Three-dimensional scene editing method, device, equipment and storage medium

Abstract

The present invention relates to the technical field of three-dimensional scene editing, and discloses a three-dimensional scene editing method, device, equipment and storage medium, which converts the natural language editing instructions input by the target editing user into step instructions and operation instructions, extracts parameter index characterization information, generates operation parameters corresponding to the operation instructions under the step instructions based on the operation parameter standard model of the target editing user, and finally constructs a three-dimensional scene based on the operation instructions and operation parameters corresponding to each step instruction. The present invention uses a large language model to replace the three-dimensional editor with a three-dimensional scene editing framework based on natural language control, converts the natural language editing instructions input by the user into the operation parameters of the corresponding operation instructions under each step instruction, introduces the operation parameter standard model to record the parameter index characterization habit characteristics of each user, improves the efficiency and flexibility of three-dimensional scene editing, and reduces the threshold of three-dimensional scene editing operations.

Description

Technical Field

[0001] The present invention relates to the technical field of three-dimensional scene editing, and in particular to a three-dimensional scene editing method, device, equipment and storage medium. Background Art

[0002] Common 3D scene editors mainly include the following: (1) View navigation and operation: Mouse control: Use the mouse to operate the view rotation, translation and zoom. Usually, the right button rotates the view, the middle button translates the view, and the wheel zooms the view. Shortcut keys: Common shortcut keys include holding down the Alt key and dragging the mouse to rotate, holding down the Ctrl key and dragging the mouse to translate, and the wheel zooming. (2) Object selection and editing: Selection tools: Usually there are multiple selection tools, such as box selection and single selection, which are used to select objects in the scene. Transformation tools: Used to move, rotate and scale the selected object, which can be achieved through the axis controller or directly in the view. (3) Scene organization and hierarchy: Hierarchy panel: Displays the hierarchy of all objects in the scene. The hierarchy panel can be used to manage the parent-child relationship of objects, show / hide objects, etc. Grouping and naming: Allows users to group multiple objects and name objects and components for better management and identification. (4) Material and texture editing: Material editor: Allows users to create, edit and apply materials, and adjust properties such as color, mapping, and reflectivity. UV mapping: used to map textures onto the surface of three-dimensional objects, and adjust the position and direction of textures by editing UV layout. (5) Real-time preview and rendering: Real-time view: provides real-time preview of scene effects, including lighting, shadows, and material reflections. Rendering settings: allows users to adjust rendering parameters such as resolution, lighting model, and post-processing effects to generate final high-quality images or animations.

[0003] The above-mentioned interaction methods and tools enable users to perform complex three-dimensional design and creation work in the scene editor, improving work efficiency and creative flexibility. However, in actual applications, there are the following objective defects: the scene editor-based method has high requirements for technical personnel and a high learning cost. Practitioners need to undergo professional learning and training to master the scene editing function. When using the existing scene editor for scene editing, it is often necessary to adjust more parameters to achieve the desired effect. This also requires practitioners to have strong professional knowledge and proficiency to master the role of all parameters and set more appropriate parameter properties. Different three-dimensional platform software has different editors and control methods. Different platform terminals of the same software also have different interaction methods. Practitioners need to learn and use new interaction forms and understand new parameters when switching between different software or platforms.

[0004] Therefore, how to improve the efficiency and flexibility of 3D scene editing and lower the threshold of 3D scene editing operations is a technical problem that needs to be solved urgently. Summary of the invention

[0005] The present invention provides a three-dimensional scene editing method, device, equipment and storage medium, aiming to solve at least one of the above technical problems.

[0006] To achieve the above object, the present invention provides a three-dimensional scene editing method, comprising the following steps:

[0007] Obtaining a natural language editing instruction input by a target editing user for a target three-dimensional scene construction task, and converting the natural language editing instruction into a step instruction set including a plurality of step instructions;

[0008] Based on each step instruction in the step instruction set, calling a pre-stored interactive operation set to determine an operation instruction corresponding to each step instruction;

[0009] Extracting parameter indicator representation information associated with each step instruction in the natural language editing instruction, and using the parameter indicator representation information to generate operation parameters corresponding to the operation instructions under the step instructions with the associated relationship based on the operation parameter standard model of the target editing user;

[0010] Based on the operation instructions and operation parameters corresponding to each step instruction in the step instruction set, a three-dimensional scene editing operation instruction set is generated and a three-dimensional scene is constructed. The natural language editing instructions supplemented by the target editing user are obtained to correct the constructed three-dimensional scene.

