A method for constructing a metaverse exhibition scene based on a 3D rendering engine

Abstract

The present invention provides a method for constructing a metaverse exhibition scene based on a 3D rendering engine, which adopts NLP technology and LLM large language model to dynamically monitor the entity data mapping under each exhibition plot in the metaverse exhibition scene, and uses NLP technology to identify the mapping relationship of the metaverse exhibition mapping data set under each exhibition plot and further construct the mapping data monitoring prompt word, so that the LLM large language model monitors the prompt word of the metaverse exhibition mapping data set under each exhibition plot, and if a keyword that does not match the mapping data monitoring prompt word appears in a certain exhibition plot, it is determined that the mapping data of the exhibition plot has changed, and the system generates a model update rendering task for the exhibition plot, so that the 3D rendering engine performs model rendering update on the metaverse exhibition mapping data set under the exhibition plot. Therefore, the difficulty of background task scheduling and the pressure of computing operation are greatly reduced, and management and control are more intelligent.

Description

Technical Field

[0001] The present invention relates to the field of virtual reality technology, and in particular to a method for constructing a metaverse exhibition scene based on a 3D rendering engine, and a device, electronic device and computer-readable storage medium for constructing a metaverse exhibition scene based on a 3D rendering engine. Background Art

[0002] The Metaverse Exhibition Center (or Metaverse Exhibition Scene) can provide users with an immersive virtual exhibition experience. In the Metaverse Exhibition Scene, users can freely shuttle between booths, interact with industry elites, obtain the latest industry information in real time, and experience the integration of technology and the future. Users can visit exhibits in 3D stereoscopic display, perform interactive operations through virtual touch, and gain in-depth understanding of product details. At the same time, the Metaverse Exhibition Scene also provides a virtual conference room that supports multi-person online collaboration and discussion, breaking geographical restrictions and allowing exhibitors from all over the world to participate conveniently.

[0003] The general method of constructing a metaverse exhibition scene includes the following steps:

[0004] 1. Model preparation and preprocessing

[0005] ‌Model optimization‌: Before importing into the 3D rendering engine, the model of the exhibition scene is optimized as necessary, including reducing the number of polygons, merging materials, optimizing textures, etc., to improve rendering efficiency.

[0006] ‌Bone binding and animation setting‌: For models that need to be displayed dynamically (such as virtual characters, exhibit animations, etc.), perform bone binding and set the corresponding animation sequence.

[0007] 2. Rendering engine configuration

[0008] ‌Select rendering engine‌: Choose a suitable 3D rendering engine according to project requirements, such as Unity, Unreal Engine, etc. These engines provide rich APIs and tool sets to facilitate dynamic mapping control.

[0009] ‌Scene construction‌: Building exhibition scenes in the rendering engine, including arranging booths, setting lights, adjusting materials, etc., to ensure the realism and atmosphere of the scene.

[0010] 3. Dynamic mapping technology

[0011] ‌Real-time data mapping‌: Use the rendering engine’s scripts or APIs to map real-time data from external data sources (such as databases, sensors, etc.) to the scene model to dynamically update the model. For example, adjust the actions or expressions of virtual characters based on the audience’s behavior data.

[0012] ‌Interaction logic design‌: Design the interaction logic between users and scene models, such as clicking on exhibits to display detailed information, touching virtual characters to trigger conversations, etc. These interactions require real-time response and update of model status.

[0013] 4. Control and Optimization

[0014] ‌Performance monitoring and tuning‌: Real-time monitoring of rendering engine performance indicators, such as frame rate, memory usage, etc., and necessary optimization based on the monitoring results, such as adjusting rendering parameters, optimizing script logic, etc.

[0015] Dynamic loading and unloading: For large exhibition scenes, a dynamic loading and unloading mechanism for models is implemented. Based on the user's position and line of sight, nearby models are dynamically loaded and distant models are unloaded to improve rendering efficiency and user experience.

[0016] Therefore, the virtual interactive control of the metaverse exhibition scene is mainly achieved through the data set mapping interaction between the physical space and the virtual space, so as to realize the dynamic control and display of the metaverse exhibition scene.

[0017] However, when the sensor data of the physical space is collected into the virtual space and mapped in real time, the real-time mapping of the data will call upon and occupy huge computer graphics, vision, rendering and other computing resources, and require the driving system (such as the digital twin application system of the digital twin platform) to continuously perform multi-task scheduling, which brings greater operating pressure to the digital twin platform.

[0018] In addition, when performing mapping control of the metaverse exhibition scene, it not only involves the data mapping management between the physical model and the virtual model in the exhibition scene, but also mainly involves the scene data mapping management of the metaverse exhibition scene. The real-time data mapping management of traditional virtual models is handed over to the digital twin application system of the digital twin platform for task allocation management. As long as the front-end sensor data is monitored to change, the back-end needs to update the mapping of the data set of the entire model, and then through cloud computing technology, the data mapping, rendering and other tasks are assigned to multiple servers to perform the corresponding mapping operations respectively, so some model mapping work will be repeated. This leads to redundant task scheduling and updating, and even requires administrators to participate in task scheduling. This process consumes more task scheduling resources and may cause scheduling errors, and its operation is not intelligent enough.

