system

Abstract

We provide the system. [Solution] Means of obtaining location information, A means for collecting relevant data based on acquired location information, A means for training a generative model using the aforementioned dataset, Means for providing the generated image and background information to the user's device, A means of selecting evaluators with diverse backgrounds and evaluating the generated images, A means for detecting bias based on evaluation data and adjusting the model, A system that includes this.

Description

Technical Field

[0001] The technology of the present disclosure relates to a system.

Background Art

[0002] Patent Document 1 discloses a persona chatbot control method performed by at least one processor, including steps of receiving a user utterance, adding the user utterance to a prompt including an instruction sentence related to an explanation of a chatbot character, encoding the prompt, and inputting the encoded prompt into a language model to generate a chatbot utterance in response to the user utterance.

Prior Art Documents

Patent Documents

[0003]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0004] In the development of generative AI technology, there is a problem of bias that produces inaccurate or biased expressions towards specific cultures, especially Asian cultures. This bias can cause cultural misunderstandings and discomfort, and may damage the reliability and acceptance of generative AI. Particularly in Japanese culture, the problem that image results that do not fully reflect its uniqueness and diversity may occur is prominent.

Means for Solving the Problems

[0005] To solve the above problems, this invention provides a system for collecting diverse datasets related to Japanese culture, preprocessing them, and using them to train a generative model. Furthermore, it verifies the presence or absence of cultural bias by selecting evaluators with diverse cultural backgrounds and having them evaluate the generated images. Based on this evaluation, the model is readjusted to correct any detected biases, thereby enabling image generation that balances cultural accuracy and fairness. Ultimately, the aim is to provide users with unbiased images and realize a highly reliable service.

[0006] A "dataset" is a collection of data that has been gathered and organized for a specific purpose, and in this case, it is used to reflect Japanese culture and its related elements.

[0007] "Preprocessing" is the process of transforming and organizing collected datasets that cannot be used directly, so that they are in a format that can be effectively used by AI models.

[0008] A "generative model" is an artificial intelligence algorithm that generates new images or information based on input data, and in this invention, it is trained with the aim of accurately reflecting Japanese culture.

[0009] An "evaluator" is an individual or group whose role is to verify the cultural accuracy and appropriateness of the generated images, and they are required to have diverse cultural backgrounds.

[0010] "Bias" refers to factors that produce biased results that deviate from the cultural or social context in which they are originally appropriate, and can be a cause of misunderstandings and inaccurate representations in generative AI.

[0011] "Feedback" refers to information and opinions provided by evaluators and users, and is important information used to modify models and improve their performance.

[0012] A "user" is the entity that receives the images and information generated using this system, and is the ultimate beneficiary of the generated results. [Brief explanation of the drawing]

[0013] [Figure 1] This is a conceptual diagram showing an example of the configuration of a data processing system according to the first embodiment. [Figure 2] This is a conceptual diagram showing an example of the essential functions of a data processing device and a smart device according to the first embodiment. [Figure 3] This is a conceptual diagram showing an example of the configuration of a data processing system according to the second embodiment. [Figure 4] This is a conceptual diagram showing an example of the main functions of a data processing device and smart glasses according to the second embodiment. [Figure 5] This is a conceptual diagram showing an example of the configuration of a data processing system according to the third embodiment. [Figure 6] This is a conceptual diagram showing an example of the main functions of a data processing device and a headset-type terminal according to the third embodiment. [Figure 7] This is a conceptual diagram showing an example of the configuration of a data processing system according to the fourth embodiment. [Figure 8] This is a conceptual diagram showing an example of the main functions of a data processing device and a robot according to the fourth embodiment. [Figure 9] This shows an emotion map where multiple emotions are mapped. [Figure 10] This shows an emotion map where multiple emotions are mapped. [Figure 11] This is a sequence diagram showing the processing flow of the data processing system in Example 1. [Figure 12] This is a sequence diagram showing the processing flow of the data processing system in Application Example 1. [Figure 13] This is a sequence diagram showing the processing flow of the data processing system in Example 2, which incorporates an emotion engine. [Figure 14]It is a sequence diagram showing the processing flow of a data processing system in Application Example 2 when a sentiment engine is combined.

Mode for Carrying Out the Invention

[0014] Hereinafter, an example of an embodiment of a system according to the technology of the present disclosure will be described with reference to the accompanying drawings.

[0015] First, the terms used in the following description will be explained.

[0016] In the following embodiments, a processor with a reference number (hereinafter simply referred to as "processor") may be a single arithmetic unit or a combination of multiple arithmetic units. Also, the processor may be a single type of arithmetic unit or a combination of multiple types of arithmetic units. Examples of arithmetic units include a CPU (Central Processing Unit), a GPU (Graphics Processing Unit), a GPGPU (General-Purpose computing on Graphics Processing Units), an APU (Accelerated Processing Unit), and the like.

[0017] In the following embodiments, a RAM (Random Access Memory) with a reference number is a memory in which information is temporarily stored and is used as a work memory by the processor.

[0018] In the following embodiments, a storage with a reference number is one or more non-volatile storage devices that store various programs and various parameters, etc. Examples of non-volatile storage devices include flash memory (SSD (Solid State Drive)), magnetic disks (e.g., hard disks), or magnetic tapes, etc.

[0019] In the following embodiments, the signed communication interface (I / F) is an interface that includes a communication processor and an antenna, etc. The communication interface manages communication between multiple computers. Examples of communication standards applicable to the communication interface include wireless communication standards such as 5G (5th Generation Mobile Communication System), Wi-Fi (registered trademark), or Bluetooth (registered trademark).

[0020] In the following embodiments, "A and / or B" is synonymous with "at least one of A and B." That is, "A and / or B" means that it may be A alone, or B alone, or a combination of A and B. Furthermore, in this specification, the same concept as "A and / or B" applies when expressing three or more things linked by "and / or."

[0021] [First Embodiment]

[0022] Figure 1 shows an example of the configuration of the data processing system 10 according to the first embodiment.

[0023] As shown in Figure 1, the data processing system 10 includes a data processing device 12 and a smart device 14. An example of the data processing device 12 is a server.

[0024] The data processing device 12 comprises a computer 22, a database 24, and a communication interface 26. The computer 22 is an example of a "computer" related to the technology of this disclosure. The computer 22 comprises a processor 28, RAM 30, and storage 32. The processor 28, RAM 30, and storage 32 are connected to a bus 34. The database 24 and the communication interface 26 are also connected to the bus 34. The communication interface 26 is connected to a network 54. An example of the network 54 is a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0025] The smart device 14 comprises a computer 36, a reception device 38, an output device 40, a camera 42, and a communication interface 44. The computer 36 comprises a processor 46, RAM 48, and storage 50. The processor 46, RAM 48, and storage 50 are connected to a bus 52. The reception device 38, output device 40, and camera 42 are also connected to the bus 52.

[0026] The reception device 38 is equipped with a touch panel 38A and a microphone 38B, etc., and receives user input. The touch panel 38A receives user input by detecting contact with an object (e.g., a pen or finger). The microphone 38B receives user input by detecting the user's voice. The control unit 46A transmits data indicating the user input received by the touch panel 38A and microphone 38B to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the data indicating the user input.

[0027] The output device 40 includes a display 40A and a speaker 40B, and presents data to the user 20 by outputting the data in a form perceptible to the user 20 (e.g., audio and / or text). The display 40A displays visible information such as text and images according to instructions from the processor 46. The speaker 40B outputs audio according to instructions from the processor 46. The camera 42 is a small digital camera equipped with an optical system such as a lens, aperture, and shutter, and an image sensor such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor.

[0028] Communication interface 44 is connected to network 54. Communication interfaces 44 and 26 are responsible for the exchange of various types of information between processor 46 and processor 28 via network 54.

[0029] Figure 2 shows an example of the main functions of the data processing device 12 and the smart device 14.

[0030] As shown in Figure 2, in the data processing device 12, a specific processing is performed by the processor 28. A specific processing program 56 is stored in the storage 32. The specific processing program 56 is an example of a "program" related to the technology of this disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 according to the specific processing program 56 executed on the RAM 30.

[0031] The storage 32 stores the data generation model 58 and the emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0032] In the smart device 14, the processor 46 performs the reception output processing. The storage 50 stores the reception output program 60. The reception output program 60 is used in conjunction with a specific processing program 56 by the data processing system 10. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output processing is realized by the processor 46 operating as a control unit 46A according to the reception output program 60 executed on the RAM 48.

[0033] Next, the specific processing performed by the specific processing unit 290 of the data processing device 12 will be described. In the following description, the data processing device 12 will be referred to as the "server" and the smart device 14 as the "terminal".

[0034] This invention provides a system for generating images that accurately reflect Japanese culture. Specific embodiments thereof will be described below.

[0035] First, the server collects a dataset containing diverse elements that constitute Japanese culture. This dataset includes traditional festivals, architectural styles, costumes, landscapes, and more. The collected data is then pre-processed, such as unifying image resolution and removing noise. This process ensures the necessary quality when generating images.

[0036] Next, the server trains an AI generative model using the preprocessed dataset. This AI model employs a transformer-type network architecture and is designed to learn the uniqueness and diversity of Japanese culture. The model is trained to generate images that reflect elements of Japanese culture.

[0037] To evaluate the cultural accuracy of images generated by a trained AI model, the device selects evaluators with diverse cultural backgrounds and presents them with the generated images. The evaluators determine whether the generated images accurately represent Japanese culture and whether they contain any elements that may cause offense. During this process, the evaluators' feedback is sent to a server.

[0038] The server detects biases in the generative model based on feedback from evaluators and readjusts the model weights as needed. For example, if an evaluator feels that a particular cultural element is not being represented correctly, the server applies feedback to the AI ​​model to correct that.

[0039] Finally, the user is provided with the final image generated by the finely tuned AI model. The user can then verify that the received image accurately reflects Japanese culture and meets their expected quality. This entire process generates and provides images that accurately represent uniquely Japanese cultural elements.

[0040] This invention enables the generation AI system to promote cultural understanding and gain user trust by providing generation results that respect and accurately reflect Japanese culture.

[0041] The following describes the processing flow.

[0042] Step 1:

[0043] The server collects diverse image data related to Japanese culture from the internet and existing databases. The collected data includes various elements that constitute Japanese culture, such as traditions, festivals, architecture, clothing, and landscapes. After collection, the image data undergoes preprocessing such as resolution standardization and noise reduction.

[0044] Step 2:

[0045] The server builds an AI generative model using the preprocessed dataset and begins training. The model used here is a transformer-type network designed to learn the standards and characteristics of Japanese culture. By repeatedly learning the dataset, the model will be able to understand and effectively represent the unique cultural features of Japan.

[0046] Step 3:

[0047] The device generates images using a model that has completed training. These generated images are presented to pre-selected evaluators with diverse cultural backgrounds to assess whether they accurately reflect elements of Japanese culture. The evaluators provide feedback on the generated images regarding their cultural appropriateness and lack of bias.

[0048] Step 4:

[0049] The server processes the evaluator's feedback sent from the terminal. Based on the feedback, it adjusts the model parameters if any bias or cultural misunderstanding exists in the generated images. The model is then fine-tuned based on this feedback.

[0050] Step 5:

[0051] The server generates the final image using the modified generative model and provides it to the user. This image is required to accurately reproduce Japanese culture and meet the user's desired quality and expectations. User feedback will be used to improve the model in the future.

[0052] (Example 1)

[0053] Next, we will describe Example 1. In the following description, the data processing device 12 will be referred to as the "server," and the smart device 14 will be referred to as the "terminal."

[0054] The problem that this invention aims to solve is to generate visual representations that accurately reflect Japanese culture from diverse perspectives using a generative AI model, while minimizing bias and inaccuracies based on cultural background.

[0055] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 1 is realized by the following means.

[0056] In this invention, the server includes means for collecting information and performing image quality uniformization and noise reduction, means for training a generative AI model based on the collected information, and means for selecting evaluators with diverse backgrounds and evaluating the generated visual representations. This makes it possible to enhance the cultural accuracy of the generated visual representations and continuously improve the performance of the generative AI model.

[0057] "Information" refers to the collective data collected to create visual representations related to Japanese culture, including traditional festivals, architectural styles, costumes, and landscapes.

[0058] A "generative AI model" is an artificial intelligence model designed to learn from collected information and generate visual representations that reflect Japanese culture.

[0059] "Visual representations" refer to visual content such as images and videos created by generative AI models that reflect Japanese culture.

[0060] "Bias" refers to the tendency for a visual representation to contain unintended cultural errors due to the generative AI model overemphasizing or ignoring certain cultural elements.

[0061] A "weight" is one of the parameters within a generative AI model, and it is a factor that determines how the model processes information and what kind of output it produces.

[0062] An "evaluator" is an individual or organization that evaluates generated visual representations from diverse cultural backgrounds and judges their accuracy and appropriateness.

[0063] "Noise reduction" is a preprocessing step that removes unnecessary data and visual defects from collected information, enabling the generative AI model to produce high-quality visual representations.

[0064] A "feedback loop" is a continuous improvement process that readjusts the model based on evaluations from evaluators to improve the output accuracy of the generated AI model.

[0065] "Cultural accuracy" refers to the appropriateness of a generated visual representation, ensuring it accurately reflects Japanese culture and avoids misunderstanding or offense.

[0066] "Users" refer to end-users who receive the final generated visual representation and evaluate its cultural accuracy and aesthetic value.

[0067] This invention is a system for generating visual representations that accurately reflect Japanese culture. Specifically, the system is constructed through the cooperation of three parties: a server, a terminal, and a user.

[0068] First, the server collects information related to Japanese culture from the internet and existing databases. This information includes visual elements such as festivals, architecture, clothing, and landscapes. The server uses image editing software and filtering techniques to equalize the image quality of the collected information and remove noise. Specifically, it can utilize open-source image processing libraries.

[0069] Next, the server trains a generative AI model based on the pre-processed information. This generative AI model consists of a transformer-type neural network and is designed to generate visual representations that reflect Japanese culture. Existing deep learning frameworks such as TENSORFLOW® and PyTorch are used in this process.

[0070] To evaluate the quality of the generated visual representations, the terminal presents them to evaluators with diverse cultural backgrounds. The evaluators assess the appropriateness of the presented visual representations and provide feedback. This evaluation is based on cultural accuracy and visual appeal.

[0071] The server receives feedback from evaluators, detects biases in the generated AI model, and readjusts the model's weights. This creates a feedback loop that further improves the quality of the generated visual representations.

[0072] Ultimately, the user receives a visual representation generated via a finely tuned generative AI model. The user can then verify that this visual representation accurately reflects Japanese culture and utilize it according to their individual needs.

[0073] As a concrete example, if a server inputs the prompt "Generate images themed on the Japanese tea ceremony" into an AI model, the AI ​​model will generate visual representations based on the tea ceremony. Through this generation process, it becomes possible to deepen our understanding of Japanese culture and provide beautiful and accurate cultural visual representations.

[0074] The flow of the specific processing in Example 1 will be explained using Figure 11.

[0075] Step 1:

[0076] The server collects information related to Japanese culture from the internet and databases. This information includes image data of traditional festivals, architectural styles, costumes, and landscapes. Raw image data is used as input, and a dataset for preprocessing is generated as output. At this stage, the volume and diversity of the collected data are checked, and sufficient data is ensured for subsequent processing.

[0077] Step 2:

[0078] The server performs preprocessing on the collected data, equalizing its resolution and removing noise. It uses the dataset obtained in Step 1 as input, unifying the resolution and filtering out unwanted noise using image editing software. The output is a high-quality, clean dataset suitable for training. Specifically, it performs resizing and filtering using image processing libraries.

[0079] Step 3:

[0080] The server uses a preprocessed dataset to train a generative AI model. The preprocessed dataset is fed as input to the AI ​​framework, which then trains the model based on a transformer-type neural network. The output is a fully trained generative AI model capable of reflecting Japanese culture. This step includes setting hyperparameters and evaluating the model's accuracy.

[0081] Step 4:

[0082] The terminal presents the evaluator with visual representations generated by a generative AI model. The input is the image generated by the AI ​​model, and the output is the evaluator's feedback. The evaluator judges whether the visual representations are culturally accurate and appropriate, and inputs their evaluation into the terminal. Specifically, the evaluation information is aggregated through the user interface.

[0083] Step 5:

[0084] The server detects bias in the generating AI model and readjusts the weights based on the evaluation results received from the terminal. It analyzes the evaluator's feedback as input to confirm the presence of bias. The output is an AI model readjusted to generate more accurate and culturally accurate images. Specifically, it adjusts the model parameters using a feedback loop.