[0011] Optionally, obtaining a natural language editing instruction input by a target editing user for a target three-dimensional scene construction task, and converting the natural language editing instruction into a step instruction set including a plurality of step instructions, specifically including:

[0012] Upon receiving a 3D scene construction request for a target 3D scene construction task transmitted by a target editing user, extracting a user identifier of the target editing user in the 3D scene construction request and building a 3D scene construction environment, and after the 3D scene construction environment is built, feeding back input guidance information to the target editing user;

[0013] The natural language editing instructions input by the target editing user in response to the input guide information are obtained, and the natural language editing instructions are converted into a step instruction set including a plurality of step instructions for the three-dimensional scene construction environment.

[0014] Optionally, obtaining a natural language editing instruction input by a target editing user in response to the input guide information, and converting the natural language editing instruction into a step instruction set including a plurality of step instructions for the three-dimensional scene construction environment, specifically includes:

[0015] Acquire a natural language editing instruction input by a target editing user in response to the input guide information, and use a large language model based on natural language processing to perform word segmentation, part-of-speech analysis, and operation-related vocabulary extraction on the natural language editing instruction to obtain an operation-related vocabulary set;

[0016] Based on the pre-stored mapping relationship table between operation-related words and three-dimensional scene construction steps, the three-dimensional scene construction steps corresponding to each operation-related word in the operation-related word set are matched to construct a step instruction set containing several step instructions for the three-dimensional scene construction environment.

[0017] Optionally, based on each step instruction in the step instruction set, calling a pre-stored interactive operation set to determine an operation instruction step corresponding to each step instruction specifically includes:

[0018] Calling the interactive operation set pre-stored in the three-dimensional scene construction environment, extracting a plurality of interactive operations recorded in the interactive operation set and the interactive effect of each interactive operation;

[0019] By using a large language model based on causal reasoning, a causal relationship is judged between each step instruction in the step instruction set and the interaction effect of each interactive operation, and the interactive operation with a causal relationship is used as the operation instruction of the corresponding step instruction.

[0020] Optionally, extracting parameter indicator representation information associated with each step instruction in the natural language editing instruction, and using the parameter indicator representation information to generate an operation parameter step corresponding to the operation instruction under the step instruction with the associated relationship based on the operation parameter standard model of the target editing user, specifically includes:

[0021] Using a large language model based on semantic analysis, the natural language editing instruction is segmented, semantically analyzed, parameter indicator representation vocabulary is extracted, and operation-related words are matched to construct parameter indicator representation information; wherein the parameter indicator representation information includes a plurality of parameter indicator representation vocabulary having a matching relationship with the operation-related vocabulary;

[0022] Each parameter indicator representation word and the operation-related words with matching relationships are input into the target editing user's operation parameter standard model to generate operation parameters corresponding to the operation instructions under the operation instructions determined by the operation-related words with matching relationships.

[0023] Optionally, based on the operation instructions and operation parameters corresponding to each step instruction in the step instruction set, a 3D scene editing operation instruction set is generated and a 3D scene is constructed, a natural language editing instruction supplemented by a target editing user is obtained, and the constructed 3D scene is corrected, specifically including:

[0024] Based on the operation instructions and operation parameters corresponding to each step instruction in the step instruction set, a 3D scene editing operation instruction set is generated, and the 3D scene editing operation instruction is input into a 3D scene construction environment to construct a 3D scene for obtaining a target 3D scene construction task;

[0025] Obtain the natural language editing instructions supplemented by the target editing user for the constructed three-dimensional scene, convert the supplemented natural language editing instructions into operation instructions and operation parameters corresponding to each step instruction to modify the constructed three-dimensional scene.