[0019] Moreover, the scene interaction, plot presentation, etc. of the Metaverse exhibition scene need to be transformed according to the scene scenario nodes, and the administrator is also required to manage the scene switching, which also requires switching control. Therefore, the management and control of the Metaverse exhibition scene is not intelligent enough. Summary of the invention

[0020] In order to solve the technical problems existing in the prior art, the present invention provides the following technical solutions:

[0021] On the one hand, a method for constructing a metaverse exhibition scene based on a 3D rendering engine is provided, the method being implemented by an electronic device, and the method comprising:

[0022] S1. Based on a 3D rendering engine, the entity data of each exhibition plot in the Metaverse exhibition scene is mapped to the virtual model of the Metaverse exhibition scene, and a Metaverse exhibition mapping data set of each exhibition plot is generated and saved;

[0023] S2. Based on NLP technology, identify the mapping relationship of the metaverse exhibition mapping data set under each exhibition plot, and bind the mapping relationship with the corresponding exhibition plot;

[0024] S3, extracting the data mapping keywords in the mapping relationship bound to the exhibition plot, constructing a mapping data monitoring prompt word based on the mapping relationship and the data mapping keywords, and inputting the prompt word into a preset LLM large language model;

[0025] S4, uploading the Metaverse exhibition entity data under each of the exhibition scenarios in real time, and identifying whether there are keywords in the Metaverse exhibition entity data that do not match the mapping data monitoring prompt words through the LLM large language model traversal:

[0026] If it appears, the exhibition plot in which the keywords that do not match the mapping data monitoring prompt words appear is marked as a mapping node, and a model update rendering task of the mapping node is generated;

[0027] Otherwise, continue monitoring;

[0028] S5. Output and send the model update rendering task of the mapping node to the 3D rendering engine, and perform model rendering update on the Metaverse exhibition mapping data set under the mapping node.

[0029] Preferably, S1, based on a 3D rendering engine, mapping entity data under each exhibition plot in the Metaverse exhibition scene to a virtual model of the Metaverse exhibition scene, generating and saving a Metaverse exhibition mapping data set under each exhibition plot, including:

[0030] According to the preset disassembly and combination method, the exhibition scene of the metaverse is disassembled into exhibition plots, and storage space corresponding to the exhibition plots is reserved in the background database;

[0031] The physical model dataset of the Metaverse exhibition scene is collected by sensing, and a virtual model of the Metaverse exhibition scene is constructed based on a 3D rendering engine and a corresponding virtual model dataset is generated, and a data mapping relationship between the physical model dataset and the virtual model dataset is established to obtain a Metaverse exhibition mapping dataset of the Metaverse exhibition scene;

[0032] According to the exhibition plot, the metaverse exhibition mapping data set is split into subsets corresponding to different exhibition plots, and the metaverse exhibition mapping data set under each exhibition plot is obtained and saved in the storage space corresponding to the exhibition plot.

[0033] Preferably, S2, based on NLP technology, identifying the mapping relationship of the metaverse exhibition mapping data set under each exhibition plot, and binding the mapping relationship with the corresponding exhibition plot, including:

[0034] Reading the Metaverse exhibition mapping data set stored in each of the storage spaces in sequence;

[0035] Parsing and preprocessing the Metaverse exhibition mapping data set to obtain a number of mapping data blocks corresponding to various models under the exhibition plot;

[0036] Using NLP technology to identify entity information, virtual information and model mapping data of each model in the mapping data block, and extracting the mapping relationship between the entity model and the virtual model in the entity information, virtual information and model mapping data based on a preset relationship extraction template;

[0037] The mapping relationship extracted by the output identification is bound to the corresponding exhibition plot.

[0038] Preferably, S3, extracting the data mapping keywords in the mapping relationship bound to the exhibition plot, constructing a mapping data monitoring prompt word based on the mapping relationship and the data mapping keywords, and inputting it into a preset LLM large language model, including:

[0039] Using NLP technology, extracting the data mapping keywords in the mapping relationship of each of the exhibition plots;

[0040] Based on the mapping logic in the mapping relationship and the data mapping keywords corresponding to the exhibition plot, construct the mapping data monitoring prompt words for training the LLM large language model;

[0041] The mapping data monitoring prompt words are input into the LLM large language model to perform model learning.

[0042] Preferably, in step S4, the real-time uploading of the Metaverse exhibition entity data under each exhibition scenario includes:

[0043] Preset sampling frequency;

[0044] Through the digital twin application system, a sampling instruction is issued to the sensor system corresponding to the exhibition plot in the physical exhibition scene according to the sampling frequency, and the metaverse exhibition entity data of the exhibition plot is collected and uploaded to the background database;

[0045] The metaverse exhibition entity data is stored in the storage space corresponding to the exhibition plot.

[0046] On the other hand, a device for constructing a metaverse exhibition scene based on a 3D rendering engine is provided, and the device for constructing a metaverse exhibition scene based on a 3D rendering engine is used to implement the above-mentioned method for constructing a metaverse exhibition scene based on a 3D rendering engine, and the device includes:

[0047] The digital twin system is used to map the entity data of each exhibition plot in the Metaverse exhibition scene to the virtual model of the Metaverse exhibition scene based on the 3D rendering engine, generate and save the Metaverse exhibition mapping data set under each exhibition plot;

[0048] A mapping processing module, used to identify the mapping relationship of the metaverse exhibition mapping data set under each exhibition plot based on NLP technology, and bind the mapping relationship with the corresponding exhibition plot;

[0049] A prompt word generation module, used to extract the data mapping keywords in the mapping relationship bound to the exhibition plot, and based on the mapping relationship and the data mapping keywords, construct a mapping data monitoring prompt word and input it into a preset LLM large language model;

[0050] The intelligent recognition module is used to upload the Metaverse exhibition entity data under each of the exhibition plots in real time, and identify whether there are keywords in the Metaverse exhibition entity data that do not match the mapping data monitoring prompt words through the LLM large language model traversal:

[0051] If it appears, the exhibition plot in which the keywords that do not match the mapping data monitoring prompt words appear is marked as a mapping node, and a model update rendering task of the mapping node is generated;

[0052] Otherwise, continue monitoring;

[0053] The rendering task scheduling module is used to output and send the model update rendering task of the mapping node to the 3D rendering engine, and perform model rendering update on the metaverse exhibition mapping data set under the mapping node.