[0085] Step 6:

[0086] The user receives the final visual representation generated using the retuned generative AI model. As input, they receive images generated from the optimized AI model, and as output, they save or use the images according to their purpose. The user may verify that the resulting images meet their expectations and provide feedback. Specific actions include downloading images and sending user feedback to the server.

[0087] (Application Example 1)

[0088] Next, we will explain Application Example 1. In the following explanation, the data processing device 12 will be referred to as the "server," and the smart device 14 will be referred to as the "terminal."

[0089] In the tourism industry, providing real-time images and information that reflect the actual context of tourist destinations is necessary to deepen visitors' cultural understanding of the areas they visit. However, conventional systems have struggled to generate images that accurately reflect the cultural elements of individual tourist destinations, resulting in a lack of sufficient cultural experience for visitors.

[0090] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 1 is realized by the following means.

[0091] In this invention, the server includes means for acquiring location information, means for collecting relevant data based on the acquired location information, and means for training a generative model using the dataset. This makes it possible to generate and provide images that reflect specific cultural elements of tourist destinations in real time.

[0092] "Location information" refers to data that indicates the geographical coordinates or specific location where the user is currently located.

[0093] "Relevant data" refers to data obtained based on location information, including information about the cultural background, traditions, and history of that place or region.

[0094] A "dataset" is a collection of diverse, pre-processed information used to train an image generation model.

[0095] A "generative model" is a system designed to generate images of specific cultural elements using AI technology.

[0096] "Real-time" refers to processing or responding immediately without delay.

[0097] An "image" is information presented in the form of a diagram or picture that is provided visually.

[0098] "User devices" refer to electronic devices used by users to receive information, such as smartphones and smart glasses.

[0099] An "evaluator" is a person with diverse cultural backgrounds who has been selected to evaluate the content and quality of the generated images.

[0100] "Feedback" refers to opinions and evaluations from users and evaluators that are used to improve system performance.

[0101] "Bias" refers to factors that cause inaccurate results that lack the expected equilibrium when a generative model has a particular tendency or bias.

[0102] This invention is a system for providing visitors seeking a deeper understanding of Japanese culture with images reflecting specific regional cultures in real time. Details for implementing this system are provided below.

[0103] First, when a user visits a tourist destination, their device acquires location information. This allows the system to identify the user's location. Next, the server uses the acquired geographical information to collect cultural data related to that region from its database. This data includes information on traditions, history, festivals, costumes, architecture, and more.

[0104] Subsequently, the server uses an AI generative model based on the collected data to generate images that accurately reflect the local culture. This generative model is based on a transformer network and is trained using frameworks such as TensorFlow and PyTorch.

[0105] The generated images and cultural background information are provided to the user through their device. This allows tourists to deepen their cultural understanding of their destination and have a richer experience.

[0106] For example, if a user visits "Kyoto," the system sends a prompt to the Transformer model saying, "Generate images that reflect Kyoto's festivals and history." This then provides the user with an image of the Gion Festival as a specific example, along with an explanation of the festival's origins and importance.

[0107] Furthermore, feedback from users and evaluators with diverse cultural backgrounds is sent to the server and used to tune the model. This allows the AI ​​model to be continuously improved, enabling more accurate and reliable image generation.

[0108] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[0109] Step 1:

[0110] The device acquires location information from users who have arrived at a tourist destination. The input is the latitude and longitude data of the current location using GPS functionality, and the output is that same data. This location information is used to collect related data in the next step.

[0111] Step 2:

[0112] The server collects relevant cultural data from a database based on the acquired location information. The input is the location information acquired in step 1, and the output is cultural background data identified based on that information. The server accesses the database using SQL queries to search for and collect information such as festivals, buildings, and historical events related to the visited location.

[0113] Step 3:

[0114] The server uses the collected cultural data to send a prompt to a generative AI model, which then generates an image. The input is the collected cultural data and the prompt, and the output is the generated image. A transformer model is used, which processes the data to create an image that reflects cultural elements. The prompt is in the format of "Generate an image that reflects the local culture for tourists visiting AA."

[0115] Step 4:

[0116] Along with the generated image, the server sends cultural background information to the terminal. The input is the image and cultural background information generated in step 3, and the output is the display on the user's terminal. The image is presented visually on the terminal and displayed in different formats so that the user can view detailed information.

[0117] Step 5:

[0118] Users send feedback on displayed images and information to the server via their device. The input is user feedback, and the output is evaluation data stored on the server. The feedback concerns the cultural accuracy of the images and the user experience, and is used to improve the system.

[0119] Step 6:

[0120] The server evaluates the bias and accuracy of the generated AI model based on the collected feedback and adjusts the model as needed. The input is user feedback data, and the output is the parameters of the adjusted model. By retraining the model, higher accuracy image generation can be expected in subsequent attempts.

[0121] Furthermore, an emotion engine that estimates the user's emotions may be incorporated. That is, the identification processing unit 290 may use the emotion identification model 59 to estimate the user's emotions and perform identification processing using the user's emotions.

[0122] This invention provides a system for generating images that accurately reflect Japanese culture and take into consideration the user's emotions. This technology combines an image generation model with an emotion engine that analyzes the user's feelings.

[0123] First, the server collects diverse image datasets related to Japanese culture and preprocesses the data. These datasets include a variety of elements representing Japanese culture, such as festivals, traditional crafts, architecture, clothing, and nature. Based on the preprocessed data, an AI generative model is built and trained to have the ability to generate images that reflect Japanese culture.

[0124] Next, the device uses an emotion engine to recognize the user's emotions in real time. While the user is viewing the generated images, the camera and sensors analyze facial expressions, tone of voice, and other factors to collect emotional data such as positive, negative, and neutral. This process helps the device understand what emotional response the images are eliciting from the user.

[0125] Once user sentiment data is collected, the server feeds this information back to the image generation model. Using the data analyzed by the sentiment engine, the model optimizes the generated images and adjusts them to elicit more emotionally positive responses from the user. For example, if a user shows a positive reaction to a particular cultural element, the model provides guidance for generating images that emphasize that element.

[0126] Ultimately, the server provides users with images that have been fine-tuned to take emotional responses into account. These images not only accurately reflect Japanese culture but also aim to provide users with a desirable emotional experience.

[0127] For example, if a user requests an image themed around a spring festival, the generative model will generate an image that includes festival floats and Japanese drums. When the emotion engine analyzes the user's response as positive, it adjusts the image to further highlight those elements. This process leads to greater user satisfaction and improves the reliability of the generative system.

[0128] The following describes the processing flow.

[0129] Step 1:

[0130] The server collects image data related to Japanese culture from the internet and existing databases. The collected data includes images of Japanese festivals, traditional clothing, architectural styles, and natural landscapes. This data undergoes pre-processing, such as standardizing image resolution and removing noise.

[0131] Step 2:

[0132] The server trains an AI generative model using a preprocessed dataset. This model employs a transformer-type network and is designed to learn the characteristics of Japanese culture. By learning this dataset, the model acquires the ability to represent cultural elements unique to Japan.

[0133] Step 3:

[0134] The user accesses the system and requests image generation based on a specific Japanese cultural theme (e.g., a spring festival). The device then uses its camera and microphone to initiate emotion recognition, collecting emotion data in real time from the user's facial expressions and voice.

[0135] Step 4:

[0136] The device analyzes the collected emotional data using an emotion engine. The emotion engine identifies the emotion the user is feeling (positive, negative, or neutral) from the input data. For example, if the user is smiling, it will be judged as positive.

[0137] Step 5:

[0138] The server receives the analyzed emotion data and uses it as feedback for the generated image. Based on this information, the AI ​​model adjusts the generation result, reinforcing elements that make the user more likely to feel positive emotions. In this way, an optimal image is generated that matches the user's emotions.

[0139] Step 6:

[0140] The server provides the user with the final, adjusted image. The user confirms that this image accurately reflects Japanese culture and evokes a positive emotional experience for them. This improves the overall user experience of the system.

[0141] (Example 2)

[0142] Next, we will describe Example 2. In the following description, the data processing device 12 will be referred to as the "server," and the smart device 14 will be referred to as the "terminal."

[0143] There are technical challenges in generating accurate and emotionally sensitive images for users using information related to Japanese culture. Conventional technologies have struggled to generate images that adequately consider cultural accuracy and the user's emotional state, resulting in a limited user experience.

[0144] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means.

[0145] In this invention, the server includes means for collecting and processing information related to Japanese culture, means for training an automatically generated model using said information, and means for acquiring emotional information using a device for analyzing the user's emotional state. This makes it possible to optimize the generated images based on the user's emotions and provide a more satisfying user experience.

[0146] "Japanese culture" refers to the totality of elements related to Japan's unique history, values, customs, art, and so on.

[0147] "Information" refers to data or content stored in digital or analog format, including text, images, and audio data.

[0148] "Processing" refers to a series of operations such as collecting, analyzing, organizing, and transforming information, with the aim of making it usable.

[0149] An "automatic generation model" refers to an algorithm or program that uses machine learning or artificial intelligence techniques to automatically generate results from input data.

[0150] A "device for analyzing emotional states" refers to a combination of hardware and software that uses cameras, microphones, sensors, etc., to recognize emotions from a person's facial expressions, voice, and posture.

[0151] "Emotional information" refers to data that represents a user's emotions and mood, and includes states such as positive, negative, and neutral, collected through emotion recognition technology.

[0152] "Generated image" refers to a digital image output by an automated generation model based on the input information.

[0153] "Optimization" refers to a series of operations that adjust or improve a system or process in order to enhance its adaptability and efficiency to a given purpose.

[0154] "User experience" refers to the overall satisfaction and emotional response that users feel when using a product or service.

[0155] This invention is a system that utilizes data related to Japanese culture to generate images adapted to the user's emotional response. To implement this invention, the system is configured as follows.

[0156] The server collects information about Japanese culture and manages it as a dataset. This information is obtained from the internet and other sources and includes images and text data. The server also preprocesses the data, performing noise reduction, formatting standardization, and labeling. This creates a dataset that is optimally suited for training generative AI models. The generative AI models are trained using software such as TensorFlow and PyTorch.

[0157] The device is equipped with a system for analyzing the user's emotional state in real time. Using a camera and microphone, it analyzes the user's facial expressions and voice tone with an emotion engine and extracts emotional information. Emotion recognition can utilize emotion analysis tools such as OpenFace or IBM Watson®.

[0158] The server optimizes its generative AI model based on emotional information received from the device to generate images tailored to the user. Because the generated images are adjusted based on the user's emotions, a more personalized experience can be provided. The server then sends the optimized images to the device for the user to receive.

[0159] For example, if a user enters a prompt requesting "images themed around a spring festival," the AI ​​model will generate images that include festival floats and Japanese drums. If the device analyzes the user's facial expression as positive at this stage, it can provide images that further emphasize those elements.

[0160] An example of a prompt message would be, "Generate an image of a Japanese spring festival, and if the user shows a positive reaction, highlight the festival floats and taiko drums." This improves user satisfaction and increases the reliability of the system.

[0161] The flow of the specific processing in Example 2 will be explained using Figure 13.

[0162] Step 1:

[0163] The server collects information about Japanese culture from the internet and other data sources. This information includes images and text. It takes raw data as input, performs noise reduction and resolution adjustment, and generates a pre-processed dataset as output. In this process, unnecessary elements are trimmed from the collected images and the sizes are standardized.

[0164] Step 2:

[0165] The server trains an automated generative model based on a preprocessed dataset. Software such as TensorFlow and PyTorch are used. By using preprocessed image data as input and learning hidden patterns and features within the data, a generative model reflecting Japanese culture is obtained as output. Here, the model is tuned to learn cultural elements such as tatami mats and traditional Japanese clothing.

[0166] Step 3:

[0167] The device analyzes facial expressions and voice through the camera and microphone while the user is viewing images. It takes real-time video and audio data as input, uses an emotion engine to analyze positive, negative, and neutral emotional states, and generates emotional information as output. For example, a smile is interpreted as a positive state.

[0168] Step 4:

[0169] The server feeds back the sentiment information received from the terminal to the generative model. It receives sentiment data and prompt text as input and generates or optimizes an image based on that feedback. The output is a customized image tailored to the user's sentiment. For example, if the user has a positive reaction to an image of cherry blossoms, the server will adjust and generate an image that emphasizes cherry blossoms.

[0170] Step 5:

[0171] The user receives the generated image provided by the server on their device and visually confirms it. The user experience is completed by viewing the generated image as output. Because this image is adjusted based on the user's own emotional information, it provides a greater emotional satisfaction.

[0172] (Application Example 2)

[0173] Next, we will explain application example 2. In the following explanation, the data processing device 12 will be referred to as a "server" and the smart device 14 as a "terminal".

[0174] This invention solves the technical challenge of generating images that reflect Japanese culture while considering user emotions and providing a culturally accurate and emotionally positive experience. Conventional technologies have found it difficult to consider user emotions when generating images based on Japanese culture, and have therefore failed to improve the quality of the user experience.

[0175] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means.

[0176] In this invention, the server includes means for analyzing the user's emotions and feeding that information back to a generation model, means for adjusting the generated image based on the user's emotions, and means for providing the generated image to the user. This makes it possible to generate images that are culturally accurate while eliciting an emotionally positive response from the user.

[0177] A "data set" is a collection of information containing diverse elements, including visual information related to Japanese culture.

[0178] A "generative model" is an algorithm that is trained using a dataset and automatically generates new images.

[0179] An "evaluator" is a person or group whose role is to evaluate the generated images and use the results to adjust the model.

[0180] "Evaluation information" refers to information that shows the evaluation results that the evaluator gave to the generated image.

[0181] "Bias" refers to the skew that occurs during the model generation process, and can arise when certain cultural elements are not accurately reflected.

[0182] "Emotional analysis" is a technology that analyzes a user's psychological state in real time through facial expressions, tone of voice, and other factors, and infers their emotional state.

[0183] "Feedback" is the act of returning information obtained through sentiment analysis to a generative model to help optimize the generative process.

[0184] "Image adjustment" is the process of correcting images generated according to the user's emotions to make them more appropriate.

[0185] A "feedback loop" is a process that repeatedly utilizes information obtained from users and evaluators in order to continuously improve the performance of a generative model.

[0186] This invention relates to a system that richly expresses Japanese culture and generates images that take into account the user's emotions. This system operates primarily through the cooperation of a server and a user terminal.

[0187] The server first collects and preprocesses a dataset containing diverse elements of Japanese culture. This preprocessed dataset is then used to train a generative AI model. Typically, machine learning libraries such as TensorFlow are used for training. This generative model has the ability to generate images that accurately reflect Japanese culture.

[0188] On the user's device, emotion analysis plays a crucial role. The device's built-in camera and microphone capture the user's facial expressions and voice tone in real time, and their emotional state is inferred using OpenCV and Google's Cloud Speech-to-Text API.

[0189] The emotion data generated by the emotion analysis engine is fed back to the server. Based on this information, the server optimizes the generative model and adjusts the image generation process. This results in the generation of images that elicit more positive emotional responses from the user.

[0190] For example, when a user selects and views a "cherry blossom patterned fan" in a virtual store, if sentiment analysis detects that the user is smiling, the image on the screen will be adjusted to include a background of cherry blossom trees in bloom. In this way, the user can enjoy a more satisfying experience.

[0191] An example of a prompt for the generation AI model is, "Adjust the background to match the smiling user holding a cherry blossom-patterned fan, and generate an image that more effectively represents Japanese culture." This program offers new possibilities for improving the user experience.

[0192] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[0193] Step 1:

[0194] The server collects data related to Japanese culture from the internet and local repositories. This includes image data of traditional events, costumes, architecture, and natural landscapes. The collected data is preprocessed, such as denoising and format conversion, to be ready for use as training data for generative AI models. The output of this step is the preprocessed dataset.

[0195] Step 2:

[0196] The server trains a generative AI model using preprocessed data. Machine learning libraries such as TensorFlow are used for training to build a model capable of generating images that accurately reflect Japanese culture. This step involves data calculations to adjust model parameters based on a large amount of data and improve prediction accuracy. The output is the trained image generation model.

[0197] Step 3:

[0198] The user terminal captures the user's facial expressions and voice tone through its camera and microphone. The acquired data is analyzed in real time using tools such as OpenCV and the Google Cloud Speech-to-Text API to infer the user's emotional state (positive, negative, or neutral). The input is the user's real-time video and audio data, and the output is the analyzed emotional data.

[0199] Step 4:

[0200] The data obtained through sentiment analysis is fed back to the server. The server uses this sentiment data to instruct the generative AI model to optimize the image generation process. Specifically, it selects elements of Japanese culture to emphasize according to the user's emotions and adjusts the generated image. The input for this step is the sentiment analysis data, and the output is a prompt message with adjustment instructions.