[0026] Optionally, before generating the operation parameter step corresponding to the operation instruction under the step instruction with the associated relationship based on the operation parameter standard model of the target editing user by using the parameter indicator characterization information, the method further includes:

[0027] Constructing an initial standard model of operating parameters; wherein the initial standard model of operating parameters stores standard operating parameters corresponding to operating instructions of each parameter indicator characterization vocabulary;

[0028] Obtain the natural language editing instructions input and supplemented by the target editing user for each historical three-dimensional scene construction task, extract the operation parameters and corresponding parameter indicator representation vocabulary with the same operation instructions in the natural language editing instructions input and supplemented in the same historical three-dimensional scene construction task, and construct them as a correction sample of the operation parameter standard model;

[0029] According to the operation parameters and corresponding parameter index representation words of the same operation instructions in the correction samples of several operation parameter standard models in each correction cycle, the operation parameters of the operation instructions corresponding to the parameter index representation words in the supplemented input natural language editing instructions are used as correction values ​​to correct the operation parameters of the operation instructions corresponding to the parameter index representation words in the input natural language editing instructions;

[0030] The operation parameter standard model is updated by using each modified parameter indicator to represent the operation parameter of the operation instruction corresponding to the vocabulary, and the updated operation parameter standard model is associated and stored with the user identification of the target editing user;

[0031] The operation parameters of the operation instructions corresponding to the parameter indicator representation vocabulary in the input natural language editing instruction are the operation parameters of the operation instructions corresponding to each parameter indicator representation vocabulary in the operation parameter standard model updated and obtained in the previous correction cycle;

[0032] Among them, the operating parameters of the operating instructions corresponding to the vocabulary represented by the parameter indicators in the natural language editing instructions input in the first correction cycle are the standard operating parameters of the operating instructions corresponding to each vocabulary represented by the parameter indicators in the initial operating parameter standard model.

[0033] In addition, in order to achieve the above-mentioned purpose, the present invention also provides a three-dimensional scene editing device, comprising:

[0034] An acquisition module is used to acquire a natural language editing instruction input by a target editing user for a target three-dimensional scene construction task, and convert the natural language editing instruction into a step instruction set including a plurality of step instructions;

[0035] A determination module, configured to call a pre-stored interactive operation set based on each step instruction in the step instruction set, and determine an operation instruction corresponding to each step instruction;

[0036] An extraction module is used to extract parameter indicator representation information associated with each step instruction in the natural language editing instruction, and generate operation parameters corresponding to the operation instructions under the step instructions with the associated relationship based on the operation parameter standard model of the target editing user by using the parameter indicator representation information;

[0037] The generation module is used to generate a three-dimensional scene editing operation instruction set and construct a three-dimensional scene based on the operation instructions and operation parameters corresponding to each step instruction in the step instruction set, obtain the natural language editing instructions supplemented by the target editing user, and correct the constructed three-dimensional scene.

[0038] In addition, in order to achieve the above-mentioned purpose, the present invention also provides a three-dimensional scene editing device, which includes: a memory, a processor, and a three-dimensional scene editing program stored in the memory and executable on the processor, and the three-dimensional scene editing program implements the steps of the three-dimensional scene editing method described above when executed by the processor.

[0039] In addition, in order to achieve the above-mentioned purpose, the present invention also provides a storage medium, on which a 3D scene editing program is stored, and when the 3D scene editing program is executed by a processor, the steps of the above-mentioned 3D scene editing method are implemented.