[0054] On the other hand, an electronic device is provided, comprising: a processor; a memory, wherein the memory stores computer-readable instructions, and when the computer-readable instructions are executed by the processor, any one of the methods for constructing a metaverse exhibition scene based on a 3D rendering engine as described above is implemented.

[0055] On the other hand, a computer-readable storage medium is provided, in which at least one instruction is stored. The at least one instruction is loaded and executed by a processor to implement any one of the above-mentioned methods for constructing a metaverse exhibition scene based on a 3D rendering engine.

[0056] The beneficial effects brought about by the technical solution provided by the embodiment of the present invention include at least:

[0057] In order to reduce the difficulty of updating the mapping of the data set of the Metaverse exhibition scene after construction, the present invention adopts NLP technology and LLM large language model to dynamically monitor the data set of the entity data mapping under each exhibition plot in the Metaverse exhibition scene, and uses NLP technology to identify the mapping relationship of the Metaverse exhibition mapping data set under each exhibition plot and further construct the mapping data monitoring prompt word, so that the LLM large language model monitors the prompt word of the Metaverse exhibition mapping data set under each exhibition plot. If a keyword that does not match the mapping data monitoring prompt word appears in a certain exhibition plot, it is determined that the mapping data of the exhibition plot has changed, and the system generates a model update rendering task for the exhibition plot, so that the 3D rendering engine performs model rendering update on the Metaverse exhibition mapping data set under the exhibition plot. Therefore, the present invention performs model rendering update of the corresponding Metaverse exhibition scene for the node that has changed through real-time data mapping monitoring of the plot node, and does not need to update the virtual model data set of the Metaverse exhibition scene as a whole, so it can greatly reduce the difficulty of task scheduling in the background and reduce the computing and operating pressure of the back end.

[0058] The present invention uses intelligent technology to realize the update and switching management of data sets under each plot, allowing the LLM large prediction model to perform real-time data monitoring under each plot node based on prompt words, and automatically generate and submit rendering tasks for plot nodes where data has changed, thereby making the management and control of the metaverse exhibition scene more intelligent. BRIEF DESCRIPTION OF THE DRAWINGS

[0059] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0060] Figure 1 It is a schematic diagram of a method implementation flow provided by an embodiment of the present invention;

[0061] Figure 2 It is a schematic diagram of the overall mapping of the data set of the Metaverse exhibition scene model using traditional technology;

[0062] Figure 3 It is a schematic diagram of using the present invention to map a data set node (plot node) to a metaverse exhibition scene model;

[0063] Figure 4 It is a flow chart of the present invention using the LLM large language model to monitor whether the mapping data set of each plot has changed;

[0064] Figure 5 It is a structural schematic diagram of an electronic device provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0065] The technical solution of the present invention is described below in conjunction with the accompanying drawings.

[0066] In the embodiments of the present invention, words such as "exemplarily" and "for example" are used to indicate examples, illustrations or explanations. Any embodiment or design described as "example" in the present invention should not be interpreted as being more preferred or more advantageous than other embodiments or designs. Specifically, the use of the word "example" is intended to present the concept in a specific way. In addition, in the embodiments of the present invention, the meaning expressed by "and / or" can be both, or it can be either of the two.

[0067] In the embodiments of the present invention, "image" and "picture" can sometimes be used interchangeably. It should be noted that when the difference between them is not emphasized, the meanings they intend to express are the same. "of", "corresponding, relevant" and "corresponding" can sometimes be used interchangeably. It should be noted that when the difference between them is not emphasized, the meanings they intend to express are the same.

[0068] In the embodiments of the present invention, sometimes the subscripts such as W 1 It may be mistakenly written as a non-subscript form such as W1. When the difference is not emphasized, the meanings they express are the same.

[0069] In order to make the technical problems, technical solutions and advantages to be solved by the present invention more clear, a detailed description will be given below with reference to the accompanying drawings and specific embodiments.

[0070] The embodiment of the present invention provides a method for constructing a metaverse exhibition scene based on a 3D rendering engine, which can be implemented by an electronic device, which can be a terminal or a server. Figure 1 The flowchart of the method for constructing a metaverse exhibition scene based on a 3D rendering engine is shown. The processing flow of the method may include the following steps:

[0071] S1. Based on a 3D rendering engine, the entity data of each exhibition plot in the Metaverse exhibition scene is mapped to the virtual model of the Metaverse exhibition scene, and a Metaverse exhibition mapping data set of each exhibition plot is generated and saved;

[0072] S2. Based on NLP technology, identify the mapping relationship of the metaverse exhibition mapping data set under each exhibition plot, and bind the mapping relationship with the corresponding exhibition plot;

[0073] S3, extracting the data mapping keywords in the mapping relationship bound to the exhibition plot, constructing a mapping data monitoring prompt word based on the mapping relationship and the data mapping keywords, and inputting the prompt word into a preset LLM large language model;

[0074] S4, uploading the Metaverse exhibition entity data under each of the exhibition scenarios in real time, and identifying whether there are keywords in the Metaverse exhibition entity data that do not match the mapping data monitoring prompt words through the LLM large language model traversal:

[0075] If it appears, the exhibition plot in which the keywords that do not match the mapping data monitoring prompt words appear is marked as a mapping node, and a model update rendering task of the mapping node is generated;

[0076] Otherwise, continue monitoring;

[0077] S5. Output and send the model update rendering task of the mapping node to the 3D rendering engine, and perform model rendering update on the Metaverse exhibition mapping data set under the mapping node.

[0078] In the present invention, the application platform of the digital twin model, such as the digital twin application system of the digital twin platform, can refer to the application of the existing platform for its deployment and application on the background server.

[0079] ‌Rendering engine selection‌: users can choose the appropriate 3D rendering engine according to project requirements, such as Unity, Unreal Engine, etc. These engines provide a rich API and tool set to facilitate dynamic mapping control. ‌Scene construction‌: build exhibition scenes in the rendering engine, including arranging booths, setting lights, adjusting materials, etc., to ensure the realism and atmosphere of the scene.