[0201] Step 5:

[0202] The server generates an image using a generative model based on adjustment instructions, applies necessary modifications, and then produces the final image. This image is optimized based on the user's emotions and cultural preferences. The generated image is sent to the user's terminal. The input is a prompt and a generative model, and the output is a customized image.

[0203] Step 6:

[0204] Users receive the final image on their device and experience it through browsing in virtual stores and galleries. User feedback may be sent to the server for further refinement, forming a feedback loop. This step provides the final user experience and provides potential feedback. The output is an evaluation of the improvement in the quality of the user experience.

[0205] The specific processing unit 290 transmits the result of the specific processing to the smart device 14. In the smart device 14, the control unit 46A causes the output device 40 to output the result of the specific processing. The microphone 38B acquires audio indicating user input for the result of the specific processing. The control unit 46A transmits the audio data indicating user input acquired by the microphone 38B to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the audio data.

[0206] Data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of data generation model 58 is ChatGPT (registered trademark) (Internet search).<URL: https: / / openai.com / blog / chatgpt> ), Gemini (registered trademark) (Internet search) <url: https: gemini.google.com ?hl="ja">Examples of generative AI include the following. The data generation model 58 is obtained by performing deep learning on a neural network. The data generation model 58 is input with prompts containing instructions, and with inference data such as audio data representing speech, text data representing text, and image data representing images. The data generation model 58 infers from the input inference data according to the instructions indicated by the prompts, and outputs the inference results in data formats such as audio data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0207] In the above embodiment, an example was given in which specific processing is performed by the data processing device 12, but the technology of this disclosure is not limited thereto, and the specific processing may also be performed by the smart device 14.

[0208] [Second Embodiment]

[0209] Figure 3 shows an example of the configuration of the data processing system 210 according to the second embodiment.

[0210] As shown in Figure 3, the data processing system 210 includes a data processing device 12 and smart glasses 214. An example of the data processing device 12 is a server.

[0211] The data processing device 12 comprises a computer 22, a database 24, and a communication interface 26. The computer 22 is an example of a "computer" related to the technology of this disclosure. The computer 22 comprises a processor 28, RAM 30, and storage 32. The processor 28, RAM 30, and storage 32 are connected to a bus 34. The database 24 and the communication interface 26 are also connected to the bus 34. The communication interface 26 is connected to a network 54. An example of the network 54 is a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0212] The smart glasses 214 include a computer 36, a microphone 238, a speaker 240, a camera 42, and a communication interface 44. The computer 36 includes a processor 46, RAM 48, and storage 50. The processor 46, RAM 48, and storage 50 are connected to a bus 52. The microphone 238, speaker 240, and camera 42 are also connected to the bus 52.

[0213] The microphone 238 receives voice signals from the user 20 and receives instructions from the user 20. The microphone 238 captures the voice signals from the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio according to the instructions from the processor 46.

[0214] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an image sensor such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the area around the user 20 (for example, an imaging range defined by a field of view equivalent to the width of a typical healthy person's field of vision).

[0215] Communication interface 44 is connected to network 54. Communication interfaces 44 and 26 are responsible for the exchange of various information between processor 46 and processor 28 via network 54. The exchange of various information between processor 46 and processor 28 using communication interfaces 44 and 26 is performed in a secure manner.

[0216] Figure 4 shows an example of the main functions of the data processing device 12 and the smart glasses 214. As shown in Figure 4, the data processing device 12 performs specific processing using the processor 28. The storage 32 stores the specific processing program 56.

[0217] The specific processing program 56 is an example of a "program" relating to the technology of this disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0218] The storage 32 stores the data generation model 58 and the emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0219] In the smart glasses 214, the processor 46 performs the reception output processing. The storage 50 stores the reception output program 60. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output processing is realized by the processor 46 operating as a control unit 46A according to the reception output program 60 executed on the RAM 48.

[0220] Next, the identification processing performed by the identification processing unit 290 of the data processing device 12 will be described. In the following description, the data processing device 12 will be referred to as the "server" and the smart glasses 214 will be referred to as the "terminal".

[0221] This invention provides a system for generating images that accurately reflect Japanese culture. Specific embodiments thereof will be described below.

[0222] First, the server collects a dataset containing diverse elements that constitute Japanese culture. This dataset includes traditional festivals, architectural styles, costumes, landscapes, and more. The collected data is then pre-processed, such as unifying image resolution and removing noise. This process ensures the necessary quality when generating images.

[0223] Next, the server trains an AI generative model using the preprocessed dataset. This AI model employs a transformer-type network architecture and is designed to learn the uniqueness and diversity of Japanese culture. The model is trained to generate images that reflect elements of Japanese culture.

[0224] To evaluate the cultural accuracy of images generated by a trained AI model, the device selects evaluators with diverse cultural backgrounds and presents them with the generated images. The evaluators determine whether the generated images accurately represent Japanese culture and whether they contain any elements that may cause offense. During this process, the evaluators' feedback is sent to a server.

[0225] The server detects biases in the generative model based on feedback from evaluators and readjusts the model weights as needed. For example, if an evaluator feels that a particular cultural element is not being represented correctly, the server applies feedback to the AI ​​model to correct that.

[0226] Finally, the user is provided with the final image generated by the finely tuned AI model. The user can then verify that the received image accurately reflects Japanese culture and meets their expected quality. This entire process generates and provides images that accurately represent uniquely Japanese cultural elements.

[0227] This invention enables the generation AI system to promote cultural understanding and gain user trust by providing generation results that respect and accurately reflect Japanese culture.

[0228] The following describes the processing flow.

[0229] Step 1:

[0230] The server collects diverse image data related to Japanese culture from the internet and existing databases. The collected data includes various elements that constitute Japanese culture, such as traditions, festivals, architecture, clothing, and landscapes. After collection, the image data undergoes preprocessing such as resolution standardization and noise reduction.

[0231] Step 2:

[0232] The server builds an AI generative model using the preprocessed dataset and begins training. The model used here is a transformer-type network designed to learn the standards and characteristics of Japanese culture. By repeatedly learning the dataset, the model will be able to understand and effectively represent the unique cultural features of Japan.

[0233] Step 3:

[0234] The device generates images using a model that has completed training. These generated images are presented to pre-selected evaluators with diverse cultural backgrounds to assess whether they accurately reflect elements of Japanese culture. The evaluators provide feedback on the generated images regarding their cultural appropriateness and lack of bias.

[0235] Step 4:

[0236] The server processes the evaluator's feedback sent from the terminal. Based on the feedback, it adjusts the model parameters if any bias or cultural misunderstanding exists in the generated images. The model is then fine-tuned based on this feedback.

[0237] Step 5:

[0238] The server generates the final image using the modified generative model and provides it to the user. This image is required to accurately reproduce Japanese culture and meet the user's desired quality and expectations. User feedback will be used to improve the model in the future.

[0239] (Example 1)

[0240] Next, we will describe Example 1. In the following description, the data processing device 12 will be referred to as the "server" and the smart glasses 214 will be referred to as the "terminal".

[0241] The problem that this invention aims to solve is to generate visual representations that accurately reflect Japanese culture from diverse perspectives using a generative AI model, while minimizing bias and inaccuracies based on cultural background.

[0242] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 1 is realized by the following means.

[0243] In this invention, the server includes means for collecting information and performing image quality uniformization and noise reduction, means for training a generative AI model based on the collected information, and means for selecting evaluators with diverse backgrounds and evaluating the generated visual representations. This makes it possible to enhance the cultural accuracy of the generated visual representations and continuously improve the performance of the generative AI model.

[0244] "Information" refers to the collective data collected to create visual representations related to Japanese culture, including traditional festivals, architectural styles, costumes, and landscapes.

[0245] A "generative AI model" is an artificial intelligence model designed to learn from collected information and generate visual representations that reflect Japanese culture.

[0246] "Visual representations" refer to visual content such as images and videos created by generative AI models that reflect Japanese culture.

[0247] "Bias" refers to the tendency for a visual representation to contain unintended cultural errors due to the generative AI model overemphasizing or ignoring certain cultural elements.

[0248] A "weight" is one of the parameters within a generative AI model, and it is a factor that determines how the model processes information and what kind of output it produces.

[0249] An "evaluator" is an individual or organization that evaluates generated visual representations from diverse cultural backgrounds and judges their accuracy and appropriateness.

[0250] "Noise reduction" is a preprocessing step that removes unnecessary data and visual defects from collected information, enabling the generative AI model to produce high-quality visual representations.

[0251] A "feedback loop" is a continuous improvement process that readjusts the model based on evaluations from evaluators to improve the output accuracy of the generated AI model.

[0252] "Cultural accuracy" refers to the appropriateness of a generated visual representation, ensuring it accurately reflects Japanese culture and avoids misunderstanding or offense.

[0253] "Users" refer to end-users who receive the final generated visual representation and evaluate its cultural accuracy and aesthetic value.

[0254] This invention is a system for generating visual representations that accurately reflect Japanese culture. Specifically, the system is constructed through the cooperation of three parties: a server, a terminal, and a user.

[0255] First, the server collects information related to Japanese culture from the internet and existing databases. This information includes visual elements such as festivals, architecture, clothing, and landscapes. The server uses image editing software and filtering techniques to equalize the image quality of the collected information and remove noise. Specifically, it can utilize open-source image processing libraries.

[0256] Next, the server trains a generative AI model based on the pre-processed information. This generative AI model consists of a transformer-type neural network and is designed to generate visual representations that reflect Japanese culture. Existing deep learning frameworks such as TensorFlow and PyTorch are used for this process.

[0257] To evaluate the quality of the generated visual representations, the terminal presents them to evaluators with diverse cultural backgrounds. The evaluators assess the appropriateness of the presented visual representations and provide feedback. This evaluation is based on cultural accuracy and visual appeal.

[0258] The server receives feedback from evaluators, detects biases in the generated AI model, and readjusts the model's weights. This creates a feedback loop that further improves the quality of the generated visual representations.

[0259] Ultimately, the user receives a visual representation generated via a finely tuned generative AI model. The user can then verify that this visual representation accurately reflects Japanese culture and utilize it according to their individual needs.

[0260] As a concrete example, if a server inputs the prompt "Generate images themed on the Japanese tea ceremony" into an AI model, the AI ​​model will generate visual representations based on the tea ceremony. Through this generation process, it becomes possible to deepen our understanding of Japanese culture and provide beautiful and accurate cultural visual representations.

[0261] The flow of the specific processing in Example 1 will be explained using Figure 11.

[0262] Step 1:

[0263] The server collects information related to Japanese culture from the internet and databases. This information includes image data of traditional festivals, architectural styles, costumes, and landscapes. Raw image data is used as input, and a dataset for preprocessing is generated as output. At this stage, the volume and diversity of the collected data are checked, and sufficient data is ensured for subsequent processing.

[0264] Step 2:

[0265] The server performs preprocessing on the collected data, equalizing its resolution and removing noise. It uses the dataset obtained in Step 1 as input, unifying the resolution and filtering out unwanted noise using image editing software. The output is a high-quality, clean dataset suitable for training. Specifically, it performs resizing and filtering using image processing libraries.

[0266] Step 3:

[0267] The server uses a preprocessed dataset to train a generative AI model. The preprocessed dataset is fed as input to the AI ​​framework, which then trains the model based on a transformer-type neural network. The output is a fully trained generative AI model capable of reflecting Japanese culture. This step includes setting hyperparameters and evaluating the model's accuracy.

[0268] Step 4:

[0269] The terminal presents the evaluator with visual representations generated by a generative AI model. The input is the image generated by the AI ​​model, and the output is the evaluator's feedback. The evaluator judges whether the visual representations are culturally accurate and appropriate, and inputs their evaluation into the terminal. Specifically, the evaluation information is aggregated through the user interface.

[0270] Step 5:

[0271] The server detects bias in the generating AI model and readjusts the weights based on the evaluation results received from the terminal. It analyzes the evaluator's feedback as input to confirm the presence of bias. The output is an AI model readjusted to generate more accurate and culturally accurate images. Specifically, it adjusts the model parameters using a feedback loop.

[0272] Step 6:

[0273] The user receives the final visual representation generated using the retuned generative AI model. As input, they receive images generated from the optimized AI model, and as output, they save or use the images according to their purpose. The user may verify that the resulting images meet their expectations and provide feedback. Specific actions include downloading images and sending user feedback to the server.

[0274] (Application Example 1)

[0275] Next, we will explain Application Example 1. In the following explanation, the data processing device 12 will be referred to as the "server," and the smart glasses 214 will be referred to as the "terminal."

[0276] In the tourism industry, providing real-time images and information that reflect the actual context of tourist destinations is necessary to deepen visitors' cultural understanding of the areas they visit. However, conventional systems have struggled to generate images that accurately reflect the cultural elements of individual tourist destinations, resulting in a lack of sufficient cultural experience for visitors.

[0277] The specific processing by the specific processing unit 290 of the data processing apparatus 12 in Application Example 1 is realized by the following means.

[0278] In this invention, the server includes means for acquiring position information, means for collecting related data based on the acquired position information, and means for training a generation model using the data set. Thereby, it becomes possible to generate and provide in real time an image reflecting specific cultural elements of a tourist destination.

[0279] "Position information" is data indicating the geographical coordinates or a specific location where the user is currently located.

[0280] "Related data" is data including information such as the cultural background, traditions, history, etc. of the place or region, acquired based on the position information.

[0281] "Data set" is a collection of various pre - processed information used for training an image generation model.

[0282] "Generation model" is a system designed to generate specific cultural elements as images using AI technology.

[0283] "Real time" refers to processing or responding immediately without delay.

[0284] "Image" is information represented in the form of a visually provided diagram or picture.

[0285] "User device" is an electronic device such as a smartphone or smart glasses that the user uses to receive information.

[0286] "Evaluator" is a person with various cultural backgrounds selected to evaluate the content and quality of the generated image.

[0287] "Feedback" refers to opinions and evaluations from users and evaluators that are used to improve system performance.

[0288] "Bias" refers to factors that cause inaccurate results that lack the expected equilibrium when a generative model has a particular tendency or bias.

[0289] This invention is a system for providing visitors seeking a deeper understanding of Japanese culture with images reflecting specific regional cultures in real time. Details for implementing this system are provided below.

[0290] First, when a user visits a tourist destination, their device acquires location information. This allows the system to identify the user's location. Next, the server uses the acquired geographical information to collect cultural data related to that region from its database. This data includes information on traditions, history, festivals, costumes, architecture, and more.

[0291] Subsequently, the server uses an AI generative model based on the collected data to generate images that accurately reflect the local culture. This generative model is based on a transformer network and is trained using frameworks such as TensorFlow and PyTorch.

[0292] The generated images and cultural background information are provided to the user through their device. This allows tourists to deepen their cultural understanding of their destination and have a richer experience.

[0293] For example, if a user visits "Kyoto," the system sends a prompt to the Transformer model saying, "Generate images that reflect Kyoto's festivals and history." This then provides the user with an image of the Gion Festival as a specific example, along with an explanation of the festival's origins and importance.

[0294] Furthermore, feedback from users and evaluators with diverse cultural backgrounds is sent to the server and used to tune the model. This allows the AI ​​model to be continuously improved, enabling more accurate and reliable image generation.

[0295] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[0296] Step 1:

[0297] The device acquires location information from users who have arrived at a tourist destination. The input is the latitude and longitude data of the current location using GPS functionality, and the output is that same data. This location information is used to collect related data in the next step.

[0298] Step 2:

[0299] The server collects relevant cultural data from a database based on the acquired location information. The input is the location information acquired in step 1, and the output is cultural background data identified based on that information. The server accesses the database using SQL queries to search for and collect information such as festivals, buildings, and historical events related to the visited location.

[0300] Step 3:

[0301] The server uses the collected cultural data to send a prompt to a generative AI model, which then generates an image. The input is the collected cultural data and the prompt, and the output is the generated image. A transformer model is used, which processes the data to create an image that reflects cultural elements. The prompt is in the format of "Generate an image that reflects the local culture for tourists visiting AA."

[0302] Step 4:

[0303] The server transmits cultural background information to the terminal together with the generated image. The input is the image generated in Step 3 and the cultural background information, and the output is the display on the user's terminal. The image is visually provided on the terminal and is displayed in different formats so that the user can view the detailed information.

[0304] Step 5:

[0305] The user transmits feedback on the displayed image and information to the server via the terminal. The input is the user's feedback, and the output is the evaluation data saved on the server. The feedback is related to the cultural accuracy of the image and the user experience, and is used to improve the system.

[0306] Step 6:

[0307] The server evaluates the bias and accuracy of the generated AI model based on the collected feedback, and adjusts the model as necessary. The input is the feedback data from the user, and the output is the parameters of the adjusted model. By retraining the model, higher-accuracy image generation can be expected from the next time onwards.

