Information processing device

Abstract

Information processing device (2) is an information processing device that outputs explanatory data that provides a description of a 3D model. Information processing device (2) includes a 3D model transmission unit (231) that acquires 3D model data, which is data representing a 3D model, and outputs the 3D model data and a prompt for generating explanatory data to the first artificial intelligence unit (11), and an explanatory data acquisition unit (232) that receives and outputs explanatory data about the 3D model generated by the first artificial intelligence unit (11) based on the prompt. Information processing device (2) has the effect that it is not necessary to create a database of 3D models to which explanatory data providing a description of the 3D model is attached in advance, and explanatory data can be attached to 3D models without burdening the user.

Description

[Technical Field]

[0001] The present disclosure relates to an information processing apparatus Place . [Background Art]

[0002] Assigning appropriate metadata to a three-dimensional (3D) model is essential for enabling users to quickly search for and acquire the required 3D model. Metadata of a 3D model includes information such as the type of the 3D model, the use of the 3D model, and tags for the 3D model. Assigning such metadata to a 3D model facilitates classification and searching of 3D models.

[0003] In a conventional 3D model management system, a person who understands the content of a 3D model manually assigns metadata to the 3D model, which enables appropriate metadata assignment to the 3D model, but requires much time and effort. To address this problem, methods for automatically assigning metadata to 3D models have been proposed to reduce the burden on humans.

[0004] Patent Document 1 describes a system that automatically assigns metadata, which is used for searching 3D models such as types and tags, to 3D models. Specifically, a 3D model to which no metadata is assigned is input as a query, a plurality of 3D models similar to the input query 3D model are extracted from a pre-constructed database, and among the metadata assigned to the extracted 3D models, metadata assigned to at least half of the extracted 3D models is assigned to the 3D model input as the query. [Prior Art Literature] [Patent Literature]

[0005] [Patent Document 1] Japanese National Publication of International Patent Application No. 2019-530062 [Summary of Invention] [Problems that the invention aims to solve]

[0006] However, the system described in Patent Document 1 had the problem that it required the prior creation of a database of 3D models with appropriate metadata, and that metadata had to be added manually until a sufficient amount of data was created in the database. Furthermore, the system described in Patent Document 1 had the problem that it could not add appropriate metadata if the database did not contain the appropriate metadata for the 3D model entered as a query.

[0007] This disclosure is made in view of the above, and aims to provide an information processing device that can add explanatory data to 3D models without requiring the prior creation of a database of 3D models with explanatory data describing the 3D models, and without burdening the user. [Means for solving the problem]

[0008] To solve the above-mentioned problems and achieve the objective, the information processing device relating to this disclosure is an information processing device that outputs explanatory data showing a description of a three-dimensional model. The information processing device includes a three-dimensional model transmission unit that acquires three-dimensional model data, which is data showing a three-dimensional model, and outputs the three-dimensional model data and a prompt for generating explanatory data to a first artificial intelligence unit, and an explanatory data acquisition unit that receives and outputs explanatory data about the three-dimensional model generated by the first artificial intelligence unit based on the prompt. The 3D model transmission unit includes an image extraction unit that acquires 3D model data and extracts an image of the 3D model from the 3D model data, and an image conversion unit that outputs the image extracted by the image extraction unit to the first artificial intelligence unit. The explanation data acquisition unit includes an explanation data determination unit that acquires explanation data output from the first artificial intelligence unit, and a 3D model operation unit that edits the 3D model data. When the explanation data determination unit receives a determination result from the first artificial intelligence unit that determines that the 3D model is covered with a cover, it outputs an editing command to the 3D model operation unit instructing the deletion of the cover data from the 3D model data. The 3D model operation unit performs editing on the 3D model data to delete the cover data based on the editing command, and outputs the edited 3D model data to the image extraction unit. [Effects of the Invention]

[0009] According to this disclosure, it is not necessary to create a database of 3D models with explanatory data attached that provides descriptions of the 3D models in advance, and explanatory data can be attached to 3D models without burdening the user. [Brief explanation of the drawing]

[0010] [Figure 1] A diagram showing an example configuration of the information processing system according to Embodiment 1. [Figure 2] This figure shows an example of the configuration of the information processing device included in the information processing system according to Embodiment 1. [Figure 3] This figure shows an example of the configuration of the first server included in the information processing system according to Embodiment 1. [Figure 4] This figure shows an example of the configuration of a terminal device included in the information processing system according to Embodiment 1. [Figure 5] A diagram showing the main configurations related to the operation of the information processing system according to Embodiment 1. [Figure 6] A flowchart illustrating an example of the operation procedure of the 3D model transmission unit, explanatory data acquisition unit, and database included in the information processing device according to Embodiment 1. [Figure 7] A flowchart showing an example of the operation procedure of the first artificial intelligence unit provided in the first server according to Embodiment 1. [Figure 8] A flowchart showing an example of the operation procedure of the search unit included in the information processing device according to Embodiment 1. [Figure 9] This figure shows the main configurations related to the operation of an information processing system, which is a modified version of the information processing system according to Embodiment 1. [Figure 10] A flowchart showing an example of the operation procedure of the image extraction unit, image conversion unit, explanatory data determination unit, 3D model operation unit, and database of an information processing device, which is a modified version of the information processing device according to Embodiment 1. [Figure 11] A flowchart showing an example of the operation procedure of the first artificial intelligence unit included in the information processing system, which is a modified version of the information processing system according to Embodiment 1. [Figure 12] A flowchart showing an example of the operation procedure of the search unit of the information processing device included in the information processing system, which is a modified version of the information processing system according to Embodiment 1. [Figure 13] A flowchart showing an example of the operation procedure of the second artificial intelligence unit of the second server, which is a modified version of the information processing system according to Embodiment 1. [Figure 14] Diagram illustrating the configuration of an information processing system according to Embodiment 2 [Figure 15] Diagram illustrating the main configuration related to the operation of an information processing apparatus of the information processing system according to Embodiment 2 [Figure 16] Diagram illustrating an example configuration of a vector search unit provided in the information processing apparatus according to Embodiment 2 [Figure 17] Diagram illustrating an example configuration of a character string search unit provided in the information processing apparatus according to Embodiment 2 [Figure 18] Flowchart illustrating an example operation procedure of a model information search unit provided in a search unit of a processing unit of the information processing system according to Embodiment 2 [Figure 19] Diagram illustrating an example of a search top screen of a search screen displayed on a terminal device according to Embodiment 2 [Figure 20] Diagram illustrating an example of a search result display screen of a search screen displayed on a terminal device according to Embodiment 2 [Figure 21] Diagram illustrating an example of a detailed information screen of a search screen displayed on a terminal device according to Embodiment 2 [Figure 22] Flowchart illustrating an example operation procedure of a third artificial intelligence unit provided in the information processing system according to Embodiment 2 [Figure 23] Flowchart illustrating an example operation procedure of a fourth artificial intelligence unit provided in the information processing system according to Embodiment 2 [Figure 24] Flowchart illustrating an example operation procedure of the terminal device in Embodiment 2 [Figure 25] Diagram illustrating the main configuration related to the operation of an information processing apparatus of the information processing system according to Embodiment 3 [Figure 26] Diagram illustrating an example of a prompt creation assistance screen displayed on a display unit of a terminal device according to Embodiment 3 [Figure 27] Diagram explaining a pull-down menu in a device name setting input area on the prompt creation assistance screen according to Embodiment 3 [Figure 28] Diagram illustrating a state where prompt creation conditions have been input and set on the prompt creation assistance screen according to Embodiment 3 [Figure 29]This figure shows an example of a prompt created by the prompt creation support unit according to Embodiment 3 being displayed in the prompt input area of ​​the search screen. [Figure 30] A flowchart showing an example of the operation procedure of the prompt input unit of the information processing device in the information processing system according to Embodiment 3. [Figure 31] This diagram shows the configuration in which each function of the control unit according to Embodiments 1 to 3 is implemented in hardware. [Figure 32] This diagram shows the configuration in which each function of the control unit according to Embodiments 1 to 3 is implemented in software. [Modes for carrying out the invention]

[0011] The following describes the information processing system according to the embodiment. Place I will explain in detail based on the drawings.

[0012] Embodiment 1. Figure 1 is a diagram showing an example configuration of the information processing system 100 according to Embodiment 1. Figure 1 also shows a terminal device 3 and a network 200. The information processing system 100 is a system that associates and registers descriptive data, which is data used to describe 3D model data, with 3D model data. 3D model data is data that represents a 3D model. Descriptive data, which is data used to describe 3D model data, that is, descriptive data for 3D model data, is metadata for 3D model data. Descriptive data for describing 3D model data can be rephrased as descriptive data for describing a 3D model. Descriptive data is data used when a user searches for a 3D model. Descriptive data consists of descriptive text that describes the 3D model data.

[0013] A user of the information processing system 100 according to Embodiment 1 is, for example, an administrator who manages the production of products using production equipment, but the user is not limited to this.

[0014] As shown in Figure 1, the information processing system 100 comprises a first server 1, which is an artificial intelligence processing device having a first artificial intelligence unit 11, and an information processing device 2.

[0015] The first server 1 is composed of, for example, one or more cloud servers. A cloud server is a server built in a cloud environment that includes computer resources provided on a cloud service platform. The information processing device 2 and the first server 1 are connected to each other via a network 200 such as a VPN (Virtual Private Network). The information processing device 2 and the first server 1 send and receive information to each other through communication via the network 200.

[0016] The information processing device 2 sends 3D model data, which is data representing a 3D model, to the first artificial intelligence unit 11. The information processing device 2 obtains explanatory data from the first artificial intelligence unit 11, which is data for explaining the 3D model. In this way, the information processing device 2 exchanges information with the first artificial intelligence unit 11.

[0017] The information processing device 2 sends 3D model data, which is data indicating the 3D model that the user wants to register, to the first artificial intelligence unit 11 of the first server 1. The information processing device 2 then obtains explanatory data from the first artificial intelligence unit 11, which is data that explains the 3D model that it sent to the first artificial intelligence unit 11. In other words, the information processing device 2 obtains explanatory data from the first artificial intelligence unit 11, which is data that explains the 3D model data that the user wants to register. In this way, the information processing device 2 exchanges information with the first artificial intelligence unit 11.

[0018] 3D model descriptive data is data used to describe the 3D model, and is data used to describe the 3D model represented by the 3D model data.

[0019] Figure 2 shows an example of the configuration of the information processing device 2 included in the information processing system 100 according to Embodiment 1. The information processing device 2 comprises a communication unit 21, a storage unit 22, a processing unit 23, and a control unit 24. Information can be exchanged between each component of the information processing device 2.

[0020] The communication unit 21 is connected to the network 200 and communicates with external devices of the information processing device 2 via the network 200. The communication unit 21 also communicates with the first server 1 and the terminal device 3 via the network 200.

[0021] The memory unit 22 stores various types of information, such as information and explanatory data used to control the information processing device 2. The memory unit 22 includes a database (Data Base: DB) 221 that stores explanatory data acquired from the first artificial intelligence unit 11 of the first server 1.

[0022] DB221 stores and remembers descriptive data, which is data used to describe 3D model data. For example, DB221 stores descriptive data by importing descriptive text that describes 3D model data. DB221 also stores the 3D model data of the 3D model corresponding to the descriptive data, or the file path of the 3D model corresponding to the descriptive data, in association with the descriptive data. Alternatively, DB221 may store vectors generated from the descriptive data, in association with the 3D model data of the 3D model, or the file path of the 3D model corresponding to the descriptive data. If DB221 has established an association between the 3D model file path and the descriptive data, the 3D model data after the descriptive data generation may be stored outside the information processing device 2.

[0023] The processing unit 23 registers description data for 3D model data that the user wants to register, and searches for description data for any 3D model data that the user wants to search for.

[0024] The processing unit 23 includes a 3D model transmission unit 231, an explanatory data acquisition unit 232, and a search unit 233.

[0025] The 3D model transmission unit 231 acquires 3D model data, which is data indicating a 3D model that the user wants to register. The 3D model transmission unit 231 outputs the acquired 3D model data to the first artificial intelligence unit 11 of the first server 1. The 3D model data is input to a terminal device 3 by the user, for example, and transmitted from the terminal device 3 to the 3D model transmission unit 231. However, the method of inputting 3D model data to the 3D model transmission unit 231 is not limited to this. The 3D model data includes a 3D model file. In Embodiment 1, "user" refers to a user of the information processing device 2. An example of a user of the information processing device 2 is a designer of production equipment.

[0026] In Embodiment 1, the 3D model data is information represented by a file. The file stores the 3D model data in a file format such as STEP format (Standard for the Exchange of Product Data). The 3D model transmission unit 231 acquires the 3D model data represented by the file. For example, the 3D model transmission unit 231 acquires the 3D model data stored in the file by taking in a STEP format file.

[0027] The 3D model transmission unit 231 prompts the first artificial intelligence unit 11 to output text describing the function of the 3D model in addition to the 3D model data. For example, the 3D model transmission unit 231 inputs the prompt "Infer the function of the 3D model from its shape" to the first artificial intelligence unit 11. The first artificial intelligence unit 11 outputs descriptive data, including a description of the inferred function of the 3D model, to the information processing device 2.

[0028] When the above prompt is input to the first artificial intelligence unit 11, descriptive data including a description of the function of the predicted 3D model, such as text indicating "This robot arm can perform welding work to join metal parts," is output from the first artificial intelligence unit 11 to the information processing device 2. Prompts are predetermined and stored in the 3D model transmission unit 231. Prompts may also be stored in the storage unit 22.

[0029] The 3D model transmission unit 231 transmits the 3D model data along with design information related to the 3D model to the first artificial intelligence unit 11. This design information does not include structural information of the 3D model. Examples of design information related to the 3D model include bill of materials (BOM) information, project plans, and specifications. The 3D model corresponds to the structural information of the product. Information related to the design of the 3D model is considered design information related to the 3D model. This design information is input by the user into the terminal device 3 and transmitted from the terminal device 3 to the 3D model transmission unit 231.

[0030] The first artificial intelligence unit 11 can generate explanatory data based on the 3D model data and design information related to the 3D model, and information that is difficult to obtain from the 3D model data obtained from the design information related to the 3D model. Examples of information that is difficult to obtain from the 3D model data include the development history of the product corresponding to the 3D model, the purpose of the product corresponding to the 3D model, hierarchical relationships between 3D models such as the old model of the 3D model or the source from which the 3D model's design was reused, the supplier of the product corresponding to the 3D model, and the specific components of the product corresponding to the 3D model. Information on the development history of the product corresponding to the 3D model is included in the project plan. Information on hierarchical relationships between 3D models such as the old model of the 3D model or the source from which the 3D model's design was reused is included in the specifications. Information on the supplier of the product corresponding to the 3D model and the specific components of the product corresponding to the 3D model is included in the BOM information.

[0031] The information processing system 100 can search for 3D models based on a wider range of information by using the explanatory data generated as described above.

[0032] The 3D model transmission unit 231 outputs one or more prompts to the first artificial intelligence unit 11 along with the 3D model data, for extracting features of the 3D model from the 3D model data and outputting explanatory data. For example, the 3D model transmission unit 231 outputs prompts to the first artificial intelligence unit 11 such as "This is an AI assistant that is knowledgeable about 3D models in general," or "This is an AI assistant that is knowledgeable about factory equipment lines."

[0033] The prompts here specify how the first artificial intelligence unit 11 should behave. The prompt "This is an AI assistant that is knowledgeable about 3D models in general" is a prompt that requests the first artificial intelligence unit 11 to behave as an AI assistant that is knowledgeable about 3D models in general. The prompt "This is an AI assistant that is knowledgeable about factory equipment lines" is a prompt that requests the first artificial intelligence unit 11 to behave as an AI assistant that is knowledgeable about factory equipment lines.

[0034] The first artificial intelligence unit 11 can generate explanatory data focusing on specific features by acquiring one 3D model data and prompts for extracting features of the 3D model shown in the 3D model data. Furthermore, the first artificial intelligence unit 11 can generate explanatory data focusing on specific features for each prompt by acquiring one 3D model data and multiple prompts for extracting features of the 3D model shown in the 3D model data.

[0035] The first artificial intelligence unit 11 basically generates one explanatory data for each of the above prompts. However, since it is conceivable that there may be some overlapping content in the explanatory data, the first artificial intelligence unit 11 may combine multiple explanatory data into a single explanatory data and output it.

[0036] The information processing system 100 can search for 3D models based on a wider range of information by using the explanatory data generated as described above.

[0037] The 3D model transmission unit 231 outputs a prompt to the first artificial intelligence unit 11 along with the 3D model data, which includes an example of input / output. An example of a prompt that includes an example of input is a prompt with an image attached.