[0080] The sensor data of the Metaverse exhibition scene can be sampled by the sensor system in the real scene (such as space cameras, visual systems, etc.). All sensor devices are controlled by the on-site terminal or industrial computer, and the industrial computer communicates with the background server where the platform is located to realize data interaction and communication control. For details, please refer to the virtual reality interaction control principle of the existing virtual server and application server.

[0081] For the steps of constructing the virtual model of the Metaverse exhibition scene, please refer to the description in the background technology.

[0082] The real-time data mapping process of the virtual model can be referred to:

[0083] 1. Real-time rendering is the key to virtual model data mapping, which quickly converts 3D model data into 2D images to ensure that users see a coherent video.

[0084] This process involves complex calculations such as coordinate transformation, lighting, and color processing, which are all efficiently completed by the CPU and GPU.

[0085] 2. Data Mapping Process

[0086] ‌Data Acquisition‌: Acquire dynamic data from external data sources in real time.

[0087] ‌Data Mapping‌: Map the acquired data to the virtual model and update the model status.

[0088] ‌Rendering‌: Through real-time rendering technology, the updated model data is converted into 2D images and presented to the user.

[0089] In order to facilitate the scene interaction control and scene switching of the metaverse exhibition scene and to manage data according to the nodes corresponding to the plot, the present invention processes the scene data set under each plot according to the exhibition plot on the system.

[0090] As attached Figure 2 and Figure 3 As shown: Figure 2 This is a traditional virtual model mapping diagram. Figure 3It is a virtual mapping schematic diagram of the Metaverse exhibition scene model of the present invention. By comparison, it can be seen that: the present application uses AI to identify the nodes where the data set has changed and issues the model update rendering tasks of the corresponding nodes, so that the data set of the scene nodes where the data set has changed can be updated in a targeted manner, and the model nodes with data set changes can be dynamically rendered based on the 3D rendering engine. Therefore, there is no need to process and render the data set of the entire Metaverse exhibition model as a whole, which greatly reduces the computing pressure on the back end, and can update the mapping data set of the model for each plot in the exhibition scene in a targeted manner, and can manage the model data set of the plot nodes according to the scene nodes, reducing the difficulty of processing and managing the Metaverse exhibition scene model.

[0091] As for the various exhibition plot nodes in the metaverse exhibition scene, they can be divided according to the construction process of the metaverse exhibition scene or the component nodes of the metaverse exhibition scene. For example, booth A is divided into a node where an exhibition plot is located. The subsequent model mapping dataset under this node is based on the mapping dataset A of the physical model (booth A) where the node is located in the virtual space. The system where the digital twin is located can generate a virtual model A of booth A based on the mapping dataset A, and control the dynamic display of virtual model A based on the mapping relationship in A.

[0092] Based on the above-mentioned node-based management, AI technology is then used to monitor the mapping data sets of each node to determine which nodes’ mapping data have changed, and then targeted digital twin control is performed.

[0093] Preferably, S1, based on a 3D rendering engine, mapping entity data under each exhibition plot in the Metaverse exhibition scene to a virtual model of the Metaverse exhibition scene, generating and saving a Metaverse exhibition mapping data set under each exhibition plot, including:

[0094] According to the preset disassembly and combination method, the exhibition scene of the metaverse is disassembled into exhibition plots, and storage space corresponding to the exhibition plots is reserved in the background database;

[0095] The physical model dataset of the Metaverse exhibition scene is collected by sensing, and a virtual model of the Metaverse exhibition scene is constructed based on a 3D rendering engine and a corresponding virtual model dataset is generated, and a data mapping relationship between the physical model dataset and the virtual model dataset is established to obtain a Metaverse exhibition mapping dataset of the Metaverse exhibition scene;

[0096] According to the exhibition plot, the metaverse exhibition mapping data set is split into subsets corresponding to different exhibition plots, and the metaverse exhibition mapping data set under each exhibition plot is obtained and saved in the storage space corresponding to the exhibition plot.

[0097] The exhibition plot of the Metaverse exhibition scene can be disassembled in an orderly manner according to the node combination method of the Metaverse exhibition scene, and the Metaverse exhibition scene can be disassembled into the corresponding parts of the exhibition plot, and the dynamic data set (including entity data and the mapping data set generated after the virtual modeling of the part) of each exhibition plot (Metaverse exhibition scene part, that is, node model) is stored in the background database. Through node-based splitting and space reservation, it is convenient to perform data management and scene node management according to the modeling nodes of the Metaverse exhibition scene.

[0098] As for sensing and collecting the physical model dataset of the Metaverse exhibition scene, building a virtual model of the Metaverse exhibition scene based on a 3D rendering engine and generating a corresponding virtual model dataset, and establishing a data mapping relationship between the physical model dataset and the virtual model dataset, you can refer to the previous description process of digital twin modeling, which will not be repeated here.

[0099] Preferably, S2, based on NLP technology, identifying the mapping relationship of the metaverse exhibition mapping data set under each exhibition plot, and binding the mapping relationship with the corresponding exhibition plot, including:

[0100] Reading the Metaverse exhibition mapping data set stored in each of the storage spaces in sequence;

[0101] Parsing and preprocessing the Metaverse exhibition mapping data set to obtain a number of mapping data blocks corresponding to various models under the exhibition plot;

[0102] Using NLP technology to identify entity information, virtual information and model mapping data of each model in the mapping data block, and extracting the mapping relationship between the entity model and the virtual model in the entity information, virtual information and model mapping data based on a preset relationship extraction template;

[0103] The mapping relationship extracted by the output identification is bound to the corresponding exhibition plot.