[0308] Furthermore, an emotion engine for estimating the user's emotion may be combined. That is, the specific processing unit 290 may estimate the user's emotion using the emotion recognition model 59 and perform specific processing using the user's emotion.

[0309] The present invention provides a system for generating an image that accurately reflects Japanese culture and takes into account the user's emotion. This technology combines an image generation model and an emotion engine for analyzing the user's emotion.

[0310] First, the server collects diverse image datasets related to Japanese culture and preprocesses the data. These datasets include a variety of elements representing Japanese culture, such as festivals, traditional crafts, architecture, clothing, and nature. Based on the preprocessed data, an AI generative model is built and trained to have the ability to generate images that reflect Japanese culture.

[0311] Next, the device uses an emotion engine to recognize the user's emotions in real time. While the user is viewing the generated images, the camera and sensors analyze facial expressions, tone of voice, and other factors to collect emotional data such as positive, negative, and neutral. This process helps the device understand what emotional response the images are eliciting from the user.

[0312] Once user sentiment data is collected, the server feeds this information back to the image generation model. Using the data analyzed by the sentiment engine, the model optimizes the generated images and adjusts them to elicit more emotionally positive responses from the user. For example, if a user shows a positive reaction to a particular cultural element, the model provides guidance for generating images that emphasize that element.

[0313] Ultimately, the server provides users with images that have been fine-tuned to take emotional responses into account. These images not only accurately reflect Japanese culture but also aim to provide users with a desirable emotional experience.

[0314] For example, if a user requests an image themed around a spring festival, the generative model will generate an image that includes festival floats and Japanese drums. When the emotion engine analyzes the user's response as positive, it adjusts the image to further highlight those elements. This process leads to greater user satisfaction and improves the reliability of the generative system.

[0315] The following describes the processing flow.

[0316] Step 1:

[0317] The server collects image data related to Japanese culture from the internet and existing databases. The collected data includes images of Japanese festivals, traditional clothing, architectural styles, and natural landscapes. This data undergoes pre-processing, such as standardizing image resolution and removing noise.

[0318] Step 2:

[0319] The server trains an AI generative model using a preprocessed dataset. This model employs a transformer-type network and is designed to learn the characteristics of Japanese culture. By learning this dataset, the model acquires the ability to represent cultural elements unique to Japan.

[0320] Step 3:

[0321] The user accesses the system and requests image generation based on a specific Japanese cultural theme (e.g., a spring festival). The device then uses its camera and microphone to initiate emotion recognition, collecting emotion data in real time from the user's facial expressions and voice.

[0322] Step 4:

[0323] The device analyzes the collected emotional data using an emotion engine. The emotion engine identifies the emotion the user is feeling (positive, negative, or neutral) from the input data. For example, if the user is smiling, it will be judged as positive.

[0324] Step 5:

[0325] The server receives the analyzed emotion data and uses it as feedback for the generated image. Based on this information, the AI ​​model adjusts the generation result, reinforcing elements that make the user more likely to feel positive emotions. In this way, an optimal image is generated that matches the user's emotions.

[0326] Step 6:

[0327] The server provides the user with the final, adjusted image. The user confirms that this image accurately reflects Japanese culture and evokes a positive emotional experience for them. This improves the overall user experience of the system.

[0328] (Example 2)

[0329] Next, we will describe Example 2. In the following description, the data processing device 12 will be referred to as the "server" and the smart glasses 214 will be referred to as the "terminal".

[0330] There are technical challenges in generating accurate and emotionally sensitive images for users using information related to Japanese culture. Conventional technologies have struggled to generate images that adequately consider cultural accuracy and the user's emotional state, resulting in a limited user experience.

[0331] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means.

[0332] In this invention, the server includes means for collecting and processing information related to Japanese culture, means for training an automatically generated model using said information, and means for acquiring emotional information using a device for analyzing the user's emotional state. This makes it possible to optimize the generated images based on the user's emotions and provide a more satisfying user experience.

[0333] "Japanese culture" refers to the totality of elements related to Japan's unique history, values, customs, art, and so on.

[0334] "Information" refers to data or content stored in digital or analog format, including text, images, and audio data.

[0335] "Processing" refers to a series of operations such as collecting, analyzing, organizing, and transforming information, with the aim of making it usable.

[0336] An "automatic generation model" refers to an algorithm or program that uses machine learning or artificial intelligence techniques to automatically generate results from input data.

[0337] A "device for analyzing emotional states" refers to a combination of hardware and software that uses cameras, microphones, sensors, etc., to recognize emotions from a person's facial expressions, voice, and posture.

[0338] "Emotional information" refers to data that represents a user's emotions and mood, and includes states such as positive, negative, and neutral, collected through emotion recognition technology.

[0339] "Generated image" refers to a digital image output by an automated generation model based on the input information.

[0340] "Optimization" refers to a series of operations that adjust or improve a system or process in order to enhance its adaptability and efficiency to a given purpose.

[0341] "User experience" refers to the overall satisfaction and emotional response that users feel when using a product or service.

[0342] This invention is a system that utilizes data related to Japanese culture to generate images adapted to the user's emotional response. To implement this invention, the system is configured as follows.

[0343] The server collects information about Japanese culture and manages it as a dataset. This information is obtained from the internet and other sources and includes images and text data. The server also preprocesses the data, performing noise reduction, formatting standardization, and labeling. This creates a dataset that is optimally suited for training generative AI models. The generative AI models are trained using software such as TensorFlow and PyTorch.

[0344] The device is equipped with a system for analyzing the user's emotional state in real time. Using a camera and microphone, it analyzes the user's facial expressions and voice tone with an emotion engine to extract emotional information. Emotion recognition can utilize emotion analysis tools such as OpenFace or IBM Watson.

[0345] The server optimizes its generative AI model based on emotional information received from the device to generate images tailored to the user. Because the generated images are adjusted based on the user's emotions, a more personalized experience can be provided. The server then sends the optimized images to the device for the user to receive.

[0346] For example, if a user enters a prompt requesting "images themed around a spring festival," the AI ​​model will generate images that include festival floats and Japanese drums. If the device analyzes the user's facial expression as positive at this stage, it can provide images that further emphasize those elements.

[0347] An example of a prompt message would be, "Generate an image of a Japanese spring festival, and if the user shows a positive reaction, highlight the festival floats and taiko drums." This improves user satisfaction and increases the reliability of the system.

[0348] The flow of the specific processing in Example 2 will be explained using Figure 13.

[0349] Step 1:

[0350] The server collects information about Japanese culture from the internet and other data sources. This information includes images and text. It takes raw data as input, performs noise reduction and resolution adjustment, and generates a pre-processed dataset as output. In this process, unnecessary elements are trimmed from the collected images and the sizes are standardized.

[0351] Step 2:

[0352] The server trains an automated generative model based on a preprocessed dataset. Software such as TensorFlow and PyTorch are used. By using preprocessed image data as input and learning hidden patterns and features within the data, a generative model reflecting Japanese culture is obtained as output. Here, the model is tuned to learn cultural elements such as tatami mats and traditional Japanese clothing.

[0353] Step 3:

[0354] The device analyzes facial expressions and voice through the camera and microphone while the user is viewing images. It takes real-time video and audio data as input, uses an emotion engine to analyze positive, negative, and neutral emotional states, and generates emotional information as output. For example, a smile is interpreted as a positive state.

[0355] Step 4:

[0356] The server feeds back the sentiment information received from the terminal to the generative model. It receives sentiment data and prompt text as input and generates or optimizes an image based on that feedback. The output is a customized image tailored to the user's sentiment. For example, if the user has a positive reaction to an image of cherry blossoms, the server will adjust and generate an image that emphasizes cherry blossoms.

[0357] Step 5:

[0358] The user receives the generated image provided by the server on their device and visually confirms it. The user experience is completed by viewing the generated image as output. Because this image is adjusted based on the user's own emotional information, it provides a greater emotional satisfaction.

[0359] (Application Example 2)

[0360] Next, we will explain application example 2. In the following explanation, the data processing device 12 will be referred to as the "server" and the smart glasses 214 as the "terminal".

[0361] This invention solves the technical challenge of generating images that reflect Japanese culture while considering user emotions and providing a culturally accurate and emotionally positive experience. Conventional technologies have found it difficult to consider user emotions when generating images based on Japanese culture, and have therefore failed to improve the quality of the user experience.

[0362] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means.

[0363] In this invention, the server includes means for analyzing the user's emotions and feeding that information back to a generation model, means for adjusting the generated image based on the user's emotions, and means for providing the generated image to the user. This makes it possible to generate images that are culturally accurate while eliciting an emotionally positive response from the user.

[0364] A "data set" is a collection of information containing diverse elements, including visual information related to Japanese culture.

[0365] A "generative model" is an algorithm that is trained using a dataset and automatically generates new images.

[0366] An "evaluator" is a person or group whose role is to evaluate the generated images and use the results to adjust the model.

[0367] "Evaluation information" refers to information that shows the evaluation results that the evaluator gave to the generated image.

[0368] "Bias" refers to the skew that occurs during the model generation process, and can arise when certain cultural elements are not accurately reflected.

[0369] "Emotional analysis" is a technology that analyzes a user's psychological state in real time through facial expressions, tone of voice, and other factors, and infers their emotional state.

[0370] "Feedback" is the act of returning information obtained through sentiment analysis to a generative model to help optimize the generative process.

[0371] "Image adjustment" is the process of correcting images generated according to the user's emotions to make them more appropriate.

[0372] A "feedback loop" is a process that repeatedly utilizes information obtained from users and evaluators in order to continuously improve the performance of a generative model.

[0373] This invention relates to a system that richly expresses Japanese culture and generates images that take into account the user's emotions. This system operates primarily through the cooperation of a server and a user terminal.

[0374] The server first collects and preprocesses a dataset containing diverse elements of Japanese culture. This preprocessed dataset is then used to train a generative AI model. Typically, machine learning libraries such as TensorFlow are used for training. This generative model has the ability to generate images that accurately reflect Japanese culture.

[0375] On the user's device, sentiment analysis plays a crucial role. The device's built-in camera and microphone capture the user's facial expressions and voice tone in real time, and their emotional state is inferred using OpenCV or the Google Cloud Speech-to-Text API.

[0376] The emotion data generated by the emotion analysis engine is fed back to the server. Based on this information, the server optimizes the generative model and adjusts the image generation process. This results in the generation of images that elicit more positive emotional responses from the user.

[0377] For example, when a user selects and views a "cherry blossom patterned fan" in a virtual store, if sentiment analysis detects that the user is smiling, the image on the screen will be adjusted to include a background of cherry blossom trees in bloom. In this way, the user can enjoy a more satisfying experience.

[0378] An example of a prompt for the generation AI model is, "Adjust the background to match the smiling user holding a cherry blossom-patterned fan, and generate an image that more effectively represents Japanese culture." This program offers new possibilities for improving the user experience.

[0379] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[0380] Step 1:

[0381] The server collects data related to Japanese culture from the internet and local repositories. This includes image data of traditional events, costumes, architecture, and natural landscapes. The collected data is preprocessed, such as denoising and format conversion, to be ready for use as training data for generative AI models. The output of this step is the preprocessed dataset.

[0382] Step 2:

[0383] The server trains a generative AI model using preprocessed data. Machine learning libraries such as TensorFlow are used for training to build a model capable of generating images that accurately reflect Japanese culture. This step involves data calculations to adjust model parameters based on a large amount of data and improve prediction accuracy. The output is the trained image generation model.

[0384] Step 3:

[0385] The user terminal captures the user's facial expressions and voice tone through its camera and microphone. The acquired data is analyzed in real time using tools such as OpenCV and the Google Cloud Speech-to-Text API to infer the user's emotional state (positive, negative, or neutral). The input is the user's real-time video and audio data, and the output is the analyzed emotional data.

[0386] Step 4:

[0387] The data obtained through sentiment analysis is fed back to the server. The server uses this sentiment data to instruct the generative AI model to optimize the image generation process. Specifically, it selects elements of Japanese culture to emphasize according to the user's emotions and adjusts the generated image. The input for this step is the sentiment analysis data, and the output is a prompt message with adjustment instructions.

[0388] Step 5:

[0389] The server generates an image using a generative model based on adjustment instructions, applies necessary modifications, and then produces the final image. This image is optimized based on the user's emotions and cultural preferences. The generated image is sent to the user's terminal. The input is a prompt and a generative model, and the output is a customized image.

[0390] Step 6:

[0391] Users receive the final image on their device and experience it through browsing in virtual stores and galleries. User feedback may be sent to the server for further refinement, forming a feedback loop. This step provides the final user experience and provides potential feedback. The output is an evaluation of the improvement in the quality of the user experience.

[0392] The specific processing unit 290 transmits the result of the specific processing to the smart glasses 214. In the smart glasses 214, the control unit 46A causes the speaker 240 to output the result of the specific processing. The microphone 238 acquires audio indicating user input for the result of the specific processing. The control unit 46A transmits the audio data indicating user input acquired by the microphone 238 to the data processing unit 12. In the data processing unit 12, the specific processing unit 290 acquires the audio data.

[0393] Data generation model 58 is a type of so-called generative AI (Artificial Intelligence). One example of data generation model 58 is ChatGPT (Internet search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search) <url: https: gemini.google.com ?hl="ja">Examples of generative AI include the following. The data generation model 58 is obtained by performing deep learning on a neural network. The data generation model 58 is input with prompts containing instructions, and with inference data such as audio data representing speech, text data representing text, and image data representing images. The data generation model 58 infers from the input inference data according to the instructions indicated by the prompts, and outputs the inference results in data formats such as audio data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0394] In the above embodiment, an example was given in which specific processing is performed by the data processing device 12, but the technology of this disclosure is not limited thereto, and the specific processing may also be performed by the smart glasses 214.

[0395] [Third Embodiment]

[0396] Figure 5 shows an example of the configuration of the data processing system 310 according to the third embodiment.

[0397] As shown in Figure 5, the data processing system 310 includes a data processing device 12 and a headset terminal 314. An example of the data processing device 12 is a server.

[0398] The data processing device 12 comprises a computer 22, a database 24, and a communication interface 26. The computer 22 is an example of a "computer" related to the technology of this disclosure. The computer 22 comprises a processor 28, RAM 30, and storage 32. The processor 28, RAM 30, and storage 32 are connected to a bus 34. The database 24 and the communication interface 26 are also connected to the bus 34. The communication interface 26 is connected to a network 54. An example of the network 54 is a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0399] The headset terminal 314 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication interface 44, and a display 343. The computer 36 includes a processor 46, RAM 48, and storage 50. The processor 46, RAM 48, and storage 50 are connected to a bus 52. The microphone 238, speaker 240, camera 42, and display 343 are also connected to the bus 52.

[0400] The microphone 238 receives voice signals from the user 20 and receives instructions from the user 20. The microphone 238 captures the voice signals from the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio according to the instructions from the processor 46.

[0401] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an image sensor such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the area around the user 20 (for example, an imaging range defined by a field of view equivalent to the width of a typical healthy person's field of vision).

[0402] Communication interface 44 is connected to network 54. Communication interfaces 44 and 26 are responsible for the exchange of various information between processor 46 and processor 28 via network 54. The exchange of various information between processor 46 and processor 28 using communication interfaces 44 and 26 is performed in a secure manner.

[0403] Figure 6 shows an example of the main functions of the data processing device 12 and the headset terminal 314. As shown in Figure 6, the data processing device 12 performs specific processing using the processor 28. The storage 32 stores the specific processing program 56.

[0404] The specific processing program 56 is an example of a "program" relating to the technology of this disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0405] The storage 32 stores the data generation model 58 and the emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0406] In the headset terminal 314, the processor 46 performs the reception output processing. The storage 50 stores the reception output program 60. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output processing is realized by the processor 46 operating as a control unit 46A according to the reception output program 60 executed on the RAM 48.

[0407] Next, the specific processing performed by the specific processing unit 290 of the data processing device 12 will be described. In the following description, the data processing device 12 will be referred to as the "server" and the headset terminal 314 will be referred to as the "terminal".

[0408] This invention provides a system for generating images that accurately reflect Japanese culture. Specific embodiments thereof will be described below.

[0409] First, the server collects a dataset containing diverse elements that constitute Japanese culture. This dataset includes traditional festivals, architectural styles, costumes, landscapes, and more. The collected data is then pre-processed, such as unifying image resolution and removing noise. This process ensures the necessary quality when generating images.

[0410] Next, the server trains an AI generative model using the preprocessed dataset. This AI model employs a transformer-type network architecture and is designed to learn the uniqueness and diversity of Japanese culture. The model is trained to generate images that reflect elements of Japanese culture.

