device that suggests clothes for a user to wear. In many cases, the suggestions made by the information processing device reflect the sensibilities of a professional stylist. However, the sensibilities of a professional stylist do not necessarily coincide with the sensibilities of a user. Therefore, the suggestions made by conventional information processing devices may deviate from the clothes that the user actually wants to wear.

[0005] Therefore, the present disclosure proposes an information processing device, an information processing method, and a program that enable highly accurate suggestions.

[0006] It should be noted that the above problem or object is merely one of multiple problems or objects that can be solved or achieved by multiple embodiments disclosed in this specification.

[0007] In order to solve the above problem, an information processing device according to one embodiment of the present disclosure includes a generation unit that generates suggested information regarding clothing selection for a user based on information indicating characteristics regarding the user's clothing selection or a learning model that has learned the characteristics regarding the user's clothing selection.

[0008] 1 is a diagram for explaining an overview of an embodiment; FIG. 1 is a diagram for explaining a user interface included in an information processing device according to an embodiment; FIG. 2 is a diagram for explaining a configuration example of an information processing system according to an embodiment of the present disclosure; FIG. 3 is a diagram for explaining a configuration example of a server according to an embodiment of the present disclosure; FIG. 4 is a diagram for explaining a configuration example of a terminal device according to an embodiment of the present disclosure; FIG. 5 is a diagram for explaining an example of an external appearance of a terminal device; FIG. 1 is a diagram for explaining an overview of an operation of an information processing system; FIG. 2 is a diagram for explaining an example of a selection screen for a user image; FIG. 3 is a diagram for explaining an example of an input screen for clothing information; FIG. 4 is a diagram for explaining an example of an acquisition screen for evaluation information; FIG. 5 is a diagram for explaining an example of a proposal screen for selecting clothing; FIG. 6 is a diagram for explaining an example of a chat screen; FIG. 6 is a diagram for explaining an example of a wearing history confirmation screen; FIG. 7 is a diagram for explaining a wearing history confirmation screen in which a calendar is hidden; FIG. 7 is a diagram for explaining an operation of an information processing system according to Example 1;

[0009] Hereinafter, embodiments of the present disclosure will be described in detail with reference to the drawings. In the following embodiments, the same components are designated by the same reference numerals, and redundant description will be omitted.

[0010] Additionally, in this description / specification, the phrase "at least one of" following a list of elements is understood to mean that the listed elements are optional. For example, "at least one of A, B, and C" means "(A), (B), (C), (A and B), (A and C), (B and C), or (A, B, and C)." "At least one of A, B, or C" and "at least one of A, B, and / or C" are similar to "at least one of A, B, and C." Here, A, B, and C are all arbitrary expressions (e.g., words, phrases, clauses, terms, or items).

[0011] In addition, in this specification and drawings, multiple components having substantially the same functional configuration may be distinguished by adding different numbers to the same reference numeral. For example, multiple components having substantially the same functional configuration may be distinguished by adding different numbers to the same reference numerals to the server 10 as needed. １ , and 10 ２ However, when there is no need to particularly distinguish between multiple components having substantially the same functional configuration, only the same reference numerals are used. For example, the server 10 １ , and 10 ２ When there is no need to particularly distinguish between them, they will be simply referred to as the server 10.

[0012] One or more embodiments (including examples and variations) described below can be implemented independently. However, at least a portion of the embodiments described below may be implemented in appropriate combination with at least a portion of another embodiment. These embodiments may include novel features that are different from one another. Therefore, these embodiments may contribute to solving different purposes or problems and may produce different effects.

[0013] The present disclosure will be described in the following order: 1. Overview of the present disclosure 2. Configuration of information processing system 2-1. Example of server configuration 2-2. Example of terminal device configuration 2-3. Learning model 3. Operation of information processing system 3-1. Overview of operation of information processing system 3-2. Functions of terminal device 3-2-1. Registration function 3-2-2. Suggestion function 3-2-3. Confirmation function 3-2-4. Analysis function 3-3. Specific examples of operation of information processing system 3-3-1. Example 1 3-3-2. Example 2 3-3-3. Example 3 4. Modified examples 4-1. Modified example related to output format 4-2. Modified example related to output information 4-3. Modified example related to learning 4-4. Modified example related to functions of terminal device 4-5. Modified example related to proposed image 4-6. Other modified examples 5. Conclusion

[0014] <<1. Overview of the Present Disclosure>> First, an overview of the present disclosure will be described.

[0015] With the recent development of information processing technology, users are increasingly deciding their own actions based on suggestions from information processing devices. For example, users may decide what clothes to wear based on suggestions from information processing devices. However, it is difficult for conventional information processing devices to provide highly accurate suggestions that users truly desire.

[0016] For example, suppose an information processing device is a device that suggests clothes for a user to wear. In many cases, the suggestions made by the information processing device reflect the sensibilities of a professional stylist. However, the sensibilities of the professional stylist do not necessarily match the sensibilities of the user. In other words, suggestions made by conventional information processing devices do not reflect the characteristics of a user's clothing selection (e.g., preferences, style, or clothing selection logic). Therefore, suggestions made by conventional information processing devices may deviate from the clothes that a user actually wants to wear in their daily clothing selection.

[0017] Furthermore, conventional information processing devices do not suggest clothes to users according to clothing selection conditions (e.g., time, place, or occasion). Therefore, suggestions from conventional information processing devices may be inconsistent in some cases.

[0018] Furthermore, conventional information processing devices require a user to perform a large amount of registration work. In this case, the burden on the user until receiving suggestions is large, and it is expected that the registered content will be insufficient and / or inaccurate. In this case, the information processing device will not be able to provide highly accurate suggestions. If the user gives up on registration, the information processing device will not be able to provide suggestions at all, even before providing highly accurate suggestions.

[0019] Therefore, in this embodiment, the above problem is solved as follows.

[0020] 1 is a diagram illustrating an overview of this embodiment. The information processing device of this embodiment makes suggestions regarding clothing selection to the user using an AI engine that has learned characteristics regarding the user's clothing selection. For example, the information processing device makes suggestions regarding clothing selection to the user (e.g., coordination suggestions) using a learning model that has learned characteristics regarding the user's clothing selection.

[0021] Here, the learning model may be a model that learns characteristics related to a user's clothing selection based on learning data that includes at least information indicating the user's past clothing selection results according to clothing selection conditions. Here, clothing selection conditions are conditions that are considered in clothing selection. For example, clothing selection conditions include at least one of weather, temperature, humidity, season, date, time, purpose, and location. Of course, clothing selection conditions may also include other conditions (e.g., the date of last wear). Furthermore, clothing selection conditions may also include user attribute information (e.g., at least one of gender, age, physical information (e.g., height and / or weight), and income level).

[0022] The training data may include, as information indicating the clothing selection result, clothing information extracted from a user image of the user and a context corresponding to the user image. Here, the context is information corresponding to the clothing selection conditions described above. For example, the context is information that is assumed to be considered by the user when selecting clothing. For example, the context may be at least one of weather, temperature, humidity, season, date, time, purpose, and location. Of course, the context may also include information other than these (e.g., the last date of wearing). The context may also include user attribute information (e.g., at least one of gender, age, physical information (e.g., height and / or weight), and income level). The device acquiring the training data may acquire the context based on metadata attached to the user image. Here, the device acquiring the training data may be an information processing device that generates suggested information or a learning device that trains a learning model. The learning device may be an information processing device that generates suggested information (e.g., a terminal device owned by the user) or another information processing device (e.g., a server on a network).

[0023] The information processing device then generates suggested information for selecting clothing for the user based on a learning model that has learned characteristics related to the user's clothing selection. For example, the learning model may be a model that outputs clothing information suggested to the user when clothing selection conditions corresponding to the context are input. In this case, the information processing device may use an image that matches the clothing information output from the learning model (e.g., "white coat" and "gray skirt") as suggested information (suggested image).

[0024] The information processing device then outputs the generated suggested information to the user (or a terminal device owned by the user). In the example of Fig. 1, the information processing device outputs, as suggested information, information on clothing selection conditions (in the example of Fig. 1, purpose ("shopping"), date ("Sunday, February 3rd"), weather ("sunny, then cloudy"), temperature ("5°C"), and location ("Tokyo")), as well as a suggested image showing tomorrow's outfit (for example, an image of a person wearing a white coat and a gray skirt).

[0025] This makes it possible to make suggestions that are in line with the characteristics of the user's clothing choices.

[0026] The information processing device may use feature information indicating features related to the user's clothing selection to make suggestions regarding clothing selection to the user. For example, the information processing device may acquire, as feature information, information that associates and accumulates user images of the user with contexts corresponding to the user images. The information processing device may then acquire, from the feature information, user images associated with contexts similar to the input clothing selection conditions. The information processing device may then use the acquired user images as suggestion information.

[0027] This also makes it possible to make suggestions that are in line with the characteristics of the user's clothing selection.

[0028] The information processing device of this embodiment may be a terminal device owned by a user or a server on a network. When the information processing device is a terminal device, the information processing device may have a user interface for inputting and / or outputting various information related to clothing selection. FIG. 2 is a diagram showing a user interface of the information processing device of this embodiment. FIG. 2 shows a registration screen, a main screen, and a calendar screen as user interfaces. The registration screen is a screen for registering user information (gender and owned clothing in the example of FIG. 2). The main screen is a screen for displaying clothing selection conditions (location, weather, temperature, and purpose in the example of FIG. 2) and suggested information (suggested text and suggested images in the example of FIG. 2). The calendar screen is a screen for displaying past information (images showing suggestion history / wearing history in the example of FIG. 2).

[0029] This allows the user to easily input registration information and obtain suggested information.

[0030] The technology of this embodiment is not limited to proposing clothes in the real world, but can also be applied to proposing clothes to be worn by characters in a game or avatars in an xR space (for example, a VR space).

[0031] The outline of this embodiment has been described above, and the information processing system 1 including the information processing device of this embodiment will be described in detail below.

[0032] <<2. Configuration of Information Processing System>> First, the configuration of the information processing system 1 will be described.

[0033] 3 is a diagram illustrating an example of the configuration of the information processing system 1 according to an embodiment of the present disclosure. The information processing system 1 is a system for making suggestions regarding clothing selection to a user.

[0034] The information processing system 1 includes one or more information processing devices. In the example of Fig. 3, the information processing system 1 includes a server 10 and a terminal device 20. Note that the devices in the figure may be considered devices in a logical sense. In other words, some of the devices in the figure may be realized by a virtual machine (VM), a container, a docker, or the like, and these may be physically implemented on the same hardware.

[0035] The server 10 and the terminal device 20 may each have a communication function. The server 10 and the terminal device 20 may be connected via a network N. In this case, the server 10 and the terminal device 20 can be referred to as communication devices. Although only one network N is shown in the example of Fig. 3, multiple networks N may exist.

[0036] Here, the network N is a communication network such as a LAN (Local Area Network), a WAN (Wide Area Network), a cellular network, a fixed telephone network, a regional IP (Internet Protocol) network, or the Internet. The network N may include a wired network or a wireless network. The network N may also include a core network. The core network is, for example, an EPC (Evolved Packet Core) or a 5GC (5G Core network). The network N may also include a data network other than the core network. The data network may be a service network of a telecommunications carrier, for example, an IMS (IP Multimedia Subsystem) network. The data network may also be a private network such as an in-house network.

[0037] The following describes in detail the configuration of each device that makes up the information processing system 1. Note that the configuration of each device shown below is merely an example. The configuration of each device may be different from the configuration shown below.

[0038] 2-1. Example of Server Configuration First, the configuration of the server 10 will be described.

[0039] The server 10 is an information processing device (computer) that provides various services to the terminal device 20 .

[0040] Any type of computer can be used as the server 10. The server 10 may be an application server or a web server. The server 10 may be a cloud server or an edge server. The server 10 may be a PC server, a mid-range server, or a mainframe server. The server 10 may be an information processing device (computer) that performs data processing (edge ​​processing) near a user or a terminal. For example, the server 10 may be an information processing device (computer) attached to or built into a base station or a roadside device. The server 10 may be an information processing device (computer) that performs cloud computing.

[0041] 4 is a diagram illustrating an example configuration of the server 10 according to an embodiment of the present disclosure. The server 10 includes a communication unit 11, a storage unit 12, and a control unit 13. Note that the configuration illustrated in FIG. 4 is a functional configuration, and the hardware configuration may be different from this. Furthermore, the functions of the server 10 may be distributed and implemented in multiple physically separated configurations. For example, the server 10 may be configured by multiple server devices.

[0042] The communication unit 11 is a communication interface for communicating with other devices. For example, the communication unit 11 is a LAN (Local Area Network) interface such as a NIC (Network Interface Card). The communication unit 11 may be a wired interface or a wireless interface. The communication unit 11 communicates with, for example, the terminal device 20 or another server 10 under the control of the control unit 13.

[0043] The storage unit 12 is a data readable / writable storage device such as a dynamic random access memory (DRAM), a static random access memory (SRAM), a flash memory, or a hard disk. The storage unit 12 stores, for example, image data and a learning model. There may be multiple pieces of image data and multiple learning models. The storage unit 12 may not store all or part of the image data and learning models.