[0038] The first artificial intelligence unit 11 can acquire 3D model data and prompts to which images are attached, find common patterns between the images attached to the prompts in the input example and the 3D model data, and generate more appropriate explanatory data.

[0039] Examples of prompts that provide output examples include: "Please output a description of the 3D model. Output example 1: This production equipment is an automated pallet warehouse. Output example 2: This production equipment is a picking robot cell. Output example 3: This production equipment is a painting system."

[0040] The first artificial intelligence unit 11 can reduce variations in the output format of explanatory data by acquiring 3D model data and prompts describing examples of output.

[0041] The information processing system 100 improves the search accuracy of 3D model description data by using the description data generated as described above.

[0042] The information processing device 2 stores descriptive data, including a description of the function of the 3D model inferred by the first artificial intelligence unit 11, in its DB 221. This allows the information processing system 100 to search for 3D models not only by their shape but also by the functions they possess.

[0043] The explanation data acquisition unit 232 receives explanation data from the first artificial intelligence unit 11 that describes the 3D model data sent to the first artificial intelligence unit 11 by the 3D model transmission unit 231, and outputs it to DB 221. In other words, the explanation data acquisition unit 232 acquires explanation data from the first artificial intelligence unit 11, which is data to describe the 3D model that the user wants to register. As an example, the explanation data acquisition unit 232 acquires explanation data, which is text data, by taking in the explanation text that describes the 3D model data. The explanation data acquisition unit 232 also acquires 3D model data from the 3D model transmission unit 231.

[0044] Here, when the explanation data acquisition unit 232 outputs explanation data, it includes not only outputting explanation data to DB 221, but also outputting explanation data externally as search results via DB 221 and the search unit 233. For example, if DB 221 is located on an external device to the information processing device 2, the explanation data acquisition unit 232 may output explanation data to that external device. In other words, the explanation data acquisition unit 232 only needs to have the function of outputting explanation data to other components, and the output destination of the explanation data can be set arbitrarily.

[0045] The explanation data acquisition unit 232 associates the explanation data acquired from the first artificial intelligence unit 11 with the 3D model data transmitted by the 3D model transmission unit 231 to the first artificial intelligence unit 11, and outputs and stores it together with the 3D model data in DB 221. Specifically, the explanation data acquisition unit 232 associates the explanation data acquired from the first artificial intelligence unit 11 with at least one of the metadata of the 3D model data sent by the 3D model transmission unit 231 to the first artificial intelligence unit 11, and outputs and stores it together with the 3D model data in DB 221. An example of metadata for 3D model data is the file name of the 3D model data.

[0046] The search unit 233 searches for explanatory data and 3D models that have a relatively high similarity to the content of the query, based on the multiple explanatory data stored in DB 221 and the query input from terminal device 3, and outputs the search results to terminal device 3. Specifically, the search unit 233 compares all the explanatory data stored in DB 221 with the query input from terminal device 3 and determines the explanatory data that has a relatively high similarity to the content of the query from the multiple explanatory data stored in DB 221. Then, the search unit 233 searches DB 221 for and retrieves the 3D model that is determined to be the explanatory data with the highest similarity to the content of the query, associates the explanatory data with the 3D model data, and outputs the explanatory data and the 3D model data as search results to terminal device 3.

[0047] Descriptive data that has a relatively high degree of similarity to the query content is, for example, the descriptive data with the highest degree of similarity to the query content, or a predetermined number of descriptive data with a high degree of similarity. In other words, the search unit 233 compares the query input to the information processing device 2 with the descriptive data stored in DB 221 to search for a predetermined number of descriptive data and 3D models that have a relatively high degree of similarity to the query content, and outputs the search results to the terminal device 3.

[0048] As an example, the search unit 233 calculates the similarity between all the descriptive data stored in DB 221 and the query input from terminal device 3 using a known similarity calculation method, such as cosine similarity calculation using vectors, and determines the descriptive data with the highest similarity to the content of the query. The search unit 233 determines that the 3D model corresponding to the descriptive data with the highest similarity among the multiple descriptive data stored in DB 221 is the 3D model with the highest similarity to the content of the query. The query contains content that identifies the 3D model, such as "search for a transport device".

[0049] In Embodiment 1, the query is information that specifies the search target to be searched by the search unit 233, and is expressed in natural language. The search unit 233 acquires the query expressed in natural language. As an example, the search unit 233 acquires the query, which is text data, by taking in text input about the content of the search request. The manner in which the search unit 233 acquires the query is not limited to that described herein and is arbitrary. For example, the search unit 233 may acquire the query, which is audio data.

[0050] After the search, the search unit 233 associates the descriptive data with the highest similarity to the query content with the 3D model data of the 3D model corresponding to that descriptive data, and outputs the descriptive data and the 3D model data to the terminal device 3 as search results. Alternatively, after the search, the search unit 233 may associate the descriptive data with the highest similarity to the query content with the file path of the 3D model corresponding to that descriptive data, and output the descriptive data and the file path of the 3D model to the terminal device 3 as search results. The 3D model data or the file path of the 3D model is stored in DB 221 in association with the 3D model. The search unit 233 only needs to have the function to output search results to other components, and the output destination of the search results may be set arbitrarily.

[0051] The control unit 24 controls the entire information processing device 2.

[0052] The information processing device 2 stores information processing application software in its storage unit 22. This application software is used to execute the functions of the information processing system 100 by performing the functions of each component of the information processing device 2 described above. The information processing application software includes information processing programs that cause the computer to perform the functions of each component of the information processing device 2 described above. These information processing programs are installed in the storage unit 22.

[0053] Figure 3 shows an example of the configuration of the first server 1 provided in the information processing system 100 according to Embodiment 1. The first server 1 comprises a first artificial intelligence unit 11, a first server communication unit 12, a first server storage unit 13, and a first server control unit 14. Information can be exchanged between each of the components of the first server 1.

[0054] The first server 1 includes a first artificial intelligence unit 11, which is an artificial intelligence unit having artificial intelligence functions. In Embodiment 1, the artificial intelligence unit is an artificial intelligence having intelligent functions such as reasoning or judgment, and its operating environment.

[0055] The first artificial intelligence unit 11 comprises a first model control unit 111 and a first trained model storage unit 112. The first artificial intelligence unit 11 is a model and its operating environment configured to output explanatory data corresponding to 3D model data when 3D model data is input. When 3D model data is input from the 3D model transmission unit 231 of the information processing device 2, the first artificial intelligence unit 11 outputs explanatory data to the information processing device 2 based on the 3D model data and the trained model. Specifically, the first artificial intelligence unit 11 can output explanatory data to the information processing device 2 based on the shape of the 3D model shown in the 3D model data and the trained model.

[0056] If the first artificial intelligence unit 11 receives a prompt from the 3D model transmission unit 231 instructing it to output text describing the function of the 3D model, it can output descriptive data, including a description of the function of the predicted 3D model, to the information processing device 2.

[0057] The first trained model storage unit 112 stores the trained model. The trained model includes model information, which will be described later. The trained model may also include model parameters, which are information that defines the behavior of the model, such as constraints, weighting variables, and evaluation functions.

[0058] The model may be, for example, a model called NN (Neural Network), CNN (Convolutional Neural Network), RNN (Recurrent Neural Network), VAE (Variational Autoencoder), GAN (Generative Adversarial Networks), Diffusion Model, Transformer, LLM (Large Language Model), VLM (Visual Language Model), BERT (Bidirectional Encoder Representations from Transformers), GPT (Registered Trademark) (Generative Pre-trained Transformer), or CLIP (Contrastive Language Image Pre-training).

[0059] In the first artificial intelligence unit 11, a language learning model such as an LLM that takes natural language as input and produces output results can be applied as a model for obtaining explanatory data from 3D model data.

[0060] When the first model control unit 111 acquires 3D model data from the 3D model transmission unit 231 of the information processing device 2, it generates explanatory data corresponding to the 3D model data using the model indicated by the trained model and outputs it to the explanatory data acquisition unit 232 of the information processing device 2.

[0061] The first model control unit 111 outputs the necessary information to the search unit 233 of the information processing device 2. The search unit 233 then performs a search using the descriptive data corresponding to the 3D model data. For this reason, the first model control unit 111 associates the 3D model data with the descriptive data corresponding to the 3D model data and outputs it to the descriptive data acquisition unit 232.

[0062] Furthermore, the first model control unit 111 may output the file path of the 3D model data to the explanation data acquisition unit 232 instead of the 3D model data itself. That is, the first model control unit 111 may associate the file path of the 3D model data with the explanation data corresponding to the 3D model data and output this to the explanation data acquisition unit 232. In DB221, as long as the file path of the 3D model data and the explanation data corresponding to the 3D model data are associated, the 3D model data after the generation of the explanation data may be located outside the information processing system 100.

[0063] The trained model and other information used by the first artificial intelligence unit 11 may be pre-prepared, or they may be acquired via the network 200 as needed.

[0064] The first server communication unit 12 connects to the network 200 and communicates with external devices of the first server 1 via the network 200.

[0065] The first server storage unit 13 stores information used for controlling the first server 1.

[0066] The first server control unit 14 controls the entire first server 1.

[0067] Figure 4 shows an example of the configuration of a terminal device 3 included in the information processing system 100 according to Embodiment 1. As shown in Figure 4, the terminal device 3 includes an operation unit 31, a display unit 32, a terminal communication unit 33, a terminal storage unit 34, and a terminal control unit 35. Information can be exchanged between each component of the terminal device 3.

[0068] The operation unit 31 is a functional unit used when operating the information processing system 100. The user performs various operations using the operation unit 31 on the operation screen of the information processing system 100 displayed on the display unit 32. The operation unit 31 receives information input from the user and transmits operation information corresponding to that information to the terminal control unit 35.

[0069] The display unit 32 displays information about operations performed on the operation unit 31 and various information related to the information processing system 100. The display unit 32 is a display device such as an LCD (Liquid Crystal Display) or an organic EL (Electro Luminescence) display.

[0070] The terminal communication unit 33 communicates with the communication unit 21 of the information processing device 2 via the network 200.

[0071] The terminal storage unit 34 is a storage unit that stores information used in the information processing system 100, and stores various types of information such as information used to control the terminal device 3, information input at the operation unit 31, and information acquired from the information processing device 2.

[0072] The terminal control unit 35 is a control unit that controls the processing of the entire terminal device 3. The terminal control unit 35 transmits the operation information received from the operation unit 31 to the processing unit 23 of the information processing device 2. The terminal control unit 35 transmits operation information corresponding to the operation performed by the user on the operation screen of the information processing system 100 displayed on the display unit 32 of the terminal device 3 to the processing unit 23 of the information processing device 2, retrieves the information corresponding to the user's operation from the processing unit 23 of the information processing device 2, and displays it on the display unit 32.

[0073] Next, the operation of the 3D model transmission unit 231, the explanatory data acquisition unit 232, and the DB 221 of the information processing device 2 will be described. Figure 5 is a diagram showing the main configurations related to the operation of the information processing system 100 according to Embodiment 1. Figure 6 is a flowchart showing an example of the operation procedure of the 3D model transmission unit 231, the explanatory data acquisition unit 232, and the DB 221 provided in the information processing device 2 according to Embodiment 1.

[0074] In step S110, the 3D model transmission unit 231 of the processing unit 23 of the information processing device 2 acquires 3D model data. The 3D model data acquired by the 3D model transmission unit 231 is information represented by a file.

[0075] When the information processing system 100 is started, a screen requesting input of 3D model data is displayed on the display unit 32 of the terminal device 3. The user inputs the 3D model data of the 3D model they wish to register into the terminal device 3 from the input request screen. The terminal device 3 transmits the 3D model data to the 3D model transmission unit 231 of the information processing device 2. As a result, the 3D model transmission unit 231 acquires the 3D model data.

[0076] In step S120, the 3D model transmission unit 231 outputs the acquired 3D model data to the first artificial intelligence unit 11 of the first server 1.

[0077] In step S130, the explanation data acquisition unit 232 of the processing unit 23 of the information processing device 2 receives and acquires explanation data corresponding to the 3D model data transmitted in step S110 from the first artificial intelligence unit 11. The explanation data acquisition unit 232 transmits the acquired explanation data together with the 3D model data to the DB 221 of the storage unit 22 of the information processing device 2.

[0078] In step S140, DB221 receives and acquires the explanatory data and 3D model data obtained in step S130 from the explanatory data acquisition unit 232. DB221 stores and stores the acquired explanatory data and 3D model data. With this, the information processing device 2 completes its operation according to the procedure shown in Figure 6.

[0079] Next, the operation of the first artificial intelligence unit 11 of the first server 1 will be described. Figure 7 is a flowchart showing an example of the operation procedure of the first artificial intelligence unit 11 provided in the first server 1 according to Embodiment 1.

[0080] In step S210, the first artificial intelligence unit 11 acquires the 3D model data transmitted from the 3D model transmission unit 231 of the information processing device 2.

[0081] In step S220, the first artificial intelligence unit 11 generates explanatory data based on the acquired 3D model data. The first artificial intelligence unit 11 generates explanatory data corresponding to the 3D model data based on the 3D model data and the trained model.

[0082] In step S230, the first artificial intelligence unit 11 outputs the explanatory data generated in step S220 to the explanatory data acquisition unit 232 of the processing unit 23 of the information processing device 2. With this, the first artificial intelligence unit 11 of the first server 1 completes its operation according to the procedure shown in Figure 7.

[0083] Next, the operation of the search unit 233 of the information processing device 2 will be described. Figure 8 is a flowchart showing an example of the operation procedure of the search unit 233 provided in the information processing device 2 according to Embodiment 1.

[0084] In step S310, the search unit 233 obtains a query. The query obtained by the search unit 233 is information expressed in natural language. The query is transmitted from the terminal device 3 to the search unit 233 via the network 200.

[0085] In step S320, the search unit 233 retrieves all the description data stored in DB221.

[0086] In step S330, the search unit 233 searches for descriptive data with a high similarity to the content of the query from the descriptive data based on the acquired query.

[0087] In step S340, the search unit 233 outputs the search results, which it has retrieved from the explanatory data based on the acquired query, to the terminal device 3. The destination for outputting the search results may be set arbitrarily. With this, the search unit 233 completes its operation according to the procedure shown in Figure 8.

[0088] Next, an information processing system 100a, which is a modified version of the information processing system 100 according to Embodiment 1, will be described. Figure 9 is a diagram showing the main configurations related to the operation of the information processing system 100a, which is a modified version of the information processing system 100 according to Embodiment 1. In the following, the same reference numerals are used for components that are the same as those in the information processing system 100 described above, and the configurations that differ from the information processing system 100 will be mainly described. The modified information processing system 100a differs from the information processing system 100 described above in that it includes an information processing device 2a instead of an information processing device 2, and further includes a second server 4 equipped with a second artificial intelligence unit 41.

[0089] The modified information processing device 2a includes an image extraction unit 2311 and an image conversion unit 2312 in the 3D model transmission unit 231 of the processing unit 23.

[0090] Furthermore, the modified information processing device 2a includes an explanatory data acquisition unit 232 in the processing unit 23, which comprises an explanatory data determination unit 2321 and a 3D model operation unit 2322.

[0091] First, we will explain the image extraction unit 2311 and the image conversion unit 2312 of the 3D model transmission unit 231.

[0092] The image extraction unit 2311 acquires 3D model data acquired by the 3D model transmission unit 231. The image extraction unit 2311 performs image processing on the acquired 3D model data to extract images of the 3D model. Specifically, the image extraction unit 2311 photographs the 3D model from the acquired 3D model data and extracts images of the 3D model. The image extraction unit 2311 extracts one or more images of the 3D model. The image extraction unit 2311 transmits the extracted images to the image conversion unit 2312. The image extraction unit 2311 may also transmit the extracted images and the 3D model data to the image conversion unit 2312. Furthermore, if the image extraction unit 2311 receives an image extraction command (described later) and edited 3D model data (described later) from the 3D model operation unit 2322, it extracts images of the edited 3D model from the edited 3D model data based on the image extraction command. Furthermore, if the image extraction unit 2311 receives only an image extraction command from the 3D model operation unit 2322 and has not received the edited 3D model data from the 3D model operation unit 2322, it extracts an image of the 3D model from the 3D model data based on the image extraction command.

[0093] Since the image of a 3D model has a smaller file size than the 3D model data, the processing load on the information processing device 2a and the first artificial intelligence unit 11 is reduced.