[0411] To evaluate the cultural accuracy of images generated by a trained AI model, the device selects evaluators with diverse cultural backgrounds and presents them with the generated images. The evaluators determine whether the generated images accurately represent Japanese culture and whether they contain any elements that may cause offense. During this process, the evaluators' feedback is sent to a server.

[0412] The server detects biases in the generative model based on feedback from evaluators and readjusts the model weights as needed. For example, if an evaluator feels that a particular cultural element is not being represented correctly, the server applies feedback to the AI ​​model to correct that.

[0413] Finally, the user is provided with the final image generated by the finely tuned AI model. The user can then verify that the received image accurately reflects Japanese culture and meets their expected quality. This entire process generates and provides images that accurately represent uniquely Japanese cultural elements.

[0414] This invention enables the generation AI system to promote cultural understanding and gain user trust by providing generation results that respect and accurately reflect Japanese culture.

[0415] The following describes the processing flow.

[0416] Step 1:

[0417] The server collects diverse image data related to Japanese culture from the internet and existing databases. The collected data includes various elements that constitute Japanese culture, such as traditions, festivals, architecture, clothing, and landscapes. After collection, the image data undergoes preprocessing such as resolution standardization and noise reduction.

[0418] Step 2:

[0419] The server builds an AI generative model using the preprocessed dataset and begins training. The model used here is a transformer-type network designed to learn the standards and characteristics of Japanese culture. By repeatedly learning the dataset, the model will be able to understand and effectively represent the unique cultural features of Japan.

[0420] Step 3:

[0421] The device generates images using a model that has completed training. These generated images are presented to pre-selected evaluators with diverse cultural backgrounds to assess whether they accurately reflect elements of Japanese culture. The evaluators provide feedback on the generated images regarding their cultural appropriateness and lack of bias.

[0422] Step 4:

[0423] The server processes the evaluator's feedback sent from the terminal. Based on the feedback, it adjusts the model parameters if any bias or cultural misunderstanding exists in the generated images. The model is then fine-tuned based on this feedback.

[0424] Step 5:

[0425] The server generates the final image using the modified generative model and provides it to the user. This image is required to accurately reproduce Japanese culture and meet the user's desired quality and expectations. User feedback will be used to improve the model in the future.

[0426] (Example 1)

[0427] Next, we will describe Example 1. In the following description, the data processing device 12 will be referred to as the "server," and the headset-type terminal 314 will be referred to as the "terminal."

[0428] The problem that this invention aims to solve is to generate visual representations that accurately reflect Japanese culture from diverse perspectives using a generative AI model, while minimizing bias and inaccuracies based on cultural background.

[0429] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 1 is realized by the following means.

[0430] In this invention, the server includes means for collecting information and performing image quality uniformization and noise reduction, means for training a generative AI model based on the collected information, and means for selecting evaluators with diverse backgrounds and evaluating the generated visual representations. This makes it possible to enhance the cultural accuracy of the generated visual representations and continuously improve the performance of the generative AI model.

[0431] "Information" refers to the collective data collected to create visual representations related to Japanese culture, including traditional festivals, architectural styles, costumes, and landscapes.

[0432] A "generative AI model" is an artificial intelligence model designed to learn from collected information and generate visual representations that reflect Japanese culture.

[0433] "Visual representations" refer to visual content such as images and videos created by generative AI models that reflect Japanese culture.

[0434] "Bias" refers to the tendency for a visual representation to contain unintended cultural errors due to the generative AI model overemphasizing or ignoring certain cultural elements.

[0435] A "weight" is one of the parameters within a generative AI model, and it is a factor that determines how the model processes information and what kind of output it produces.

[0436] An "evaluator" is an individual or organization that evaluates generated visual representations from diverse cultural backgrounds and judges their accuracy and appropriateness.

[0437] "Noise reduction" is a preprocessing step that removes unnecessary data and visual defects from collected information, enabling the generative AI model to produce high-quality visual representations.

[0438] A "feedback loop" is a continuous improvement process that readjusts the model based on evaluations from evaluators to improve the output accuracy of the generated AI model.

[0439] "Cultural accuracy" refers to the appropriateness of a generated visual representation, ensuring it accurately reflects Japanese culture and avoids misunderstanding or offense.

[0440] "Users" refer to end-users who receive the final generated visual representation and evaluate its cultural accuracy and aesthetic value.

[0441] This invention is a system for generating visual representations that accurately reflect Japanese culture. Specifically, the system is constructed through the cooperation of three parties: a server, a terminal, and a user.

[0442] First, the server collects information related to Japanese culture from the internet and existing databases. This information includes visual elements such as festivals, architecture, clothing, and landscapes. The server uses image editing software and filtering techniques to equalize the image quality of the collected information and remove noise. Specifically, it can utilize open-source image processing libraries.

[0443] Next, the server trains a generative AI model based on the pre-processed information. This generative AI model consists of a transformer-type neural network and is designed to generate visual representations that reflect Japanese culture. Existing deep learning frameworks such as TensorFlow and PyTorch are used for this process.

[0444] To evaluate the quality of the generated visual representations, the terminal presents them to evaluators with diverse cultural backgrounds. The evaluators assess the appropriateness of the presented visual representations and provide feedback. This evaluation is based on cultural accuracy and visual appeal.

[0445] The server receives feedback from evaluators, detects biases in the generated AI model, and readjusts the model's weights. This creates a feedback loop that further improves the quality of the generated visual representations.

[0446] Ultimately, the user receives a visual representation generated via a finely tuned generative AI model. The user can then verify that this visual representation accurately reflects Japanese culture and utilize it according to their individual needs.

[0447] As a concrete example, if a server inputs the prompt "Generate images themed on the Japanese tea ceremony" into an AI model, the AI ​​model will generate visual representations based on the tea ceremony. Through this generation process, it becomes possible to deepen our understanding of Japanese culture and provide beautiful and accurate cultural visual representations.

[0448] The flow of the specific processing in Example 1 will be explained using Figure 11.

[0449] Step 1:

[0450] The server collects information related to Japanese culture from the internet and databases. This information includes image data of traditional festivals, architectural styles, costumes, and landscapes. Raw image data is used as input, and a dataset for preprocessing is generated as output. At this stage, the volume and diversity of the collected data are checked, and sufficient data is ensured for subsequent processing.

[0451] Step 2:

[0452] The server performs preprocessing on the collected data, equalizing its resolution and removing noise. It uses the dataset obtained in Step 1 as input, unifying the resolution and filtering out unwanted noise using image editing software. The output is a high-quality, clean dataset suitable for training. Specifically, it performs resizing and filtering using image processing libraries.

[0453] Step 3:

[0454] The server uses a preprocessed dataset to train a generative AI model. The preprocessed dataset is fed as input to the AI ​​framework, which then trains the model based on a transformer-type neural network. The output is a fully trained generative AI model capable of reflecting Japanese culture. This step includes setting hyperparameters and evaluating the model's accuracy.

[0455] Step 4:

[0456] The terminal presents the evaluator with visual representations generated by a generative AI model. The input is the image generated by the AI ​​model, and the output is the evaluator's feedback. The evaluator judges whether the visual representations are culturally accurate and appropriate, and inputs their evaluation into the terminal. Specifically, the evaluation information is aggregated through the user interface.

[0457] Step 5:

[0458] The server detects bias in the generating AI model and readjusts the weights based on the evaluation results received from the terminal. It analyzes the evaluator's feedback as input to confirm the presence of bias. The output is an AI model readjusted to generate more accurate and culturally accurate images. Specifically, it adjusts the model parameters using a feedback loop.

[0459] Step 6:

[0460] The user receives the final visual representation generated using the retuned generative AI model. As input, they receive images generated from the optimized AI model, and as output, they save or use the images according to their purpose. The user may verify that the resulting images meet their expectations and provide feedback. Specific actions include downloading images and sending user feedback to the server.

[0461] (Application Example 1)

[0462] Next, we will explain Application Example 1. In the following explanation, the data processing device 12 will be referred to as the "server," and the headset-type terminal 314 will be referred to as the "terminal."

[0463] In the tourism industry, providing real-time images and information that reflect the actual context of tourist destinations is necessary to deepen visitors' cultural understanding of the areas they visit. However, conventional systems have struggled to generate images that accurately reflect the cultural elements of individual tourist destinations, resulting in a lack of sufficient cultural experience for visitors.

[0464] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 1 is realized by the following means.

[0465] In this invention, the server includes means for acquiring location information, means for collecting relevant data based on the acquired location information, and means for training a generative model using the dataset. This makes it possible to generate and provide images that reflect specific cultural elements of tourist destinations in real time.

[0466] "Location information" refers to data that indicates the geographical coordinates or specific location where the user is currently located.

[0467] "Relevant data" refers to data obtained based on location information, including information about the cultural background, traditions, and history of that place or region.

[0468] A "dataset" is a collection of diverse, pre-processed information used to train an image generation model.

[0469] A "generative model" is a system designed to generate images of specific cultural elements using AI technology.

[0470] "Real-time" refers to processing or responding immediately without delay.

[0471] An "image" is information presented in the form of a diagram or picture that is provided visually.

[0472] "User devices" refer to electronic devices used by users to receive information, such as smartphones and smart glasses.

[0473] An "evaluator" is a person with diverse cultural backgrounds who has been selected to evaluate the content and quality of the generated images.

[0474] "Feedback" refers to opinions and evaluations from users and evaluators that are used to improve system performance.

[0475] "Bias" refers to factors that cause inaccurate results that lack the expected equilibrium when a generative model has a particular tendency or bias.

[0476] This invention is a system for providing visitors seeking a deeper understanding of Japanese culture with images reflecting specific regional cultures in real time. Details for implementing this system are provided below.

[0477] First, when a user visits a tourist destination, their device acquires location information. This allows the system to identify the user's location. Next, the server uses the acquired geographical information to collect cultural data related to that region from its database. This data includes information on traditions, history, festivals, costumes, architecture, and more.

[0478] Subsequently, the server uses an AI generative model based on the collected data to generate images that accurately reflect the local culture. This generative model is based on a transformer network and is trained using frameworks such as TensorFlow and PyTorch.

[0479] The generated images and cultural background information are provided to the user through their device. This allows tourists to deepen their cultural understanding of their destination and have a richer experience.

[0480] For example, if a user visits "Kyoto," the system sends a prompt to the Transformer model saying, "Generate images that reflect Kyoto's festivals and history." This then provides the user with an image of the Gion Festival as a specific example, along with an explanation of the festival's origins and importance.

[0481] Furthermore, feedback from users and evaluators with diverse cultural backgrounds is sent to the server and used to tune the model. This allows the AI ​​model to be continuously improved, enabling more accurate and reliable image generation.

[0482] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[0483] Step 1:

[0484] The device acquires location information from users who have arrived at a tourist destination. The input is the latitude and longitude data of the current location using GPS functionality, and the output is that same data. This location information is used to collect related data in the next step.

[0485] Step 2:

[0486] The server collects relevant cultural data from a database based on the acquired location information. The input is the location information acquired in step 1, and the output is cultural background data identified based on that information. The server accesses the database using SQL queries to search for and collect information such as festivals, buildings, and historical events related to the visited location.

[0487] Step 3:

[0488] The server uses the collected cultural data to send a prompt to a generative AI model, which then generates an image. The input is the collected cultural data and the prompt, and the output is the generated image. A transformer model is used, which processes the data to create an image that reflects cultural elements. The prompt is in the format of "Generate an image that reflects the local culture for tourists visiting AA."

[0489] Step 4:

[0490] Along with the generated image, the server sends cultural background information to the terminal. The input is the image and cultural background information generated in step 3, and the output is the display on the user's terminal. The image is presented visually on the terminal and displayed in different formats so that the user can view detailed information.

[0491] Step 5:

[0492] Users send feedback on displayed images and information to the server via their device. The input is user feedback, and the output is evaluation data stored on the server. The feedback concerns the cultural accuracy of the images and the user experience, and is used to improve the system.

[0493] Step 6:

[0494] The server evaluates the bias and accuracy of the generated AI model based on the collected feedback and adjusts the model as needed. The input is user feedback data, and the output is the parameters of the adjusted model. By retraining the model, higher accuracy image generation can be expected in subsequent attempts.

[0495] Furthermore, an emotion engine that estimates the user's emotions may be incorporated. That is, the identification processing unit 290 may use the emotion identification model 59 to estimate the user's emotions and perform identification processing using the user's emotions.

[0496] This invention provides a system for generating images that accurately reflect Japanese culture and take into consideration the user's emotions. This technology combines an image generation model with an emotion engine that analyzes the user's feelings.

[0497] First, the server collects diverse image datasets related to Japanese culture and preprocesses the data. These datasets include a variety of elements representing Japanese culture, such as festivals, traditional crafts, architecture, clothing, and nature. Based on the preprocessed data, an AI generative model is built and trained to have the ability to generate images that reflect Japanese culture.

[0498] Next, the device uses an emotion engine to recognize the user's emotions in real time. While the user is viewing the generated images, the camera and sensors analyze facial expressions, tone of voice, and other factors to collect emotional data such as positive, negative, and neutral. This process helps the device understand what emotional response the images are eliciting from the user.

[0499] Once user sentiment data is collected, the server feeds this information back to the image generation model. Using the data analyzed by the sentiment engine, the model optimizes the generated images and adjusts them to elicit more emotionally positive responses from the user. For example, if a user shows a positive reaction to a particular cultural element, the model provides guidance for generating images that emphasize that element.

[0500] Ultimately, the server provides users with images that have been fine-tuned to take emotional responses into account. These images not only accurately reflect Japanese culture but also aim to provide users with a desirable emotional experience.

[0501] For example, if a user requests an image themed around a spring festival, the generative model will generate an image that includes festival floats and Japanese drums. When the emotion engine analyzes the user's response as positive, it adjusts the image to further highlight those elements. This process leads to greater user satisfaction and improves the reliability of the generative system.

[0502] The following describes the processing flow.

[0503] Step 1:

[0504] The server collects image data related to Japanese culture from the internet and existing databases. The collected data includes images of Japanese festivals, traditional clothing, architectural styles, and natural landscapes. This data undergoes pre-processing, such as standardizing image resolution and removing noise.

[0505] Step 2:

[0506] The server trains an AI generative model using a preprocessed dataset. This model employs a transformer-type network and is designed to learn the characteristics of Japanese culture. By learning this dataset, the model acquires the ability to represent cultural elements unique to Japan.

[0507] Step 3:

[0508] The user accesses the system and requests image generation based on a specific Japanese cultural theme (e.g., a spring festival). The device then uses its camera and microphone to initiate emotion recognition, collecting emotion data in real time from the user's facial expressions and voice.

[0509] Step 4:

[0510] The device analyzes the collected emotional data using an emotion engine. The emotion engine identifies the emotion the user is feeling (positive, negative, or neutral) from the input data. For example, if the user is smiling, it will be judged as positive.

[0511] Step 5:

[0512] The server receives the analyzed emotion data and uses it as feedback for the generated image. Based on this information, the AI ​​model adjusts the generation result, reinforcing elements that make the user more likely to feel positive emotions. In this way, an optimal image is generated that matches the user's emotions.

[0513] Step 6:

[0514] The server provides the user with the final, adjusted image. The user confirms that this image accurately reflects Japanese culture and evokes a positive emotional experience for them. This improves the overall user experience of the system.

[0515] (Example 2)

[0516] Next, we will describe Example 2. In the following description, the data processing device 12 will be referred to as the "server," and the headset-type terminal 314 will be referred to as the "terminal."

[0517] There are technical challenges in generating accurate and emotionally sensitive images for users using information related to Japanese culture. Conventional technologies have struggled to generate images that adequately consider cultural accuracy and the user's emotional state, resulting in a limited user experience.

[0518] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means.

[0519] In this invention, the server includes means for collecting and processing information related to Japanese culture, means for training an automatically generated model using said information, and means for acquiring emotional information using a device for analyzing the user's emotional state. This makes it possible to optimize the generated images based on the user's emotions and provide a more satisfying user experience.

[0520] "Japanese culture" refers to the totality of elements related to Japan's unique history, values, customs, art, and so on.

[0521] "Information" refers to data or content stored in digital or analog format, including text, images, and audio data.

[0522] "Processing" refers to a series of operations such as collecting, analyzing, organizing, and transforming information, with the aim of making it usable.

[0523] An "automatic generation model" refers to an algorithm or program that uses machine learning or artificial intelligence techniques to automatically generate results from input data.

[0524] A "device for analyzing emotional states" refers to a combination of hardware and software that uses cameras, microphones, sensors, etc., to recognize emotions from a person's facial expressions, voice, and posture.