[0044] The control unit 13 is a controller that controls each component of the server 10. The control unit 13 may be implemented by a processor such as a central processing unit (CPU) or a micro processing unit (MPU). Specifically, the control unit 13 may be implemented by a processor executing various programs stored in a storage device within the server 10 using a random access memory (RAM) or the like as a work area. The control unit 13 may be implemented by an integrated circuit such as an application specific integrated circuit (ASIC) or a field programmable gate array (FPGA). The control unit 13 may also be implemented by a graphics processing unit (GPU). A CPU, an MPU, an ASIC, an FPGA, and a GPU can all be considered controllers. The control unit 13 may be configured by multiple physically separated entities. For example, the control unit 13 may be configured by multiple semiconductor chips.

[0045] The control unit 13 includes at least one block selected from the group consisting of a recognition unit 131, a selection unit 132, a learning unit 133, a generation unit 134, a dialogue unit 135, an output control unit 136, and a discrimination unit 137. Each block constituting the control unit 13 (e.g., the recognition unit 131 to the discrimination unit 137) is a functional block that indicates the function of the control unit 13. These functional blocks may be software blocks or hardware blocks. For example, each of the above-described functional blocks may be a software module realized by software (including a microprogram), or may be a circuit block on a semiconductor chip (die). Of course, each functional block may be a processor or an integrated circuit. The control unit 13 may be configured with functional units different from the above-described functional blocks. The method of configuring the functional blocks is arbitrary.

[0046] The control unit 13 may be configured with functional units different from the above-described functional blocks. Also, some or all of the operations of each block (e.g., the recognition unit 131 to the discrimination unit 137) constituting the control unit 13 may be performed by another device. For example, some or all of the operations of each block constituting the control unit 13 may be performed by the control unit 23 of the terminal device 20 or the control unit 13 of another server 10.

[0047] <2-2. Example of the Configuration of the Terminal Device> Next, an example of the configuration of the terminal device 20 will be described.

[0048] 5 is a diagram illustrating a configuration example of the terminal device 20 according to an embodiment of the present disclosure. The terminal device 20 is an information processing device (computer) owned by a user.

[0049] The terminal device 20 is typically a smart device (smartphone or tablet), but is not limited to a smart device. Any type of information processing device (computer) can be used as the terminal device 20. For example, the terminal device 20 may be a mobile terminal such as a mobile phone, a smart device, a PDA (Personal Digital Assistant), or a notebook PC. The terminal device 20 may also be a wearable device such as a smart watch. Alternatively, the terminal device 20 may be a portable IoT (Internet of Things) device.

[0050] The terminal device 20 may be an xR device such as an AR (Augmented Reality) device, a VR (Virtual Reality) device, or an MR (Mixed Reality) device. In this case, the xR device may be a glasses-type device such as AR glasses or MR glasses, or a head-mounted device such as a VR head-mounted display. When the terminal device 20 is an xR device, the terminal device 20 may be a standalone device consisting only of a part worn by a user (e.g., a glasses part). Furthermore, the terminal device 20 may be a terminal-linked device consisting of a part worn by a user (e.g., a glasses part) and a terminal part (e.g., a smart device) linked to the part worn by a user.

[0051] As shown in Fig. 5, the terminal device 20 includes a communication unit 21, a storage unit 22, a control unit 23, an input unit 24, and an output unit 25. Note that the configuration shown in Fig. 5 is a functional configuration, and the hardware configuration may be different from this. Furthermore, the functions of the terminal device 20 may be distributed and implemented in multiple physically separated configurations.

[0052] The communication unit 21 is a communication interface for communicating with other devices. For example, the communication unit 21 is a LAN (Local Area Network) interface such as a NIC (Network Interface Card). The communication unit 21 may be a wired interface or a wireless interface. The communication unit 21 communicates with, for example, the server 10 or another terminal device 20 under the control of the control unit 23.

[0053] When the communication unit 21 has a wireless interface, the communication unit 21 may be configured to connect to a network or other communication devices using radio access technologies (RATs) such as LTE (Long Term Evolution), NR (New Radio), Wi-Fi, Bluetooth (registered trademark), etc. In this case, the communication device may be configured to be able to use different radio access technologies. For example, the communication device may be configured to be able to use NR and Wi-Fi. The communication device may also be configured to be able to use different cellular communication technologies (e.g., LTE and NR). LTE and NR are types of cellular communication technologies that enable mobile communication for communication devices by arranging multiple areas covered by base stations in a cell-like manner. Additionally, the terminal device 20 may be able to connect to a network or other communication devices using radio access technologies other than LTE, NR, Wi-Fi, and Bluetooth.

[0054] The storage unit 22 is a data readable / writable storage device such as a dynamic random access memory (DRAM), a static random access memory (SRAM), a flash memory, or a hard disk. The storage unit 22 functions as a storage means of the terminal device 20. The storage unit 22 stores, for example, image data and a learning model. There may be multiple pieces of image data and multiple learning models. Furthermore, the storage unit 22 may not store all or some of the image data and the learning model.

[0055] The control unit 23 is a controller that controls each unit of the terminal device 20. The control unit 23 may be realized by a processor such as a CPU or an MPU. In particular, the control unit 23 may be realized by a processor executing various programs stored in a storage device inside the terminal device 20 using RAM or the like as a work area. The control unit 23 may be realized by an integrated circuit such as an ASIC or an FPGA. The control unit 23 may also be realized by a GPU. A CPU, an MPU, an ASIC, an FPGA, and a GPU can all be considered controllers. The control unit 23 may be composed of multiple physically separated objects. For example, the control unit 23 may be composed of multiple semiconductor chips.

[0056] The control unit 23 includes at least one block selected from the group consisting of a recognition unit 231, a selection unit 232, a learning unit 233, a generation unit 234, a dialogue unit 235, an output control unit 236, and a discrimination unit 237. Each block constituting the control unit 23 (e.g., the recognition unit 231 to the discrimination unit 237) is a functional block that indicates a function of the control unit 23. These functional blocks may be software blocks or hardware blocks. For example, each of the above-described functional blocks may be a software module implemented by software (including a microprogram), or may be a circuit block on a semiconductor chip (die). Of course, each functional block may be a processor or an integrated circuit. The control unit 23 may be configured with functional units different from the above-described functional blocks. The method of configuring the functional blocks is arbitrary.

[0057] The control unit 23 may be configured with functional units different from the above-described functional blocks. Also, some or all of the operations of the blocks (e.g., the recognition unit 231 to the discrimination unit 237) constituting the control unit 23 may be performed by another device. For example, some or all of the operations of the blocks constituting the control unit 23 may be performed by the control unit 13 of the server 10 or the control unit 23 of another terminal device 20.

[0058] The input unit 24 is an input device that accepts various inputs from the outside. For example, the input unit 24 is an operation device such as a keyboard, a mouse, or operation keys that allows the user to perform various operations. If a touch panel is employed in the terminal device 20, the touch panel is also included in the input unit 24. In this case, the user performs various operations by touching the screen with a finger or a stylus.

[0059] The terminal device 20 may be provided with a touch sensor for operation in addition to the touch panel. For example, the terminal device 20 may be provided with a sensor area along the longitudinal direction on the outer side of the display screen that can detect contact with a user's finger. The user can perform various operations on the terminal device 20 by sliding, for example, their thumb across the area.

[0060] The output unit 25 is a device that outputs various types of information to the outside, such as sound, light, vibration, and images. The output unit 25 outputs various types of information to the user under the control of the control unit 23. The output unit 25 includes a display device that displays various types of information. The display device is, for example, a liquid crystal display or an organic electroluminescence display (OLED). In the following description, the display device included in the terminal device 20 or the screen formed by the display device may be referred to as a display screen 251. The display screen 251 may be a touch panel display device. In this case, the display screen 251 may be considered to be an integrated configuration with the input unit 24.

[0061] FIG. 6 is a diagram illustrating an example of the external appearance of the terminal device 20. The display screen 251 may be, for example, a vertically long screen as shown in FIG. 6. An upper area D1 of the display screen 251 is an area for displaying general information (e.g., time, radio wave status, battery status, etc.). A lower area D3 of the display screen 251 is an area for displaying various keys (e.g., back key, home key, menu / multitasking key). A central area D2 sandwiched between these areas is a display area for an application screen. Note that the display area for the application screen is not limited to the central area D2. For example, the terminal device 20 may use an area that combines the central area D2 with the upper area D1 and / or lower area D3 as the display area for the application screen.

[0062] If the terminal device 20 is an xR device (e.g., AR / MR glasses), the output unit 25 may be a transparent device that projects an image onto the glasses, or a retinal projection device that projects an image directly onto the user's retina. The terminal device 20 may have a means for projecting an image directly onto the user's retina, and may not have a screen as an object (e.g., a display, a screen projected onto a panel, etc.). Even in this case, if the user can recognize the screen, the screen can be considered the display screen 251 of this embodiment.

[0063] If the terminal device 20 is an xR device, the terminal device 20 may detect an action of the user attempting to touch the screen based on information from a sensor that detects the movement of the user's hand. In this case, the action of the user attempting to touch the screen may be regarded as a touch operation, and the following embodiment may be applied.

[0064] In the following description, the display screen 251 of the terminal device 20 is assumed to be a touch panel display, for example.

[0065] <2-3. Learning Model> Next, the learning model will be described.

[0066] The information processing device of this embodiment (for example, at least one of the server 10 and the terminal device 20) uses a learning model to generate information regarding suggestions to users. Note that the learning model can also be referred to as an AI (Artificial Intelligence) model, an ML (Machine Learning) model, a trained model, an inference model, or a prediction model. In the following description, the learning model may be simply referred to as a model.

[0067] In this embodiment, a program that includes a learning model in part is also referred to as a learning model. For example, a program (i.e., a generative AI) that generates information (e.g., at least one of an image, text, video, and music) using a learning model (generative AI model) is also a type of learning model.

[0068] A learning model is a machine learning model generated by a learning process using machine learning. Machine learning is, for example, one of the artificial intelligence techniques that allows a computer to perform recognition, judgment, or estimation similar to humans. The processes performed by machine learning include two processes: a learning process and a determination process for recognition, judgment, or estimation.

[0069] In the learning process, processing is performed to learn a learning model (e.g., training using learning data). In the learning process, an information processing device that performs the learning (hereinafter referred to as a learning device) uses the learning data to learn a learning model (e.g., a neural network model). For example, in the case of a neural network model, the weight coefficient of each edge is optimized during the learning process. The model extracted as a result of the learning process is sometimes referred to as a trained model.

[0070] The learning device is, for example, at least one of the server 10 and the terminal device 20, or two or more devices that cooperate with each other.

[0071] The determination process is a process for recognition, judgment, or estimation. In the determination process, for example, an information processing device that performs the determination (hereinafter referred to as a determination device) inputs data related to unknown data into a learning model (trained model). The learning model outputs a result of recognition, judgment, or estimation for the unknown data as a result of calculation processing. The determination device may be the same device as the learning device, or may be a different device.

[0072] The learning model is, for example, a neural network model. A neural network model is composed of layers called an input layer, a hidden layer (or intermediate layer), and an output layer, each of which includes a plurality of nodes, and each node is connected via an edge. Each layer has a function called an activation function, and each edge is weighted. The learning model has one or more intermediate layers (or hidden layers). When the learning model is a neural network model, learning the learning model means, for example, setting the number of hidden layers (or intermediate layers), the number of nodes in each layer, or the weight of each edge.

[0073] Here, the neural network model may be a deep learning model. By using a trained deep learning model that has been trained using a huge amount of data, the accuracy of recognition, judgment, or estimation is improved. Common algorithms used in deep learning include, for example, the following. The learning model algorithm used in this embodiment may be at least one of the following:

[0074] ・Deep Neural Network (DNN) ・Convolution Neural Network (CNN) ・Recurrent Neural Network (RNN) ・Fully Connected Neural Network (FCN) ・Long Short-Term Memory (LSTM) ・Autoencoder

[0075] A DNN has two or more hidden layers, which allows the DNN (Deep Neural Network) to achieve higher accuracy in recognition, judgment, or estimation than a conventional neural network with one hidden layer.

[0076] In a CNN, the hidden layer is composed of layers called convolution layers and pooling layers. In the convolution layers, filtering is performed using convolution operations, and data called feature maps, for example, are extracted. In the pooling layers, the information in the feature maps output from the convolution layers is compressed and down-sampling is performed.

[0077] The RNN has a network structure in which values ​​of hidden layers are recursively input to hidden layers, and is used to process, for example, short-term time-series data.

[0078] In a fully connected neural network, all of the intermediate layers are fully connected layers. In other words, in a fully connected neural network, all nodes between the intermediate layers are connected to each other. Fully connected neural networks have been widely applied, mainly in the field of speech recognition.

[0079] In LSTM, parameters called memory cells that hold the state of the hidden layer are introduced into the output of the hidden layer of the RNN. This allows LSTM to retain the influence of outputs from the distant past. In other words, LSTM processes time-series data over a longer period than RNN.

[0080] Autoencoders extract low-dimensional features that can reproduce input data through unsupervised learning. Autoencoders are useful for noise removal and dimensionality reduction.

[0081] Alternatively, the neural network model may be a Transformer.

[0082] The learning model may also be a model based on reinforcement learning, in which behaviors (settings) that maximize value are learned through trial and error. Alternatively, the learning model may be a logistic regression model.