[0094] The image extraction unit 2311 uses, for example, the capture function of a 3D CAD (Three Dimensional Computer Aided Design) program to capture images of the 3D model from the acquired 3D model data. The 3D CAD capture function allows for the acquisition of images of the 3D model by capturing the rendered 3D model data, and the 3D model can be captured from coordinates that allow for the entire 3D model to be captured. The image extraction unit 2311 uses all the functions of the 3D CAD to capture the 3D model from all possible directions. It is also possible for the user to specify the shooting direction of the 3D model to the image extraction unit 2311. In other words, the image extraction unit 2311 can use the 3D CAD capture function to capture the 3D model data from multiple angles and extract multiple images.

[0095] The image extraction unit 2311 uses the 3DCAD's shooting function to capture the 3D model. For example, by obtaining the camera coordinates of the shooting function and multiplying each element of the coordinates by a negative value, the 3D model can be photographed from various directions. In this case, the 3D model is assumed to be centered on the origin of the camera coordinates. It is also assumed that the entire 3D model fits within the field of view at the first shooting point (x,y,z) of the camera coordinates. Then, by changing the shooting point to one of the other seven points ((-x,y,z), (x,-y,z), (x,y,-z), (-x,-y,-z), (x,-y,-z), (-x,-y,-z), and shooting towards the origin, the image extraction unit 2311 can obtain multiple images of the 3D model that contain the entire 3D model.

[0096] The image extraction unit 2311 extracts more features from the 3D model into the image, thereby preventing the loss of features from the 3D model when generating explanatory data.

[0097] The image conversion unit 2312 transmits the image extracted by the image extraction unit 2311 to the first artificial intelligence unit 11. At the same time, the 3D model transmission unit 231 transmits the 3D model data to the explanatory data acquisition unit 232. The multiple images extracted by the image extraction unit 2311 may, for example, be transmitted to the first artificial intelligence unit 11 as individual images, or they may be combined into a single image and transmitted to the first artificial intelligence unit 11.

[0098] The image conversion unit 2312 prompts the first artificial intelligence unit 11 to output text representing the function of the 3D model, in addition to the image, similar to the 3D model transmission unit 231 of the information processing device 2 described above.

[0099] The image conversion unit 2312 transmits the image along with design information related to the 3D model to the first artificial intelligence unit 11, similar to the 3D model transmission unit 231 of the information processing device 2 described above.

[0100] The image conversion unit 2312 outputs one or more prompts to the first artificial intelligence unit 11 along with the image, similar to the 3D model transmission unit 231 of the information processing device 2 described above, for extracting features of the 3D model from the image and outputting explanatory data.

[0101] The image conversion unit 2312 outputs a prompt to the first artificial intelligence unit 11, along with the image, which includes an example of input / output, similar to the 3D model transmission unit 231 of the information processing device 2 described above.

[0102] Next, we will explain the explanation data determination unit 2321 and the 3D model manipulation unit 2322 of the explanation data acquisition unit 232.

[0103] The explanation data determination unit 2321 acquires the explanation data output from the first artificial intelligence unit 11 from the explanation data acquisition unit 232 and acquires the 3D model data from the 3D model transmission unit 231. The explanation data determination unit 2321 transmits the acquired 3D model data and explanation data to the first artificial intelligence unit 11.

[0104] In the information processing system 100a, the first artificial intelligence unit 11 only has the function of outputting data based on input to the first artificial intelligence unit 11, and does not have the function of inputting the data output by the first artificial intelligence unit 11 back into the first artificial intelligence unit 11 itself based on the data output by the first artificial intelligence unit 11. For this reason, the first artificial intelligence unit 11 first transmits the explanatory data to the explanatory data determination unit 2321, and then the explanatory data determination unit 2321 processes the explanatory data and inputs it back into the first artificial intelligence unit 11.

[0105] Furthermore, if the first artificial intelligence unit 11 has a function to input the data output by the first artificial intelligence unit 11 back into the first artificial intelligence unit 11 based on the data output by the first artificial intelligence unit 11, then the process of the first artificial intelligence unit 11 sending the explanatory data to the explanatory data determination unit 2321 becomes unnecessary.

[0106] In the information processing system 100a, when an image is input from the image conversion unit 2312, the first artificial intelligence unit 11 generates explanatory data corresponding to the image based on the image and transmits the generated explanatory data to the explanatory data acquisition unit 232.

[0107] The first artificial intelligence unit 11, upon acquiring the image from the image conversion unit 2312 and the 3D model data and explanatory data from the explanatory data determination unit 2321, generates determination data corresponding to the 3D model data and explanatory data, and transmits the determination data to the explanatory data determination unit 2321. The explanatory data determination unit 2321 receives and acquires the determination data transmitted from the first artificial intelligence unit 11. Here, the case where the first artificial intelligence unit 11 acquires the 3D model data from the explanatory data determination unit 2321 is shown, but the first artificial intelligence unit 11 may also acquire the 3D model data from the 3D model transmission unit 231. That is, the first artificial intelligence unit 11 may generate determination data based on the image acquired from the image conversion unit 2312, the 3D model data acquired from the 3D model transmission unit 231, and the explanatory data acquired from the explanatory data determination unit 2321.

[0108] The judgment data basically includes the judgment result, 3D model editing commands, and image extraction commands. If the judgment result included in the judgment data is a judgment result that has not been converted to text as described later, the explanation data judgment unit 2321 sends the 3D model editing command and the image extraction command to the 3D model operation unit 2322. If the judgment result included in the judgment data is a judgment result that has been converted to text, the explanation data judgment unit 2321 associates the 3D model data and explanation data and sends them to DB221, and stores the 3D model data and explanation data in DB221.

[0109] The judgment data includes two types of judgment results: a first judgment result and a second judgment result.

[0110] In other words, the first artificial intelligence unit 11 performs two types of determinations, a first determination and a second determination, based on the image acquired from the image conversion unit 2312 and the 3D model data and explanation data acquired from the explanation data determination unit 2321. However, the determinations performed by the first artificial intelligence unit 11 based on the image, 3D model data and explanation data are not limited to these.

[0111] The first artificial intelligence unit 11 performs a first determination based on the image obtained from the image conversion unit 2312 and the 3D model data and descriptive data obtained from the descriptive data determination unit 2321, to determine whether the name of the device shown by the 3D model data is expressed in text. In other words, the first determination determines whether the text conversion, that is, whether the text conversion of the name of the device shown by the 3D model data has been completed.

[0112] "The name of the device shown by the 3D model data" refers to the name of the device that indicates the function of the device, and, using production equipment as an example, what kind of production equipment "the device shown by the 3D model data" is.

[0113] Furthermore, the first artificial intelligence unit 11 performs a second determination based on the image acquired from the image conversion unit 2312 and the 3D model data and descriptive data acquired from the descriptive data determination unit 2321, determining whether the 3D model shown by the 3D model data is covered by a cover. Taking production equipment as an example, the second determination is whether the 3D model shown by the 3D model data is covered by a cover such as a protective part, a transparent dustproof and soundproof cover, or an intrusion prevention cover.

[0114] A cover is the outer shell or enclosure in a 3D model represented by 3D model data. In other words, a cover is the outer shell or enclosure in a device represented by a 3D model represented by 3D model data. A 3D model represented by 3D model data may include components that are not directly related to the device itself represented by the 3D model. In this case, the cover may include, for example, fences and screens, or it may include surrounding buildings such as roofs, beams, and walls that are not directly related to the device itself represented by the 3D model represented by 3D model data.

[0115] The first artificial intelligence unit 11 then includes the first judgment result, which is the result of the first judgment, and the second judgment result, which is the result of the second judgment, in the judgment data as the judgment result and transmits it to the explanation data judgment unit 2321.

[0116] The descriptive data determination unit 2321 determines that the text conversion is complete if it determines that the features of the 3D model shown by the 3D model data, including the parts covered by the cover, are expressed in text. The descriptive data determination unit 2321 determines that the text conversion is incomplete if it determines that the features of the 3D model shown by the 3D model data, including the parts covered by the cover, are not expressed in text.

[0117] The 3D model editing command is an editing command that instructs the deletion of the cover from the 3D model, that is, an editing command that instructs the deletion of the cover data from the 3D model data. The first artificial intelligence unit 11 needs a 3D model to generate the 3D model editing command. For this reason, the explanation data determination unit 2321 sends the 3D model data to the first artificial intelligence unit 11.

[0118] The image extraction command is a command that instructs the system to extract images again from the 3D model data. The image extraction command includes instruction information such as the shooting angle and magnification when extracting images again from the 3D model data.

[0119] When the explanation data determination unit 2321 transmits the 3D model data and explanation data to the first artificial intelligence unit 11, it also transmits a prompt to the first artificial intelligence unit 11 along with the data.

[0120] The description data determination unit 2321 sends a prompt to the first artificial intelligence unit 11, along with the 3D model data and the description data, instructing it to perform a first determination. Specifically, the description data determination unit 2321 sends a prompt to the first artificial intelligence unit 11 instructing it to determine whether or not the name of the device shown by the 3D model shown by the 3D model data is expressed in text. The prompt may contain, for example, "Determine whether or not the name of the device shown by the 3D model shown by the 3D model data can be expressed in text. If the name of the device shown by the 3D model shown by the 3D model data cannot be expressed in text, output the 3D model editing command and the image extraction command."

[0121] The first artificial intelligence unit 11 can extract the names of the devices shown by the 3D models shown by the 3D model data without omission by acquiring the above prompts, 3D model data, and explanatory data, thereby reducing the variability of the generated explanatory data. One of the factors causing variability in the explanatory data generated by the first artificial intelligence unit 11 is noise contained in the 3D model. When the 3D model contains noise, the explanatory data may not include necessary explanations or may include explanations that are not actually necessary. One example of noise is cover.

[0122] The first artificial intelligence unit 11 acquires the above prompt, 3D model data, and explanatory data, thereby reducing the influence of noise on the generated explanatory data. As a result, the search unit 233 can search for 3D models corresponding to explanatory data with a high similarity to the query content, improving the search accuracy of the explanatory data for 3D models.

[0123] The description data determination unit 2321 sends a prompt to the first artificial intelligence unit 11, along with the 3D model data and the description data, instructing it to perform a second determination. Specifically, the description data determination unit 2321 sends a prompt to the first artificial intelligence unit 11 instructing it to determine whether or not the 3D model shown by the 3D model data is covered by a cover. The prompt may contain, for example, "Determine whether or not the 3D model shown by the 3D model data is covered by a cover. If the 3D model shown by the 3D model data is covered by a cover, output an editing command instructing the removal of the cover from the 3D model."

[0124] Taking production equipment as an example, the description data determination unit 2321 sends a prompt to the first artificial intelligence unit 11 instructing it to determine whether the 3D model indicated by the 3D model data is covered by a cover such as a protective part, a transparent dustproof and soundproof cover, or an intrusion prevention cover. The prompt may contain a message such as, "Determine whether the 3D model indicated by the 3D model data is covered by a cover such as a protective part, a transparent dustproof and soundproof cover, or an intrusion prevention cover. If the 3D model indicated by the 3D model data is covered by a cover, output an editing command to remove the cover from the 3D model."

[0125] The first artificial intelligence unit 11 can extract the internal structural features of the 3D model covered by the cover by acquiring the above prompt, 3D model data, and descriptive data, and can generate descriptive data about the internal structural features of the 3D model, thereby reducing the variability of the generated descriptive data. As a result, the search unit 233 can use the descriptive data about the internal structural features of the 3D model to search for 3D models that correspond to descriptive data with a high similarity to the query content, thereby improving the search accuracy of the 3D model descriptive data.

[0126] The 3D model operation unit 2322 obtains a 3D model editing command and an image extraction command from the explanatory data determination unit 2321. The 3D model operation unit 2322 obtains 3D model data from the 3D model transmission unit 231. Alternatively, the 3D model operation unit 2322 may obtain the 3D model data from the explanatory data determination unit 2321. The 3D model operation unit 2322 uses the functions of 3DCAD to execute a 3D model editing command on the 3D model data and edits the 3D model data. After editing the 3D model data, the 3D model operation unit 2322 sends the edited 3D model data and the image extraction command to the image extraction unit 2311.

[0127] The 3D model operation unit 2322 needs to obtain 3D model data from the 3D model transmission unit 231 or the explanation data determination unit 2321 in order to edit the 3D model data. In other words, the 3D model operation unit 2322 needs to obtain 3D model data for editing. When the first artificial intelligence unit 11 obtains 3D model data in step S450, which will be described later, the 3D model data held by the explanation data determination unit 2321 and the 3D model data held by the 3D model transmission unit 231 are the same. For this reason, the 3D model operation unit 2322 may obtain the 3D model data from the 3D model transmission unit 231 or from the explanation data determination unit 2321.

[0128] After the search, the search unit 233 associates the descriptive data with the highest similarity to the query content with the 3D model data of the 3D model corresponding to that descriptive data, and outputs the descriptive data and the 3D model data as search results to the second artificial intelligence unit 41. Alternatively, after the search, the search unit 233 may associate the descriptive data with the highest similarity to the query content with the file path of the 3D model corresponding to that descriptive data and output this as a search result. The 3D model data or the file path of the 3D model is stored in DB 221 in association with the 3D model.

[0129] The search unit 233 sends a query, search results, and a prompt containing a request for information to narrow down the 3D model to the second artificial intelligence unit 41 of the second server 4. The prompt containing the request might read, for example, "The user entered a query and searched in the search system where the 3D model was registered. Please provide information that the user should enter into the search system to further narrow down the 3D model. #Query entered by the user: Find production equipment using a robotic arm."

[0130] The search unit 233 may, as a search result, transmit to the second artificial intelligence unit 41 the result of determining the similarity between the query content and the descriptive data stored in DB 221, associated with the file path of the 3D model corresponding to the descriptive data. The data that associates the descriptive data with the highest similarity to the query content with the 3D model data of the 3D model corresponding to the descriptive data, and the descriptive data with the highest similarity to the query content, can be said to be the result of determining the similarity between the query content and the descriptive data stored in DB 221 in a broad sense.

[0131] When the second artificial intelligence unit 41 receives a query, search results, and a prompt containing a request statement from the search unit 233, it outputs information necessary for narrowing down the 3D model to the search unit 233 as suggestion text corresponding to the query and search results. The information necessary for narrowing down the model may include, for example, "To further narrow down the 3D model, the user may want to input the following information: application and function: for example, specific work content such as welding, assembly, or transport."

[0132] The search unit 233 transmits the information necessary for narrowing down the 3D models to the terminal device 3 and presents it to the user. By presenting the user with information useful for narrowing down the 3D models, the search unit 233 makes it easier for users with little knowledge or experience in narrowing down 3D models to enter the next query to narrow down the 3D models.

[0133] The modified information processing system 100a includes a second server 4 equipped with a second artificial intelligence unit 41. The second server 4 has the same configuration as the first server 1. The second server 4 includes a second artificial intelligence unit 41, a second server communication unit (not shown), a second server storage unit (not shown), and a second server control unit (not shown). Information can be exchanged between the components of the second server 4.

[0134] The second server 4 includes a second artificial intelligence unit 41, which is an artificial intelligence unit having artificial intelligence functions. The second artificial intelligence unit 41 includes a second model control unit 411 and a second trained model storage unit 412. The second artificial intelligence unit 41 is a model and its operating environment configured to output information necessary for narrowing down the 3D model as suggested text corresponding to the query and search results when a query, search results, and a prompt containing a request statement requesting information to narrow down the 3D model are input to it. When the second artificial intelligence unit 41 receives a query, search results, and a prompt containing a request statement requesting information to narrow down the 3D model from the search unit 233 of the information processing device 2a, it outputs information necessary for narrowing down the 3D model as suggested text corresponding to the query and search results to the search unit 233, based on the query, search results, prompt and trained model.

[0135] The second trained model storage unit 412 stores the trained models. The form of the trained models is the same as that of the trained models stored in the first trained model storage unit 112.

[0136] Furthermore, since it is not necessary to use the same artificial intelligence unit for acquiring explanatory data and the same artificial intelligence unit for acquiring information necessary for narrowing down 3D models, the information processing system 100a uses the second artificial intelligence unit 41 to acquire information necessary for narrowing down 3D models. The first artificial intelligence unit 11 may also be configured to perform the functions of the second artificial intelligence unit 41.

[0137] Next, the operation of the image extraction unit 2311, the image conversion unit 2312, the explanatory data determination unit 2321, the 3D model operation unit 2322, and the DB 221 will be described. Figure 10 is a flowchart showing an example of the operation procedure of the image extraction unit 2311, the image conversion unit 2312, the explanatory data determination unit 2321, the 3D model operation unit 2322, and the DB 221, which are included in the information processing device 2a, a modified example of the information processing device 2 according to Embodiment 1.