[0525] "Emotional information" refers to data that represents a user's emotions and mood, and includes states such as positive, negative, and neutral, collected through emotion recognition technology.

[0526] "Generated image" refers to a digital image output by an automated generation model based on the input information.

[0527] "Optimization" refers to a series of operations that adjust or improve a system or process in order to enhance its adaptability and efficiency to a given purpose.

[0528] "User experience" refers to the overall satisfaction and emotional response that users feel when using a product or service.

[0529] This invention is a system that utilizes data related to Japanese culture to generate images adapted to the user's emotional response. To implement this invention, the system is configured as follows.

[0530] The server collects information about Japanese culture and manages it as a dataset. This information is obtained from the internet and other sources and includes images and text data. The server also preprocesses the data, performing noise reduction, formatting standardization, and labeling. This creates a dataset that is optimally suited for training generative AI models. The generative AI models are trained using software such as TensorFlow and PyTorch.

[0531] The device is equipped with a system for analyzing the user's emotional state in real time. Using a camera and microphone, it analyzes the user's facial expressions and voice tone with an emotion engine to extract emotional information. Emotion recognition can utilize emotion analysis tools such as OpenFace or IBM Watson.

[0532] The server optimizes its generative AI model based on emotional information received from the device to generate images tailored to the user. Because the generated images are adjusted based on the user's emotions, a more personalized experience can be provided. The server then sends the optimized images to the device for the user to receive.

[0533] For example, if a user enters a prompt requesting "images themed around a spring festival," the AI ​​model will generate images that include festival floats and Japanese drums. If the device analyzes the user's facial expression as positive at this stage, it can provide images that further emphasize those elements.

[0534] An example of a prompt message would be, "Generate an image of a Japanese spring festival, and if the user shows a positive reaction, highlight the festival floats and taiko drums." This improves user satisfaction and increases the reliability of the system.

[0535] The flow of the specific processing in Example 2 will be explained using Figure 13.

[0536] Step 1:

[0537] The server collects information about Japanese culture from the internet and other data sources. This information includes images and text. It takes raw data as input, performs noise reduction and resolution adjustment, and generates a pre-processed dataset as output. In this process, unnecessary elements are trimmed from the collected images and the sizes are standardized.

[0538] Step 2:

[0539] The server trains an automated generative model based on a preprocessed dataset. Software such as TensorFlow and PyTorch are used. By using preprocessed image data as input and learning hidden patterns and features within the data, a generative model reflecting Japanese culture is obtained as output. Here, the model is tuned to learn cultural elements such as tatami mats and traditional Japanese clothing.

[0540] Step 3:

[0541] The device analyzes facial expressions and voice through the camera and microphone while the user is viewing images. It takes real-time video and audio data as input, uses an emotion engine to analyze positive, negative, and neutral emotional states, and generates emotional information as output. For example, a smile is interpreted as a positive state.

[0542] Step 4:

[0543] The server feeds back the sentiment information received from the terminal to the generative model. It receives sentiment data and prompt text as input and generates or optimizes an image based on that feedback. The output is a customized image tailored to the user's sentiment. For example, if the user has a positive reaction to an image of cherry blossoms, the server will adjust and generate an image that emphasizes cherry blossoms.

[0544] Step 5:

[0545] The user receives the generated image provided by the server on their device and visually confirms it. The user experience is completed by viewing the generated image as output. Because this image is adjusted based on the user's own emotional information, it provides a greater emotional satisfaction.

[0546] (Application Example 2)

[0547] Next, we will explain application example 2. In the following explanation, the data processing device 12 will be referred to as the "server," and the headset-type terminal 314 will be referred to as the "terminal."

[0548] This invention solves the technical challenge of generating images that reflect Japanese culture while considering user emotions and providing a culturally accurate and emotionally positive experience. Conventional technologies have found it difficult to consider user emotions when generating images based on Japanese culture, and have therefore failed to improve the quality of the user experience.

[0549] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means.

[0550] In this invention, the server includes means for analyzing the user's emotions and feeding that information back to a generation model, means for adjusting the generated image based on the user's emotions, and means for providing the generated image to the user. This makes it possible to generate images that are culturally accurate while eliciting an emotionally positive response from the user.

[0551] A "data set" is a collection of information containing diverse elements, including visual information related to Japanese culture.

[0552] A "generative model" is an algorithm that is trained using a dataset and automatically generates new images.

[0553] An "evaluator" is a person or group whose role is to evaluate the generated images and use the results to adjust the model.

[0554] "Evaluation information" refers to information that shows the evaluation results that the evaluator gave to the generated image.

[0555] "Bias" refers to the skew that occurs during the model generation process, and can arise when certain cultural elements are not accurately reflected.

[0556] "Emotional analysis" is a technology that analyzes a user's psychological state in real time through facial expressions, tone of voice, and other factors, and infers their emotional state.

[0557] "Feedback" is the act of returning information obtained through sentiment analysis to a generative model to help optimize the generative process.

[0558] "Image adjustment" is the process of correcting images generated according to the user's emotions to make them more appropriate.

[0559] A "feedback loop" is a process that repeatedly utilizes information obtained from users and evaluators in order to continuously improve the performance of a generative model.

[0560] This invention relates to a system that richly expresses Japanese culture and generates images that take into account the user's emotions. This system operates primarily through the cooperation of a server and a user terminal.

[0561] The server first collects and preprocesses a dataset containing diverse elements of Japanese culture. This preprocessed dataset is then used to train a generative AI model. Typically, machine learning libraries such as TensorFlow are used for training. This generative model has the ability to generate images that accurately reflect Japanese culture.

[0562] On the user's device, sentiment analysis plays a crucial role. The device's built-in camera and microphone capture the user's facial expressions and voice tone in real time, and their emotional state is inferred using OpenCV or the Google Cloud Speech-to-Text API.

[0563] The emotion data generated by the emotion analysis engine is fed back to the server. Based on this information, the server optimizes the generative model and adjusts the image generation process. This results in the generation of images that elicit more positive emotional responses from the user.

[0564] For example, when a user selects and views a "cherry blossom patterned fan" in a virtual store, if sentiment analysis detects that the user is smiling, the image on the screen will be adjusted to include a background of cherry blossom trees in bloom. In this way, the user can enjoy a more satisfying experience.

[0565] An example of a prompt for the generation AI model is, "Adjust the background to match the smiling user holding a cherry blossom-patterned fan, and generate an image that more effectively represents Japanese culture." This program offers new possibilities for improving the user experience.

[0566] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[0567] Step 1:

[0568] The server collects data related to Japanese culture from the internet and local repositories. This includes image data of traditional events, costumes, architecture, and natural landscapes. The collected data is preprocessed, such as denoising and format conversion, to be ready for use as training data for generative AI models. The output of this step is the preprocessed dataset.

[0569] Step 2:

[0570] The server trains a generative AI model using preprocessed data. Machine learning libraries such as TensorFlow are used for training to build a model capable of generating images that accurately reflect Japanese culture. This step involves data calculations to adjust model parameters based on a large amount of data and improve prediction accuracy. The output is the trained image generation model.

[0571] Step 3:

[0572] The user terminal captures the user's facial expressions and voice tone through its camera and microphone. The acquired data is analyzed in real time using tools such as OpenCV and the Google Cloud Speech-to-Text API to infer the user's emotional state (positive, negative, or neutral). The input is the user's real-time video and audio data, and the output is the analyzed emotional data.

[0573] Step 4:

[0574] The data obtained through sentiment analysis is fed back to the server. The server uses this sentiment data to instruct the generative AI model to optimize the image generation process. Specifically, it selects elements of Japanese culture to emphasize according to the user's emotions and adjusts the generated image. The input for this step is the sentiment analysis data, and the output is a prompt message with adjustment instructions.

[0575] Step 5:

[0576] The server generates an image using a generative model based on adjustment instructions, applies necessary modifications, and then produces the final image. This image is optimized based on the user's emotions and cultural preferences. The generated image is sent to the user's terminal. The input is a prompt and a generative model, and the output is a customized image.

[0577] Step 6:

[0578] Users receive the final image on their device and experience it through browsing in virtual stores and galleries. User feedback may be sent to the server for further refinement, forming a feedback loop. This step provides the final user experience and provides potential feedback. The output is an evaluation of the improvement in the quality of the user experience.

[0579] The specific processing unit 290 transmits the result of the specific processing to the headset terminal 314. In the headset terminal 314, the control unit 46A causes the speaker 240 and display 343 to output the result of the specific processing. The microphone 238 acquires audio indicating user input for the result of the specific processing. The control unit 46A transmits the audio data indicating user input acquired by the microphone 238 to the data processing unit 12. In the data processing unit 12, the specific processing unit 290 acquires the audio data.

[0580] Data generation model 58 is a type of so-called generative AI (Artificial Intelligence). One example of data generation model 58 is ChatGPT (Internet search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search) <url: https: gemini.google.com ?hl="ja">Examples of generative AI include the following. The data generation model 58 is obtained by performing deep learning on a neural network. The data generation model 58 is input with prompts containing instructions, and with inference data such as audio data representing speech, text data representing text, and image data representing images. The data generation model 58 infers from the input inference data according to the instructions indicated by the prompts, and outputs the inference results in data formats such as audio data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0581] In the above embodiment, an example was given in which specific processing is performed by the data processing device 12, but the technology of this disclosure is not limited thereto, and specific processing may also be performed by the headset terminal 314.

[0582] [Fourth Embodiment]

[0583] Figure 7 shows an example of the configuration of the data processing system 410 according to the fourth embodiment.

[0584] As shown in Figure 7, the data processing system 410 includes a data processing device 12 and a robot 414. An example of the data processing device 12 is a server.

[0585] The data processing device 12 comprises a computer 22, a database 24, and a communication interface 26. The computer 22 is an example of a "computer" related to the technology of this disclosure. The computer 22 comprises a processor 28, RAM 30, and storage 32. The processor 28, RAM 30, and storage 32 are connected to a bus 34. The database 24 and the communication interface 26 are also connected to the bus 34. The communication interface 26 is connected to a network 54. An example of the network 54 is a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0586] The robot 414 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication interface 44, and a controlled object 443. The computer 36 includes a processor 46, RAM 48, and storage 50. The processor 46, RAM 48, and storage 50 are connected to a bus 52. The microphone 238, speaker 240, camera 42, and controlled object 443 are also connected to the bus 52.

[0587] The microphone 238 receives voice signals from the user 20 and receives instructions from the user 20. The microphone 238 captures the voice signals from the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio according to the instructions from the processor 46.

[0588] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an image sensor such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the area around the user 20 (for example, an imaging range defined by a field of view equivalent to the width of a typical healthy person's field of vision).

[0589] Communication interface 44 is connected to network 54. Communication interfaces 44 and 26 are responsible for the exchange of various information between processor 46 and processor 28 via network 54. The exchange of various information between processor 46 and processor 28 using communication interfaces 44 and 26 is performed in a secure manner.

[0590] The controlled object 443 includes a display device, LEDs in the eyes, and motors that drive the arms, hands, and feet. The posture and gestures of the robot 414 are controlled by controlling the motors of the arms, hands, and feet. Some of the robot 414's emotions can be expressed by controlling these motors. Furthermore, the robot 414's facial expressions can also be expressed by controlling the illumination state of the LEDs in its eyes.

[0591] Figure 8 shows an example of the main functions of the data processing device 12 and the robot 414. As shown in Figure 8, the data processing device 12 performs specific processing using the processor 28. The storage 32 stores the specific processing program 56.

[0592] The specific processing program 56 is an example of a "program" relating to the technology of this disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0593] The storage 32 stores the data generation model 58 and the emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0594] In robot 414, the processor 46 performs the reception output processing. The storage 50 stores the reception output program 60. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output processing is realized by the processor 46 operating as a control unit 46A according to the reception output program 60 executed on the RAM 48.

[0595] Next, the specific processing performed by the specific processing unit 290 of the data processing device 12 will be described. In the following description, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[0596] This invention provides a system for generating images that accurately reflect Japanese culture. Specific embodiments thereof will be described below.

[0597] First, the server collects a dataset containing diverse elements that constitute Japanese culture. This dataset includes traditional festivals, architectural styles, costumes, landscapes, and more. The collected data is then pre-processed, such as unifying image resolution and removing noise. This process ensures the necessary quality when generating images.

[0598] Next, the server trains an AI generative model using the preprocessed dataset. This AI model employs a transformer-type network architecture and is designed to learn the uniqueness and diversity of Japanese culture. The model is trained to generate images that reflect elements of Japanese culture.

[0599] To evaluate the cultural accuracy of images generated by a trained AI model, the device selects evaluators with diverse cultural backgrounds and presents them with the generated images. The evaluators determine whether the generated images accurately represent Japanese culture and whether they contain any elements that may cause offense. During this process, the evaluators' feedback is sent to a server.

[0600] The server detects biases in the generative model based on feedback from evaluators and readjusts the model weights as needed. For example, if an evaluator feels that a particular cultural element is not being represented correctly, the server applies feedback to the AI ​​model to correct that.

[0601] Finally, the user is provided with the final image generated by the finely tuned AI model. The user can then verify that the received image accurately reflects Japanese culture and meets their expected quality. This entire process generates and provides images that accurately represent uniquely Japanese cultural elements.

[0602] This invention enables the generation AI system to promote cultural understanding and gain user trust by providing generation results that respect and accurately reflect Japanese culture.

[0603] The following describes the processing flow.

[0604] Step 1:

[0605] The server collects diverse image data related to Japanese culture from the internet and existing databases. The collected data includes various elements that constitute Japanese culture, such as traditions, festivals, architecture, clothing, and landscapes. After collection, the image data undergoes preprocessing such as resolution standardization and noise reduction.

[0606] Step 2:

[0607] The server builds an AI generative model using the preprocessed dataset and begins training. The model used here is a transformer-type network designed to learn the standards and characteristics of Japanese culture. By repeatedly learning the dataset, the model will be able to understand and effectively represent the unique cultural features of Japan.

[0608] Step 3:

[0609] The device generates images using a model that has completed training. These generated images are presented to pre-selected evaluators with diverse cultural backgrounds to assess whether they accurately reflect elements of Japanese culture. The evaluators provide feedback on the generated images regarding their cultural appropriateness and lack of bias.

[0610] Step 4:

[0611] The server processes the evaluator's feedback sent from the terminal. Based on the feedback, it adjusts the model parameters if any bias or cultural misunderstanding exists in the generated images. The model is then fine-tuned based on this feedback.

[0612] Step 5:

[0613] The server generates the final image using the modified generative model and provides it to the user. This image is required to accurately reproduce Japanese culture and meet the user's desired quality and expectations. User feedback will be used to improve the model in the future.

[0614] (Example 1)

[0615] Next, we will describe Example 1. In the following description, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[0616] The problem that this invention aims to solve is to generate visual representations that accurately reflect Japanese culture from diverse perspectives using a generative AI model, while minimizing bias and inaccuracies based on cultural background.

[0617] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 1 is realized by the following means.

[0618] In this invention, the server includes means for collecting information and performing image quality uniformization and noise reduction, means for training a generative AI model based on the collected information, and means for selecting evaluators with diverse backgrounds and evaluating the generated visual representations. This makes it possible to enhance the cultural accuracy of the generated visual representations and continuously improve the performance of the generative AI model.

[0619] "Information" refers to the collective data collected to create visual representations related to Japanese culture, including traditional festivals, architectural styles, costumes, and landscapes.

[0620] A "generative AI model" is an artificial intelligence model designed to learn from collected information and generate visual representations that reflect Japanese culture.

[0621] "Visual representations" refer to visual content such as images and videos created by generative AI models that reflect Japanese culture.

[0622] "Bias" refers to the tendency for a visual representation to contain unintended cultural errors due to the generative AI model overemphasizing or ignoring certain cultural elements.

[0623] A "weight" is one of the parameters within a generative AI model, and it is a factor that determines how the model processes information and what kind of output it produces.

[0624] An "evaluator" is an individual or organization that evaluates generated visual representations from diverse cultural backgrounds and judges their accuracy and appropriateness.

[0625] "Noise reduction" is a preprocessing step that removes unnecessary data and visual defects from collected information, enabling the generative AI model to produce high-quality visual representations.

[0626] A "feedback loop" is a continuous improvement process that readjusts the model based on evaluations from evaluators to improve the output accuracy of the generated AI model.

[0627] "Cultural accuracy" refers to the appropriateness of a generated visual representation, ensuring it accurately reflects Japanese culture and avoids misunderstanding or offense.

[0628] "Users" refer to end-users who receive the final generated visual representation and evaluate its cultural accuracy and aesthetic value.

[0629] This invention is a system for generating visual representations that accurately reflect Japanese culture. Specifically, the system is constructed through the cooperation of three parties: a server, a terminal, and a user.