[0083] Furthermore, the learning model may be composed of multiple models. For example, the learning model may be composed of multiple neural network models. More specifically, the learning model may be composed of multiple neural network models selected from the multiple neural network models described above (e.g., DNN, CNN, RNN, FCNN, LSTM, autoencoder, transformer, etc.). When the learning model is composed of multiple neural network models, these multiple neural network models may be in a subordinate relationship or a parallel relationship.

[0084] The learning model of this embodiment (hereinafter also referred to as learning model M) is a learning model (trained model) that has been trained using, for example, information indicating a user's past clothing selection results (for example, clothing information extracted from a user image and a context corresponding to the user image) as input data and / or correct labels (teaching data).

[0085] When the server 10 or the terminal device 20 inputs condition information (e.g., information that serves as a clothing selection condition) into the learning model M, the learning model outputs information regarding suggestions to the user (e.g., suggestion information or clothing information).

[0086] In this case, the learning model M may include an input layer that inputs condition information (e.g., information that serves as a clothing selection condition), an output layer that outputs information regarding suggestions to the user (e.g., suggestion information or clothing information), a first element that belongs to any layer between the input layer and the output layer other than the output layer, and a second element whose value is calculated based on the first element and the weight of the first element, and may be a learning model that causes a computer to function so that, for the information input to the input layer, each element belonging to each layer other than the output layer is used as the first element, and by performing calculations based on the first element and the weight of the first element (i.e., connection coefficients), the computer outputs information regarding suggestions to the user from the output layer in accordance with the condition information input to the input layer.

[0087] Here, it is assumed that the learning model M is realized by a neural network having one or more intermediate layers, such as a DNN. In this case, the first element included in the learning model corresponds to any node in the input layer or the intermediate layer. The second element corresponds to a node in the next stage, which is a node to which a value is transmitted from the node corresponding to the first element. The weight of the first element corresponds to a connection coefficient, which is a weight taken into account for the value transmitted from the node corresponding to the first element to the node corresponding to the second element.

[0088] Also, assume that the learning model M is realized as a regression model expressed as "y = a1 * x1 + a2 * x2 + ... + ai * xi". In this case, the first element included in the learning model M corresponds to input data (xi) such as x1 or x2. The weight of the first element corresponds to the coefficient ai corresponding to xi. Here, the regression model can be regarded as a simple perceptron having an input layer and an output layer. When each model is regarded as a simple perceptron, the first element corresponds to one of the nodes in the input layer, and the second element can be regarded as a node in the output layer.

[0089] The server 10 or the terminal device 20 calculates the information to be output using a model having any structure, such as a neural network or a regression model. Specifically, the learning model M has coefficients set so that, when condition information is input, information regarding suggestions to the user is output. For example, the server 10 or the terminal device 20 sets the coefficients based on the similarity between clothing information extracted from a user image of the user and a value obtained by inputting context corresponding to the user image into the learning model. The server 10 or the terminal device 20 uses such a learning model M to generate information regarding suggestions to the user from the condition information.

[0090] In the above example, a model that outputs information regarding suggestions to a user when condition information is input is shown as an example of the learning model M. However, the learning model M according to the embodiment may be a model that is generated based on results obtained by repeatedly inputting and outputting data to the learning model. Here, the learning model M may be generated or updated by online learning. Furthermore, the learning model M may be generated or updated by mini-batch learning.

[0091] Furthermore, when the server 10 or the terminal device 20 performs learning or generates output information using GAN (Generative Adversarial Networks), the learning model may be a model that constitutes part of the GAN.

[0092] The learning device that learns the learning model M may be the server 10 or the terminal device 20. For example, assume that the server 10 learns the learning model M. In this case, the server 10 learns the learning model M and stores the learned learning model M in a memory unit. More specifically, the server 10 sets the connection coefficients of the learning model M so that when condition information is input to the learning model M, the learning model outputs information regarding suggestions to the user.

[0093] For example, an information processing device (e.g., the server 10 or the terminal device 20) inputs past network quality information to a node in the input layer of the learning model M, and propagates the data through each intermediate layer to the output layer of the learning model M, causing the learning model M to output future network quality information. The information processing device then corrects the connection coefficients of the learning model M based on the difference between the value actually output by the learning model M and the value designated as the correct label (teaching data). At this time, the information processing device may correct the connection coefficients using a technique such as backpropagation. At this time, the information processing device may correct the connection coefficients based on the cosine similarity between a vector indicating the input value and a vector indicating the value actually output by the learning model.

[0094] Any learning algorithm may be used for the learning. For example, the information processing device may perform learning of the learning model using a learning algorithm such as a neural network, a support vector machine, clustering, reinforcement learning, transfer learning, fine tuning, distillation, a decision tree, ensemble learning, or a random forest.

[0095] Although the method for generating the learning model M has been described above, the above embodiment can also be applied to learning models other than the learning model M.

[0096] The learning algorithm used in this embodiment may be one that is learned by a single information processing device (for example, the server 10 or the terminal device 20) alone, or one that is learned by a plurality of information processing devices (for example, a plurality of devices selected from the server 10 and the terminal device 20) in cooperation with each other. Here, federated learning is an example of a learning algorithm in which a plurality of information processing devices learn in cooperation with each other.

[0097] <<3. Operation of Information Processing System>> The configuration of the information processing system 1 has been described above, and the operation of the information processing system 1 will now be described.

[0098] The operations of the information processing system 1 described below may be performed by one information processing device, or may be performed by multiple information processing devices working together. Here, the information processing device may be a server 10 or a terminal device 20. In the following description, the processing is performed by one information processing device, but the processing may also be performed by multiple information processing devices working together. The term "information processing device" described below or above may be replaced with "multiple information processing devices" as appropriate.

[0099] <3-1. Overview of Operation of Information Processing System> First, an overview of operation of the information processing system 1 will be described.

[0100] 7 is a diagram for explaining an outline of the operation of the information processing system 1. The operation of the information processing system 1 can be roughly divided into the following operations (A1) to (A3).

[0101] (A1) Accumulation of User Data The information processing device accumulates user data indicating the user's past clothing selection results. For example, the information processing device associates a context with each of a plurality of user images stored in the terminal device 20. Then, the information processing device accumulates the plurality of user images associated with the context as user data indicating the user's past clothing selection results. The context is, for example, information that is assumed to be considered by the user when selecting clothing. For example, the context is at least one of weather, temperature, humidity, season, date, time, purpose, and location.

[0102] Here, the weather may be, for example, the weather at a location where the user has visited (hereinafter referred to as the destination). The temperature may be the temperature at the destination (for example, at least one of the maximum temperature, minimum temperature, and average temperature). The humidity may be the humidity at the destination. The date and time may be the date and time when the user image was created. The season may be a season identified based on the date when the user image was created. The purpose may be, for example, the purpose of the user's visit. The location may be, for example, the location (destination) where the user has visited. Of course, the context may include information other than the above (for example, the date of last wear). The information processing device that accumulates the user data may be an information processing device that performs the processing of (A2) described below and / or a device different from the information processing device that performs the processing of (A3) described below.

[0103] (A2) Generation of Feature Information / Learning Model The information processing device generates feature information indicating the characteristics of a user's clothing selection based on accumulated user data. The information processing device may use the user data as feature information. Alternatively, the information processing device generates a learning model that learns the characteristics of a user's clothing selection based on the accumulated user data. For example, the information processing device uses the user data as learning data (teaching data) to train the learning model. Note that the information processing device that generates the feature information (including the information processing device that acquires the feature information) may be a device different from the device that performs the processing of (A1) above and / or the device that performs the processing of (3) described below. Similarly, the information processing device that generates the learning model (including the information processing device that trains the learning model) may be a device different from the information processing device that performs the processing of (A1) above and / or the information processing device that performs the processing of (A3) described below.

[0104] (A3) Suggestion to the User For example, the information processing device makes a suggestion according to the condition data based on the feature information and / or the learning model. Here, the condition data is, for example, data indicating clothing selection conditions corresponding to the above-mentioned context. The clothing selection conditions are conditions that are considered when selecting clothing. Here, the clothing selection conditions may include at least one of weather, temperature, humidity, season, date, time, purpose, and location. Of course, the clothing selection conditions may also include conditions other than these (e.g., the date of last wear). Note that the information processing device that makes a suggestion to the user may be an information processing device that performs the above-mentioned process (A1) and / or a device different from the information processing device that performs the above-mentioned process (A2).

[0105] <3-2. Functions of the Terminal Device> Next, the functions of the terminal device 20 will be described.

[0106] As described above, the terminal device 20 is an information processing device owned by a user. The terminal device 20 provides the user with various services related to clothing selection. The terminal device 20 has a function for registering user information, a function for suggesting clothing selection, a function for checking clothing wearing history, and a function for analyzing user characteristics related to clothing selection. Note that the terminal device 20 does not necessarily have to have all of these functions.

[0107] In the following description, it is assumed that the processes for realizing the above functions are executed by the terminal device 20 itself. However, some or all of the processes described below may be executed by another information processing device (e.g., the server 10 and / or another terminal device 20). For example, the server 10 may perform various processes based on a request from the terminal device 20, and the terminal device 20 may output information to a screen based on the processing results. For the registration function, the server 10 may perform a user image selection process, and the terminal device 20 may output the selected user image to a screen. For the suggestion function, the server 10 may generate suggestion information, and the terminal device 20 may output the generated suggestion information to a screen. For the analysis function, the server 10 may perform an analysis process, and the terminal device 20 may output information that is the analysis result to a screen. Of course, other processes may also be executed by other information processing devices.

[0108] The functions of the terminal device 20 will be described below.

[0109] <3-2-1. Registration Function> First, the registration function will be described. As described above, conventional information processing devices impose a heavy burden on the user when registering user information. In this embodiment, the terminal device 20 is equipped with the following functions to reduce the burden on the user when registering user information.

[0110] (1) User Image Selection Function In order to understand the characteristics related to a user's clothing selection (for example, to generate feature information / learning models), a large amount of user data is required. In this embodiment, the characteristics related to a user's clothing selection are understood from the user image, so a large number of user images are required. Although it is possible to require the user to register a user image, this is expected to be a very large amount of work. Therefore, in this embodiment, the terminal device 20 is equipped with an automatic user image selection function, thereby reducing the burden on the user for registering user information.

[0111] FIG. 8 is a diagram illustrating an example of a user image selection screen. The terminal device 20 acquires one or more images for user recognition from among multiple images stored in a storage device. In the example of FIG. 8, image E1 is the image for user recognition. Here, the storage device may be the terminal device's own storage unit 22 or a storage unit of another device connected via the network N (e.g., the storage unit 22 of another terminal device 20 and / or the storage unit 12 of the server 10). In this case, the terminal device 20 may acquire, as the image for user recognition, an image that is highly likely to be a photograph of the user. For example, the terminal device 20 may acquire, as the image for user recognition, a profile image of the user's SNS account. Of course, the terminal device 20 may also request the user to register an image for user recognition. Then, the terminal device 20 performs user recognition (e.g., facial recognition) based on one or more images for user recognition.

[0112] The terminal device 20 may be configured to allow the user to register an image of another person (for example, a family member, an acquaintance, a model, or another user) as an image for user recognition. In this case, the terminal device 20 may perform user recognition based on the image of the other person. When there are multiple images as images for user recognition, all of the multiple images may be images of the other person, or some of the multiple images may be images of the other person. Of course, all of the multiple images may be images of the user. When some of the multiple images are images of the other person, the terminal device 20 may recognize each of the multiple different people as a user.

[0113] The image for user recognition is not limited to an image of a person in the real world. The image for user recognition may be an image of a virtual character. The virtual character is, for example, a character in a game and / or an avatar in an xR space (for example, a VR space). In this case, the terminal device 20 may be capable of registering an image of the virtual character as an image for user recognition. Then, the terminal device 20 may perform user recognition based on the image of the virtual character. In this case, the terminal device 20 may recognize the virtual character as a user.

[0114] Next, the terminal device 20 selects a user image from among multiple images stored in the storage device based on the user recognition result. As described above, the storage device may be its own storage unit 22, or may be a storage unit of another device connected via the network N. In the example of FIG. 8, images E2 to E5 are each a user image. A user image is typically a photograph of a user. However, if an image of another user is used as the image for user recognition, the image of the other user may also be considered as the user image. Furthermore, if an image of a virtual character is used as the image for user recognition, the image of the virtual character may also be considered as the user image.

[0115] It should be noted that the terminal device 20 does not necessarily need to select all images that match the user recognition result as the user image. For example, the terminal device 20 may select, as the user image, an image that is easily visible to the user from among a plurality of images that match the user recognition result. For example, the terminal device 20 may select, as the user image, a photo that shows more than a predetermined percentage of the user's body (preferably a photo that shows the user's entire body), or an image in which the user's display area occupies more than a predetermined percentage of the entire image. Furthermore, the terminal device 20 may select, as the user image, an image whose image quality meets a predetermined standard. Of course, the terminal device 20 may select, as the user image, an image other than these.