[0138] In step S410, the image extraction unit 2311 acquires 3D model data. The image extraction unit 2311 acquires the 3D model data acquired by the 3D model transmission unit 231. The 3D model data acquired by the image extraction unit 2311 is information indicated by a file.

[0139] When the information processing system 100a is started, a screen requesting input of 3D model data is displayed on the display unit 32 of the terminal device 3. The user inputs the 3D model data of the 3D model they wish to register into the terminal device 3 from the input request screen. The terminal device 3 transmits the 3D model data to the 3D model transmission unit 231 of the information processing device 2a. As a result, the 3D model transmission unit 231 acquires the 3D model data.

[0140] In step S420, the image extraction unit 2311 extracts an image from the acquired 3D model data. The image extraction unit 2311 then transmits the extracted image to the image conversion unit 2312.

[0141] In step S430, the image conversion unit 2312 outputs the acquired image to the first artificial intelligence unit 11. The image conversion unit 2312 also sends one or more prompts to the first artificial intelligence unit 11 for extracting features from the image. In addition, the image conversion unit 2312 outputs a prompt to the first artificial intelligence unit 11 along with the image, which includes an example of input and output.

[0142] In step S440, the explanation data determination unit 2321 obtains explanation data corresponding to the image output in step S430 from the first artificial intelligence unit 11.

[0143] In step S450, the explanation data determination unit 2321 outputs the 3D model data and explanation data to the first artificial intelligence unit 11. The explanation data determination unit 2321 also outputs a prompt to the first artificial intelligence unit 11 instructing it to perform a first determination and a prompt to perform a second determination.

[0144] In step S460, the explanation data determination unit 2321 obtains determination data from the first artificial intelligence unit 11 that corresponds to the 3D model data and explanation data output in step S450. The determination data basically includes the determination result, 3D model editing commands, and image extraction commands.

[0145] In step S470, the descriptive data determination unit 2321 determines whether or not the text conversion is complete. Specifically, the descriptive data determination unit 2321 determines whether the first determination result included in the determination result is a determination result indicating that "the text conversion of the name of the device indicated by the 3D model indicated by the 3D model data is complete." The descriptive data determination unit 2321 also determines whether the second determination result included in the determination result is a determination result indicating that "the 3D model indicated by the 3D model data is covered by a cover." If the descriptive data determination unit 2321 determines that the features of the 3D model indicated by the 3D model data, including the part covered by the cover, are expressed in text, it determines that the text conversion is complete. If the descriptive data determination unit 2321 determines that the features of the 3D model indicated by the 3D model data, including the part covered by the cover, are not expressed in text, it determines that the text conversion is incomplete.

[0146] If it is determined that the text conversion is complete, that is, if the first determination result included in the determination result is a determination result stating that "the text conversion of the name of the device indicated by the 3D model data indicated by the 3D model data is complete" (Yes in step S470), the explanation data determination unit 2321 associates the explanation data obtained from the first artificial intelligence unit 11 with the 3D model data obtained by the 3D model transmission unit 231 and transmits it to DB221 together with the 3D model data. Then, the process proceeds to step S480.

[0147] If it is determined that the text conversion is not complete, that is, if the first determination result included in the determination result is a determination result stating that "the text conversion of the name of the device indicated by the 3D model data indicated by the 3D model data is not complete" (No in step S470), the process proceeds to step S490.

[0148] In step S480, when DB221 receives the associated explanatory data and 3D model data transmitted in step S450, it stores the explanatory data and 3D model data. With this, the modified information processing system 100a completes its operation according to the procedure shown in Figure 10.

[0149] In step S490, if the first determination result included in the determination result is that "the text conversion of the name of the device indicated by the 3D model indicated by the 3D model data has not been completed" (No in step S470), the description data determination unit 2321 sends the 3D model data and the 3D model editing command and image extraction command included in the determination data to the 3D model operation unit 2322. Note that the determination data includes at least the image extraction command from among the 3D model editing command and image extraction command. In this case, the description data determination unit 2321 sends the command included in the determination data from among the 3D model editing command and image extraction command to the 3D model operation unit 2322.

[0150] In step S500, the 3D model operation unit 2322 determines whether or not the command sent from the explanation data determination unit 2321 contains a 3D model editing command.

[0151] If the command sent from the description data determination unit 2321 includes a 3D model editing command (Yes in step S500), in step S510, the 3D model operation unit 2322 edits the 3D model data using the 3D model editing command and proceeds to step S520.

[0152] If the command sent from the explanation data determination unit 2321 does not contain a 3D model editing command (No in step S500), the 3D model operation unit 2322 proceeds to step S520.

[0153] In step S520, the 3D model operation unit 2322 sends an image extraction command to the image extraction unit 2311. If the 3D model data has been edited, the 3D model operation unit 2322 also sends the edited 3D model data to the image extraction unit 2311. The process then returns to step S420.

[0154] If the image extraction unit 2311 receives both an image extraction command and the edited 3D model data from the 3D model operation unit 2322, it extracts an image of the edited 3D model from the edited 3D model data based on the image extraction command. If the image extraction unit 2311 receives only an image extraction command from the 3D model operation unit 2322 and not the edited 3D model data from the 3D model operation unit 2322, it extracts an image of the 3D model from the 3D model data based on the image extraction command. The procedure shown in Figure 10 continues thereafter.

[0155] Next, the operation of the first artificial intelligence unit 11 of the first server 1 will be described. Figure 11 is a flowchart showing an example of the operation procedure of the first artificial intelligence unit 11 provided in the information processing system 100a, which is a modified example of the information processing system 100 according to Embodiment 1.

[0156] In step S610, the first artificial intelligence unit 11 acquires the image transmitted from the image conversion unit 2312 of the 3D model transmission unit 231 of the information processing device 2a.

[0157] In step S620, the first artificial intelligence unit 11 generates explanatory data based on the acquired image. The first artificial intelligence unit 11 generates explanatory data corresponding to the image based on the image and the trained model.

[0158] In step S630, the first artificial intelligence unit 11 outputs the explanatory data generated in step S620 to the explanatory data determination unit 2321 of the explanatory data acquisition unit 232 of the processing unit 23 of the information processing device 2a.

[0159] In step S640, the first artificial intelligence unit 11 acquires 3D model data and explanatory data from the explanatory data determination unit 2321. As a result, the first artificial intelligence unit 11 acquires an image from the image conversion unit 2312 and acquires 3D model data and explanatory data from the explanatory data determination unit 2321.

[0160] In step S650, the first artificial intelligence unit 11 generates determination data corresponding to the 3D model data and the explanatory data, based on the image, the 3D model data, and the explanatory data.

[0161] In step S660, the first artificial intelligence unit 11 transmits the judgment data to the explanatory data judgment unit 2321. With this, the first artificial intelligence unit 11 of the first server 1 completes its operation according to the procedure shown in Figure 11.

[0162] Next, the operation of the search unit 233 will be described. Figure 12 is a flowchart showing an example of the operation procedure of the search unit 233 of the information processing device 2a of the information processing system 100a, which is a modified example of the information processing system 100 according to Embodiment 1.

[0163] In step S710, the search unit 233 obtains a query. The query obtained by the search unit 233 is information expressed in natural language. The query is transmitted from the terminal device 3 to the search unit 233 via the network 200.

[0164] In step S720, the search unit 233 retrieves all the description data stored in DB221.

[0165] In step S730, the search unit 233 searches for descriptive data with a high similarity to the content of the query from the descriptive data based on the acquired query.

[0166] In step S740, the search unit 233 outputs the search results and the query to the second artificial intelligence unit 41.

[0167] In step S750, the search unit 233 obtains suggested text corresponding to the search results and query output in step S740 from the second artificial intelligence unit 41.

[0168] In step S760, the search unit 233 outputs the search results and suggested text to the terminal device 3. That is, the search unit 233 outputs the suggested text along with the search results to the terminal device 3. The output destination for the search results may be set arbitrarily. With this, the search unit 233 of the information processing device 2a, which is a modified version of the information processing device 2, completes its operation according to the procedure shown in Figure 12.

[0169] Next, the operation of the second artificial intelligence unit 41 of the second server 4 will be described. Figure 13 is a flowchart showing an example of the operation procedure of the second artificial intelligence unit 41 of the second server 4, which is included in the information processing system 100a, a modified example of the information processing system 100 according to Embodiment 1.

[0170] In step S810, the second artificial intelligence unit 41 obtains the query, search results, and prompt containing the request statement transmitted from the search unit 233 of the information processing device 2a.

[0171] In step S820, the second artificial intelligence unit 41 generates suggested text based on the acquired query, search results, and prompt. The second artificial intelligence unit 41 generates suggested text corresponding to the query and search results based on the query, search results, prompt, and trained model.

[0172] In step S830, the second artificial intelligence unit 41 outputs the proposed text generated in step S820 to the search unit 233 of the information processing device 2a. With this, the second artificial intelligence unit 41 of the second server 4 completes its operation according to the procedure shown in Figure 13.

[0173] According to Embodiment 1 described above, an information processing device is provided that outputs explanatory data showing a description of a three-dimensional model, comprising: a three-dimensional model transmission unit that acquires three-dimensional model data which is data showing a three-dimensional model, and outputs the three-dimensional model data and a prompt for generating explanatory data to a first artificial intelligence unit; and an explanatory data acquisition unit that receives and outputs explanatory data about the three-dimensional model generated by the first artificial intelligence unit based on the prompt.

[0174] According to Embodiment 1 described above, an information processing system is provided which comprises an information processing device and an artificial intelligence processing device having an artificial intelligence unit and capable of communicating with the information processing device, wherein the information processing device includes a 3D model transmission unit that acquires 3D model data, which is data representing a 3D model, and outputs the 3D model data to the artificial intelligence unit, and an explanatory data acquisition unit that receives and outputs explanatory data about the 3D model output from the artificial intelligence unit.

[0175] According to Embodiment 1 described above, an information processing method is provided for an information processing device that outputs explanatory data describing a three-dimensional model, the method comprising the steps of: acquiring three-dimensional model data which is data representing a three-dimensional model, outputting the three-dimensional model data and a prompt for generating explanatory data to the artificial intelligence unit; and receiving and outputting explanatory data about the three-dimensional model generated by the artificial intelligence unit based on the prompt.

[0176] In the information processing system 100 according to the above-described embodiment 1, the 3D model transmission unit 231 acquires 3D model data, which is data indicating a 3D model that the user wants to register, and outputs the acquired 3D model data to the first artificial intelligence unit 11 of the first server 1. The explanation data acquisition unit 232 receives explanation data from the first artificial intelligence unit 11, which is explanation data that explains the 3D model data that the 3D model transmission unit 231 sent to the first artificial intelligence unit 11. The explanation data acquisition unit 232 then associates the explanation data acquired from the first artificial intelligence unit 11 with the 3D model data that the 3D model transmission unit 231 sent to the first artificial intelligence unit 11 and stores it in DB 221.

[0177] Since the information processing system 100 is configured as described above, it can automatically add explanatory data to 3D models without having to create a database of 3D models with appropriate explanatory data in advance.

[0178] Therefore, the information processing system 100 according to Embodiment 1 described above has the effect of allowing users to add explanatory data to 3D models without burdening them, as it does not require the user to create a database of 3D models with explanatory data describing the 3D models in advance.

[0179] Embodiment 2. Embodiment 2 describes an information processing system 100b, which is a modified version of the information processing system 100a described in Embodiment 1. The information processing system 100b according to Embodiment 2 can retrieve 3D models corresponding to prompts entered by the user using a graphical user interface (GUI) tool from the DB221 described above, and present the search results to the user using the GUI tool.

[0180] In Embodiment 2, the prompts input by the user using the GUI tool are prompts requesting a search for a 3D model, and can be described as search request prompts that request the Generating Artificial Intelligence Unit (Generating AI) to search for a 3D model from DB221.

[0181] Figure 14 is a diagram showing the main configuration related to the operation of the information processing system 100b according to Embodiment 2. Figure 15 is a diagram showing the main configuration related to the operation of the information processing device 2b of the information processing system 100b according to Embodiment 2. In Figure 15, the terminal device 3 is also shown. In Figure 14, the same reference numerals are used for components that are the same as those in the information processing system 100 according to Embodiment 1 described above, and detailed explanations are omitted. Below, the configuration of the information processing device 2b that differs from that of the information processing system 100 will be mainly described. The information processing system 100b according to Embodiment 2 differs from the information processing system 100 described above in that it includes an information processing device 2b instead of an information processing device 2.

[0182] The information processing device 2b according to Embodiment 2 has the configuration and functions of the information processing device 2 described above. Furthermore, the information processing device 2b's search unit 233 of the processing unit 23 further comprises a model information search unit 241.

[0183] The model information retrieval unit 241 searches for a 3D model corresponding to a given descriptive data, based on a set of descriptive data stored in DB 221 and prompts input from outside the information processing device 2b, using a generation artificial intelligence unit (generation AI) based on natural language.

[0184] The model information retrieval unit 241 comprises a prompt input unit 242, a search processing unit 243, and a search result output unit 244. A prompt is input from the terminal device 3 to the prompt input unit 242, the search processing unit 243 searches for a 3D model from DB 221 based on the prompt, and the search result output unit 244 outputs the search results from the search processing unit 243 to the terminal device 3.

[0185] The prompt input unit 242 receives prompts from outside the information processing device 2b. Specifically, the prompt input unit 242 acquires prompts that are input from the user to the terminal device 3 and output from the terminal device 3 via the communication unit 21 of the information processing device 2b. The prompt input unit 242 then transmits the acquired prompts to the search processing unit 243.

[0186] Furthermore, when a user searches for a 3D model on the information processing device 2b, the prompt input unit 242 creates search screen data to display the top search screen of the 3D model search screen on the terminal device 3, and outputs this search screen data to the terminal device 3 via the communication unit 21 of the information processing device 2b. This top search screen is a GUI screen that accepts prompt setting inputs from the user. Therefore, the prompt input unit 242 can be described as a screen data creation unit that creates and outputs screen data to display a GUI screen that accepts setting inputs for search request prompts.

[0187] The search processing unit 243 searches for 3D models corresponding to the content of a prompt using a vector search with a generative AI, based on multiple descriptive data stored in DB 221 and a prompt input from the prompt input unit 242. Specifically, the search processing unit 243 compares the search request prompt input to the information processing device 2b, which requests the search for a 3D model, with the descriptive data stored in DB 221, which is the storage unit, using a generative AI for 3D model searching in natural language. It then searches for a predetermined number of descriptive data and 3D models that have a relatively high similarity to the content of the search request prompt and outputs the search results.

[0188] Specifically, the search processing unit 243 performs a vector search to find explanatory data with a relatively high vector similarity to the content of a prompt, based on multiple explanatory data stored in DB 221 and the prompt input from the prompt input unit 242. Then, the search processing unit 243 extracts 3D models corresponding to the explanatory data with a relatively high vector similarity to the content of the prompt, which were found by the vector search, from multiple 3D models stored in DB 221. In other words, the search processing unit 243 searches for 3D models by evaluating the closeness between the natural language of the prompt and the explanatory data inside DB 221. The search processing unit 243 performs the above processing using generative AI.

[0189] Furthermore, the search processing unit 243 searches for a 3D model corresponding to the content of a prompt, based on multiple descriptive data stored in DB 221 and the prompt input from the prompt input unit 242, using a string search with a generation AI.

[0190] Specifically, the search processing unit 243 performs a string search to find descriptive data that exactly or partially matches the string contained in the prompt, based on the multiple descriptive data stored in DB 221 and the prompt input from the prompt input unit 242. Then, the search processing unit 243 extracts 3D models corresponding to the descriptive data that exactly or partially matches the string contained in the prompt, which was found by the string search, from among the multiple 3D models stored in DB 221. In other words, the search processing unit 243 searches for 3D models by evaluating the degree of similarity between the natural language of the prompt and the descriptive data inside DB 221. The search processing unit 243 performs the above processing using a generative AI. The string search performed by the search processing unit 243 can be said to be a search for descriptive data and 3D models that have a relatively high degree of similarity to the content of the prompt.

[0191] The search processing unit 243 performs vector searches using the generating AI and string searches using the generating AI in parallel. The search processing unit 243 can also perform only one of these operations. Furthermore, the search processing unit 243 may perform one operation first and the other later, rather than processing the vector and string searches in parallel.

[0192] After searching for 3D models, the search processing unit 243 associates the found descriptive data with the 3D model data of the 3D model corresponding to the descriptive data, and outputs the descriptive data and the 3D model data as search results to the search results output unit 244. Alternatively, after searching for 3D models, the search processing unit 243 may associate the found descriptive data with the file path of the 3D model corresponding to the descriptive data, and output the descriptive data and the file path of the 3D model as search results to the search results output unit 244.