[0104] The Metaverse exhibition mapping data set under each exhibition plot, that is, the Metaverse exhibition mapping data set of each part / node model, needs to be managed one-to-one. In order to facilitate the LLM large language model to monitor the model mapping data set of each node model, it is necessary to input the mapping keywords of the corresponding node model into the LLM large language model, so that the large language model can monitor and identify the data set changes of the model mapping data set of the node model.

[0105] Here, we first use NLP technology to identify the mapping relationship of the metaverse exhibition mapping data set under each exhibition plot, and bind the mapping relationship with the corresponding exhibition plot, so as to obtain the mapping relationship between the physical model and the virtual model in each node model mapping data and the corresponding entity and other information.

[0106] Because the Metaverse Exhibition Mapping Dataset under each node model contains several sub-models (for example, Booth A is constructed by several sub-models), it is necessary to parse and pre-process the Metaverse Exhibition Mapping Dataset to obtain several mapping data blocks corresponding to each model under the exhibition plot. The mapping data block is the mapping data set of the sub-model in each node model. The node model is composed of several sub-models, and the Metaverse Exhibition Mapping Dataset corresponding to the node model contains the mapping and aggregation of the corresponding sub-model, that is, it contains the corresponding mapping data block, which contains the detailed mapping relationship between the various physical model parts that constitute the node model and the physical model and the virtual model.

[0107] Here, NLP technology is used to identify the entity information, virtual information and model mapping data of each model in the mapping data block: Data set change supervision, NLP technology can be used to identify changes in the data set. This can be achieved by comparing data sets at different time points, for example, using text similarity algorithms (such as cosine similarity, Jaccard similarity, etc.) to measure the difference between two texts. However, technologies such as named entity recognition (NER) can be used to extract key entities in the data set and track the changes of these entities over time.

[0108] The named entity recognition (NER) technology is used to identify key entities in the data set, such as names of people, places, organizations, etc. These entities are usually used as the starting point and end point of the mapping relationship. After identifying the entities, the relationship between the entities needs to be extracted. This can be achieved by analyzing the verbs, prepositions and other words in the text, which usually represent some kind of connection between entities. After identifying the entities and relationships, a mapping relationship graph between the entities can be constructed. This is usually a directed graph in which nodes represent entities and edges represent relationships. The mapping relationship graph can be used for subsequent data analysis, knowledge graph construction and other tasks. The relationship extraction template can be customized by the user, and the relationship extraction conditions / rules in the template can be customized by the user.

[0109] After the mapping relationship is identified, it is bound to the corresponding exhibition plot, so that it is convenient to understand and control the model control of each exhibition plot from the mapping relationship of the node model.

[0110] like Figure 4As shown, preferably, S3, extracting the data mapping keywords in the mapping relationship bound to the exhibition plot, constructing a mapping data monitoring prompt word based on the mapping relationship and the data mapping keyword, and inputting it into a preset LLM large language model, including:

[0111] Using NLP technology, extracting the data mapping keywords in the mapping relationship of each of the exhibition plots;

[0112] Based on the mapping logic in the mapping relationship and the data mapping keywords corresponding to the exhibition plot, construct the mapping data monitoring prompt words for training the LLM large language model;

[0113] The mapping data monitoring prompt words are input into the LLM large language model to perform model learning.

[0114] Here, the LLM large language model is used to monitor the changes of the mapping data sets of each node model based on the prompt words related to mapping monitoring, so as to identify the scene nodes (node ​​models) with data changes, and generate the model update rendering tasks corresponding to the node models with data changes and hand them over to the 3D rendering engine, and perform model rendering update on the node models with data changes. Based on the real-time uploaded entity model data of the node model, the data rendering update of the original mapping data set of the node model is performed to realize the rendering update of the corresponding node model, and there is no need to render other node models in vain, thereby improving rendering efficiency.

[0115] Specifically, first, the LLM large language model needs to be trained to let the large language model learn the mapping data keywords and mapping logic of each scene plot, so it is necessary to build prompt words based on keywords and logic. In this way, after the LLM large language model is deployed, the large language model can supervise whether the mapping data set of each node model triggers the prompt word. NLP technology is used here to extract the data mapping keywords in the mapping relationship of each exhibition plot. For details, see the NLP extraction entity and other processes.

[0116] Administrators can combine the mapping keywords and mapping logic of each node model into prompt words in advance and input them into the LLM large language model for model training and learning. Collect data from multiple groups of models in advance and train them. When the training meets the standards, the LLM large language model can be put into use. Here, you can directly log in to the third party through the API, call the third party's LLM large prediction model to perform model training and learning, and call the model for subsequent data set monitoring and recognition work.

[0117] The LLM large language model can be a user-defined or third-party model, such as Baidu's Wenxin Yiyan.

[0118] After the large language model has been trained and learned with the prompt words of different node models, it can be used to monitor the dynamic changes of the dataset of the Metaverse exhibition scene model. If the entity dataset of that node model changes later, its large language model reads the dataset uploaded in real time and identifies the mapping keywords and mapping logic therein, so as to compare it with the information in the learned prompt words. If inconsistency is found, it is determined that the entity model of the corresponding scene node (node ​​model) has changed. At this time, the corresponding judgment result needs to be output: that is, it is identified that the keywords that do not match the mapping data monitoring prompt words appear in the Metaverse exhibition entity data. At this time, the large language model outputs the recognition result of the node model, and the system sends the model update rendering task of the corresponding node model to the 3D rendering engine. The task includes the real-time reported dataset of the node model (it can even include the location of the dataset that has changed identified by the large language model, etc., so that the 3D rendering engine can replace and change the initial mapping dataset of the node model based on the real-time dataset according to the data change, and dye it into the corresponding model rendering update).

[0119] The 3D rendering engine performs model rendering and updating on the Metaverse exhibition mapping data set under the mapping node. This can be understood in conjunction with the rendering process of mapping the physical model to the virtual model, and will not be described in detail here.