[0116] Furthermore, the terminal device 20 may be configured to allow the user to manually select a user image. For example, the terminal device 20 may display a check box for each of multiple images selected as user images (images E2 to E5 in the example of FIG. 8). Note that the images for which check boxes are displayed are not limited to the images selected by the terminal device 20 as user images. The terminal device 20 may display multiple images stored in a storage device and display a check box for each of the multiple images. Then, the terminal device 20 may acquire one or more images selected by the user using the check box as the user image. Alternatively, the terminal device 20 may remove one or more images selected by the user using the check box from the candidate user images.

[0117] Furthermore, the terminal device 20 may associate each user image with a context corresponding to the user image at the time of selecting the user image as a user image. For example, the terminal device 20 may tag each user image with a context corresponding to the user image. The context may be, for example, at least one of weather, temperature, humidity, season, date, time, purpose, and location. In this case, the terminal device 20 may estimate the context using AI. Furthermore, the terminal device 20 may acquire / estimate the context (e.g., at least one of date, time, and location) from metadata assigned to the user image. Alternatively, the terminal device 20 may acquire / estimate the context from information accessible to the terminal device 20. For example, the terminal device 20 may acquire / estimate the context (e.g., purpose and / or location) from the user's schedule information on the day the user image was captured, or may acquire / estimate the context (e.g., weather and / or temperature) from weather information on the day the user image was captured.

[0118] The user image automatically selected by the terminal device 20 is not limited to the above. The terminal device 20 can select any image as the user image as long as it can grasp the characteristics of the user's clothing selection. For example, the terminal device 20 may select an image identified from the user's EC (electronic commerce) purchase history (e.g., an image of clothing purchased by the user) as the user image. The terminal device 20 may also select a user image based on a virtual character created or selected by the user (e.g., a character in a game and / or an avatar in an xR space). For example, the terminal device 20 may select an image of a virtual character selected by the user or an image of clothing worn by the virtual character by the user as the user image.

[0119] When button B1 is pressed, the terminal device 20 executes registration of the user image. Note that the user image acquisition function of the terminal device 20 is not limited to the above. For example, the terminal device 20 may have a function for acquiring a user image by photographing it with a camera. Alternatively, the terminal device 20 may acquire a user image by linking with an external service.

[0120] (2) Clothing Information Registration Function User information is not limited to a user image. The terminal device 20 may have a function for registering information about clothing owned by the user (hereinafter also referred to as clothing information). Figure 9 is a diagram showing an example of an input screen for clothing information.

[0121] FIG. 9 illustrates a simplified clothing registration screen that allows color selection for each item, omitting image registration. A user can easily register the colors of their clothing by checking a checkbox. The terminal device 20 may automatically select colors based on a user image and / or EC purchase history. In the example of FIG. 9 , the terminal device 20 is configured to allow selection of a color for each item. The terminal device 20 may be configured to allow selection of, for example, pattern, shape, or brand, in addition to color. The terminal device 20 may also allow selection of a style (e.g., at least one of skin exposure, tightness, and open collar). In the example of FIG. 9 , the terminal device 20 displays tops and bottoms as items. The terminal device 20 may also display, for example, coordinating items (e.g., outerwear) and / or personal items (e.g., accessories and / or bags) in addition to tops and bottoms.

[0122] The terminal device 20 may be configured to allow the user's gender to be input as user information. The terminal device 20 may also be configured to allow the user's personal color, skeletal type, date of birth, and physical information to be input. The physical information may be, for example, height, weight, body size (e.g., at least one of waist size, leg length, arm length, and shoulder width), or the size of clothing the user normally wears. The terminal device 20 may also be configured to allow the user's preferred clothing type to be input.

[0123] (3) Evaluation Information Acquisition Function The user information may be information on the user's evaluation of clothing (hereinafter referred to as evaluation information). FIG. 10 is a diagram showing an example of an evaluation information acquisition screen. For example, the terminal device 20 displays at least one image to register the user's preferences regarding clothing (coordination). For example, the terminal device 20 displays an image including elements that can learn the user's preferences regarding clothing (coordination). For example, the terminal device 20 displays an image of a person wearing the clothing to be evaluated. The terminal device 20 also displays a selection button for a positive evaluation ("LIKE" and a heart mark in the example of FIG. 10) and a selection button for a negative evaluation ("NOPE" and an X mark in the example of FIG. 10). The terminal device 20 then acquires the user's selection result (evaluation result) as evaluation information. This allows the terminal device 20 to automatically learn the user's preferences without requiring the user to register.

[0124] <3-2-2. Suggestion Function> Next, the suggestion function will be described. The terminal device 20 of this embodiment has a suggestion function for selecting clothes. In this embodiment, the terminal device 20 displays information on the clothing selection conditions in addition to the suggestion information on the screen, thereby creating a sense of satisfaction and / or a clear overview.

[0125] FIG. 11 is a diagram showing an example of a suggestion screen for selecting clothes. In the example of FIG. 11, as clothing selection conditions, a display F1 indicating a location, a display F2 indicating at least one of a date, a day of the week, weather, and temperature, and a display F3 indicating a purpose, a schedule, or a situation are displayed. The temperature is, for example, at least one of a current temperature, a maximum temperature, a minimum temperature, and an average temperature. The temperature may be the temperature at any time. The terminal device 20 may acquire weather information based on location information and a date. In this case, the terminal device 20 may display weather information for one or more days, or may display weather information for each time period within a day.

[0126] The suggestion screen may be configured to allow the user to operate a user interface (e.g., a slider) for selecting a time period. The terminal device 20 may then display the weather for the time period selected by the user using the user interface (e.g., the slider). In this case, the terminal device 20 may also change the display of information other than weather information (e.g., the display of clothing selection conditions and / or suggested information) according to the time period selected by the user. For example, the suggestion screen may be configured so that, when the user changes the time period using the slider, suggestions for clothes optimal for each time period can be viewed. This allows the user to know suggestions tailored to the time period (e.g., information such as recommended clothes for daytime and recommended clothes for nighttime).

[0127] The terminal device 20 may also select a plan (e.g., purpose and / or situation) from one or more types of schedule information. The terminal device 20 may also predict a plan using AI. For example, the terminal device 20 may predict a plan based on at least one of the user's arbitrary plan, calendar, and image location information. The terminal device 20 may also acquire a plan input by the user using a proposal screen. In the example of FIG. 11 , the proposal screen is configured to allow the user to select a plan from multiple options (e.g., "daily life," "business," etc.).

[0128] The terminal device 20 may change the background image (wallpaper image) of the proposal screen depending on the weather and / or schedule. For example, if the weather is sunny on that day, the terminal device 20 may set the background image to, for example, an image of sunny weather, and if the weather is rainy on that day, the background image may set the background image to, for example, an image of a rainy day. Furthermore, if the plan for that day is shopping, the terminal device 20 may set the background image to, for example, an image of a shopping mall, and if the plan for that day is business, the background image may set the background image to, for example, an image of a conference room. This allows the user to judge the weather and / or schedule based on a visual impression. Furthermore, the user can judge whether the weather and / or schedule are appropriate for the outfit based on the proposal image and background image.

[0129] In the example of FIG. 11 , a suggested text F4 corresponding to the clothing selection conditions is displayed as the suggested information. The suggested text may also include a reason for the suggestion. For example, the terminal device 20 may display the suggested text as, "It's warm during the day, but it gets chilly in the mornings, so bring a jacket." When the clothing selection conditions are changed, the terminal device 20 replaces the suggested image displayed on the screen. Note that there may be cases where the user is not satisfied with the suggestion. Therefore, the terminal device 20 may be configured to allow the user to input a comment on the suggested text. For example, an interface for displaying a chat screen (in the example of FIG. 11 , a ">" mark) may be displayed in the display area of ​​the suggested text. The terminal device 20 may then display the chat screen when this interface is tapped. FIG. 12 illustrates an example of a chat screen. When a user enters a comment in the text field, the suggested content is interactively changed by the text generation AI. For example, when a chat is started, the user can use text to set detailed schedules, request other suggestions, add items, adjust clothing weight / weight, etc. The user may specify to the generating AI (for example, detailed schedule settings, requests for other suggestions, adding items, adjusting clothing thickness / thinness, etc.) using images instead of or in addition to text. The terminal device 20 displays new comments and new suggested images in accordance with the user's input.

[0130] In the example of FIG. 11 , a suggested image F5 corresponding to the clothing selection conditions is displayed as the suggested information. The terminal device 20 may be configured to not display the same suggested image for a certain period of time since the last presentation, or for a certain number of times under the same conditions. The suggested image may be an image prepared in advance by the service. The suggested image may be the user's own image, an image of another user, or a favorite image imported from an external service such as a social networking service. The suggested image may also be the image with the most heart marks or other identifiers under the same conditions. The suggested image may also be an image listed for sale on an e-commerce site. The suggested image may also be an image generated by a generation AI. The suggested image may also be an image that has been processed in a predetermined manner. For example, the suggested image may be an image that has been resized to fit the display area, or may have its brightness and / or color adjusted. The suggested image may also be an image processed by a generation AI. For example, the suggested image may be an image that has been processed by the generation AI to fit the suggestion, or may be an image that has been processed by the generation AI to create a full-body image.

[0131] There may be multiple suggested images as suggested information. In this case, some of the suggested images may be displayed on the suggestion screen. In the example of FIG. 11 , a part of suggested image F6 is displayed in addition to suggested image F5. This allows the terminal device 20 to suggest to the user that there are multiple suggestions. When the user swipes the suggested image horizontally, the multiple suggested images scroll horizontally. This allows the user to view other suggested images. Note that the terminal device 20 replaces the suggested images displayed on the screen when the clothing selection conditions are changed.

[0132] The terminal device 20 may present clothing information to the user by expressing a combination of one or more colored clothing parts. In the example of FIG. 11 , an indication F7 indicating a combination of a gray top and white bottoms is added to the suggested image F5. The terminal device 20 may present not only coordination but also information about personal belongings. When a clothing item is tapped, the terminal device 20 may crop or highlight the item portion of the suggested image, or display similar images with the same conditions. The terminal device 20 may have a function for purchasing or subscribing to a product (or similar product) corresponding to the clothing information.

[0133] The terminal device 20 may have a function to add an identifier such as a heart mark to a proposed image. In the example of FIG. 11 , a heart mark F8 is added to the proposed image F5. By tapping the heart mark F8, the user can add an identifier indicating that the proposed image F5 is a favorite image (or an image of clothes actually worn). The terminal device 20 may add a display indicating the date on which the same image was registered (e.g., the most recent date) to the proposed image. The terminal device 20 may also add a display indicating the date on which the same outfit combination was registered (e.g., the most recent date) to the proposed image. The terminal device 20 may also add a display indicating the number of times the same image and / or outfit combination has been registered to the proposed image. The terminal device 20 may have a function to post photos and coordination information to a social networking service. When the user is sharing information with other users, the terminal device 20 may display the number of times the heart mark F8 has been pressed.

[0134] The image displayed when making a suggestion may include, in addition to the clothing, a suggestion on how to wear it. For example, as shown in FIG. 11 , the terminal device 20 may display a button F9 for advice on how to wear it on the screen. When the user taps the button F9, the terminal device 20 may display advice on how to wear it. The terminal device 20 may indicate the corresponding part of each item in the image with an arrow and display a comment. In the example of FIG. 11 , a comment F10 recommending that the user roll up their sleeves is displayed on the screen. １ Comment F10 recommends rolling up the hem.２ The comment may be a sentence such as "We recommend rolling up the hem" or "We recommend rolling up the sleeves."

[0135] <3-2-3. Confirmation Function> Next, the confirmation function will be described. The terminal device 20 of this embodiment has a function for confirming the history of wearing clothes. In this embodiment, the terminal device 20 displays a plurality of past outfits (for example, information about clothes worn by the user in the past) on the screen, enabling the user to avoid wearing the same outfit or to select clothes that match past outfits.

[0136] Fig. 13 is a diagram showing an example of a wearing history confirmation screen. In the example of Fig. 13, the terminal device 20 displays, as the wearing history confirmation screen, a calendar on which images indicating the wearing history (e.g., images of clothes worn by the user in the past) are added for each date. In the example of Fig. 13, the terminal device 20 displays a list of the wearing history from March 10, 2024 to April 6, 2024. This allows the user to check outfits for a certain period (e.g., one month) at a glance.

[0137] Note that no image may be displayed for days on which the user did not make a selection. In the example of FIG. 13 , an image is displayed in the image display area G1 for the 10th, but no image is displayed in the image display area for the 16th. Note that the terminal device 20 may display, in the image display area, an image to which the user has attached an identifier such as a heart mark on the proposal screen. If the user is sharing information with other users, the terminal device 20 may display, in the image display area, the image to which the most identifiers such as heart marks have been attached. Furthermore, the terminal device 20 may display, in the image display area, a favorite image (or a scene from a video) imported from an external service such as an SNS.

[0138] A schedule (e.g., a purpose and / or a scene) may be displayed for each date. For example, the terminal device 20 may add a display indicating a schedule such as "Outing," "Business," or "Daily Life" to each date. In the example of FIG. 13 , a display G3 indicating a schedule is added to the bottom of the image display area G1 for the 10th. The schedule may be selected on the proposal screen described above, or may be predicted by AI from calendar information. The display added to each date is not limited to a display indicating a schedule, and may also be, for example, a display indicating a location. Furthermore, the information added to each date may be information obtained through collaboration with an external calendar service, or information input as a calendar service.