[0193] In Embodiment 2, the descriptive data stored in DB221 includes information such as the name of the 3D model, an overview of the 3D model, the category of the 3D model, a detailed description of the 3D model, and the file name of the 3D model, as content describing the 3D model. In other words, in Embodiment 2, the first artificial intelligence unit 11 generates descriptive data that associates information such as the name of the 3D model, an overview of the 3D model, the category of the 3D model, a detailed description of the 3D model, and the file name of the 3D model, and outputs it to the descriptive data acquisition unit 232 of the information processing device 2b.

[0194] The name of the 3D model is the name determined by the first artificial intelligence unit 11 for the 3D model, and may include information about the category of the 3D model, such as device, unit, part, or system. The overview of the 3D model is an overview of the configuration or function of the device of the 3D model. The category of the 3D model is the type of 3D model, which is one of device, unit, part, or system. The detailed description of the 3D model is a detailed description of the configuration or function of the device of the 3D model. The file name of the 3D model is the file name of the 3D model data.

[0195] The explanatory data acquisition unit 232 then receives explanatory data about the 3D model output from the first artificial intelligence unit 11 and outputs it to DB221. In other words, the names of the 3D models included in the explanatory data stored in DB221 are the names of the 3D models determined and generated by the first artificial intelligence unit 11.

[0196] Therefore, the search result description data output by the search processing unit 243 to the search result output unit 244 includes information describing the 3D model, such as the name of the 3D model, an overview of the 3D model, and a detailed description of the 3D model.

[0197] In Embodiment 2, the prompt is information that instructs the search processing unit 243 on the search conditions, and is a search request prompt that requests the AI ​​generated by the search processing unit 243 to search for 3D models from DB221, and is expressed in natural language. The search processing unit 243 acquires the prompt expressed in natural language. As an example, the search processing unit 243 acquires the prompt, which is text data, by taking text input about the content of the search request. Note that the manner in which the search processing unit 243 acquires the prompt is not limited to that described herein and is arbitrary. For example, the search processing unit 243 may acquire the prompt, which is audio data.

[0198] The search processing unit 243 comprises a vector search unit 2431, which is a search processing unit that performs 3D model searches using vector searches; a string search unit 2432, which is a search processing unit that performs 3D model searches using string searches; and a database control unit 2433.

[0199] The vector search unit 2431 is a search processing unit that searches for 3D models corresponding to the content of a prompt using vector search with generation AI, based on multiple descriptive data stored in DB 221 and prompts input from the prompt input unit 242. In other words, the vector search unit 2431 is a search processing unit that searches for 3D models corresponding to the content of a prompt from multiple 3D models stored in DB 221 using vector search.

[0200] Specifically, the vector search unit 2431 searches DB 221 for explanatory data with a relatively high vector similarity to the content of the prompt, based on multiple explanatory data stored in DB 221 and the prompt input from the prompt input unit 242. The vector search unit 2431 then extracts 3D models corresponding to the explanatory data with a relatively high vector similarity to the content of the prompt, which were found by the vector search, from multiple 3D models stored in DB 221. The vector search unit 2431 then associates the explanatory data with the 3D model data and outputs the explanatory data and the 3D model data as search results to the search result output unit 244. The vector search unit 2431 performs the above processing using a generation AI.

[0201] In Embodiment 2, DB221 stores, in addition to the explanatory data, vectors generated from the explanatory data, associating them with the 3D model data of the 3D model or the file path of the 3D model corresponding to the explanatory data. As long as DB221 can associate the explanatory data or vectors generated from the explanatory data with the 3D model data or the file path of the 3D model, the explanatory data and the 3D model data after the vectors have been generated may be stored outside the information processing device 2b.

[0202] In Embodiment 2, the first model control unit 111 of the first artificial intelligence unit 11 outputs necessary information to the model information retrieval unit 241 of the information processing device 2b. For this purpose, the first model control unit 111 generates a vector corresponding to the generated explanatory data from the explanatory data. The first model control unit 111 then associates the 3D model data, the explanatory data corresponding to the 3D model data, and the vector information corresponding to the explanatory data, and outputs it to the explanatory data acquisition unit 232. The explanatory data acquisition unit 232 associates the 3D model data, the explanatory data corresponding to the 3D model data, and the vector information corresponding to the explanatory data, and stores it in the DB 221.

[0203] The vector search unit 2431 generates a vector representing the content of the prompt input from the terminal device 3. The vector search unit 2431 compares the vector representing the content of the prompt with the vectors of all the explanatory data stored in DB 221 and determines the explanatory data from among the multiple explanatory data stored in DB 221 that have a relatively high vector similarity score to the content of the prompt. The vector search unit 2431 obtains the 3D model corresponding to the explanatory data that has been determined to have a relatively high vector similarity to the content of the prompt. Then, the vector search unit 2431 associates the explanatory data with the 3D model data and outputs a predetermined number of such explanatory data and 3D model data as search results to the search result output unit 244.

[0204] Explanatory data with a relatively high vector similarity score to the prompt content are, for example, the explanatory data with the highest vector similarity score to the prompt content, or a predetermined number of explanatory data with a relatively high vector similarity score. In other words, the vector search unit 2431 compares the prompt input to the vector search unit 2431 with the explanatory data stored in DB 221 to search for a predetermined number of explanatory data and 3D models with a relatively high vector similarity score to the prompt content, and outputs the search results to the search result output unit 244.

[0205] As an example, the vector search unit 2431 calculates the similarity between all the descriptive data stored in DB221 and the prompts input from terminal device 3 using known vector similarity calculation methods, such as calculation using cosine similarity with vectors, and determines four descriptive data with relatively high vector similarity scores from the content of the prompt. The vector search unit 2431 determines that the four 3D models corresponding to the four descriptive data with relatively high vector similarity scores from all the descriptive data stored in DB221 are the four 3D models with relatively high vector similarity scores from the content of the prompt. The prompt contains content that identifies the 3D model, for example, "Search for transport device".

[0206] Figure 16 shows an example of the configuration of the vector search unit 2431 included in the information processing device 2b according to Embodiment 2. The vector search unit 2431 includes a third artificial intelligence unit 51.

[0207] The third artificial intelligence unit 51 is an artificial intelligence unit having artificial intelligence functions, and is a generating AI for 3D model searching. In Embodiment 2, the artificial intelligence unit is an artificial intelligence equipped with intelligent functions such as reasoning or judgment, and its operating environment. The third artificial intelligence unit 51 is a generating AI for vector searching.

[0208] The third artificial intelligence unit 51 comprises a third model control unit 52, a third trained model storage unit 53, and a reference information storage unit 54.

[0209] The third artificial intelligence unit 51 is a model and its operating environment configured to output explanatory data corresponding to a prompt and 3D model data corresponding to the explanatory data when a prompt requesting the search of a 3D model is input via the prompt input unit 242. When a prompt requesting the search of a 3D model is input to the third artificial intelligence unit 51, it outputs explanatory data and 3D model data corresponding to the explanatory data as search results for the vector search unit 2431, based on the prompt and the trained model. Specifically, the third artificial intelligence unit 51 can output a predetermined number of explanatory data items with a relatively high vector similarity score to the prompt content and 3D model data corresponding to the explanatory data as search results for the vector search unit 2431, based on a vector indicating the content of the prompt requesting the search of a 3D model and the trained model.

[0210] The third trained model storage unit 53 stores the trained models. The form of the trained models is the same as that of the trained models stored in the first trained model storage unit 112.

[0211] Here, in the third artificial intelligence unit 51, a language learning model such as an LLM that takes natural language as input and produces output results can be applied as a model for obtaining explanatory data and corresponding 3D model data from prompts indicated in natural language.

[0212] When the third model control unit 52 receives a prompt via the prompt input unit 242, it retrieves all the information stored in DB 221 via the database control unit 2433. Then, using the model shown by the trained model, the third model control unit 52 searches all the information stored in DB 221 for a predetermined number of explanatory data items with relatively high vector similarity scores to the prompt content, and for 3D model data corresponding to that explanatory data, and outputs these as search results from the vector search unit 2431. The third model control unit 52 associates the explanatory data corresponding to the prompt with the 3D model data corresponding to that explanatory data and outputs it as a search result.

[0213] The vector similarity score is a numerical value that indicates how highly relevant the explanatory data is to the prompt in terms of vector similarity.

[0214] The trained model and other information used by the third model control unit 52 may be pre-prepared, or they may be acquired from outside the information processing device 2b via the network 200 as needed.

[0215] The third artificial intelligence unit 51 described above can calculate vector similarity between the input prompt characters and the vector of explanatory data stored in DB221, for example, by inference using LLM. The third artificial intelligence unit 51 may also generate a vector corresponding to the explanatory data stored in DB221.

[0216] Furthermore, the third artificial intelligence unit 51 can calculate vector similarity between the input prompt characters and the vector of explanatory data stored in DB221, for example, by inference using RAG (Retrieval Augmented Generation) in LLM.

[0217] The third artificial intelligence unit 51 includes a reference information storage unit 54, as shown in Figure 16. The third artificial intelligence unit 51 can calculate vector similarity by RAG inference by referring to the reference information stored in the reference information storage unit 54. The third artificial intelligence unit 51 may also be a model and its operating environment configured to output explanatory data corresponding to the prompt and 3D model data corresponding to the explanatory data, based on the prompt, other information that can be referenced by the third artificial intelligence unit 51, and a trained model, when a prompt is input.

[0218] The reference information storage unit 54 stores reference information. The reference information storage unit 54 may consist of multiple databases connected via a network.

[0219] Reference information is information that the third model control unit 52 refers to in order to output explanatory data corresponding to a prompt and 3D model data corresponding to said explanatory data. In other words, reference information is information that the third model control unit 52 refers to when calculating vector similarity by inference using RAG by LLM, for example. Reference information is, for example, information that the trained model of the third artificial intelligence unit 51 has not learned, or information that contains only correct information.

[0220] Reference information includes, for example, information related to prompts, information related to expressions contained in prompts, information related to concepts contained in prompts, information on the history of prompts previously entered, information on the history of explanatory data output by the third artificial intelligence unit 51 in the past, information on the history of 3D model data corresponding to explanatory data output by the third artificial intelligence unit 51 in the past, information that associates features contained in previously entered prompts with features contained in the output from the third artificial intelligence unit 51 for those prompts, information that evaluates the results output by the third artificial intelligence unit 51 for previously entered prompts, so-called grounding information, non-public internal information, design rules, and past trouble cases.

[0221] When a prompt is input, the third model control unit 52 searches for and outputs explanatory data corresponding to the prompt and 3D model data corresponding to the explanatory data, based on the prompt and reference information. For example, when a prompt is input, the third model control unit 52 searches the reference information storage unit 54 and obtains reference information corresponding to the prompt. The third model control unit 52 searches for explanatory data corresponding to the prompt and 3D model data corresponding to the explanatory data, based on the prompt, the trained model, and the reference information.

[0222] The third artificial intelligence unit 51 may also include a search engine for retrieving reference information or an interface to said search engine, instead of the reference information storage unit 54. In such a case, the search range of said search engine may be an external network or a specific network. Here, as one of the external network or a specific network, the database provided by the information processing device 2b of this disclosure, for example, the storage unit 22, can be used.

[0223] The string search unit 2432 is a search processing unit that searches for a 3D model corresponding to the content of a prompt using a string search generated by AI, based on multiple descriptive data stored in DB221 and a prompt input from the prompt input unit 242. In other words, the string search unit 2432 is a search processing unit that searches for a 3D model corresponding to the content of a prompt from multiple 3D models stored in DB221 using a string search.

[0224] Specifically, the string search unit 2432 uses a generating AI to perform a string search based on multiple descriptive data stored in DB 221 and prompts input from the prompt input unit 242. The search searches for descriptive data that exactly or partially matches the string contained in the prompt, and for the 3D model corresponding to that descriptive data.

[0225] Specifically, the string search unit 2432 compares all the description data stored in DB 221 with the string included in the prompt input from the prompt input unit 242, and searches for and retrieves description data that exactly or partially matches the string included in the prompt. The string search unit 2432 searches for and retrieves 3D models corresponding to the description data that was determined to exactly or partially match the string included in the prompt from among multiple 3D models stored in DB 221. Then, the string search unit 2432 associates the retrieved description data with the 3D model data and outputs the description data and the 3D model data as search results to the search result output unit 244. The string search unit 2432 performs the above processing using a generation AI.

[0226] In other words, the string search unit 2432 compares the prompt input to the information processing device 2b with the description data stored in the DB 221, searches for a predetermined number of description data entries that at least partially match the string contained in the prompt, and the corresponding 3D models, and outputs the search results to the search result output unit 244.

[0227] Figure 17 shows an example of the configuration of the string search unit 2432 included in the information processing device 2b according to Embodiment 2. The string search unit 2432 includes a fourth artificial intelligence unit 61.

[0228] The fourth artificial intelligence unit 61 is an artificial intelligence unit with artificial intelligence functions, and is a generating AI for 3D model searching. The fourth artificial intelligence unit 61 is a generating AI for string searching.

[0229] The fourth artificial intelligence unit 61 comprises a fourth model control unit 62 and a fourth trained model storage unit 63.

[0230] The fourth artificial intelligence unit 61 is a model and its operating environment configured to output explanatory data corresponding to a prompt and 3D model data corresponding to the explanatory data when a prompt requesting the search of a 3D model is input via the prompt input unit 242. When the fourth artificial intelligence unit 61 receives a prompt requesting the search of a 3D model, it outputs explanatory data and 3D model data corresponding to the explanatory data as search results of the string search unit 2432 based on the prompt and the trained model. Specifically, the fourth artificial intelligence unit 61 can output explanatory data and 3D model data corresponding to the explanatory data as search results of the string search unit 2432 based on the string of the prompt requesting the search of a 3D model and the trained model.

[0231] The fourth trained model storage unit 63 stores the trained models. The form of the trained models is the same as that of the trained models stored in the first trained model storage unit 112.

[0232] Here, in the fourth artificial intelligence unit 61, a language learning model such as an LLM that takes natural language as input and produces output results can be applied as a model for obtaining explanatory data and corresponding 3D model data from prompts indicated in natural language.

[0233] When the fourth model control unit 62 receives a prompt via the prompt input unit 242, it retrieves all information stored in DB 221 via the database control unit 2433. Then, using the model indicated by the trained model, the fourth model control unit 62 searches all the information stored in DB 221 for descriptive data that exactly or partially matches the string contained in the prompt, and 3D model data corresponding to that descriptive data, and outputs these as search results from the string search unit 2432. The fourth model control unit 62 outputs a predetermined number of descriptive data that exactly or partially matches the string contained in the prompt, and whose string matching score to the string contained in the prompt is relatively high, and 3D model data corresponding to that descriptive data. The fourth model control unit 62 associates the descriptive data corresponding to the prompt with the 3D model data corresponding to that descriptive data and outputs it as a search result.

[0234] The string match score is a numerical value that indicates how relevant the explanatory data is to the prompt in terms of string match.

[0235] The fourth model control unit 62 is variable in how many characters of the string included in the prompt must be exact matches for the search. That is, the fourth model control unit 62 can use a trained model to search for descriptive data that partially matches the string included in the prompt from all the information stored in DB221. For example, if the prompt contains a string of 10 characters, the fourth model control unit 62 can search for descriptive data that partially matches the string included in the prompt by performing a search for exact matches of 4 characters of that 10-character string.

[0236] The trained model and other information used by the fourth model control unit 62 may be pre-prepared, or they may be acquired from outside the information processing device 2b via the network 200 as needed.

[0237] Furthermore, the string search unit 2432 may be implemented by a search unit that performs string searches using a rule-based method without using artificial intelligence, instead of the fourth artificial intelligence unit 61 which is a generating AI.

[0238] As described above, in Embodiment 2, the configuration is such that a vector search is performed using the third artificial intelligence unit 51 and a string search is performed using the fourth artificial intelligence unit 61. However, the third artificial intelligence unit 51 for vector search and the fourth artificial intelligence unit 61 for string search may be implemented by a single artificial intelligence unit.

[0239] Furthermore, the third artificial intelligence unit 51 for vector search and the fourth artificial intelligence unit 61 for string search can also be located on an external cloud server or local server outside the information processing device 2b. In this case as well, the third artificial intelligence unit 51 and the fourth artificial intelligence unit 61 are configured to be connected to the database control unit 2433.