[0120] Therefore, the present invention can perform plot-based rendering updates and scene switching control according to the scene nodes of the metaverse exhibition scene, thereby realizing intelligent scene management and creation control.

[0121] Preferably, in step S4, the real-time uploading of the Metaverse exhibition entity data under each exhibition scenario includes:

[0122] Preset sampling frequency;

[0123] Through the digital twin application system, a sampling instruction is issued to the sensor system corresponding to the exhibition plot in the physical exhibition scene according to the sampling frequency, and the metaverse exhibition entity data of the exhibition plot is collected and uploaded to the background database;

[0124] The metaverse exhibition entity data is stored in the storage space corresponding to the exhibition plot.

[0125] The system can control the sensor system to sample the physical model data of each node model according to the preset sampling frequency, and report it to the system and store it in the corresponding storage space after being processed by the on-site industrial computer or terminal, so as to facilitate the orderly recognition of the large language model and monitor whether the data set of each node has changed, avoiding redundant fusion of data sets and affecting data processing efficiency.

[0126] On the other hand, a device for constructing a metaverse exhibition scene based on a 3D rendering engine is provided, and the device for constructing a metaverse exhibition scene based on a 3D rendering engine is used to implement the above-mentioned method for constructing a metaverse exhibition scene based on a 3D rendering engine, and the device includes:

[0127] The digital twin system is used to map the entity data of each exhibition plot in the Metaverse exhibition scene to the virtual model of the Metaverse exhibition scene based on the 3D rendering engine, generate and save the Metaverse exhibition mapping data set under each exhibition plot;

[0128] A mapping processing module, used to identify the mapping relationship of the metaverse exhibition mapping data set under each exhibition plot based on NLP technology, and bind the mapping relationship with the corresponding exhibition plot;

[0129] A prompt word generation module, used to extract the data mapping keywords in the mapping relationship bound to the exhibition plot, and based on the mapping relationship and the data mapping keywords, construct a mapping data monitoring prompt word and input it into a preset LLM large language model;

[0130] The intelligent recognition module is used to upload the Metaverse exhibition entity data under each of the exhibition plots in real time, and identify whether there are keywords in the Metaverse exhibition entity data that do not match the mapping data monitoring prompt words through the LLM large language model traversal:

[0131] If it appears, the exhibition plot in which the keywords that do not match the mapping data monitoring prompt words appear is marked as a mapping node, and a model update rendering task of the mapping node is generated;

[0132] Otherwise, continue monitoring;

[0133] The rendering task scheduling module is used to output and send the model update rendering task of the mapping node to the 3D rendering engine, and perform model rendering update on the metaverse exhibition mapping data set under the mapping node.

[0134] Please understand and implement the functions and interaction processes of the digital twin system and each module in the above virtual device in combination with the above method steps, which will not be repeated here.

[0135] Figure 5 is a schematic diagram of the structure of an electronic device provided by an embodiment of the present invention, such as Figure 5 As shown, the electronic device 410 may include a first processor 2001 .

[0136] Optionally, the electronic device 410 may further include a memory 2002 and a transceiver 2003 .

[0137] The first processor 2001, the memory 2002 and the transceiver 2003 may be connected via a communication bus.

[0138] Combine the following Figure 5 The components of the electronic device 410 are described in detail:

[0139] The first processor 2001 is the control center of the electronic device 410, and may be a processor or a general term for multiple processing elements. For example, the first processor 2001 is one or more central processing units (CPUs), or may be application specific integrated circuits (ASICs), or may be one or more integrated circuits configured to implement the embodiments of the present invention, such as one or more microprocessors (digital signal processors, DSPs), or one or more field programmable gate arrays (field programmable gate arrays, FPGAs).

[0140] Optionally, the first processor 2001 can execute various functions of the electronic device 410 by running or executing a software program stored in the memory 2002 and calling data stored in the memory 2002 .

[0141] In a specific implementation, as an embodiment, the first processor 2001 may include one or more CPUs, such as Figure 5 CPU0 and CPU1 are shown in FIG.

[0142] In a specific implementation, as an embodiment, the electronic device 410 may also include multiple processors, such as Figure 5 The first processor 2001 and the second processor 2004 are shown in FIG. Each of these processors may be a single-core processor (single-CPU) or a multi-core processor (multi-CPU). The processor here may refer to one or more devices, circuits, and / or processing cores for processing data (e.g., computer program instructions).

[0143] The memory 2002 is used to store the software program for executing the solution of the present invention, and is controlled to be executed by the first processor 2001. The specific implementation method can refer to the above method embodiment, which will not be repeated here.

[0144] Optionally, the memory 2002 may be a read-only memory (ROM) or other types of static storage devices that can store static information and instructions, a random access memory (RAM) or other types of dynamic storage devices that can store information and instructions, or an electrically erasable programmable read-only memory (EEPROM), a compact disc read-only memory (CD-ROM) or other optical disc storage, optical disc storage (including compressed optical disc, laser disc, optical disc, digital versatile disc, Blu-ray disc, etc.), a magnetic disk storage medium or other magnetic storage device, or any other medium that can be used to carry or store the desired program code in the form of instructions or data structures and can be accessed by a computer, but is not limited thereto. The memory 2002 may be integrated with the first processor 2001, or may exist independently and access the first processor 2001 through the interface circuit ( Figure 5 (not shown) is coupled to the first processor 2001, which is not specifically limited in this embodiment of the present invention.

[0145] The transceiver 2003 is used to communicate with a network device or a terminal device.

[0146] Optionally, the transceiver 2003 may include a receiver and a transmitter ( Figure 5 The receiver is used to implement a receiving function, and the transmitter is used to implement a sending function.