[0139] The wearing history confirmation screen may be configured to allow the user to select a date. When the user selects a date, the terminal device 20 may highlight the selected date. In the example of FIG. 13 , the display of the 3rd (G4 shown in FIG. 13 ) is highlighted. When the user selects a date, the terminal device 20 may notify the user that there is an overlap in the outfits. For example, the terminal device 20 may highlight images of days whose outfits overlap with the selected date. In the example of FIG. 13 , the image of the 24th (G6 shown in FIG. 13 ) is highlighted. The terminal device 20 may display the cause of the overlap using an icon and / or text. Examples of causes of the overlap include wearing the same clothes as the last time the user went to the same place, wearing the same clothes as the last time the user was with a specific person (e.g., a friend; there may be multiple people), or wearing the same clothes as a certain number of days ago. The terminal device 20 may display at least one of the following as the cause of the overlap: the name of the place, the name of the person, and a number indicating the number of days.

[0140] Furthermore, when the user selects a date, the terminal device 20 may display detailed information about the selected date on the screen. In the example of Fig. 13, the detailed information about the selected date is displayed in a display area (G5 shown in Fig. 13) at the bottom of the screen. At this time, the terminal device 20 may display at least one piece of information from the suggestion-related information as the detailed information. The suggestion-related information may include, for example, at least one of the weather, schedule, clothing items, and the selected image.

[0141] The wearing history confirmation screen may be configured to display / hide the calendar by swiping the calendar layer up or down. In the example of FIG. 13 , when the user swipes up the bar G7 at the bottom of the calendar layer, the terminal device 20 hides the calendar. FIG. 14 is a diagram showing the wearing history confirmation screen with the calendar hidden. The terminal device 20 displays detailed information in chronological order in a scrollable manner. By displaying the wearing history confirmation screen as a calendar, the user can check a list of outfits on a monthly basis. Furthermore, by hiding the calendar, the user can check detailed information for several consecutive days. The wearing history confirmation screen may be configured to allow the user to create outfits for a week by tapping a future date.

[0142] <3-2-4. Analysis Function> Next, the analysis function will be described. The terminal device 20 of this embodiment has a function for analyzing characteristics related to the user's clothing selection. In this embodiment, the terminal device 20 displays the analysis results of the characteristics related to the user's clothing selection on the screen, allowing the user to know their own characteristics.

[0143] FIG. 15 is a diagram illustrating an example of an analysis screen. In the example of FIG. 15, a fashion tree diagram is displayed as the analysis screen. The terminal device 20 analyzes where the user fits in the fashion tree based on the registered clothing. The terminal device 20 then displays the analysis results on the tree diagram. In the example of FIG. 15, the terminal device 20 adds a display U1 indicating the user to the corresponding location on the tree diagram. This allows the user to know where their outfit fits in the fashion tree. The user can also learn about outfits similar to their own and / or outfits that are different from their own. In addition, the user can learn what items are needed to create an outfit that is different from their current outfit. The terminal device 20 may also make suggestions to the user based on the analysis results. The terminal device 20 may also be configured to suggest outfits that suit the user based on the user's bone structure and / or personal color.

[0144] If the user shares information with other users, the terminal device 20 may display the analysis results of the other users on the system diagram. In the example of Fig. 15, the terminal device 20 adds indicators U2 and U3 indicating the other users to the relevant locations on the system diagram. This allows the user to know their positional relationship with the other users.

[0145] FIG. 16 is a diagram illustrating an example of an analysis screen. In the example of FIG. 16 , the positional relationship of each outfit is visualized around the user's outfit (G1 in FIG. 16 ). The terminal device 20 may display the analysis results of the user's coordination style (G2 in FIG. 16 ). The terminal device 20 may also display items needed to achieve a desired style (e.g., the number of items) (G3 in FIG. 16 ). In this case, the terminal device 20 may have a function to link with an EC site so that the user can purchase necessary items that the user does not own via an EC site. For example, the analysis screen may display thumbnail images of necessary clothing items that the user does not own. When the user taps a thumbnail image, the terminal device 20 may transition the screen to a purchase screen on an EC site where the clothing item corresponding to the tapped thumbnail image can be purchased. The analysis screen may also include a purchase button. When the user taps the purchase button, the terminal device 20 may execute a process for the user to purchase the clothing item without transitioning to the EC site. The terminal device 20 may also display coordination items owned by the user (G4 in FIG. 16 ).

[0146] <3-3. Specific Example of Operation of Information Processing System> Hereinafter, the operation of the information processing system 1 will be specifically described.

[0147] <3-3-1. First Example> First, the operation of the information processing system 1 according to the first example will be described.

[0148] 17 is a diagram illustrating the operation of the information processing system 1 according to Example 1. The operation of the information processing system 1 can be broadly divided into a process of acquiring feature information indicating features related to the user's clothing selection, and a process of generating suggested information for the user based on the feature information.

[0149] First, the information processing device included in the information processing system 1 acquires user feature information. As described above, the feature information is information indicating features related to the user's clothing selection. For example, the feature information is information indicating what kind of clothing the user tends to select under specific clothing selection conditions. In the first embodiment, the feature information is information accumulated by linking a user image with a context corresponding to the user image. The context may include at least one of weather, temperature, humidity, season, date, time, purpose, and location.

[0150] A user image is typically a photograph of the user. However, as long as the characteristics of the user's clothing selection can be known, the user image is not necessarily limited to a photograph of the user. For example, the user image may be an image of a virtual character (e.g., an in-game character and / or an avatar in an xR space) generated or selected by the user. The user image may also be an image of the user himself / herself generated using a generation AI. The user image may also be an image of someone other than the user (e.g., a celebrity the user wants to use as reference for coordinating outfits). Here, the person other than the user may be a family member, an acquaintance, a model, another user, or a fictional user generated by the generation AI.

[0151] Various methods can be adopted for generating feature information. For example, the information processing device acquires a learning model that outputs a context corresponding to an image when the image is input. This learning model may be a model trained using data of multiple other users (data pairs of images and contexts) as learning data. The information processing device then inputs a user image into this learning model to generate a context corresponding to the user image. As described above, the context may include at least one of weather, temperature, humidity, season, date, time, purpose, and location. The information processing device then associates the context with the user image. The information processing device associates each of the multiple user images with a corresponding context. The information processing device acquires the data accumulated in this manner as feature information.

[0152] Next, the information processing device generates suggested information regarding clothing selection for the user based on the feature information. The information processing device that generates the suggested information may be a device different from the information processing device that acquires the feature information, or may be the same device.

[0153] Various methods can be adopted for generating the suggested information. For example, the information processing device acquires information on the user's clothing selection conditions. The clothing selection conditions may be conditions corresponding to the context. In this case, the clothing selection conditions may include at least one of weather, temperature, humidity, season, date, time, purpose, and location. The clothing selection conditions may be the conditions directly input by the user to the terminal device 20, or may be conditions estimated using AI from the user's schedule information. The information processing device then calculates the similarity between the acquired clothing selection conditions and each of multiple contexts included in the feature information. The information processing device then acquires, as suggested information, user images associated with contexts similar to the clothing selection conditions.

[0154] Next, the processes executed by the information processing device will be specifically described with reference to flowcharts. The processes executed by the information processing device are divided into a learning process, a feature information generation process, and a proposal information generation process.

[0155] First, the learning process will be described. The learning process is a process for learning a learning model used to generate feature information. FIG. 18 is a flowchart showing the learning process according to Example 1. The information processing device that executes the learning process may be the server 10 or the terminal device 20. The information processing device that executes the learning process may be a device different from the information processing device that executes the feature information generation process and / or the proposal information generation process, or may be the same device.

[0156] First, the information processing device acquires learning data (step S101). For example, the information processing device acquires data (data of pairs of images and contexts) of multiple other users as learning data. The learning data may include not only the data of other users but also the user's own data.

[0157] The information processing device then performs learning of a learning model based on the learning data acquired in step S101 (step S102). The learning model generated here is a model that has been trained to output a context corresponding to an input image when the input image is received.

[0158] When the learning of the learning model is completed, the information processing device stores the learned learning model in the storage unit (step S103). When the storage is completed, the information processing device ends the learning process.

[0159] Next, the feature information generation process will be described. The feature information generation process is a process for generating feature information using a learning model. FIG. 19 is a flowchart showing the feature information generation process according to Example 1. The information processing device that executes the feature information generation process may be the server 10 or the terminal device 20. The information processing device that executes the feature information generation process may be a device different from the information processing device that executes the learning process and / or the proposal information generation process, or may be the same device.

[0160] The information processing device acquires the learning model generated in the above-described learning process (step S201).

[0161] Next, the information processing device acquires one image from the multiple user images (step S202). The information processing device generates a context corresponding to the user image by inputting the user image into the learning model acquired in step S201 (step S203). Then, the information processing device associates the context with the user image and stores it in a storage unit (step S204). The storage unit may be a storage unit of the information processing device or a storage unit of another device connected via a network.

[0162] Next, the information processing device determines whether the context linking for all user images has been completed (step S205). If the context linking has not been completed (step S205: No), the information processing device returns to step S202. If the context linking has been completed (step S205: Yes), the information processing device ends the feature information generation process.

[0163] Next, the proposed information generation process will be described. This is a process for generating proposed information based on feature information. FIG. 20 is a flowchart showing the proposed information generation process according to Example 1. The information processing device that executes the proposed information generation process may be the server 10 or the terminal device 20. The information processing device that executes the proposed information generation process may be a device different from the information processing device that executes the learning process and / or the feature information generation process, or may be the same device.

[0164] First, the information processing device acquires information on the user's clothing selection conditions (step S301).

[0165] The information processing device then searches for a context similar to the clothing selection condition from among the multiple contexts included in the feature information (step S302). Various methods can be used to search for a context. For example, the information processing device may search for a context based on a simple degree of match, may search for a context based on a search algorithm that takes into account the similarity between contexts, or may search for a context by combining text generation AI and search expansion generation (RAG). The information processing device then acquires a user image associated with a context similar to the clothing selection condition (step S303).

[0166] The information processing device generates suggested information based on the acquired user image (step S304). The information processing device may use the user image as suggested information (suggested image) as is. In this case, the process of acquiring the user image from the storage unit (database) may be considered as generating suggested information (suggested image). Note that the suggested information generated by the information processing device may include suggested text about the clothes suggested to the user. In this case, the information processing device may generate suggested text based on a context associated with the suggested image.

[0167] Then, the information processing device outputs the proposal information to the user. For example, if the information processing device that generates the proposal information is the user's terminal device 20, the information processing device outputs the proposal information to the display screen 251. Note that if the information processing device that generates the proposal information is not the user's terminal device 20 (for example, if the information processing device is the server 10), the information processing device may output the proposal information to the user via the terminal device 20 connected via a network. In other words, if the information processing device is not the user's terminal device 20, the transmission of the proposal information to the terminal device 20 may be considered as the output of the proposal information to the user.

[0168] When the output of the proposed information is completed, the information processing device ends the feature information generation process.

[0169] According to the first embodiment, the information processing device generates feature information indicating the features of the user's clothing selection based on the user image. Then, the information processing device generates suggestion information based on the feature information. This makes it possible to make suggestions that are in line with the features of the user's clothing selection.

[0170] <3-3-2. Example 2> Next, the operation of the information processing system 1 according to Example 2 will be described. In Example 1, the information processing device generates the suggested information based on feature information indicating features related to the user's clothing selection. In Example 2, the information processing device generates the suggested information based on output from a learning model that has learned features related to the user's clothing selection.

[0171] 21 is a diagram illustrating the operation of the information processing system 1 according to Example 2. The operation of the information processing system 1 can be broadly divided into a process of learning characteristics related to a user's clothing selection and a process of generating suggested information for the user based on the learning result.

[0172] First, the information processing device included in the information processing system 1 acquires user data. For example, the information processing device acquires, as the user data, information linking a user image with a context corresponding to the user image. The context may include at least one of weather, temperature, humidity, season, date, time, purpose, and location.

[0173] A user image is typically a photograph of the user. However, as long as the characteristics of the user's clothing selection can be known, the user image is not necessarily limited to a photograph of the user. For example, the user image may be an image of a virtual character (e.g., an in-game character and / or an avatar in an xR space) generated or selected by the user. The user image may also be an image of the user himself / herself generated using a generation AI. The user image may also be an image of someone other than the user (e.g., a celebrity the user wants to use as reference for coordinating outfits). Here, the person other than the user may be a family member, an acquaintance, a model, another user, or a fictional user generated by the generation AI.

[0174] Various methods can be used to acquire a user image. For example, the information processing device may select an image from a plurality of images based on the results of user recognition (e.g., user face recognition) as the user image. Alternatively, the information processing device may select an image using the registration function (user image selection function) shown in <3-2-1> as the user image. Of course, the information processing device may also use an image specified by the user as the user image.

[0175] Various methods can be adopted to acquire a context to be linked to a user image. For example, the information processing device may use the context obtained by inputting the user image into the learning model described in Example 1 as the context to be linked to the user image. Alternatively, the information processing device may use information input by the user to the terminal device 20 as the context to be linked to the user image. The information processing device may generate a context based on metadata (e.g., date and / or location information) added to the user image. Furthermore, the information processing device may generate a context to be linked to a user image based on user schedule information (e.g., the user's schedule on the day the user image was captured). The information processing device may generate a context to be linked to a user image based on information obtained from another device via a network (e.g., weather information on the day the user image was captured).