[0630] First, the server collects information related to Japanese culture from the internet and existing databases. This information includes visual elements such as festivals, architecture, clothing, and landscapes. The server uses image editing software and filtering techniques to equalize the image quality of the collected information and remove noise. Specifically, it can utilize open-source image processing libraries.

[0631] Next, the server trains a generative AI model based on the pre-processed information. This generative AI model consists of a transformer-type neural network and is designed to generate visual representations that reflect Japanese culture. Existing deep learning frameworks such as TensorFlow and PyTorch are used for this process.

[0632] To evaluate the quality of the generated visual representations, the terminal presents them to evaluators with diverse cultural backgrounds. The evaluators assess the appropriateness of the presented visual representations and provide feedback. This evaluation is based on cultural accuracy and visual appeal.

[0633] The server receives feedback from evaluators, detects biases in the generated AI model, and readjusts the model's weights. This creates a feedback loop that further improves the quality of the generated visual representations.

[0634] Ultimately, the user receives a visual representation generated via a finely tuned generative AI model. The user can then verify that this visual representation accurately reflects Japanese culture and utilize it according to their individual needs.

[0635] As a concrete example, if a server inputs the prompt "Generate images themed on the Japanese tea ceremony" into an AI model, the AI ​​model will generate visual representations based on the tea ceremony. Through this generation process, it becomes possible to deepen our understanding of Japanese culture and provide beautiful and accurate cultural visual representations.

[0636] The flow of the specific processing in Example 1 will be explained using Figure 11.

[0637] Step 1:

[0638] The server collects information related to Japanese culture from the internet and databases. This information includes image data of traditional festivals, architectural styles, costumes, and landscapes. Raw image data is used as input, and a dataset for preprocessing is generated as output. At this stage, the volume and diversity of the collected data are checked, and sufficient data is ensured for subsequent processing.

[0639] Step 2:

[0640] The server performs preprocessing on the collected data, equalizing its resolution and removing noise. It uses the dataset obtained in Step 1 as input, unifying the resolution and filtering out unwanted noise using image editing software. The output is a high-quality, clean dataset suitable for training. Specifically, it performs resizing and filtering using image processing libraries.

[0641] Step 3:

[0642] The server uses a preprocessed dataset to train a generative AI model. The preprocessed dataset is fed as input to the AI ​​framework, which then trains the model based on a transformer-type neural network. The output is a fully trained generative AI model capable of reflecting Japanese culture. This step includes setting hyperparameters and evaluating the model's accuracy.

[0643] Step 4:

[0644] The terminal presents the evaluator with visual representations generated by a generative AI model. The input is the image generated by the AI ​​model, and the output is the evaluator's feedback. The evaluator judges whether the visual representations are culturally accurate and appropriate, and inputs their evaluation into the terminal. Specifically, the evaluation information is aggregated through the user interface.

[0645] Step 5:

[0646] The server detects bias in the generating AI model and readjusts the weights based on the evaluation results received from the terminal. It analyzes the evaluator's feedback as input to confirm the presence of bias. The output is an AI model readjusted to generate more accurate and culturally accurate images. Specifically, it adjusts the model parameters using a feedback loop.

[0647] Step 6:

[0648] The user receives the final visual representation generated using the retuned generative AI model. As input, they receive images generated from the optimized AI model, and as output, they save or use the images according to their purpose. The user may verify that the resulting images meet their expectations and provide feedback. Specific actions include downloading images and sending user feedback to the server.

[0649] (Application Example 1)

[0650] Next, we will explain Application Example 1. In the following explanation, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[0651] In the tourism industry, providing real-time images and information that reflect the actual context of tourist destinations is necessary to deepen visitors' cultural understanding of the areas they visit. However, conventional systems have struggled to generate images that accurately reflect the cultural elements of individual tourist destinations, resulting in a lack of sufficient cultural experience for visitors.

[0652] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 1 is realized by the following means.

[0653] In this invention, the server includes means for acquiring location information, means for collecting relevant data based on the acquired location information, and means for training a generative model using the dataset. This makes it possible to generate and provide images that reflect specific cultural elements of tourist destinations in real time.

[0654] "Location information" refers to data that indicates the geographical coordinates or specific location where the user is currently located.

[0655] "Relevant data" refers to data obtained based on location information, including information about the cultural background, traditions, and history of that place or region.

[0656] A "dataset" is a collection of diverse, pre-processed information used to train an image generation model.

[0657] A "generative model" is a system designed to generate images of specific cultural elements using AI technology.

[0658] "Real-time" refers to processing or responding immediately without delay.

[0659] An "image" is information presented in the form of a diagram or picture that is provided visually.

[0660] "User devices" refer to electronic devices used by users to receive information, such as smartphones and smart glasses.

[0661] An "evaluator" is a person with diverse cultural backgrounds who has been selected to evaluate the content and quality of the generated images.

[0662] "Feedback" refers to opinions and evaluations from users and evaluators that are used to improve system performance.

[0663] "Bias" refers to factors that cause inaccurate results that lack the expected equilibrium when a generative model has a particular tendency or bias.

[0664] This invention is a system for providing visitors seeking a deeper understanding of Japanese culture with images reflecting specific regional cultures in real time. Details for implementing this system are provided below.

[0665] First, when a user visits a tourist destination, their device acquires location information. This allows the system to identify the user's location. Next, the server uses the acquired geographical information to collect cultural data related to that region from its database. This data includes information on traditions, history, festivals, costumes, architecture, and more.

[0666] Subsequently, the server uses an AI generative model based on the collected data to generate images that accurately reflect the local culture. This generative model is based on a transformer network and is trained using frameworks such as TensorFlow and PyTorch.

[0667] The generated images and cultural background information are provided to the user through their device. This allows tourists to deepen their cultural understanding of their destination and have a richer experience.

[0668] For example, if a user visits "Kyoto," the system sends a prompt to the Transformer model saying, "Generate images that reflect Kyoto's festivals and history." This then provides the user with an image of the Gion Festival as a specific example, along with an explanation of the festival's origins and importance.

[0669] Furthermore, feedback from users and evaluators with diverse cultural backgrounds is sent to the server and used to tune the model. This allows the AI ​​model to be continuously improved, enabling more accurate and reliable image generation.

[0670] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[0671] Step 1:

[0672] The device acquires location information from users who have arrived at a tourist destination. The input is the latitude and longitude data of the current location using GPS functionality, and the output is that same data. This location information is used to collect related data in the next step.

[0673] Step 2:

[0674] The server collects relevant cultural data from a database based on the acquired location information. The input is the location information acquired in step 1, and the output is cultural background data identified based on that information. The server accesses the database using SQL queries to search for and collect information such as festivals, buildings, and historical events related to the visited location.

[0675] Step 3:

[0676] The server uses the collected cultural data to send a prompt to a generative AI model, which then generates an image. The input is the collected cultural data and the prompt, and the output is the generated image. A transformer model is used, which processes the data to create an image that reflects cultural elements. The prompt is in the format of "Generate an image that reflects the local culture for tourists visiting AA."

[0677] Step 4:

[0678] Along with the generated image, the server sends cultural background information to the terminal. The input is the image and cultural background information generated in step 3, and the output is the display on the user's terminal. The image is presented visually on the terminal and displayed in different formats so that the user can view detailed information.

[0679] Step 5:

[0680] Users send feedback on displayed images and information to the server via their device. The input is user feedback, and the output is evaluation data stored on the server. The feedback concerns the cultural accuracy of the images and the user experience, and is used to improve the system.

[0681] Step 6:

[0682] The server evaluates the bias and accuracy of the generated AI model based on the collected feedback and adjusts the model as needed. The input is user feedback data, and the output is the parameters of the adjusted model. By retraining the model, higher accuracy image generation can be expected in subsequent attempts.

[0683] Furthermore, an emotion engine that estimates the user's emotions may be incorporated. That is, the identification processing unit 290 may use the emotion identification model 59 to estimate the user's emotions and perform identification processing using the user's emotions.

[0684] This invention provides a system for generating images that accurately reflect Japanese culture and take into consideration the user's emotions. This technology combines an image generation model with an emotion engine that analyzes the user's feelings.

[0685] First, the server collects diverse image datasets related to Japanese culture and preprocesses the data. These datasets include a variety of elements representing Japanese culture, such as festivals, traditional crafts, architecture, clothing, and nature. Based on the preprocessed data, an AI generative model is built and trained to have the ability to generate images that reflect Japanese culture.

[0686] Next, the device uses an emotion engine to recognize the user's emotions in real time. While the user is viewing the generated images, the camera and sensors analyze facial expressions, tone of voice, and other factors to collect emotional data such as positive, negative, and neutral. This process helps the device understand what emotional response the images are eliciting from the user.

[0687] Once user sentiment data is collected, the server feeds this information back to the image generation model. Using the data analyzed by the sentiment engine, the model optimizes the generated images and adjusts them to elicit more emotionally positive responses from the user. For example, if a user shows a positive reaction to a particular cultural element, the model provides guidance for generating images that emphasize that element.

[0688] Ultimately, the server provides users with images that have been fine-tuned to take emotional responses into account. These images not only accurately reflect Japanese culture but also aim to provide users with a desirable emotional experience.

[0689] For example, if a user requests an image themed around a spring festival, the generative model will generate an image that includes festival floats and Japanese drums. When the emotion engine analyzes the user's response as positive, it adjusts the image to further highlight those elements. This process leads to greater user satisfaction and improves the reliability of the generative system.

[0690] The following describes the processing flow.

[0691] Step 1:

[0692] The server collects image data related to Japanese culture from the internet and existing databases. The collected data includes images of Japanese festivals, traditional clothing, architectural styles, and natural landscapes. This data undergoes pre-processing, such as standardizing image resolution and removing noise.

[0693] Step 2:

[0694] The server trains an AI generative model using a preprocessed dataset. This model employs a transformer-type network and is designed to learn the characteristics of Japanese culture. By learning this dataset, the model acquires the ability to represent cultural elements unique to Japan.

[0695] Step 3:

[0696] The user accesses the system and requests image generation based on a specific Japanese cultural theme (e.g., a spring festival). The device then uses its camera and microphone to initiate emotion recognition, collecting emotion data in real time from the user's facial expressions and voice.

[0697] Step 4:

[0698] The device analyzes the collected emotional data using an emotion engine. The emotion engine identifies the emotion the user is feeling (positive, negative, or neutral) from the input data. For example, if the user is smiling, it will be judged as positive.

[0699] Step 5:

[0700] The server receives the analyzed emotion data and uses it as feedback for the generated image. Based on this information, the AI ​​model adjusts the generation result, reinforcing elements that make the user more likely to feel positive emotions. In this way, an optimal image is generated that matches the user's emotions.

[0701] Step 6:

[0702] The server provides the user with the final, adjusted image. The user confirms that this image accurately reflects Japanese culture and evokes a positive emotional experience for them. This improves the overall user experience of the system.

[0703] (Example 2)

[0704] Next, we will describe Example 2. In the following description, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[0705] There are technical challenges in generating accurate and emotionally sensitive images for users using information related to Japanese culture. Conventional technologies have struggled to generate images that adequately consider cultural accuracy and the user's emotional state, resulting in a limited user experience.

[0706] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means.

[0707] In this invention, the server includes means for collecting and processing information related to Japanese culture, means for training an automatically generated model using said information, and means for acquiring emotional information using a device for analyzing the user's emotional state. This makes it possible to optimize the generated images based on the user's emotions and provide a more satisfying user experience.

[0708] "Japanese culture" refers to the totality of elements related to Japan's unique history, values, customs, art, and so on.

[0709] "Information" refers to data or content stored in digital or analog format, including text, images, and audio data.

[0710] "Processing" refers to a series of operations such as collecting, analyzing, organizing, and transforming information, with the aim of making it usable.

[0711] An "automatic generation model" refers to an algorithm or program that uses machine learning or artificial intelligence techniques to automatically generate results from input data.

[0712] A "device for analyzing emotional states" refers to a combination of hardware and software that uses cameras, microphones, sensors, etc., to recognize emotions from a person's facial expressions, voice, and posture.

[0713] "Emotional information" refers to data that represents a user's emotions and mood, and includes states such as positive, negative, and neutral, collected through emotion recognition technology.

[0714] "Generated image" refers to a digital image output by an automated generation model based on the input information.

[0715] "Optimization" refers to a series of operations that adjust or improve a system or process in order to enhance its adaptability and efficiency to a given purpose.

[0716] "User experience" refers to the overall satisfaction and emotional response that users feel when using a product or service.

[0717] This invention is a system that utilizes data related to Japanese culture to generate images adapted to the user's emotional response. To implement this invention, the system is configured as follows.

[0718] The server collects information about Japanese culture and manages it as a dataset. This information is obtained from the internet and other sources and includes images and text data. The server also preprocesses the data, performing noise reduction, formatting standardization, and labeling. This creates a dataset that is optimally suited for training generative AI models. The generative AI models are trained using software such as TensorFlow and PyTorch.

[0719] The device is equipped with a system for analyzing the user's emotional state in real time. Using a camera and microphone, it analyzes the user's facial expressions and voice tone with an emotion engine to extract emotional information. Emotion recognition can utilize emotion analysis tools such as OpenFace or IBM Watson.

[0720] The server optimizes its generative AI model based on emotional information received from the device to generate images tailored to the user. Because the generated images are adjusted based on the user's emotions, a more personalized experience can be provided. The server then sends the optimized images to the device for the user to receive.

[0721] For example, if a user enters a prompt requesting "images themed around a spring festival," the AI ​​model will generate images that include festival floats and Japanese drums. If the device analyzes the user's facial expression as positive at this stage, it can provide images that further emphasize those elements.

[0722] An example of a prompt message would be, "Generate an image of a Japanese spring festival, and if the user shows a positive reaction, highlight the festival floats and taiko drums." This improves user satisfaction and increases the reliability of the system.

[0723] The flow of the specific processing in Example 2 will be explained using Figure 13.

[0724] Step 1:

[0725] The server collects information about Japanese culture from the internet and other data sources. This information includes images and text. It takes raw data as input, performs noise reduction and resolution adjustment, and generates a pre-processed dataset as output. In this process, unnecessary elements are trimmed from the collected images and the sizes are standardized.

[0726] Step 2:

[0727] The server trains an automated generative model based on a preprocessed dataset. Software such as TensorFlow and PyTorch are used. By using preprocessed image data as input and learning hidden patterns and features within the data, a generative model reflecting Japanese culture is obtained as output. Here, the model is tuned to learn cultural elements such as tatami mats and traditional Japanese clothing.

[0728] Step 3:

[0729] The device analyzes facial expressions and voice through the camera and microphone while the user is viewing images. It takes real-time video and audio data as input, uses an emotion engine to analyze positive, negative, and neutral emotional states, and generates emotional information as output. For example, a smile is interpreted as a positive state.

[0730] Step 4:

[0731] The server feeds back the sentiment information received from the terminal to the generative model. It receives sentiment data and prompt text as input and generates or optimizes an image based on that feedback. The output is a customized image tailored to the user's sentiment. For example, if the user has a positive reaction to an image of cherry blossoms, the server will adjust and generate an image that emphasizes cherry blossoms.

[0732] Step 5:

[0733] The user receives the generated image provided by the server on their device and visually confirms it. The user experience is completed by viewing the generated image as output. Because this image is adjusted based on the user's own emotional information, it provides a greater emotional satisfaction.

[0734] (Application Example 2)

[0735] Next, we will explain application example 2. In the following explanation, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[0736] This invention solves the technical challenge of generating images that reflect Japanese culture while considering user emotions and providing a culturally accurate and emotionally positive experience. Conventional technologies have found it difficult to consider user emotions when generating images based on Japanese culture, and have therefore failed to improve the quality of the user experience.

[0737] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means.

[0738] In this invention, the server includes means for analyzing the user's emotions and feeding that information back to a generation model, means for adjusting the generated image based on the user's emotions, and means for providing the generated image to the user. This makes it possible to generate images that are culturally accurate while eliciting an emotionally positive response from the user.

[0739] A "data set" is a collection of information containing diverse elements, including visual information related to Japanese culture.

[0740] A "generative model" is an algorithm that is trained using a dataset and automatically generates new images.

[0741] An "evaluator" is a person or group whose role is to evaluate the generated images and use the results to adjust the model.

[0742] "Evaluation information" refers to information that shows the evaluation results that the evaluator gave to the generated image.

[0743] "Bias" refers to the skew that occurs during the model generation process, and can arise when certain cultural elements are not accurately reflected.

[0744] "Emotional analysis" is a technology that analyzes a user's psychological state in real time through facial expressions, tone of voice, and other factors, and infers their emotional state.