[0147] Optionally, the transceiver 2003 may be integrated with the first processor 2001, or may exist independently and communicate with the first processor 2001 through the interface circuit ( Figure 5 (not shown) is coupled to the first processor 2001, which is not specifically limited in this embodiment of the present invention.

[0148] It should be noted that Figure 5 The structure of the electronic device 410 shown in the figure does not constitute a limitation on the router, and the actual knowledge structure recognition device may include more or fewer components than those shown in the figure, or combine certain components, or arrange the components differently.

[0149] In addition, the technical effects of the electronic device 410 can refer to the technical effects of the method for constructing a metaverse exhibition scene based on a 3D rendering engine described in the above method embodiment, and will not be repeated here.

[0150] It should be understood that the first processor 2001 in the embodiment of the present invention may be a central processing unit (CPU), and the processor may also be other general-purpose processors, digital signal processors (DSP), application specific integrated circuits (ASIC), field programmable gate arrays (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor may be a microprocessor or the processor may also be any conventional processor, etc.

[0151] It should also be understood that the memory in the embodiments of the present invention may be a volatile memory or a non-volatile memory, or may include both volatile and non-volatile memories. Among them, the non-volatile memory may be a read-only memory (ROM), a programmable read-only memory (PROM), an erasable programmable read-only memory (EPROM), an electrically erasable programmable read-only memory (EEPROM), or a flash memory. The volatile memory may be a random access memory (RAM), which is used as an external cache. By way of example and not limitation, many forms of random access memory (RAM) are available, such as static RAM (SRAM), dynamic random access memory (DRAM), synchronous DRAM (SDRAM), double data rate synchronous dynamic random access memory (DDR SDRAM), enhanced synchronous dynamic random access memory (ESDRAM), synchronous link DRAM (SLDRAM), and direct rambus RAM (DR RAM).

[0152] The above embodiments can be implemented in whole or in part by software, hardware (such as circuits), firmware or any other combination. When implemented by software, the above embodiments can be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions or computer programs. When the computer instructions or computer programs are loaded or executed on a computer, the process or function described in the embodiment of the present invention is generated in whole or in part. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium, or transmitted from one computer-readable storage medium to another computer-readable storage medium. For example, the computer instructions can be transmitted from one website, computer, server or data center to another website, computer, server or data center by wired (such as infrared, wireless, microwave, etc.). The computer-readable storage medium can be any available medium that can be accessed by a computer or a data storage device such as a server or data center that contains one or more available media sets. The available medium can be a magnetic medium (for example, a floppy disk, a hard disk, a tape), an optical medium (for example, a DVD), or a semiconductor medium. The semiconductor medium can be a solid-state hard disk.

[0153] It should be understood that the term "and / or" in this article is only a description of the association relationship of associated objects, indicating that there can be three relationships. For example, A and / or B can represent: A exists alone, A and B exist at the same time, and B exists alone. A and B can be singular or plural. In addition, the character " / " in this article generally indicates that the associated objects before and after are in an "or" relationship, but it may also indicate an "and / or" relationship. Please refer to the context for specific understanding.

[0154] In the present invention, "at least one" means one or more, and "plurality" means two or more. "At least one of the following" or similar expressions refers to any combination of these items, including any combination of single items or plural items. For example, at least one of a, b, or c can mean: a, b, c, ab, ac, bc, or abc, where a, b, and c can be single or multiple.

[0155] It should be understood that in various embodiments of the present invention, the size of the serial numbers of the above-mentioned processes does not mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present invention.

[0156] Those skilled in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of the present invention.

[0157] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the above-described equipment, devices and units can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0158] In the several embodiments provided by the present invention, it should be understood that the disclosed devices, apparatuses and methods can be implemented in other ways. For example, the device embodiments described above are only schematic. For example, the division of the units is only a logical function division. There may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another device, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms.

[0159] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0160] In addition, each functional unit in each embodiment of the present invention may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit.

[0161] If the functions are implemented in the form of software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, or the part of the technical solution, can be embodied in the form of a software product. The computer software product is stored in a storage medium, including several instructions for enabling a computer device (which can be a personal computer, server, or network device, etc.) to perform all or part of the steps of the methods described in various embodiments of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM), random access memory (RAM), disk or optical disk, and other media that can store program codes.

[0162] The above is only a specific embodiment of the present invention, but the protection scope of the present invention is not limited thereto. Any person skilled in the art can easily think of changes or substitutions within the technical scope disclosed by the present invention, which should be included in the protection scope of the present invention. Therefore, the protection scope of the present invention should be based on the protection scope of the claims.

Claims

1. A method for constructing a metaverse exhibition scene based on a 3D rendering engine, characterized in that: The method comprises: S1. Based on a 3D rendering engine, the entity data of each exhibition plot in the Metaverse exhibition scene is mapped to the virtual model of the Metaverse exhibition scene, and a Metaverse exhibition mapping data set of each exhibition plot is generated and saved; S2. Based on NLP technology, identifying the mapping relationship of the metaverse exhibition mapping data set under each exhibition plot, and binding the mapping relationship with the corresponding exhibition plot, including: Reading the Metaverse exhibition mapping data set stored in each storage space in sequence; Parsing and preprocessing the Metaverse exhibition mapping data set to obtain a number of mapping data blocks corresponding to various models under the exhibition plot; Using NLP technology to identify the entity information, virtual information and model mapping data of each model in the mapping data block, and based on a preset relationship extraction template, extract the mapping relationship between the entity model and the virtual model in the entity information, virtual information and model mapping data and bind it with the corresponding exhibition plot; S3, extracting the data mapping keywords in the mapping relationship bound to the exhibition plot, and constructing mapping data monitoring prompt words based on the mapping relationship and the data mapping keywords and inputting them into a preset LLM large language model; S4, uploading the Metaverse exhibition entity data under each of the exhibition scenarios in real time, and identifying whether there are keywords in the Metaverse exhibition entity data that do not match the mapping data monitoring prompt words through the LLM large language model traversal: If it appears, the exhibition plot in which the keywords that do not match the mapping data monitoring prompt words appear is marked as a mapping node, and a model update rendering task of the mapping node is generated; Named Entity Recognition (NER) technology is used to extract key entities in the data set. By analyzing the verbs and prepositions in the text and tracking the changes of these entities over time, after identifying the entities and relationships, a mapping relationship diagram between the entities is constructed. The LLM large language model is used to monitor the changes of the mapping data set of each node model based on the prompt words related to mapping monitoring, so as to identify the scene nodes with data changes, and generate the model update rendering task corresponding to the node model with data changes and hand it over to the 3D rendering engine, and perform model rendering update on the node model with data changes. Based on the real-time uploaded entity model data of the node model, the original mapping data set of the node model is updated to achieve the rendering update of the corresponding node model. Otherwise, continue monitoring; S5. Output and send the model update rendering task of the mapping node to the 3D rendering engine, and perform model rendering update on the Metaverse exhibition mapping data set under the mapping node.