[0176] The information processing device then verbalizes (semantizes) the user data acquired in this manner. The information processing device then stores the verbalized (semantized) information in a database as information indicating the user's past clothing selection results according to clothing selection conditions (hereinafter referred to as past information). For example, the information processing device stores information linking coordination information (also referred to as clothing information) with a context corresponding to a user image in the database as past information. The context may be the context included in the user data itself. The coordination information is information about one or more pieces of clothing extracted from the user image. The clothing information included in the coordination information is information obtained by verbalizing (semantizing) images of clothing included in the user image, such as a "black blouse" and a "skirt." In the following description, coordination information may also be referred to as clothing information. The coordination information (clothing information) may include information about the style of the clothing other than the clothing items themselves (e.g., at least one of skin exposure, tightness, and open collar). The coordination information (clothing information) may also include information about how the clothing is combined. Furthermore, the coordination information (clothing information) may include information on items other than clothing (for example, at least one of accessories, bags, shoes, and hats).

[0177] Next, the information processing device causes the learning model to learn features related to the user's clothing selection based on past information. For example, the information processing device trains the learning model based on training data that includes at least past information. The learning model generated here is a model that outputs coordination information (clothing information) to be suggested to the user when clothing selection conditions corresponding to the context are input.

[0178] Note that the past information used as learning data (information indicating the user's past clothing selection results) is not limited to information linking coordination information (clothing information) extracted from a user image with a context corresponding to the user image. The learning data may also include user evaluation information on clothing images as past information. Here, the evaluation information may be information acquired using an evaluation information acquisition function (e.g., the function described in <3-2-1> above) included in the terminal device 20. Furthermore, the learning data may also include information acquired from a virtual character created or selected by the user (e.g., information about the clothing worn by the virtual character) as past information.

[0179] Next, the information processing device generates suggested information regarding clothing selection for the user based on the learning model. The information processing device that generates the suggested information may be a device different from the information processing device that trains the learning model, or may be the same device.

[0180] Various methods can be employed to generate the suggested information. For example, the information processing device acquires information on the user's clothing selection conditions. The clothing selection conditions may be conditions corresponding to the context. In this case, the clothing selection conditions may include at least one of weather, temperature, humidity, season, date, time, purpose, and location. The clothing selection conditions may be the conditions directly input by the user to the terminal device 20, or conditions estimated using AI from the user's schedule information. The information processing device then inputs the clothing selection conditions (or information obtained by converting the clothing selection conditions into a predetermined format) into a learning model to acquire coordination information corresponding to the clothing selection conditions. The information processing device then searches for coordination information similar to the coordination information output by the learning model from multiple pieces of coordination information contained in the database. The information processing device then acquires a user image associated with the searched coordination information as suggested information.

[0181] Next, the process executed by the information processing device will be specifically described with reference to a flowchart. The process executed by the information processing device is divided into a learning process and a proposal information generation process.

[0182] First, the learning process will be described. The learning process is a process for learning characteristics related to a user's clothing selection. FIG. 22 is a flowchart showing the learning process according to Example 2. The information processing device that executes the learning process may be the server 10 or the terminal device 20. Furthermore, a plurality of information processing devices may cooperate to execute the learning process. The information processing device that executes the learning process may be a device different from the information processing device that executes the proposal information generation process, or may be the same device.

[0183] First, the information processing device acquires one image from a plurality of user images (step S401). Then, the information processing device generates coordination information (clothing information) based on the user image (step S402). For example, the information processing device extracts one or more images of people included in the user image using semantic segmentation. Then, the information processing device identifies the user's image from the extracted one or more images of people using user recognition (e.g., facial recognition). Then, the information processing device identifies to which category of clothing the clothes included in the identified image (i.e., the clothes worn by the user) belong using image classification. Then, the information processing device acquires the identification result as coordination information (clothing information).

[0184] Next, the information processing device generates a context corresponding to the user image (step S403). The information processing device may generate the context based on metadata (e.g., date and / or location information) attached to the user image. For example, the information processing device may generate the context (e.g., at least one of date, time, and season) based on date and time information attached to the user image (e.g., the date and time the user image was taken). The information processing device may generate the context (e.g., location) based on location information attached to the user image (e.g., GPS information). The information processing device may also generate a context (e.g., purpose and / or location) to be associated with the user image based on user schedule information (e.g., the user's schedule on the day the user image was taken). The information processing device may also generate a context (e.g., weather and / or temperature) to be associated with the user image based on information obtained from another device via a network (e.g., weather information on the day the user image was taken). Alternatively, the information processing device may use, as the context to be linked to the user image, a context obtained by inputting a user image into the learning model shown in Example 1. Of course, the information processing device may use information input by the user to the terminal device 20 as the context to be linked to the user image.

[0185] The information processing device then associates the context generated in step S403 with the coordination information (clothing information) acquired in step S402 and stores the associated context in a storage unit (step S404). The information stored in the storage unit here represents the results of the user's past clothing selections according to the clothing selection conditions. The information processing device may associate the context generated in step S403 and / or the coordination information (clothing information) acquired in step S402 with the user image acquired in step S401. The storage unit may be a storage unit of the information processing device or a storage unit of another device connected via a network.

[0186] Next, the information processing apparatus determines whether the context linking has been completed for all user images (step S405). If the context linking has not been completed (step S405: No), the information processing apparatus returns the process to step S401.

[0187] If the process has been completed (step S405: Yes), the information processing device acquires the information stored in the storage unit in step S404 as training data (step S406). Then, the information processing device causes a training model to learn features related to the user's clothing selection based on the acquired training data (step S407). The training model generated here is a model that outputs coordination information (clothing information) to be suggested to the user when clothing selection conditions corresponding to the context are input. Note that the device that generates the training data (e.g., the device that performs the processes of steps S401 to S405) and the device that performs the training (e.g., the device that performs the processes of steps S406 to S407) may be different devices.

[0188] When the learning of the learning model is completed, the information processing device stores the learned learning model in the storage unit (step S408). When the storage is completed, the information processing device ends the learning process.

[0189] Next, the proposed information generation process will be described. This is a process for generating proposed information based on feature information. FIG. 23 is a flowchart showing the proposed information generation process according to Example 2. The information processing device that executes the proposed information generation process may be the server 10 or the terminal device 20. Furthermore, a plurality of information processing devices may cooperate to execute the proposed information generation process. The information processing device that executes the proposed information generation process may be a device different from the information processing device that executes the learning process, or may be the same device.

[0190] First, the information processing device acquires the learning model generated in the learning process (step S501). Note that the information processing device can also have another information processing device (e.g., a server on a network) process the learning model. In this case, the information processing device may skip step S501.

[0191] Next, the information processing device acquires information on the user's clothing selection conditions (step S502). The information processing device then inputs the clothing selection conditions into a learning model to acquire coordination information corresponding to the clothing selection conditions (step S503). Note that if another information processing device is executing the processing of the learning model, the information processing device may input the clothing selection conditions into the learning model via a network. As described above, the learning model is a model that outputs coordination information (clothing information) suggested to the user when clothing selection conditions corresponding to the context are input. The information processing device then searches for coordination information similar to the coordination information acquired in step S503 from the multiple pieces of coordination information saved in step S404. The information processing device then acquires a user image associated with the searched coordination information (step S504).

[0192] The information processing device generates suggested information based on the acquired user image (step S505). The information processing device may use the user image as suggested information (suggested image) as is. In this case, the process of acquiring the user image from the storage unit (database) may be considered as generating suggested information (suggested image).

[0193] The proposal information generated by the information processing device may include proposal text about the clothes proposed to the user. In this case, the information processing device may generate the proposal text based on coordination information output from the learning model (or coordination information linked to the proposal image). For example, the information processing device may use information about the type and / or color of clothes, such as a "black blouse" and / or a "skirt," as the proposal text.

[0194] When the generation of the proposal information is completed, the information processing device outputs the proposal information to the user (step S506). If the information processing device that generates the proposal information is the user's terminal device 20, the information processing device may output the proposal information to the display screen 251. If the information processing device that generates the proposal information is not the user's terminal device 20 (for example, if the information processing device is the server 10), the information processing device may output the proposal information to the user via the terminal device 20. In other words, if the information processing device is not the user's terminal device 20, the transmission of the proposal information to the terminal device 20 may be considered as the output of the proposal information to the user.

[0195] When the output of the proposed information is completed, the information processing device ends the proposed information generation process.

[0196] According to the second embodiment, the information processing device trains a learning model to learn the characteristics of a user's clothing selection based on a user image. Then, the information processing device generates suggested information based on the learning model. This makes it possible to make suggestions that are in line with the characteristics of the user's clothing selection.

[0197] <3-3-3. Example 3> Next, the operation of the information processing system 1 according to Example 3 will be described. In Example 2, the information processing device sets a user image selected from a plurality of user images stored in a storage unit as suggested information for a user. In Example 3, the information processing device generates suggested information using a generation AI.

[0198] 24 is a diagram illustrating the operation of the information processing system 1 according to Example 3. The operation of the information processing system 1 can be broadly divided into a process of learning characteristics related to a user's clothing selection and a process of generating suggested information for the user based on the learning result.

[0199] First, the information processing device included in the information processing system 1 acquires user data. For example, the information processing device acquires, as the user data, information linking a user image with a context corresponding to the user image. The context may include at least one of weather, temperature, humidity, season, date, time, purpose, and location.

[0200] A user image is typically a photograph of the user. However, as long as the characteristics of the user's clothing selection can be known, the user image is not necessarily limited to a photograph of the user. For example, the user image may be an image of a virtual character (e.g., an in-game character and / or an avatar in an xR space) generated or selected by the user. The user image may also be an image of the user himself / herself generated using a generation AI. The user image may also be an image of someone other than the user (e.g., a celebrity the user wants to use as reference for coordinating outfits). Here, the person other than the user may be a family member, an acquaintance, a model, another user, or a fictional user generated by the generation AI.

[0201] Various methods can be used to acquire a user image. For example, the information processing device may select an image from a plurality of images based on the results of user recognition (e.g., user face recognition) as the user image. Alternatively, the information processing device may select an image using the registration function (user image selection function) shown in <3-2-1> as the user image. Of course, the information processing device may also use an image specified by the user as the user image.

[0202] Various methods can be adopted to acquire a context to be linked to a user image. For example, the information processing device may use a context obtained by inputting a user image into the learning model described in Example 1 as the context to be linked to the user image. Alternatively, the information processing device may use information input by a user to the terminal device 20 as the context to be linked to the user image. The information processing device may generate a context based on metadata (e.g., date and / or location information) added to the user image. Furthermore, the information processing device may generate a context to be linked to a user image based on user schedule information (e.g., the user's schedule on the day the user image was captured). The information processing device may generate a context to be linked to a user image based on information obtained via a network (e.g., weather information on the day the user image was captured).

[0203] The information processing device then verbalizes (semantizes) the user data. The information processing device then stores the verbalized (semantized) information in a database as information indicating the results of the user's past clothing selections in accordance with the clothing selection conditions (hereinafter referred to as past information). For example, the information processing device stores information linking coordination information (also referred to as clothing information) with a context corresponding to the user image in the database as past information. The context may be the context included in the user data as is. The coordination information is information on one or more pieces of clothing extracted from the user image. The coordination information (clothing information) may also include information on items other than clothing (e.g., accessories and / or bags).

[0204] Next, the information processing device causes the generative AI to learn characteristics related to the user's clothing selection based on the past information. For example, the information processing device trains the generative AI model based on learning data that includes at least the past information.

[0205] A generative AI model is a learning model (e.g., an image generation model and / or a text generation model) that generates new data having the characteristics of data. A generative AI model may also be referred to as a generative model. In the following description, a generative AI model may simply be referred to as a generative AI.

[0206] The generative AI may be a variational autoencoder (VAE), a generative adversarial network (GAN), a diffusion model, or a flow-based generative model. Of course, the generative AI is not limited to these. Various types of models can be adopted for the generative AI. Furthermore, the information generated by the generative AI is not limited to images and text, and may be, for example, video or music.

[0207] In Example 3, when information prompting a clothing selection condition corresponding to a context is input, the generation AI generates at least one of a suggested image and suggested text. The suggested image is an image of content related to clothing suggested to the user. The suggested text is text of content related to clothing suggested to the user.

[0208] Next, the information processing device generates suggested information regarding clothing selection for the user using the generation AI. The information processing device that generates the suggested information may be a device different from the information processing device that trains the learning model, or may be the same device.

[0209] Various methods can be adopted for generating the suggested information. For example, the information processing device acquires information on the user's clothing selection conditions. The clothing selection conditions may be the same as the clothing selection conditions described in Example 2. Then, the information processing device inputs information in the form of prompts for the clothing selection conditions into the generation AI, thereby acquiring suggested information (e.g., suggested images and / or suggested text) for the user.

[0210] Next, the process executed by the information processing device will be specifically described with reference to a flowchart. The process executed by the information processing device is divided into a learning process and a proposal information generation process.