[0240] The database control unit 2433 is connected to the vector search unit 2431 and the string search unit 2432, and to DB221. The database control unit 2433 retrieves information stored in DB221, such as descriptive data and 3D model data, from DB221 and outputs it to the vector search unit 2431 and the string search unit 2432. The information stored in DB221 includes 3D model data and descriptive data for the 3D models shown in the 3D model data. The 3D model data and descriptive data for the 3D models stored in DB221 are the information that the search processing unit 243 searches. Hereinafter, the information stored in DB221 that is the target of the search processing unit 243 may be referred to as search target information.

[0241] The search result output unit 244 outputs the search results from the search processing unit 243 to the terminal device 3. Specifically, the search result output unit 244 creates screen data to display the search results from the search processing unit 243 on the display unit 32 of the terminal device 3, and outputs this screen data to the terminal device 3. In other words, the search result output unit 244 outputs the search results from the search processing unit 243 to the GUI. The terminal control unit 35 of the terminal device 3 uses the screen data obtained from the search result output unit 244 to display the search results on the display unit 32. The output destination for the search results may be set arbitrarily.

[0242] The search result output unit 244 can be described as a screen data creation unit that creates and outputs screen data for displaying a GUI screen that presents the search results from the search processing unit 243 to the user.

[0243] The search result output unit 244 includes a vector search output unit 2441 and a string search output unit 2442.

[0244] The vector search output unit 2441 outputs the search results from the vector search unit 2431 to the terminal device 3. Specifically, when the vector search output unit 2441 obtains the search results of the vector search from the vector search unit 2431, it creates screen data of the vector search results to display the search results on the display unit 32 of the terminal device 3.

[0245] The string search output unit 2442 outputs the search results from the string search unit 2432 to the terminal device 3. Specifically, when the string search output unit 2442 obtains the search results from the string search unit 2432, it creates screen data of the search results for the string search to display on the display unit 32 of the terminal device 3.

[0246] The search result output unit 244 creates screen data to display the search results on the search screen described later, based on the screen data of the vector search results created by the vector search output unit 2441 and the screen data of the string search results created by the string search output unit 2442, and outputs the screen data to the terminal device 3.

[0247] Next, the operation of the model information retrieval unit 241 will be described. Figure 18 is a flowchart showing an example of the operation procedure of the model information retrieval unit 241 provided in the retrieval unit 233 of the processing unit 23 of the information processing system 100b according to Embodiment 2.

[0248] In step S910, the prompt input unit 242 of the model information retrieval unit 241 acquires the prompt output from the terminal device 3. The prompt input unit 242 outputs the acquired prompt to the search processing unit 243.

[0249] When a user searches for a 3D model on the information processing device 2b, the prompt input unit 242 outputs search screen data to the terminal device 3 via the communication unit 21 of the information processing device 2b, which is used to display the top search screen of the 3D model search screen on the terminal device 3. The terminal control unit 35 of the terminal device 3 uses this search screen data to display the top search screen of the search screen on the display unit 32.

[0250] The search screen is a screen that displays information related to searching for 3D models in the information processing device 2b. The search screen has GUI functionality. The search screen allows input of prompt information. Prompt information includes the prompt and information for determining the prompt. The user can input or set prompt information and various instructions on the search screen using the operation unit 31 of the terminal device 3. The information input or set on the search screen is input to the terminal control unit 35 of the terminal device 3 and output from the terminal control unit 35 to the prompt input unit 242. The search screen can also display output information of the information processing device 2b, including search results output from the search result output unit 244 of the information processing device 2b. In other words, the search screen has the function of a search result output unit that displays search results output from the search result output unit 244 of the information processing device 2b.

[0251] Figure 19 shows an example of the search top screen 501 of the search screen displayed on the terminal device 3 according to Embodiment 2. The search top screen 501 is the initial screen of the search screen when performing a search for 3D models using the information processing device 2b. The search top screen 501 shown in Figure 19 is divided into multiple areas. The search top screen 501 displays a dialogue area 502, a vector search result display area 503, and a string search result display area 504.

[0252] The dialogue area 502 is an area for the user to interact with the artificial intelligence unit of the search processing unit 243, and displays the content of the conversation between the user and the artificial intelligence unit of the search processing unit 243. The dialogue area 502 is provided with a prompt input area 505. The artificial intelligence unit of the search processing unit 243 consists of the third artificial intelligence unit 51 and the fourth artificial intelligence unit 61.

[0253] The prompt input area 505 is an area where the user inputs information for a desired prompt using the operation unit 31 of the terminal device 3. The prompt information entered in the prompt input area 505 is displayed in the dialogue area 502 as user input information. The prompt information entered in the prompt input area 505 is also output to the prompt input unit 242 of the model information retrieval unit 241 of the information processing device 2b via the terminal communication unit 33 of the terminal device 3. As a result, in step S910, the prompt input unit 242 can acquire the prompt.

[0254] The vector search result display area 503 is an area where the search results for 3D models from DB221, performed by the vector search unit 2431 based on the prompt entered in the prompt input area 505, are displayed.

[0255] The string search result display area 504 is an area where the search results for 3D models from DB221, performed by the string search unit 2432 based on the prompt entered in the prompt input area 505, are displayed.

[0256] Note that the search screen shown in Figure 19 is an example of a search screen that the information processing device 2b displays on the terminal device 3. The search screen may have a different configuration than the one shown in Figure 19.

[0257] In step S920, the search processing unit 243 retrieves all search target information stored in DB221. That is, when a prompt is input from the prompt input unit 242, the search processing unit 243 requests the database control unit 2433 to transfer all search target information stored in DB221. The database control unit 2433 retrieves all search target information stored in DB221 and transfers this search target information to the search processing unit 243. As a result, the search processing unit 243 can retrieve all search target information stored in DB221.

[0258] In step S930, the prompts obtained in steps S910 and S920, along with all the search target information, are input to the third artificial intelligence unit 51 of the vector search unit 2431 and the fourth artificial intelligence unit 61 of the string search unit 2432.

[0259] In step S940, the vector search output unit 2441 obtains search results from the third artificial intelligence unit 51 corresponding to the prompt input to the third artificial intelligence unit 51 in step S930. In addition, the string search output unit 2442 obtains search results from the fourth artificial intelligence unit 61 corresponding to the prompt input to the fourth artificial intelligence unit 61 in step S930.

[0260] In step S950, the search result output unit 244 outputs the search results from the search processing unit 243 to the terminal device 3. That is, the search result output unit 244 creates screen data to display the search results from the search processing unit 243 that were input from the search processing unit 243, and transmits this screen data to the terminal device 3.

[0261] Specifically, the vector search output unit 2441 creates screen data to display the search results obtained from the vector search unit 2431 on the display unit 32 of the terminal device 3. The string search output unit 2442 also creates screen data to display the search results obtained from the string search unit 2432 on the display unit 32 of the terminal device 3. The search result output unit 244 uses the screen data created by the vector search output unit 2441 and the screen data created by the string search output unit 2442 to create screen data to display the search results from the search processing unit 243 on the display unit 32 of the terminal device 3, and outputs this screen data to the terminal device 3.

[0262] Figure 20 shows an example of the search results display screen 506 of the search screen displayed on the terminal device 3 according to Embodiment 2. The search results display screen 506 is a screen that displays the search results for 3D models in the search processing unit 243. In the search results display screen 506 shown in Figure 20, as shown in Figure 19, the search results for 3D models in the search processing unit 243 based on the prompt "Is there a transport device?" entered in the prompt input area 505 are displayed.

[0263] In the search results display screen 506 shown in Figure 20, the dialogue area 502 displays the content of the prompt entered by the user 502a and the content of the 3D model search results 502b from the vector search unit 2431 based on that prompt. In the search results display screen 506 shown in Figure 20, a predetermined icon is displayed next to the 3D model search results 502b, making it easy to understand that the 3D model search results 502b are search results generated by the AI.

[0264] The search results for 3D models, 502b, include, for example, the file name of the searched 3D model and the vector similarity score of the searched 3D model. However, the contents of the search results for 3D models, 502b, are not limited to these. Although not shown in the search results display screen 506 of Figure 20, the search results for 3D models in the prompt-based string search unit 2432 are also displayed in the dialogue area 502.

[0265] Furthermore, in the search results display screen 506 shown in Figure 20, the search results 503a of the 3D model from the vector search unit 2431 based on the prompt entered by the user are displayed in the vector search results display area 503. The search results 503a include, for example, the file name of the 3D model and a summary of the 3D model. The search results 503a also display an external view 503c of the 3D model.

[0266] By viewing the search results 503a, the user can confirm the file name and overview of the 3D model corresponding to the prompt. Furthermore, by viewing the external view 503c of the 3D model, the user can visually understand what kind of device each 3D model in the search results represents. In the vector search results display area 503, the user can view detailed information about the selected 3D model by selecting the display area 503b of each displayed 3D model. Note that the content of the search results 503a is not limited to this.

[0267] Furthermore, in the search results display screen 506 shown in Figure 20, the search results 504a of 3D models from the string search unit 2432 based on the prompt entered by the user are displayed in the string search results display area 504. The search results 504a display the file name of the 3D model, the candidate device name of the 3D model, and the category of the 3D model. Operation buttons 504b are also provided in the string search results display area 504. Operation buttons 504b are used to display detailed information of each 3D model displayed in the string search results display area 504. The user can view detailed information of the selected 3D model by selecting the operation button 504b of each displayed 3D model. Note that the content of the search results 504a is not limited to this.

[0268] Furthermore, the string search result display area 504 is provided with a keyword display area 504c. The keyword display area 504c automatically displays the keywords used in the 3D model search in the string search unit 2432. In the search result display screen 506 shown in Figure 20, "transportation device" is displayed in the keyword display area 504c. This allows the user to easily understand that the keyword used in the 3D model search in the vector search unit 2431 is "transportation device".

[0269] FIG. 21 is a diagram illustrating an example of a detailed information screen 511 of a search screen displayed on the terminal device 3 according to the second embodiment. The detailed information screen 511 is displayed, for example, in a superimposed manner on the search result display screen 506 shown in FIG. 20. The detailed information screen 511 is a detailed information display section that displays detailed information about the 3D model of the search results searched by the search processing section 243.

[0270] On the detailed information screen 511, for example, information including the file name, device name candidate, overview, category, details, and an external view 503c of the 3D model corresponding to the selected operation button 504b is displayed. The device name candidate is a name determined for the 3D model by the first artificial intelligence section 11, which is included in the description data stored in the DB 221.

[0271] Furthermore, additional information about the 3D model is displayed on the detailed information screen 511. For a 3D model registered in the DB 221, a user can add additional information about the 3D model to the DB 221 from the terminal device 3. The additional information is stored in the DB 221 in association with the corresponding 3D model, the description data of the 3D model, and the detailed information about the 3D model.

[0272] The additional information includes, for example, information on the sales of products corresponding to the 3D model, specifically information of customers, country names, regions, and prices. A customer is a customer who has purchased the product corresponding to the 3D model. The country name is the name of the country where the customer purchased the product corresponding to the 3D model. The region is the region within the country where the customer purchased the product corresponding to the 3D model. The price is the price at which the customer purchased the product corresponding to the 3D model.

[0273] By viewing the detailed information screen 511, the user can obtain various types of information about the 3D model displayed on the search screen and understand the details of the 3D model. Furthermore, by viewing the detailed information screen 511, the user can obtain various types of information about the sales of the product corresponding to the 3D model.

[0274] As a result, the model information retrieval unit 241 of the information processing device 2b terminates its operation according to the procedure shown in Figure 18.

[0275] Next, the operation of the third artificial intelligence unit 51 will be described. Figure 22 is a flowchart showing an example of the operation procedure of the third artificial intelligence unit 51 provided in the information processing system 100b according to Embodiment 2.

[0276] In step S1010, the third artificial intelligence unit 51 acquires the prompt input from the prompt input unit 242 and all search target information stored in DB 221.

[0277] In step S1020, the third artificial intelligence unit 51 searches for a 3D model by vector search based on the prompt and the explanatory data contained in the search target information. Specifically, the third artificial intelligence unit 51 performs a vector search based on the prompt vector, the vectors of all the explanatory data contained in the search target information, and the trained model, searches for explanatory data that has a relatively high vector similarity to the content of the prompt, and searches for the 3D model corresponding to that explanatory data.

[0278] In step S1030, the third artificial intelligence unit 51 outputs the search results to the search result output unit 244. The third artificial intelligence unit 51 outputs a predetermined number of explanatory data items that have a relatively high vector similarity to the prompt content, and associates these explanatory data items with the corresponding 3D models.

[0279] As a result, the third artificial intelligence unit 51 of the information processing device 2b terminates its operation according to the procedure shown in Figure 22.

[0280] Next, the operation of the fourth artificial intelligence unit 61 will be described. Figure 23 is a flowchart showing an example of the operation procedure of the fourth artificial intelligence unit 61 provided in the information processing system 100b according to Embodiment 2.

[0281] In step S1110, the fourth artificial intelligence unit 61 acquires the prompt input from the prompt input unit 242 and all search target information stored in DB 221.

[0282] In step S1120, the fourth artificial intelligence unit 61 searches for a 3D model by string search based on the prompt and the descriptive data contained in the search target information. Specifically, the fourth artificial intelligence unit 61 performs a string search based on the prompt, all the descriptive data contained in the search target information, and the trained model, searches for descriptive data that exactly or partially matches the string contained in the prompt, and searches for the 3D model corresponding to that descriptive data.

[0283] In step S1130, the fourth artificial intelligence unit 61 outputs search results to the search result output unit 244. The fourth artificial intelligence unit 61 outputs a predetermined number of descriptive data that exactly or partially matches the string contained in the prompt content, and associates these descriptive data with the corresponding 3D models.

[0284] As a result, the fourth artificial intelligence unit 61 of the information processing device 2b terminates its operation according to the procedure shown in Figure 23.

[0285] Next, we will explain the display operation in which the terminal device 3 displays the search results for the 3D model in the information processing device 2b. Figure 24 is a flowchart showing an example of the operation procedure of the terminal device 3 in Embodiment 2.

[0286] In step S1210, the terminal control unit 35 of the terminal device 3 displays a search screen on the display unit 32 of the terminal device 3. Specifically, the terminal control unit 35 obtains screen data for displaying the top search screen from the prompt input unit 242 of the information processing device 2b, and uses this screen data to display the top search screen on the display unit 32.

[0287] In step S1220, the terminal control unit 35 acquires a prompt. The terminal control unit 35 acquires information about the prompt entered by the user on the search screen from the operation unit 31 of the terminal device 3.

[0288] In step S1230, the terminal control unit 35 outputs the prompt information acquired in step S1220 to the prompt input unit 242 of the information processing device 2b via the terminal communication unit 33 of the terminal device 3.

[0289] In step S1240, the terminal control unit 35 acquires the search results for the 3D model in the information processing device 2b. Specifically, the terminal control unit 35 acquires screen data that displays the search results for the 3D model in the information processing device 2b, which is output from the search result output unit 244 of the information processing device 2b.

[0290] In step S1250, the display unit 32 of the terminal device 3 displays the search results for the 3D model in the information processing device 2b. That is, the terminal control unit 35 uses the screen data acquired in step S1240 to display the search results for the 3D model in the information processing device 2b on the display unit 32.

[0291] As a result, terminal device 3 completes its operation according to the procedure shown in Figure 24.

[0292] The model information retrieval unit 241 of the information processing device 2b according to the above-described embodiment 2 retrieves a 3D model corresponding to a given description data, based on a plurality of description data stored in DB221 and a prompt requesting the retrieval of a 3D model, using a generating artificial intelligence unit (generating AI) based on natural language. As a result, the user can easily retrieve the desired 3D model corresponding to the prompt from the 3D models stored in DB221 by interacting with the generating AI of the information processing device 2b using natural language.

[0293] Further, the model information search unit 241 performs vector search using the third artificial intelligence unit 51, which is generative AI, and character string search using the fourth artificial intelligence unit 61, which is generative AI. That is, the search processing unit 243 searches for a 3D model based on the vector similarity between the natural language content of the prompt and the description data, and searches for a 3D model based on the character string matching degree between the natural language character string of the prompt and the character string of the description data.

[0294] Accordingly, the model information search unit 241 can perform high-accuracy 3D model search through vector search of 3D models based on the context of the natural language of the prompt using generative AI. In addition to vector search of 3D models, the search processing unit 243 can improve the accuracy of 3D model search by performing character string search based on the natural language character string of the prompt using generative AI.

[0295] Further, the prompt input unit 242 of the model information search unit 241 functions as a screen data creation unit that creates and outputs screen data for displaying a GUI screen that accepts setting input of a prompt. In addition, the search result output unit 244 of the model information search unit 241 functions as a screen data creation unit that creates and outputs screen data for displaying a GUI screen that presents the search results of the description data and the 3D model. As described above, in the information processing apparatus 2b, a user can easily search for a desired 3D model corresponding to a prompt from among the 3D models stored in the DB 221 by interacting with the generative AI of the information processing apparatus 2b using a GUI.