[0745] "Feedback" is the act of returning information obtained through sentiment analysis to a generative model to help optimize the generative process.

[0746] "Image adjustment" is the process of correcting images generated according to the user's emotions to make them more appropriate.

[0747] A "feedback loop" is a process that repeatedly utilizes information obtained from users and evaluators in order to continuously improve the performance of a generative model.

[0748] This invention relates to a system that richly expresses Japanese culture and generates images that take into account the user's emotions. This system operates primarily through the cooperation of a server and a user terminal.

[0749] The server first collects and preprocesses a dataset containing diverse elements of Japanese culture. This preprocessed dataset is then used to train a generative AI model. Typically, machine learning libraries such as TensorFlow are used for training. This generative model has the ability to generate images that accurately reflect Japanese culture.

[0750] On the user's device, sentiment analysis plays a crucial role. The device's built-in camera and microphone capture the user's facial expressions and voice tone in real time, and their emotional state is inferred using OpenCV or the Google Cloud Speech-to-Text API.

[0751] The emotion data generated by the emotion analysis engine is fed back to the server. Based on this information, the server optimizes the generative model and adjusts the image generation process. This results in the generation of images that elicit more positive emotional responses from the user.

[0752] For example, when a user selects and views a "cherry blossom patterned fan" in a virtual store, if sentiment analysis detects that the user is smiling, the image on the screen will be adjusted to include a background of cherry blossom trees in bloom. In this way, the user can enjoy a more satisfying experience.

[0753] An example of a prompt for the generation AI model is, "Adjust the background to match the smiling user holding a cherry blossom-patterned fan, and generate an image that more effectively represents Japanese culture." This program offers new possibilities for improving the user experience.

[0754] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[0755] Step 1:

[0756] The server collects data related to Japanese culture from the internet and local repositories. This includes image data of traditional events, costumes, architecture, and natural landscapes. The collected data is preprocessed, such as denoising and format conversion, to be ready for use as training data for generative AI models. The output of this step is the preprocessed dataset.

[0757] Step 2:

[0758] The server trains a generative AI model using preprocessed data. Machine learning libraries such as TensorFlow are used for training to build a model capable of generating images that accurately reflect Japanese culture. This step involves data calculations to adjust model parameters based on a large amount of data and improve prediction accuracy. The output is the trained image generation model.

[0759] Step 3:

[0760] The user terminal captures the user's facial expressions and voice tone through its camera and microphone. The acquired data is analyzed in real time using tools such as OpenCV and the Google Cloud Speech-to-Text API to infer the user's emotional state (positive, negative, or neutral). The input is the user's real-time video and audio data, and the output is the analyzed emotional data.

[0761] Step 4:

[0762] The data obtained through sentiment analysis is fed back to the server. The server uses this sentiment data to instruct the generative AI model to optimize the image generation process. Specifically, it selects elements of Japanese culture to emphasize according to the user's emotions and adjusts the generated image. The input for this step is the sentiment analysis data, and the output is a prompt message with adjustment instructions.

[0763] Step 5:

[0764] The server generates an image using a generative model based on adjustment instructions, applies necessary modifications, and then produces the final image. This image is optimized based on the user's emotions and cultural preferences. The generated image is sent to the user's terminal. The input is a prompt and a generative model, and the output is a customized image.

[0765] Step 6:

[0766] Users receive the final image on their device and experience it through browsing in virtual stores and galleries. User feedback may be sent to the server for further refinement, forming a feedback loop. This step provides the final user experience and provides potential feedback. The output is an evaluation of the improvement in the quality of the user experience.

[0767] The specific processing unit 290 transmits the result of the specific processing to the robot 414. In the robot 414, the control unit 46A causes the speaker 240 and the controlled object 443 to output the result of the specific processing. The microphone 238 acquires audio indicating user input for the result of the specific processing. The control unit 46A transmits the audio data indicating user input acquired by the microphone 238 to the data processing unit 12. In the data processing unit 12, the specific processing unit 290 acquires the audio data.

[0768] Data generation model 58 is a type of so-called generative AI (Artificial Intelligence). One example of data generation model 58 is ChatGPT (Internet search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search) <url: https: gemini.google.com ?hl="ja">Examples of generative AI include the following. The data generation model 58 is obtained by performing deep learning on a neural network. The data generation model 58 is input with prompts containing instructions, and with inference data such as audio data representing speech, text data representing text, and image data representing images. The data generation model 58 infers from the input inference data according to the instructions indicated by the prompts, and outputs the inference results in data formats such as audio data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0769] In the above embodiment, an example was given in which the specific processing is performed by the data processing device 12, but the technology of this disclosure is not limited thereto, and the specific processing may also be performed by the robot 414.

[0770] Furthermore, the emotion identification model 59, acting as an emotion engine, may determine the user's emotion according to a specific mapping. Specifically, the emotion identification model 59 may determine the user's emotion according to a specific mapping, which is an emotion map (see Figure 9). Similarly, the emotion identification model 59 may also determine the robot's emotion, and the identification processing unit 290 may perform identification processing using the robot's emotion.

[0771] Figure 9 shows an emotion map 400 in which multiple emotions are mapped. In the emotion map 400, emotions are arranged in concentric circles radiating from the center. The closer to the center of the concentric circles, the more primitive the emotions are located. Further out of the concentric circles, emotions representing states and actions arising from mental states are located. Emotion is a concept that includes feelings and mental states. On the left side of the concentric circles, emotions that are generally generated from reactions occurring in the brain are located. On the right side of the concentric circles, emotions that are generally induced by situational judgment are located. Above and below the concentric circles, emotions that are generally generated from reactions occurring in the brain and induced by situational judgment are located. In addition, the emotion of "pleasure" is located on the upper side of the concentric circles, and the emotion of "displeasure" is located on the lower side. Thus, in the emotion map 400, multiple emotions are mapped based on the structure in which emotions arise, and emotions that are likely to occur simultaneously are mapped close together.

[0772] These emotions are distributed at the 3 o'clock position on the Emotion Map 400, and usually fluctuate between feelings of security and anxiety. In the right half of the Emotion Map 400, situational awareness takes precedence over internal feelings, resulting in a calm impression.

[0773] The inside of the Emotion Map 400 represents inner thoughts, while the outside represents actions. Therefore, the further you go from the outside of the Emotion Map 400, the more visible (expressed in actions) your emotions become.

[0774] Here, human emotions are based on various balances, such as posture and blood sugar levels. When these balances deviate from the ideal, it results in discomfort, and when they approach the ideal, it results in pleasure. Similarly, in robots, cars, motorcycles, etc., emotions can be created based on various balances, such as posture and battery level. When these balances deviate from the ideal, it results in discomfort, and when they approach the ideal, it results in pleasure. The emotion map can be generated, for example, based on Dr. Mitsuyoshi's emotion map (Research on a system for analyzing brain physiological signals of speech emotion recognition and emotion, Tokushima University, doctoral dissertation: https: / / ci.nii.ac.jp / naid / 500000375379). The left half of the emotion map contains emotions belonging to a region called "response," where sensation is dominant. The right half of the emotion map contains emotions belonging to a region called "situation," where situational awareness is dominant.

[0775] The emotion map defines two emotions that promote learning. One is the emotion around the middle of the negative "repentance" and "reflection" on the situation side. In other words, it is when the robot experiences negative emotions such as "I never want to feel this way again" or "I don't want to be scolded again." The other is the emotion around the positive "desire" on the reaction side. In other words, it is when the robot has positive feelings such as "I want more" or "I want to know more."

[0776] The emotion identification model 59 inputs user input into a pre-trained neural network, obtains emotion values ​​representing each emotion shown in the emotion map 400, and determines the user's emotion. This neural network is pre-trained based on multiple training data sets, which are combinations of user input and emotion values ​​representing each emotion shown in the emotion map 400. Furthermore, this neural network is trained so that emotions located close together have similar values, as shown in the emotion map 900 in Figure 10. Figure 10 shows an example where multiple emotions such as "reassured," "calm," and "confident" have similar emotion values.

[0777] The above description primarily focuses on the functions of the data processing device 12 in relation to this disclosure. However, the system related to this disclosure is not necessarily implemented on a server. The system related to this disclosure may be implemented as a general information processing system. This disclosure may be implemented, for example, as a software program that runs on a personal computer or as an application that runs on a smartphone. The method related to this disclosure may be provided to users in SaaS (Software as a Service) format.

[0778] In the above embodiment, an example was given in which a specific process is performed by a single computer 22. However, the technology of this disclosure is not limited thereto, and a distributed processing of the specific process may be performed by multiple computers, including computer 22. For example, a data generation model 58 may be provided in an external device of the data processing device 12, and the external device may generate data according to the input data.

[0779] In the above embodiment, an example was given in which the specific processing program 56 is stored in the storage 32, but the technology of this disclosure is not limited thereto. For example, the specific processing program 56 may be stored in a portable, computer-readable, non-temporary storage medium such as a USB (Universal Serial Bus) memory. The specific processing program 56 stored in the non-temporary storage medium is installed in the computer 22 of the data processing device 12. The processor 28 executes specific processing according to the specific processing program 56.

[0780] Alternatively, the specific processing program 56 may be stored in a storage device such as a server connected to the data processing device 12 via the network 54, and the specific processing program 56 may be downloaded and installed on the computer 22 in response to a request from the data processing device 12.

[0781] Furthermore, it is not necessary to store the entirety of the specific processing program 56 in a storage device such as a server connected to the data processing device 12 via the network 54, or to store the entirety of the specific processing program 56 in the storage 32; it is acceptable to store only a portion of the specific processing program 56.

[0782] The following types of processors can be used as hardware resources to perform specific processing. Examples of processors include a CPU, a general-purpose processor that functions as a hardware resource to perform specific processing by executing software, i.e., a program. Other examples of processors include dedicated electrical circuits, such as FPGAs (Field-Programmable Gate Arrays), PLDs (Programmable Logic Devices), or ASICs (Application Specific Integrated Circuits), which have circuit configurations specifically designed to perform specific processing. All of these processors have built-in or connected memory, and all of them perform specific processing by using memory.

[0783] The hardware resource that performs a specific process may consist of one of these various processors, or it may consist of a combination of two or more processors of the same or different types (for example, a combination of multiple FPGAs, or a combination of a CPU and an FPGA). Alternatively, the hardware resource that performs a specific process may consist of a single processor.

[0784] Examples of configurations using a single processor include, firstly, a configuration in which one or more CPUs and software are combined to form a single processor, and this processor functions as a hardware resource that performs a specific process. Secondly, there is a configuration using a processor that realizes the functions of the entire system, including multiple hardware resources that perform a specific process, on a single IC chip, as exemplified by SoCs (System-on-a-chip). In this way, a specific process is realized using one or more of the above types of processors as hardware resources.

[0785] Furthermore, the hardware structure of these various processors can more specifically utilize electrical circuits that combine circuit elements such as semiconductor devices. Also, the specific processing described above is merely an example. Therefore, it goes without saying that unnecessary steps can be deleted, new steps added, or the processing order rearranged, as long as it does not deviate from the main purpose.

[0786] The descriptions and illustrations presented above are detailed explanations of the technical aspects of this disclosure and are merely examples of the technical aspects. For example, the above descriptions of the structure, function, operation, and effect are examples of the structure, function, operation, and effect of the technical aspects of this disclosure. Therefore, it goes without saying that you may delete unnecessary parts, add new elements, or replace elements in the descriptions and illustrations presented above, as long as you do not deviate from the essence of the technical aspects of this disclosure. Furthermore, in order to avoid confusion and facilitate understanding of the technical aspects of this disclosure, explanations of common technical knowledge and the like that do not require special explanation to enable the implementation of the technical aspects of this disclosure have been omitted from the descriptions and illustrations presented above.

[0787] All documents, patent applications, and technical standards described herein are incorporated by reference to the same extent as if each individual document, patent application, and technical standard were specifically and individually noted to be incorporated by reference.

[0788] The following is further disclosed regarding the embodiments described above.

[0789] (Claim 1)

[0790] A means for collecting and preprocessing datasets related to Japanese culture,

[0791] A means for training a generative model using the said dataset,

[0792] A means of selecting evaluators with diverse backgrounds and evaluating the generated images,

[0793] A means for detecting bias based on evaluation data and adjusting the model,

[0794] Means for providing the generated image to the user,

[0795] A system that includes this.

[0796] (Claim 2)

[0797] The system according to claim 1, characterized in that it performs an analysis to detect culturally inaccurate elements based on the aforementioned evaluation data.

[0798] (Claim 3)

[0799] The system according to claim 1, further comprising a feedback loop for readjusting the model, and characterized in that it continuously improves the performance of the model based on user feedback.

[0800] "Example 1"

[0801] (Claim 1)

[0802] A means for collecting information related to Japanese culture and for equalizing image quality and removing noise,

[0803] A method for training a generative AI model based on collected information,

[0804] A means of selecting evaluators with diverse backgrounds and evaluating the generated visual representations,

[0805] A means for detecting bias in the generated AI model based on the evaluation results and readjusting the weights,

[0806] A means of providing users with visual representations generated using a finely tuned generative AI model,

[0807] A system that includes this.

[0808] (Claim 2)

[0809] The system according to claim 1, characterized by analyzing culturally inaccurate elements based on the aforementioned evaluation results.

[0810] (Claim 3)

[0811] The system according to claim 1, further comprising a feedback loop for readjusting the generated AI model, and characterized in that it continuously improves the performance of the generated AI model based on user evaluations.

[0812] "Application Example 1"

[0813] (Claim 1)

[0814] Means of obtaining location information,

[0815] A means for collecting relevant data based on acquired location information,

[0816] A means for training a generative model using the aforementioned dataset,

[0817] Means for providing the generated image and background information to the user's device,

[0818] A means of selecting evaluators with diverse backgrounds and evaluating the generated images,

[0819] A means for detecting bias based on evaluation data and adjusting the model,

[0820] A system that includes this.

[0821] (Claim 2)

[0822] The system according to claim 1, characterized in that it performs an analysis to detect culturally inaccurate elements based on the aforementioned evaluation data.

[0823] (Claim 3)

[0824] The system according to claim 1, further comprising a feedback loop for readjusting the model, and characterized in that it continuously improves the performance of the model based on user feedback.

[0825] "Example 2 of combining an emotion engine"

[0826] (Claim 1)

[0827] A means of collecting and processing information related to Japanese culture,

[0828] A means for training an automatically generated model using the said information,

[0829] A means for acquiring emotional information using a device that analyzes the user's emotional state,

[0830] A means for optimizing a generative model based on acquired emotional information and adjusting the image,

[0831] A means of providing the user with an adjusted image,

[0832] A system that includes this.

[0833] (Claim 2)

[0834] The system according to claim 1, characterized in that it optimizes the product to enhance its emotional response based on the aforementioned emotional information.

[0835] (Claim 3)

[0836] The system according to claim 1, further comprising a feedback mechanism for adjusting the model, characterized in that it continuously improves the performance of the model based on the acquisition of emotional information.

[0837] "Application example 2 when combining with an emotional engine"

[0838] (Claim 1)

[0839] A means for collecting and pre-processing data related to Japanese culture,

[0840] A means for training a generative model using the said dataset,

[0841] A means of selecting evaluators with diverse backgrounds and evaluating the generated images,

[0842] A means for detecting bias based on evaluation information and adjusting the model,

[0843] A means of analyzing user emotions and feeding that information back into a generative model,

[0844] A means of adjusting images generated based on user emotions,

[0845] Means for providing the generated image to the user,

[0846] A system that includes this.

[0847] (Claim 2)

[0848] The system according to claim 1, characterized by detecting culturally inaccurate elements based on the aforementioned evaluation information and further optimizing the image using user sentiment analysis information.

[0849] (Claim 3)

[0850] The system according to claim 1, further comprising a feedback loop for readjusting the model, and characterized in that it continuously improves the performance of the model based on user feedback and real-time sentiment analysis. [Explanation of Symbols]

[0851] 10, 210, 310, 410 Data Processing Systems 12 Data Processing Devices 14 Smart Devices 214 Smart Glasses 314 Headset-type terminal 414 Robots< / url:> < / url:> < / url:> < / url:>

Claims

1. Means of obtaining location information, A means for collecting relevant data based on acquired location information, A means for training a generative model using the aforementioned dataset, Means for providing the generated image and background information to the user's device, A means of selecting evaluators with diverse backgrounds and evaluating the generated images, A means for detecting bias based on evaluation data and adjusting the model, A system that includes this.

2. The system according to claim 1, characterized in that it performs an analysis to detect culturally inaccurate elements based on the aforementioned evaluation data.

3. The system according to claim 1, further comprising a feedback loop for readjusting the model, and characterized in that it continuously improves the performance of the model based on user feedback.

Images