2. The method for constructing a metaverse exhibition scene based on a 3D rendering engine according to claim 1, characterized in that: S1. Based on a 3D rendering engine, the entity data of each exhibition plot in the Metaverse exhibition scene is mapped to the virtual model of the Metaverse exhibition scene, and a Metaverse exhibition mapping data set of each exhibition plot is generated and saved, including: According to the preset disassembly and combination method, the exhibition scene of the metaverse is disassembled into exhibition plots, and storage space corresponding to the exhibition plots is reserved in the background database; The physical model dataset of the Metaverse exhibition scene is collected by sensing, and a virtual model of the Metaverse exhibition scene is constructed based on a 3D rendering engine and a corresponding virtual model dataset is generated, and a data mapping relationship between the physical model dataset and the virtual model dataset is established to obtain a Metaverse exhibition mapping dataset of the Metaverse exhibition scene; According to the exhibition plot, the metaverse exhibition mapping data set is split into subsets corresponding to different exhibition plots, and the metaverse exhibition mapping data set under each exhibition plot is obtained and saved in the storage space corresponding to the exhibition plot.

3. The method for constructing a metaverse exhibition scene based on a 3D rendering engine according to claim 1, characterized in that: S3, extracting the data mapping keywords in the mapping relationship bound to the exhibition plot, and constructing the mapping data monitoring prompt words based on the mapping relationship and the data mapping keywords and inputting them into the preset LLM large language model, including: Using NLP technology, extracting the data mapping keywords in the mapping relationship of each of the exhibition plots; Based on the mapping logic in the mapping relationship and the data mapping keywords corresponding to the exhibition plot, construct the mapping data monitoring prompt words for training the LLM large language model; The mapping data monitoring prompt words are input into the LLM large language model to perform model learning.

4. The method for constructing a metaverse exhibition scene based on a 3D rendering engine according to claim 3 is characterized in that: In step S4, the real-time uploading of the Metaverse exhibition entity data under each exhibition scenario includes: Preset sampling frequency; Through the digital twin application system, a sampling instruction is issued to the sensor system corresponding to the exhibition plot in the physical exhibition scene according to the sampling frequency, and the metaverse exhibition entity data of the exhibition plot is collected and uploaded to the background database; The metaverse exhibition entity data is stored in the storage space corresponding to the exhibition plot.

5. A device for constructing a metaverse exhibition scene based on a 3D rendering engine, wherein the device for constructing a metaverse exhibition scene based on a 3D rendering engine is used to implement a method for constructing a metaverse exhibition scene based on a 3D rendering engine as described in any one of claims 1 to 4, characterized in that: The device comprises: The digital twin system is used to map the entity data of each exhibition plot in the Metaverse exhibition scene to the virtual model of the Metaverse exhibition scene based on the 3D rendering engine, generate and save the Metaverse exhibition mapping data set under each exhibition plot; A mapping processing module, for identifying the mapping relationship of the metaverse exhibition mapping data set under each of the exhibition plots based on NLP technology, and binding the mapping relationship with the corresponding exhibition plot, including: sequentially reading the metaverse exhibition mapping data set stored in each of the storage spaces; performing data analysis and preprocessing on the metaverse exhibition mapping data set to obtain a number of mapping data blocks corresponding to each model under the exhibition plot; using NLP technology to identify the entity information, virtual information and model mapping data of each model in the mapping data block, and based on a preset relationship extraction template, extracting the mapping relationship between the entity model and the virtual model in the entity information, virtual information and model mapping data and binding it with the corresponding exhibition plot; A prompt word generation module, used to extract the data mapping keywords in the mapping relationship bound to the exhibition plot, and based on the mapping relationship and the data mapping keywords, construct a mapping data monitoring prompt word and input it into a preset LLM large language model; The intelligent recognition module is used to upload the Metaverse exhibition entity data under each of the exhibition plots in real time, and identify whether there are keywords in the Metaverse exhibition entity data that do not match the mapping data monitoring prompt words through the LLM large language model traversal: If it appears, the exhibition plot in which the keywords that do not match the mapping data monitoring prompt words appear is marked as a mapping node, and a model update rendering task of the mapping node is generated; Otherwise, continue monitoring; The rendering task scheduling module is used to output and send the model update rendering task of the mapping node to the 3D rendering engine, and perform model rendering update on the metaverse exhibition mapping data set under the mapping node.

6. An electronic device, characterized in that: The electronic device comprises: processor; A memory having computer-readable instructions stored thereon, wherein when the computer-readable instructions are executed by the processor, the method according to any one of claims 1 to 4 is implemented.

7. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores program codes, which can be called by a processor to execute the method according to any one of claims 1 to 4.

Images