[0296] The information processing device 2b includes a prompt input unit 242, a third artificial intelligence unit 51 which is a generating AI, a fourth artificial intelligence unit 61 which is a generating AI, and a search result output unit 244 as a GUI tool. The prompt input unit 242 and the search result output unit 244 also function as screen data creation units that create and output screen data to display the GUI screen in order to provide the user with a display screen as a GUI. The model information retrieval unit 241 provides screen data to display a GUI screen that accepts prompt setting inputs, explanatory data, and screen data to display a GUI screen that presents the search results for the 3D model.

[0297] Therefore, the information processing system 100b according to the above-described embodiment 2 has the effect of allowing users to automatically and accurately search for desired 3D models from DB221 without burdening the user, as the user does not need to create a database of 3D models with explanatory data describing the 3D models in advance.

[0298] In the above description, we have described a case where the search processing unit 243 performs both vector search and string search, and the search results for both vector and string searches are displayed together on the terminal device 3. However, this embodiment is not limited to this example. The search processing unit 243 may perform both vector search and string search, then sum the scores from the vector search results and the string search results, and display a predetermined number of explanatory data with relatively high summed scores, along with the corresponding 3D model data, as search results on the terminal device 3. Furthermore, the search processing unit 243 can perform only one of the two searches, vector search or string search. If the search processing unit 243 performs only one of the two searches, the terminal device 3 will only display the search results for the search performed.

[0299] Embodiment 3. Embodiment 3 describes an information processing system 100c, which is a modified version of the information processing system 100b according to Embodiment 2 described above. Figure 25 is a diagram showing the main configuration related to the operation of the information processing device 2c of the information processing system 100c according to Embodiment 3. The terminal device 3 is also shown in Figure 25. In Figure 25, the same reference numerals are used for the same components as in the information processing system 100b according to Embodiment 2 described above, and detailed explanations are omitted. Below, the configuration of the information processing device 2c that differs from the information processing system 100b will be mainly described. The information processing system 100c according to Embodiment 3 differs from the information processing system 100b described above in that it includes an information processing device 2c instead of an information processing device 2b.

[0300] The information processing device 2c according to Embodiment 3 has the configuration and functions of the information processing device 2b described above. The information processing device 2c according to Embodiment 3 differs from the information processing device 2b according to Embodiment 2 in that it includes a model information retrieval unit 241c instead of the model information retrieval unit 241 according to Embodiment 2. The model information retrieval unit 241c according to Embodiment 3 differs from the model information retrieval unit 241 according to Embodiment 2 in that the prompt input unit 242 further includes a prompt creation assistance unit 2421 and a dialogue history database 2422 as its functions.

[0301] The prompt creation support unit 2421 displays a prompt creation support screen on the search screen and automatically creates prompts based on the information set on the prompt creation support screen, thereby assisting the user in creating prompts. Specifically, the prompt creation support unit 2421 creates screen data to display the prompt creation support screen and transmits it to the terminal device 3.

[0302] As shown in Figure 19, a prompt creation button 507 is provided in the dialogue area 502 of the search screen. The prompt creation button 507 is an operation button that activates the prompt creation operation by the prompt creation support unit 2421 and displays the prompt creation support screen. When the prompt creation button 507 is selected on the search screen, prompt creation button selection information indicating that the prompt creation button 507 has been selected is input from the operation unit 31 of the terminal device 3 to the terminal control unit 35 of the terminal device 3, and output from the terminal control unit 35 to the prompt creation support unit 2421 of the prompt input unit 242.

[0303] The prompt creation support unit 2421 creates screen data to display the prompt creation support screen when prompt creation button selection information is input, and outputs the screen data to the terminal device 3. The terminal control unit 35 of the terminal device 3 uses the screen data obtained from the prompt creation support unit 2421 to display the prompt creation support screen on the display unit 32.

[0304] The prompt creation assistance screen allows the user to set and input prompt creation conditions, which are the conditions for automatically creating prompts. The prompt creation conditions set and input on the prompt creation assistance screen are input from the operation unit 31 of the terminal device 3 to the terminal control unit 35 of the terminal device 3, and output from the terminal control unit 35 to the prompt creation assistance unit 2421 of the prompt input unit 242. In other words, the prompt creation assistance screen is a GUI screen that accepts prompt creation conditions from the user, which are the conditions for automatically creating prompts.

[0305] Figure 26 shows an example of a prompt creation assistance screen displayed on the display unit 32 of the terminal device 3 according to Embodiment 3. The prompt creation assistance screen 521 shown in Figure 26 displays an input area 522 for setting the device name, an input area 523 for setting the part name, an input area 524 for setting the manufacturer name, and an input area 525 for setting the number of search results.

[0306] The device name setting input area 522 is where the device name of the 3D model to be included in the prompt as a search target for 3D models is set and entered. The device name is set and entered in the device name setting input area 522 as a prompt creation condition. The system name can also be set and entered in the device name setting input area 522.

[0307] Furthermore, the prompt creation assistance screen 521 may also include a system name setting input area in addition to the device name setting input area 522, allowing the system name of the 3D model to be included as a search target for the prompt, separately from the device name. Searching for 3D models separately for devices and systems has the effect of increasing search speed and improving the accuracy of 3D model searches.

[0308] The part name setting input area 523 is where the part names of the 3D models to be included in the prompt as search targets are entered. The part names are entered in the part name setting input area 523 as prompt creation conditions.

[0309] The manufacturer name setting input area 524 is where the manufacturer name of the 3D model, which will be included in the prompt and used to search for 3D models, is entered. The manufacturer name is entered in the manufacturer name setting input area 524 as a prompt creation condition.

[0310] The search result setting input area 525 is an area where the number of 3D model search results to be displayed in the prompt is set. The search result setting input area 525 is used to set the number of search results as a prompt creation condition.

[0311] Figure 27 illustrates the pull-down menu of the device name setting input area 522 in the prompt creation assistance screen 521 according to Embodiment 3. When the device name setting input area 522 is selected in the prompt creation assistance screen 521 shown in Figure 27, the device name options to be set in the device name setting input area 522 are displayed as options for the device name prompt creation conditions, in a list below the device name setting input area 522. Figure 27 shows the state where the options "VR assembly demo device" and "toy assembly device" are displayed as part of the device name options 522a.

[0312] The device name options for the prompt creation conditions include all device names and system names of 3D models that are included in the description data of the 3D models stored in DB221.

[0313] Furthermore, the device name options in the prompt creation conditions include all device names and system names of 3D models stored in the dialogue history database 2422, which are included in the dialogue history information, which is information about past dialogue history between the artificial intelligence unit of the model information retrieval unit 241c of the information processing device 2c and the user. In other words, the device name options include all device names of 3D models stored in the dialogue history database 2422, which are included in the information about past prompts and the search results information for 3D models in the search processing unit 243 based on those prompts.

[0314] The prompt creation support unit 2421 retrieves the device names and system names of all 3D models contained in the 3D model description data stored in DB221 and creates options for device names and system names in the prompt creation conditions. The prompt creation support unit 2421 also retrieves the device names and system names of all 3D models contained in the dialogue history information stored in the dialogue history database 2422 and creates options for device names and system names in the prompt creation conditions. Furthermore, the prompt creation support unit 2421 can retrieve technical terms and other words contained in the dialogue history information stored in the dialogue history database 2422 and create options for technical terms and other words in the prompt creation conditions. The prompt creation support unit 2421 may also use generation AI to retrieve device names and system names from the 3D model description data stored in DB221 and the dialogue history information stored in the dialogue history database 2422. More specifically, for example, the prompt creation support unit 2421 uses a generation AI to extract noun phrases from the 3D model's description data and dialogue history information, and then extracts the device name and system name from the extracted noun phrases. The prompt creation support unit 2421 may also use the generation AI to obtain technical terms and other words from the dialogue history information stored in the dialogue history database 2422.

[0315] The user can easily set the prompt generation conditions for the device name in the device name setting input area 522 by selecting the desired option from the device name options displayed in the pull-down menu using the operation unit 31 of the terminal device 3. Alternatively, the device name can also be directly entered into the device name setting input area 522 using the operation unit 31 of the terminal device 3.

[0316] The input area 523 for setting the part name, the input area 524 for setting the manufacturer name, and the input area 525 for setting the number of search results also have a function to set prompt creation conditions using pull-down menus, similar to the input area 522 for setting the device name. In the pull-down menu of the input area 525 for setting the number of search results, a list of numbers to specify the number of search results is displayed.

[0317] The part name options in the prompt creation conditions include all 3D model part names contained in the 3D model description data stored in DB221. Additionally, the part name options include all 3D model part names contained in the user interaction history information stored in the interaction history database 2422.

[0318] The manufacturer name options in the prompt creation conditions include all manufacturer names of 3D models included in the 3D model description data stored in DB221. Additionally, the manufacturer name options include all manufacturer names of 3D models included in the user interaction history information stored in the interaction history database 2422.

[0319] Figure 28 shows the state in which prompt creation conditions have been set and entered in the prompt creation assistance screen 521 according to Embodiment 3. When prompt creation conditions are set and entered in either the device name setting input area 522, the part name setting input area 523, or the search count setting input area 525 in the prompt creation assistance screen 521, the prompt creation button 526 provided on the prompt creation assistance screen 521 is activated. Setting prompt creation conditions in the manufacturer name setting input area 524 is optional.

[0320] When the prompt creation button 526 is selected, the prompt creation condition information, which is information on all prompt creation conditions set and entered on the prompt creation assistance screen 521, is input from the operation unit 31 of the terminal device 3 to the terminal control unit 35 of the terminal device 3, and is output from the terminal control unit 35 to the prompt creation assistance unit 2421 of the prompt input unit 242.

[0321] When prompt creation condition information is input, the prompt creation assistant unit 2421 automatically creates a prompt based on the prompt creation conditions indicated in the prompt creation condition information. For example, the prompt creation assistant unit 2421 creates a prompt by inserting the prompt creation conditions indicated in the prompt creation condition information into a predefined phrase that is stored in the prompt creation assistant unit 2421. However, the method of creating prompts in the prompt creation assistant unit 2421 is not limited to this.

[0322] The prompt creation support unit 2421 creates screen data to display a search screen that includes the prompt created by the prompt creation support unit 2421, and outputs the screen data to the terminal device 3. The terminal control unit 35 of the terminal device 3 uses the screen data obtained from the prompt creation support unit 2421 to display the search screen that includes the prompt created by the prompt creation support unit 2421 on the display unit 32.

[0323] Figure 29 shows an example of a state in which a prompt created by the prompt creation support unit 2421 according to Embodiment 3 is displayed in the prompt input area 505 of the search screen. The prompt created by the prompt creation support unit 2421 is displayed in the prompt input area 505 of the dialogue area 502 of the search screen. This allows the user to confirm the prompt created by the prompt creation support unit 2421 based on the prompt creation conditions set and entered by the user in the prompt creation support screen. In order to make it easier to understand that the prompt displayed in the prompt input area 505 is a prompt created by the prompt creation support unit 2421, a predetermined icon may be displayed next to the prompt input area 505 for the prompt created by the prompt creation support unit 2421.

[0324] When a prompt created by the prompt creation support unit 2421 is displayed in the prompt input area 505 on the search screen, and a user posts that prompt, the information of that prompt is input to the search processing unit 243 via the prompt input unit 242, similar to the second embodiment described above. Based on that prompt, i.e., the prompt created by the prompt creation support unit 2421, the search processing unit 243 performs a search for a 3D model corresponding to the content of the prompt, similar to the case of Embodiment 2.

[0325] The dialogue history database 2422 consists of the artificial intelligence unit of the search processing unit 243 of the model information retrieval unit 241c of the information processing device 2c, and the dialogue history storage unit that stores dialogue history information, which is information about past dialogue history with the user. The artificial intelligence unit of the search processing unit 243 consists of the third artificial intelligence unit 51 and the fourth artificial intelligence unit 61, which are generating AI.

[0326] In other words, the dialogue history database 2422 stores dialogue history information by associating the information of prompts entered by the user in the prompt input area 505 of the search screen with the information of the 3D model search results in the search processing unit 243 based on those prompts. The dialogue history database 2422 also stores dialogue history information by associating the information of prompts created by the prompt creation assistance unit 2421 with the information of the 3D model search results in the search processing unit 243 based on those prompts. For example, the dialogue history database 2422 stores dialogue history information by associating the content 502a of the prompt entered by the user, shown in the dialogue area 502 of the search results display screen 506 in Figure 20, with the content 502b of the 3D model search results in the vector search unit 2431 based on those prompts.

[0327] The prompt creation support unit 2421 stores the prompt information in the dialogue history database 2422 when it acquires prompt information entered by the user on the search screen and output from the terminal device 3, and when it automatically creates a prompt. The prompt creation support unit 2421 also acquires the 3D model search result information from the search processing unit 243 based on the prompt, associates the 3D model search result information with the prompt, and stores it in the dialogue history database 2422. As a result, the dialogue history information between the artificial intelligence unit of the model information retrieval unit 241c and the user is stored in the dialogue history database 2422 for each prompt. The prompt creation support unit 2421 can also acquire the 3D model search result information from the search processing unit 243 based on the prompt from the search result output unit 244.

[0328] Next, the operation of the prompt input unit 242 will be described. Figure 30 is a flowchart showing an example of the operation procedure of the prompt input unit 242 provided in the information processing device 2c of the information processing system 100c according to Embodiment 3.

[0329] In step S1310, the prompt creation assistance unit 2421 of the prompt input unit 242 creates screen data to display the prompt creation assistance screen and outputs the screen data to the terminal device 3. Specifically, when prompt creation button selection information is input from the terminal device 3, the prompt creation assistance unit 2421 creates screen data to display the prompt creation assistance screen and outputs the screen data to the terminal device 3. The terminal control unit 35 of the terminal device 3 uses the screen data obtained from the prompt creation assistance unit 2421 to display the prompt creation assistance screen on the display unit 32.

[0330] In step S1320, the prompt creation support unit 2421 acquires prompt creation condition information output from the terminal device 3. Specifically, the prompt creation support unit 2421 acquires prompt creation condition information set and entered by the user on the prompt creation support screen from the terminal device 3.

[0331] In step S1330, the prompt creation support unit 2421 creates a prompt based on the prompt creation condition information. The prompt creation support unit 2421 creates screen data to display a search screen including the prompt created by the prompt creation support unit 2421, and outputs the screen data to the terminal device 3. The terminal control unit 35 of the terminal device 3 uses the screen data obtained from the prompt creation support unit 2421 to display the search screen including the prompt created by the prompt creation support unit 2421 on the display unit 32. The search screen including the prompt created by the prompt creation support unit 2421 is a GUI screen that presents the prompt created by the prompt creation support unit 2421 to the user. When the user posts the prompt on the search screen, the information of the prompt, i.e., the information of the prompt created by the prompt creation support unit 2421, is output from the terminal device 3 to the information processing device 2c and input to the search processing unit 243 via the prompt input unit 242.

[0332] The search processing unit 243 searches for a 3D model corresponding to the content of a prompt, in the same manner as in Embodiment 2, based on the prompt created by the prompt creation assistance unit 2421.

[0333] In step S1340, the prompt creation support unit 2421 associates the prompt information created by the prompt creation support unit 2421 with the search results information for the 3D model in the search processing unit 243 based on the prompt, and stores this information in the dialogue history database 2422.

[0334] Furthermore, when the prompt input unit 242 acquires prompt information entered by the user on the search screen and output from the terminal device 3, the prompt creation support unit 2421 associates the prompt information with the search results information for the 3D model in the search processing unit 243 based on that prompt and stores it in the dialogue history database 2422.

[0335] As a result, the prompt input unit 242 terminates its operation according to the procedure shown in Figure 30.

[0336] The information processing system 100c according to the above-described embodiment 3 includes a prompt input unit 242 of the model information search unit 241c, which includes a prompt creation support unit 2421 that automatically creates prompts based on prompt creation conditions, which are creation conditions for automatically creating prompts. The model information search unit 241c then has the prompt creation support unit 2421 automatically create prompts based on the prompt creation conditions set and entered on the prompt creation support screen, and then searches for 3D models corresponding to the content of the prompts based on the prompts created by the prompt creation support unit 2421.

[0337] According to the model information retrieval unit 241c, by automatically creating prompts using the prompt creation assistance unit 2421, the burden of creating prompts for the user is reduced, making it possible to search for 3D models from DB221. Furthermore, according to the model information retrieval unit 241c, by automatically creating prompts using the prompt creation assistance unit 2421, the model information retrieval unit 241 c By limiting the degree of freedom in the expression of the prompts entered, inconsistencies in prompt notation can be eliminated, thereby improving the accuracy of 3D model searches.