[0211] First, the learning process will be described. The learning process is a process for learning characteristics related to a user's clothing selection. FIG. 25 is a flowchart showing the learning process according to Example 3. The information processing device that executes the learning process may be the server 10 or the terminal device 20. Furthermore, a plurality of information processing devices may cooperate to execute the learning process. The information processing device that executes the learning process may be a device different from the information processing device that executes the proposal information generation process, or may be the same device.

[0212] First, the information processing device acquires one image from a plurality of user images (step S601). Then, the information processing device generates coordination information (clothing information) based on the user image (step S602). Then, the information processing device generates a context corresponding to the user image (step S603). Then, the information processing device associates the context generated in step S603 with the user image acquired in step S601 and stores the context in a storage unit (step S604). The processing of steps S601 to S604 is the same as the processing of steps S401 to S404 in the second embodiment.

[0213] Next, the information processing apparatus determines whether the context linking has been completed for all user images (step S605). If the context linking has not been completed (step S605: No), the information processing apparatus returns the process to step S601.

[0214] If the process is complete (step S605: Yes), the information processing device acquires the information stored in the storage unit in step S604 as training data (step S606). Then, the information processing device causes the generation AI to learn characteristics related to the user's clothing selection based on the acquired training data (step S607). The generation AI generated here is a model that generates at least one of suggested images and suggested text when information prompting the clothing selection conditions corresponding to the context is input. Note that the device that generates the training data (e.g., the device that performs the processes of steps S601 to S605) and the device that performs the learning (e.g., the device that performs the processes of steps S606 to S607) may be different devices.

[0215] Note that the past information used as learning data (information indicating the user's past clothing selection results) is not limited to information linking coordination information (clothing information) extracted from a user image with a context corresponding to the user image. The learning data may also include user evaluation information on clothing images as past information. Here, the evaluation information may be information acquired using an evaluation information acquisition function (e.g., the function described in <3-2-1> above) included in the terminal device 20. Furthermore, the learning data may also include information acquired from a virtual character created or selected by the user (e.g., information about the clothing worn by the virtual character) as past information.

[0216] When the learning of the generated AI is completed, the information processing device stores the learned generated AI in the storage unit (step S608). When the storage is completed, the information processing device ends the learning process.

[0217] Next, the proposed information generation process will be described. This is a process for generating proposed information based on feature information. FIG. 26 is a flowchart showing the proposed information generation process according to Example 3. The information processing device that executes the proposed information generation process may be the server 10 or the terminal device 20. Furthermore, a plurality of information processing devices may cooperate to execute the proposed information generation process. The information processing device that executes the proposed information generation process may be a device different from the information processing device that executes the learning process, or may be the same device.

[0218] First, the information processing device acquires a generation AI that generates suggested information (step S701). The generation AI acquired here may be a generation AI generated by the above-mentioned learning process, or may be a general-purpose generation AI. Note that the information processing device can also use the generation AI via a network. In this case, the information processing device may skip step S701.

[0219] Next, the information processing device acquires information on the user's clothing selection conditions (step S702). Then, the information processing device generates a prompt based on the clothing selection conditions (step S703). The prompt is an instruction and / or question to be input to the generation AI.

[0220] Various methods can be adopted for generating prompts. For example, the information processing device may generate prompts using a learning model (text generation AI) that converts clothing selection conditions into text. In this case, the learning model that generates the prompts may be trained (e.g., fine-tuned) to suit the user. Of course, the information processing device may also generate prompts without using a learning model. For example, the information processing device may generate prompts by inserting words corresponding to the clothing selection conditions into predetermined positions in template text.

[0221] The information processing device then inputs the generated prompt to the generation AI to generate suggested information for the user (e.g., suggested image and / or suggested text) (step S704). If a server on the network has the function of the generation AI, the information processing device inputs the prompt to the generation AI via the network. As described above, the generation AI is a model that generates at least one of a suggested image and suggested text when information in the form of a prompt representing clothing selection conditions corresponding to a context is input.

[0222] When the information processing device generates a proposal image as proposal information, the generated proposal image may be an image of a person wearing the proposed clothes against a background of scenery corresponding to the clothing selection conditions. For example, assume that the clothing selection conditions include weather information. In this case, if the weather is sunny, the information processing device may set the background of the proposal image (e.g., an image of the back of the person wearing the proposed clothes) to, for example, an image of sunny weather, and if the weather is rainy, the background of the proposal image may be set to, for example, an image of rainy weather. Also, assume that the clothing selection conditions include purpose information. If the purpose is shopping, the terminal device 20 may set the background of the proposal image (e.g., an image of the back of the person wearing the proposed clothes) to, for example, an image of a shopping mall, and if the purpose is business, the background of the proposal image may be set to, for example, an image of a conference room. This allows the user to visually determine whether the clothing selection conditions and the proposal are appropriate based on the background included in the proposal image.

[0223] Then, the information processing device outputs the generated proposal information to the user (step S705). When the information processing device that generates the proposal information is the user's terminal device 20, the information processing device may output the proposal information to the display screen 251. When the information processing device that generates the proposal information is not the user's terminal device 20 (for example, when the information processing device is the server 10), the information processing device may output the proposal information to the user via the terminal device 20. In other words, when the information processing device is not the user's terminal device 20, the transmission of the proposal information to the terminal device 20 may be considered as the output of the proposal information to the user.

[0224] When the output of the proposed information is completed, the information processing device ends the proposed information generation process.

[0225] According to Example 3, the information processing device causes the generation AI to learn the characteristics of the user's clothing selection based on the user image. Then, the information processing device generates suggested information based on the generation AI. This makes it possible to make suggestions that are in line with the characteristics of the user's clothing selection.

[0226] <<4. Modifications>> The above-described embodiment is merely an example, and various modifications and applications are possible.

[0227] 4-1. Variations Related to Output Format In the above-described embodiments (e.g., Examples 1 to 3), the information processing device unilaterally outputs suggested information (suggested images and / or suggested text) to the user. However, the information processing device may also output suggested information to the user in an interactive format. In this case, the information processing device may have an interactive generation AI function (hereinafter referred to as a chat function) that enables text-based and / or image-based dialogue with the user. In a text-based dialogue, the dialogue with the generation AI (e.g., instructions / questions to the generation AI and / or responses from the generation AI) is text-based. In an image-based dialogue, the dialogue with the generation AI (e.g., instructions / questions to the generation AI and / or responses from the generation AI) includes images. The information processing device may then use the chat function to communicate with the user to understand the clothing selection conditions considered by the user. Note that the user may not be satisfied with the clothing presented as suggested information. In this case, the information processing device may also use the chat function to communicate with the user to understand the clothing selection conditions considered by the user. At this time, the information processing device may communicate with the user using the chat screen (FIG. 12) described in <3-2-2> above. The information processing device may then generate suggested information based on the clothing selection conditions that have been grasped and output the generated suggested information to the user. Here, the information processing device may make a plurality of different suggestions, for example, as shown in FIG. 12.

[0228] <4-2. Modifications Related to Output Information> In the above-described embodiments (e.g., Examples 1 to 3), the information processing device outputs suggested information (suggested images and / or suggested text) to the user. However, the information processing device may output information about clothes previously worn by the user to the user to enable the user to avoid overlapping with other users. Here, the information about clothes previously worn by the user may be information that enables the user to determine whether or not there is overlapping with other users. For example, the information about clothes previously worn by the user may include at least one of information about the most recent date the user wore the clothes indicated as suggested information, information about locations the user visited while wearing the clothes indicated as suggested information, information about people who accompanied the user when wearing the clothes indicated as suggested information in the past, and information about the number of times the user wore the clothes indicated as suggested information within a certain period of time in the past. The information processing device may output information about clothes previously worn by the user using the clothing wearing history confirmation function described in <3-2-3> above (the function described using FIG. 13 and / or FIG. 14 ).

[0229] Various methods can be employed to acquire information about clothes previously worn by the user. For example, the information processing device determines whether clothes displayed as suggested information match clothes previously worn by the user. For example, assume that a location is included in the clothing selection conditions. In this case, the information processing device may determine whether clothes displayed as suggested information match clothes previously worn by the user at the location specified as the clothing selection conditions (e.g., whether the type and / or color of the clothes match). In this case, the information processing device may determine whether the clothes match or not based on clothing information obtained by image recognition. The information processing device may then acquire information about the results of the determination as information about clothes previously worn by the user.

[0230] The information processing device may determine whether or not overlapping clothing occurs based on a predetermined rule. That is, the information processing device may determine whether or not overlapping clothing occurs on a loose basis. Here, the predetermined rule may be, for example, whether or not the clothes displayed as the suggested information match the clothes worn the most recent time at the location that is the clothing selection condition. Furthermore, the predetermined rule may be, for example, whether or not the clothes displayed as the suggested information match any of the clothes worn the most recent several times (e.g., three times) at the location that is the clothing selection condition.

[0231] Of course, the information processing device may determine whether or not overlapping clothing occurs based on learning. For example, the information processing device acquires a learning model that learns user characteristics related to overlapping clothing based on information about the user's past clothing selection results. Here, the learning model may be a model that learns user characteristics related to overlapping clothing, such as whether the user tends to avoid overlapping clothing, what clothing selection conditions the user tends to use to avoid overlapping clothing, how many recent times the user tends to avoid overlapping clothing in the same location, and how many recent times the user tends to avoid overlapping clothing with the same person with whom the user plans to be together. The information processing device may train the learning model using information indicating the user's past clothing selection results as training data. Then, the information processing device may use the learning model to determine whether or not overlapping clothing occurs (i.e., whether the user considers the clothing (coordinates) shown in the suggested information to be overlapping clothing). Note that the information processing device that trains the learning model and the information processing device that determines whether or not overlapping clothing occurs may be different devices or the same device. In addition to determining the user's characteristics regarding overlapping clothing through learning, the information processing device may also allow the user to specify, for example, how many recent instances of overlapping clothing the user wishes to avoid.

[0232] The information processing device may also exclude information about clothes that overlap with other users from the suggested information, and may output the suggested information from which the information about clothes that overlap with other users has been excluded to the user.

[0233] <4-3. Modifications Related to Learning> In the above-described embodiments (for example, Examples 1 to 3), the information processing device learned the learning model using user data as learning data. At this time, the learning data may include not only the user's own data but also data of other users. This allows a highly accurate learning model to be generated, enabling the information processing device to make more accurate suggestions.

[0234] Furthermore, the information processing device may generate a learning model used to generate feature information / suggestion information by fine-tuning a trained learning model. For example, the information processing device may generate a learning model used to generate feature information / suggestion information by retraining a part or all of a trained model trained based on user data of multiple other users with the user's data. This makes it possible to generate a learning model specialized for the user with a short learning time.

[0235] Furthermore, when the information processing device acquires new user data, the information processing device may train the learning model using the newly acquired user data, thereby obtaining a learning model with higher performance over time.

[0236] 4-4. Modifications Related to Functions of Terminal Device In the above-described embodiment, the functions of terminal device 20 are exemplified as a function for registering user information, a function for suggesting clothing selection, a function for checking clothing wearing history, and a function for analyzing user characteristics related to clothing selection. However, the functions of terminal device 20 are not limited to these, and may also include, for example, a function for notifying the user.

[0237] For example, if the clothing selection conditions that were the premise for generating the suggested information change when the user views the suggestion (for example, if the weather on that day changes from sunny to rainy), the terminal device 20 may notify the user (for example, by push notification). At this time, the terminal device 20 may output suggested information that has been generated based on the changed clothing selection conditions to the user. The suggested information may be generated by the terminal device 20 or another information processing device.

[0238] The timing at which the terminal device 20 notifies the user is not limited to the timing at which the clothing selection conditions change. For example, the terminal device 20 may notify the user (e.g., push notification) a certain time before an event scheduled by the user (e.g., an event obtained from the user's schedule information) or at a predetermined time on the day of the event (e.g., at least one of the morning, afternoon, and evening). Alternatively, the terminal device 20 may learn the timing at which the user performs a predetermined action (e.g., the timing at which the user operates the terminal device 20 to check the schedule) and notify the user at the learned timing. The terminal device 20 may notify the user together with suggested information. The suggested information may be generated by the terminal device 20 or another information processing device.

[0239] The terminal device 20 may also have a function for registering information about items owned by the user (for example, at least one of clothes, bags, and accessories). The information about the items owned by the user may include an image of the item (for example, an image of clothes). The terminal device 20 may output an image of the registered item as a suggested image. The suggested image may be generated by the terminal device 20 or another information processing device.

[0240] Furthermore, the information about the items owned by the user may include information other than images of the items. For example, the information about the clothes owned by the user may include at least one of the following information: type, color, pattern, material, price, and purchase date. The information about the items owned by the user may also include information about the storage location of the items. The information about the items owned by the user may also include the user's rating of the items (e.g., information indicating the degree to which the item is a favorite). The information about the items owned by the user may also include the current status of the items (e.g., whether or not they are being washed) and / or information about the number of times the items have been used. When an item owned by the user is selected as a suggested item, the terminal device 20 may output the information about the item owned by the user as suggested information. The suggested information may be generated by the terminal device 20 or another information processing device.