[0338] In other words, by automatically creating prompts using the prompt creation assistance unit 2421, the configuration of prompts can be restricted from a state where the user can write prompts completely freely. Then, the generation AI of the model information retrieval unit 241c searches for 3D models based on the prompts created by the prompt creation assistance unit 2421. This clarifies what the generation AI of the model information retrieval unit 241c should search for, thereby improving the accuracy of the 3D model search performed by the model information retrieval unit 241c.

[0339] Furthermore, the prompt creation support unit 2421 creates screen data to display the search top screen of the search screen, which is a GUI screen that accepts input for setting prompt creation conditions, and outputs it to the terminal device 3, providing the user with a search screen for setting prompt creation conditions. As a result, the user only needs to set and input the prompt creation conditions for searching for the desired 3D model on the search screen, without having to consider and input detailed prompts, and can easily search for the desired 3D model corresponding to the prompt from the 3D models stored in DB221.

[0340] Furthermore, the search screen that accepts input for setting prompt creation conditions is configured to accept individual input for setting at least one of the following as prompt creation conditions: the device name of the 3D model, the part name of the 3D model, the descriptive data, and the number of 3D model search results to present. This limits the search range of 3D models based on prompts and clarifies the search target of 3D models, so that the search results of 3D models based on prompts do not deviate significantly from what the user intended, and the accuracy of 3D model searches in the model information search unit 241c is improved.

[0341] Furthermore, the search screen that accepts input for setting prompt creation conditions can display the prompt creation condition options as a pull-down menu. This assists the user in inputting prompts and allows the generation AI of the model information search unit 241c, which searches for 3D models, to receive prompts that are easy for the generation AI to understand, thereby improving the accuracy of 3D model searches in the model information search unit 241.

[0342] Furthermore, the options for prompt creation conditions displayed in the pull-down menu include the names of 3D models contained in the description data stored in DB221. The names of 3D models contained in the description data stored in DB221 are the names of 3D models determined and generated by the first artificial intelligence unit 11. The first artificial intelligence unit 11 also determines the names contained in the description data stored in DB221 in conjunction with the detailed descriptions of the 3D models contained in the description data stored in DB221. In other words, the names of 3D models determined and generated by the first artificial intelligence unit 11 are consistently expressed with the detailed descriptions of the 3D models contained in the description data stored in DB221.

[0343] Therefore, the options for the prompt creation conditions displayed in the pull-down menu include the names of 3D models contained in the description data stored in DB221, and by including these names in the prompt, it becomes possible to search for 3D models from the information stored in DB221 using the names of 3D models contained in the description data stored in DB221.

[0344] Therefore, by including the names of 3D models in the description data stored in DB221 as options for prompt creation conditions displayed in the pull-down menu, the accuracy of 3D model searches in the model information search unit 241c is improved, and the user's understanding of the 3D model search results is also improved. In other words, the user will find that the accuracy of 3D model searches is higher when they select a name from the prompt creation condition options, and when the user views the search results, the name of the 3D model will match the content of the detailed description, making the name of the 3D model clearer, thus improving the user's understanding of the 3D model search results.

[0345] Furthermore, the model information retrieval unit 241c includes a dialogue history database 2422, which is a dialogue history storage unit that stores dialogue history information, which is information about the past dialogue history between the model information retrieval unit 241c's generating AI and the user.The options for prompt creation conditions include names included in the dialogue history information stored in the dialogue history database 2422.

[0346] By presenting the user with options for prompt creation conditions based on the wording included in the dialogue history information, the accuracy of 3D model searches is improved because the device name used when searching for 3D models in the past and the search results obtained at that time are linked to the prompt for the current search. Furthermore, it becomes possible to further refine the search based on the wording included in the dialogue history information, and to refine the search using specialized terminology from the description data stored in DB221, thereby improving both the search speed and the accuracy of 3D model searches.

[0347] Next, the hardware configurations of the control units 80 according to Embodiments 1 to 3 will be described. The control units 80 according to Embodiments 1 to 3 correspond to the processing units 23 and 24 of the information processing devices 2, 2a, and 2b according to Embodiments 1 to 3, the first artificial intelligence unit 11 and the first server control unit 14 of the first server 1, and the second artificial intelligence unit 41 and the second server control unit of the second server 4, respectively. The functions of the control units 80 according to Embodiments 1 to 3 are realized by processing circuits. The processing circuits may be dedicated hardware or processing devices that execute programs stored in a memory device.

[0348] When the processing circuit is dedicated hardware, the processing circuit may be a single circuit, a complex circuit, a programmed processor, a parallel programmed processor, an application-specific integrated circuit, a field-programmable gate array, or a combination thereof. Figure 31 shows a configuration in which each function of the control unit 80 according to Embodiments 1 to 3 is realized in hardware. The processing circuit 81 incorporates a logic circuit 81a that realizes the function of the control unit 80.

[0349] If the processing circuit 81 is a processing unit, the functions of the control unit 80 are realized by software, firmware, or a combination of software and firmware.

[0350] Figure 32 shows a configuration in which each function of the control unit 80 according to Embodiments 1 to 3 is implemented by software. The processing circuit 81 includes a processor 811 that executes program 81b, a random access memory 812 used by the processor 811 as a work area, and a storage device 813 that stores program 81b. The processor 811 loads program 81b stored in the storage device 813 onto the random access memory 812 and executes it, thereby realizing the functions of the control unit 80. The software or firmware is written in a programming language and stored in the storage device 813. The processor 811 can be a central processing unit, but is not limited to this. The storage device 813 can be a semiconductor memory such as RAM (Random Access Memory), ROM (Read Only Memory), flash memory, EPROM (Erasable Programmable Read Only Memory), or EEPROM (Electrically Erasable Programmable Read Only Memory). The semiconductor memory may be non-volatile memory or volatile memory. Furthermore, the storage device 813 can be a magnetic disk, flexible disk, optical disk, compact disk, minidisc, or DVD (Digital Versatile Disc) in addition to semiconductor memory. The processor 811 may output data such as calculation results to the storage device 813 for storage, or it may store such data in an auxiliary storage device (not shown) via the random access memory 812. By integrating the processor 811, random access memory 812, and storage device 813 onto a single chip, the functions of the control unit 80 can be realized by a microcomputer.

[0351] The processing circuit 81 realizes the functions of the control unit 80 by reading and executing the program 81b stored in the memory device 813. The program 81b can also be described as instructing the computer to execute the procedures and methods for realizing the functions of the control unit 80.

[0352] Furthermore, the processing circuit 81 may implement some of the functions of the control unit 80 using dedicated hardware, and some of the functions of the control unit 80 using software or firmware.

[0353] Thus, the processing circuit 81 can realize each of the above-mentioned functions through hardware, software, firmware, or a combination thereof.

[0354] The configurations shown in the above embodiments are merely examples, and it is possible to combine them with other known technologies, combine different embodiments, and omit or modify parts of the configuration without departing from the gist of the invention. [Explanation of Symbols]

[0355] 1 First Server, 2, 2a, 2b, 2c Information Processing Device, 3 Terminal Device, 4 Second Server, 11 First Artificial Intelligence Unit, 12 First Server Communication Unit, 13 First Server Storage Unit, 14 First Server Control Unit, 21 Communication Unit, 22 Storage Unit, 23 Processing Unit, 24, 80 Control Unit, 31 Operation Unit, 32 Display Unit, 33 Terminal Communication Unit, 34 Terminal Storage Unit, 35 Terminal Control Unit, 41 Second Artificial Intelligence Unit, 51 Third Artificial Intelligence Unit, 52 Third Model Control Unit, 53 Third Learned Model Storage Unit, 54 Reference Information Storage Unit, 61 Fourth Artificial Intelligence Unit, 62 Fourth Model Control Unit, 63 Fourth Learned Model Storage Unit, 81 Processing Circuit, 81a Logic Circuit, 81b Program, 100, 100a, 100b, 100c Information Processing System, 111 First Model Control Unit, 112 First Learned Model Storage Unit, 200 Network, 221 Database, 231 3D Model Transmission Unit, 232 Description Data Acquisition Unit, 233 Search Unit, 411 Second Model Control Unit, 241, 241c Model Information Search Unit, 242 Prompt Input Unit, 243 Search Processing Unit, 244 Search Result Output Unit, 412 Second Trained Model Storage Unit, 501 Search Top Screen, 502 Dialogue Area, 502a Prompt Content, 502b 3D Model Search Result Content, 503 Vector Search Result Display Area, 503a, 504a Search Results, 503b 3D Model Display Area, 503c 3D Model Outline, 504 String Search Result Display Area, 504b Operation Buttons, 504c Keyword Display Area, 505 Prompt Input Area, 506 Search Result Display Screen, 507, 526 Prompt Creation Button, 511 Detailed Information Screen, 521 Prompt Creation Assistance Screen, 522 Device name setting input area, 522a Device name selection, 523 Part name setting input area, 524 Manufacturer name setting input area, 525 Number of search results setting input area, 811 Processor, 812 Random access memory, 813 Storage device, 2311 Image extraction unit, 2312 Image conversion unit, 2321 Description data determination unit, 2322 3D model operation unit, 2421 Prompt creation assistance unit, 2422 Dialogue history database, 2431 Vector search unit, 2432 String search unit, 2433 Database control unit, 2441 Vector search output unit, 2442 String search output unit.

Claims

1. An information processing device that outputs explanatory data showing a description of a three-dimensional model, A 3D model transmission unit acquires 3D model data, which is data representing the aforementioned 3D model, and outputs the 3D model data and a prompt for generating the explanatory data to the first artificial intelligence unit. An explanatory data acquisition unit receives and outputs the explanatory data for the three-dimensional model generated by the first artificial intelligence unit based on the prompt, Equipped with, The aforementioned 3D model transmission unit is An image extraction unit that acquires the aforementioned 3D model data and extracts an image of the 3D model from the aforementioned 3D model data, An image conversion unit outputs the image extracted by the image extraction unit to the first artificial intelligence unit, Equipped with, The aforementioned explanatory data acquisition unit, An explanation data determination unit that acquires the explanation data output from the first artificial intelligence unit, A 3D model manipulation unit for editing the aforementioned 3D model data, Equipped with, The aforementioned explanatory data determination unit is: When the first artificial intelligence unit receives a determination result that the three-dimensional model is covered by a cover, it outputs an editing command to the three-dimensional model operation unit instructing it to delete the cover data from the three-dimensional model data. The 3D model manipulation unit performs editing on the 3D model data to delete the cover data based on the editing command, and outputs the edited 3D model data to the image extraction unit. An information processing device characterized by the following.

2. The image extraction unit uses the three-dimensional model data to photograph the three-dimensional model from multiple angles and extracts multiple images. The image conversion unit outputs the plurality of images extracted by the image extraction unit to the first artificial intelligence unit. The information processing apparatus according to claim 1, characterized by the following:

3. The aforementioned explanatory data determination unit is: The explanatory data output from the first artificial intelligence unit and the image extracted by the image extraction unit are output to the first artificial intelligence unit, and the editing command and the image extraction command instructing the extraction of the image again from the 3D model data are output to the 3D model operation unit until a determination result is received from the first artificial intelligence unit in which it is determined that the name of the device indicated by the 3D model indicated by the 3D model data is expressed in text. The aforementioned 3D model manipulation unit is: When the aforementioned editing command is received, the edited 3D model data and the image extraction command are output to the image extraction unit. The image extraction unit extracts an image of the edited 3D model from the edited 3D model data based on the image extraction command. An information processing apparatus according to claim 1 or 2, characterized by the above.

4. An information processing device that outputs explanatory data showing a description of a three-dimensional model, A 3D model transmission unit acquires 3D model data, which is data representing the aforementioned 3D model, and outputs the 3D model data and a prompt for generating the explanatory data to the first artificial intelligence unit. An explanatory data acquisition unit receives and outputs the explanatory data for the three-dimensional model generated by the first artificial intelligence unit based on the prompt, Equipped with, The aforementioned 3D model transmission unit is An image extraction unit that acquires the aforementioned 3D model data and extracts an image of the 3D model from the aforementioned 3D model data, An image conversion unit outputs the image extracted by the image extraction unit to the first artificial intelligence unit, Equipped with, The three-dimensional model transmission unit outputs design information relating to the three-dimensional model to the first artificial intelligence unit, in addition to the three-dimensional model data or the image extracted from the three-dimensional model data. An information processing device characterized by the following.

5. The three-dimensional model transmission unit outputs a plurality of prompts to the first artificial intelligence unit for extracting features from one three-dimensional model or one image extracted from the three-dimensional model. An information processing apparatus according to claim 1 or 4, characterized by the above.

6. The three-dimensional model transmission unit outputs a prompt to the first artificial intelligence unit that describes an example of input and output along with the three-dimensional model data or the image extracted from the three-dimensional model data. An information processing apparatus according to claim 1 or 4, characterized by the above.

7. The three-dimensional model transmission unit prompts the first artificial intelligence unit to output text representing the function of the three-dimensional model, An information processing apparatus according to claim 1 or 4, characterized by the above.

8. An information processing device that outputs explanatory data showing a description of a three-dimensional model, A 3D model transmission unit acquires 3D model data, which is data representing the aforementioned 3D model, and outputs the 3D model data and a prompt for generating the explanatory data to the first artificial intelligence unit. An explanatory data acquisition unit receives and outputs the explanatory data for the three-dimensional model generated by the first artificial intelligence unit based on the prompt, Equipped with, A storage unit that stores information of multiple three-dimensional model data and explanatory data corresponding to the three-dimensional model data in association with each other, A search unit compares a query input to the information processing device with the explanatory data stored in the storage unit, searches for a predetermined number of explanatory data and 3D models that have a relatively high similarity to the content of the query, and outputs the search results. Equipped with, The search unit outputs the query, the search results, and a prompt containing a request statement for information to narrow down the 3D model to the second artificial intelligence unit, and the second artificial intelligence unit retrieves the suggestion text corresponding to the query and the search results, which are the information output by the second artificial intelligence unit, and includes the suggestion text in the search results. An information processing device characterized by the following.

9. The search unit, based on a search request prompt input to the information processing device requesting the search of the three-dimensional model and the descriptive data stored in the storage unit, uses a generation artificial intelligence unit for searching for three-dimensional models to search for descriptive data and three-dimensional models that have a relatively high similarity to the content of the search request prompt, and outputs the search results. The information processing apparatus according to claim 8, characterized by the above.

10. The aforementioned search unit, A vector search is performed using a third artificial intelligence unit, which is a generative artificial intelligence unit, to compare the search request prompt with the explanatory data stored in the memory unit, to search for the explanatory data and the 3D model that have a relatively high vector similarity to the content of the search request prompt, and to output the search results. A string search that compares the search request prompt with the descriptive data stored in the storage unit to search for the descriptive data and the 3D model that at least partially match the string contained in the search request prompt, and outputs the search results, To do, The information processing apparatus according to claim 9, characterized by the following:

11. The search unit includes a screen data creation unit that creates and outputs screen data for displaying a graphical user interface screen that accepts setting input for the search request prompt, and screen data for displaying a graphical user interface screen that presents the explanatory data and the search results for the 3D model. The information processing apparatus according to claim 9, characterized by the following:

12. The search unit includes a prompt creation auxiliary unit that creates the search request prompt based on prompt creation conditions, which are creation conditions for automatically creating the search request prompt. The information processing apparatus according to claim 9, characterized by the following:

13. The prompt creation support unit creates and outputs screen data to display a graphical user interface screen that accepts input settings for the prompt creation conditions, and creates the search request prompt based on the prompt creation conditions received on the graphical user interface screen. The graphical user interface screen that accepts input for setting the prompt creation conditions accepts input for setting at least one of the following as prompt creation conditions: the device name of the 3D model, the part name of the 3D model, the descriptive data, and the number of search results for the 3D model. The information processing apparatus according to claim 12, characterized by the above.

14. The graphical user interface screen that accepts input for setting the prompt creation conditions displays the options for the prompt creation conditions as a pull-down menu. The information processing apparatus according to claim 13, characterized by the above.

15. The options for the prompt creation conditions include names included in the explanatory data stored in the memory unit. The information processing apparatus according to claim 14, characterized by the above.

16. It includes a dialogue history storage unit that stores dialogue history information, which is information about the past dialogue history between the user and the generating artificial intelligence unit. The options for the prompt creation conditions include the wording contained in the dialogue history information stored in the dialogue history storage unit. The information processing apparatus according to claim 14 or 15, characterized by the above.

Images