[0241] 4-5. Modifications Regarding Proposed Images In the above-described embodiments, the information processing device used a user image or an image generated by the generation AI as the proposed image. Here, the proposed image is not limited to a 2D image but may be a 3D image. In this case, the 3D image may be configured to be rotatable (e.g., horizontally and / or vertically) in response to a user's swipe operation, etc., so that the user can view the clothes (or a person wearing the clothes) from various angles. Of course, the proposed image may be a past photograph of the user. Furthermore, the proposed image is not limited to a still image but may be a video (video). In this case, sound may be added to the proposed image. Furthermore, the information processing device may output information about the clothes to the user through haptic feedback. For example, when the user touches the proposed image on the touch panel, the information processing device may provide feedback to the user about the tactile feel of the clothes by vibrating an output unit (e.g., the output unit 25 of the terminal device 20).

[0242] The proposed image may also be an image showing a person wearing the proposed clothes. In this case, the background of the proposed image (the image behind the person wearing the proposed clothes) may be an image of scenery according to the clothing selection conditions. If there is an image of a place that satisfies the clothing selection conditions, the information processing device may use that image as the background of the proposed image. The information processing device may acquire an image to be used as the background of the proposed image via a network. If there is a photo of the relevant place among multiple user images held by the user, the information processing device may use that photo as the background of the proposed image.

[0243] 4-6. Other Modifications The control device that controls the server 10 or the terminal device 20 of this embodiment may be realized by a dedicated computer system or a general-purpose computer system.

[0244] For example, a program for executing the above-described operations may be stored and distributed on a computer-readable recording medium such as an optical disk, a semiconductor memory, a magnetic tape, or a flexible disk. Then, for example, the program may be installed on a computer and the above-described process may be executed to configure a control device. In this case, the control device may be a device external to the server 10 or the terminal device 20 (e.g., a personal computer). Alternatively, the control device may be a device internal to the server 10 or the terminal device 20 (e.g., the control unit 13 or the control unit 23).

[0245] The above program may also be stored in a storage device provided in a server device on a network such as the Internet, and may be downloaded to a computer. The above functions may also be realized by cooperation between an operating system (OS) and application software. In this case, the parts other than the OS may be stored on a medium and distributed, or the parts other than the OS may be stored in a server device and may be downloaded to a computer.

[0246] Furthermore, among the processes described in the above embodiments, all or part of the processes described as being performed automatically can be performed manually, or all or part of the processes described as being performed manually can be performed automatically using a known method. In addition, the information including the processing procedures, specific names, various data, and parameters shown in the above documents and drawings can be changed as desired unless otherwise specified. For example, the various information shown in each drawing is not limited to the information shown in the drawings.

[0247] Furthermore, the components of each device shown in the figure are conceptual functional components and do not necessarily have to be physically configured as shown in the figure. In other words, the specific form of distribution and integration of each device is not limited to that shown in the figure, and all or part of them can be functionally or physically distributed and / or integrated in any unit depending on various loads, usage conditions, etc.

[0248] The above-described embodiments can be combined as appropriate within the scope of the present invention without causing any inconsistency in the processing content. The order of the steps shown in the flowcharts of the above-described embodiments can be changed as appropriate.

[0249] Furthermore, for example, the present embodiment can be implemented as any configuration constituting an apparatus or system. For example, the present embodiment can be implemented as a processor as a system LSI (Large Scale Integration), a module using multiple processors, a unit using multiple modules, or a set in which a unit further has additional functions. In other words, the present embodiment can also be implemented as a part of the configuration of an apparatus.

[0250] In this embodiment, a system refers to a collection of multiple components (devices, modules (components), etc.), regardless of whether all of the components are in the same housing. Therefore, multiple devices housed in separate housings and connected via a network, and a single device in which multiple modules are housed in a single housing, are both systems.

[0251] Furthermore, for example, this embodiment can have a cloud computing configuration in which one function is shared and processed jointly by a plurality of devices via a network.

[0252] <<5. Conclusion>> As described above, according to this embodiment, an information processing device (e.g., the server 10 and / or the terminal device 20) acquires stored information linking user images with contexts corresponding to the user images as feature information indicating features related to a user's clothing selection. The information processing device then generates suggested information based on the feature information. For example, the information processing device generates suggested information based on the user images linked with the contexts similar to clothing selection conditions.

[0253] This allows the information processing device to make suggestions that are in line with the characteristics of the user's clothing selection (for example, suggestions that are in line with the user's clothing selection logic), thereby reducing the burden on the user when selecting clothing (coordination).

[0254] Furthermore, the information processing device (e.g., the server 10 and / or the terminal device 20) acquires a learning model that learns characteristics related to the user's clothing selection based on learning data that includes at least information indicating the user's past clothing selection results according to the clothing selection conditions, and then generates suggested information based on the learning model.

[0255] This allows the information processing device to make suggestions that are in line with the characteristics of the user's clothing selection (for example, suggestions that are in line with the user's clothing selection logic), thereby reducing the burden on the user when selecting clothing (coordination).

[0256] Although the embodiments of the present disclosure have been described above, the technical scope of the present disclosure is not limited to the above-described embodiments, and various modifications are possible within the scope of the gist of the present disclosure. Furthermore, components of different embodiments and modifications may be combined as appropriate.

[0257] Furthermore, the effects of each embodiment described in this specification are merely examples and are not limiting, and other effects may also be obtained.

[0258] The present technology may also be configured as follows. (1) An information processing device comprising: a generation unit that generates suggested information regarding clothing selection for a user based on information indicating characteristics related to the user's clothing selection or a learning model that has learned the characteristics related to the user's clothing selection. (2) The information processing device described in (1), wherein the learning model is a model that has learned the characteristics related to the user's clothing selection based on learning data that includes at least information indicating the user's past clothing selection results according to clothing selection conditions, and the generation unit generates the suggested information based on the learning model. (3) The information processing device described in (2), wherein the learning data includes, as information indicating the clothing selection results, clothing information extracted from a user image and a context corresponding to the user image, and the context includes at least one of weather, temperature, humidity, season, date, time, purpose, and location. (4) The information processing device according to (3), wherein the learning model is a model that outputs clothing information to be suggested to the user when clothing selection conditions corresponding to the context are input, and the generation unit generates the suggested information based on the clothing information generated using the learning model. (5) The learning model is a generation AI that, when information in the form of prompts for the clothing selection conditions corresponding to the context is input, generates at least one of a suggested image of content related to clothing to be suggested to the user and suggested text of content related to clothing to be suggested to the user, and the generation unit acquires at least one of the suggested image and the suggested text generated using the generation AI as the suggested information. (6) The information processing device according to any one of (2) to (5), wherein the learning data includes, as information indicating the clothing selection result, evaluation information of the user on clothing images. (7) The information processing device according to any one of (2) to (6), wherein the learning data includes, as information indicating the clothing selection result, information acquired from a character generated or selected by the user. (8) The information processing device according to any one of (2) to (7), further comprising: a learning unit that performs learning of the learning model.(9) The information processing device according to (8), comprising: a recognition unit that recognizes the user in an image; and a selection unit that selects an image in which the user appears from a plurality of images based on a recognition result, wherein the learning unit learns the learning model based on the selected image. (10) The information processing device according to any one of (2) to (9), wherein the generation unit generates the proposal information according to the clothing selection conditions based on the learning model, wherein the clothing selection conditions include at least one of weather, temperature, humidity, season, date, time, purpose, and location. (11) The information processing device according to (10), wherein the generation unit generates, as the proposal information, at least one of a proposal image of content related to clothing to be proposed to the user and a proposal text of content related to clothing to be proposed to the user. (12) The information processing device according to (11), wherein the generation unit generates at least one of the proposal image and the proposal text using a generation AI. (13) The information processing device according to (11) or (12), wherein the generation unit generates, as the proposed image, an image in which a person wearing the proposed clothes is displayed against a background of scenery according to the clothing selection conditions. (14) The information processing device according to any one of (10) to (13), further comprising: a dialogue unit that has a text-based or image-based dialogue with the user regarding the clothes indicated as the proposed information. (15) The information processing device according to any one of (10) to (14), further comprising: an output control unit that displays information regarding clothes previously worn by the user. (16) The information processing device according to (15), further comprising: a determination unit that determines whether the clothes indicated as the proposed information match clothes previously worn by the user, and the output control unit outputs information regarding the result of the determination as information regarding the clothes previously worn by the user. (17) The information processing device according to (16), further comprising: a location, and the determination unit that determines whether the clothes indicated as the proposed information match clothes previously worn by the user in the location set as the clothing selection conditions.(18) The information processing device according to (1), wherein the information indicating characteristics related to the user's clothing selection is information accumulated by linking a user image with a context corresponding to the user image, the context including at least one of weather, temperature, humidity, season, date, time, purpose, and location, and the generation unit generates the user image linked to the context similar to clothing selection conditions as the suggested information. (19) An information processing method, wherein suggested information related to clothing selection for the user is generated based on information indicating characteristics related to the user's clothing selection or a learning model that has learned the characteristics related to the user's clothing selection. (20) A program for causing a computer to function as a generation unit that generates suggested information related to clothing selection for the user based on information indicating characteristics related to the user's clothing selection or a learning model that has learned the characteristics related to the user's clothing selection.

[0259] REFERENCE SIGNS LIST 1 Information processing system 10 Server 20 Terminal device 11, 21 Communication unit 12, 22 Storage unit 13, 23 Control unit 24 Input unit 25 Output unit 131, 231 Recognition unit 132, 232 Selection unit 133, 233 Learning unit 134, 234 Generation unit 135, 235 Dialogue unit 136, 236 Output control unit 137, 237 Discrimination unit 251 Display screen F4 Suggested text F5, F6 Suggested image N Network

Claims

1. An information processing device comprising: a generation unit that generates suggested information regarding clothing selection for a user based on information indicating characteristics regarding the user's clothing selection or a learning model that has learned characteristics regarding the user's clothing selection.

2. The information processing device described in claim 1, wherein the learning model is a model that learns characteristics related to the user's clothing selection based on learning data that includes at least information indicating the user's past clothing selection results in accordance with clothing selection conditions, and the generation unit generates the suggested information based on the learning model.

3. The information processing device of claim 2, wherein the learning data includes, as information indicating the clothing selection result, clothing information extracted from the user image and a context corresponding to the user image, and the context includes at least one of weather, temperature, humidity, season, date, time, purpose, and location.

4. The information processing device described in claim 3, wherein the learning model is a model that outputs clothing information to be suggested to the user when clothing selection conditions corresponding to the context are input, and the generation unit generates the suggested information based on the clothing information generated using the learning model.

5. The information processing device of claim 3, wherein the learning model is a generation AI that generates at least one of a suggested image of content related to clothes to be suggested to the user and suggested text of content related to clothes to be suggested to the user when information that prompts the clothing selection conditions corresponding to the context is input, and the generation unit acquires at least one of the suggested image and the suggested text generated using the generation AI as the suggested information.

6. The information processing device according to claim 2, wherein the learning data includes evaluation information of the user on images of clothes as information indicating the clothes selection result.

7. The information processing device according to claim 2, wherein the learning data includes information obtained from a character generated or selected by the user as information indicating the clothing selection result.

8. The information processing device according to claim 2, further comprising a learning unit that performs learning of the learning model.

9. An information processing device as described in claim 8, comprising: a recognition unit that recognizes the user in an image; and a selection unit that selects an image in which the user appears from a plurality of images based on the recognition result, wherein the learning unit learns the learning model based on the selected image.

10. The information processing device of claim 2, wherein the generation unit generates the suggested information according to the clothing selection conditions based on the learning model, and the clothing selection conditions include at least one of weather, temperature, humidity, season, date, time, purpose, and location.

11. The information processing device according to claim 10, wherein the generation unit generates, as the suggestion information, at least one of a suggested image of content related to clothes to be suggested to the user and a suggested text of content related to clothes to be suggested to the user.

12. The information processing device according to claim 11, wherein the generation unit generates at least one of the proposed image and the proposed text using a generation AI.

13. The information processing device according to claim 11, wherein the generation unit generates, as the suggested image, an image in which a person wearing the suggested clothing is displayed against a background of scenery according to the clothing selection conditions.

14. The information processing device according to claim 10, further comprising: a dialogue unit that dialogues with the user on a text or image basis regarding the clothes displayed as the suggested information.

15. The information processing device according to claim 10, further comprising an output control unit that displays information about clothes previously worn by the user.

16. An information processing device as described in claim 15, further comprising a discrimination unit that discriminates whether the clothes shown as the suggested information match clothes previously worn by the user, and the output control unit outputs information regarding the result of the discrimination as information regarding clothes previously worn by the user.

17. The information processing device according to claim 16, wherein the clothing selection conditions include a location, and the determination unit determines whether the clothing shown as the suggested information matches clothing previously worn by the user in the location set as the clothing selection conditions.

18. The information processing device of claim 1, wherein the information indicating characteristics related to the user's clothing selection is information accumulated by linking a user image with a context corresponding to the user image, the context including at least one of weather, temperature, humidity, season, date, time, purpose, and location, and the generation unit generates the user image linked to the context similar to the clothing selection conditions as the suggested information.

19. An information processing method for generating suggested information regarding clothing selection for a user based on information indicating characteristics regarding the user's clothing selection or a learning model that has learned characteristics regarding the user's clothing selection.

20. A program for causing a computer to function as a generation unit that generates suggested information regarding clothing selection for a user based on information indicating characteristics regarding the user's clothing selection or a learning model that has learned characteristics regarding the user's clothing selection.

Images