[0040] The beneficial effects of the present invention are as follows: a three-dimensional scene editing method, device, equipment and storage medium are proposed, by obtaining the natural language editing instructions input by the target editing user, converting the natural language editing instructions into step instructions and operation instructions, and extracting parameter index characterization information, using the parameter index characterization information, based on the target editing user's operation parameter standard model, generating the operation parameters corresponding to the operation instructions under the step instructions with the associated relationship, and finally generating a three-dimensional scene editing operation instruction set based on the operation instructions and operation parameters corresponding to each step instruction in the step instruction set and constructing a three-dimensional scene. The present invention uses a large language model to replace the three-dimensional editor with a three-dimensional scene editing framework based on natural language control, converts the natural language editing instructions input by the user into the operation parameters of the corresponding operation instructions under each step instruction, introduces the operation parameter standard model to record the parameter index characterization habit characteristics of each user, and by periodically optimizing the operation parameter standard model, improves the adaptability of the parameter index characterization habit characteristics and the user, improves the efficiency and flexibility of three-dimensional scene editing, and reduces the threshold of three-dimensional scene editing operations. BRIEF DESCRIPTION OF THE DRAWINGS

[0041] Figure 1 A schematic diagram of the device structure of the hardware operating environment involved in the embodiment of the present invention;

[0042] Figure 2 A schematic diagram of a flow chart of an embodiment of a three-dimensional scene editing method of the present invention;

[0043] Figure 3 It is a structural block diagram of a three-dimensional scene editing device in an embodiment of the present invention.

[0044] The realization of the purpose, functional features and advantages of the present invention will be further explained in conjunction with embodiments and with reference to the accompanying drawings. DETAILED DESCRIPTION

[0045] In order to make the purpose, technical solution and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0046] like Figure 1 As shown, Figure 1 It is a schematic diagram of the device structure of the hardware operating environment involved in the embodiment of the present invention.

[0047] like Figure 1As shown, the device may include: a processor 1001, such as a CPU, a communication bus 1002, a user interface 1003, a network interface 1004, and a memory 1005. Among them, the communication bus 1002 is used to realize the connection and communication between these components. The user interface 1003 may include a display screen (Display), an input unit such as a keyboard (Keyboard), and the optional user interface 1003 may also include a standard wired interface and a wireless interface. The network interface 1004 may optionally include a standard wired interface and a wireless interface (such as a WI-FI interface). The memory 1005 may be a high-speed RAM memory, or it may be a stable memory (non-volatile memory), such as a disk memory. The memory 1005 may also be a storage device independent of the aforementioned processor 1001.

[0048] Those skilled in the art will understand that Figure 1 The structure of the device shown in the figure does not constitute a limitation of the device, and may include more or less components than shown in the figure, or combine certain components, or arrange the components differently.

[0049] like Figure 1 As shown, the memory 1005 as a computer storage medium may include an operating system, a network communication module, a user interface module, and a three-dimensional scene editing program.

[0050] exist Figure 1 In the terminal shown, the network interface 1004 is mainly used to connect to the backend server and perform data communication with the backend server; the user interface 1003 is mainly used to connect to the client (user end) and perform data communication with the client; and the processor 1001 can be used to call the three-dimensional scene editing program stored in the memory 1005 and perform the following operations:

[0051] Obtaining a natural language editing instruction input by a target editing user for a target three-dimensional scene construction task, and converting the natural language editing instruction into a step instruction set including a plurality of step instructions;

[0052] Based on each step instruction in the step instruction set, calling a pre-stored interactive operation set to determine an operation instruction corresponding to each step instruction;

[0053] Extracting parameter indicator representation information associated with each step instruction in the natural language editing instruction, and using the parameter indicator representation information to generate operation parameters corresponding to the operation instructions under the step instructions with the associated relationship based on the operation parameter standard model of the target editing user;

[0054] Based on the operation instructions and operation parameters corresponding to each step instruction in the step instruction set, a three-dimensional scene editing operation instruction set is generated and a three-dimensional scene is constructed. The natural language editing instructions supplemented by the target editing user are obtained, and the constructed three-dimensional scene is corrected.

[0055] The specific embodiments of the present invention applied to the device are basically the same as the embodiments of the three-dimensional scene editing method described below, and will not be described in detail here.

[0056] The embodiment of the present invention provides a three-dimensional scene editing method, referring to Figure 2 , Figure 2 The figure is a flow chart of an embodiment of a three-dimensional scene editing method of the present invention.

[0057] In this embodiment, the three-dimensional scene editing method includes the following steps:

[0058] S100: Obtaining a natural language editing instruction input by a target editing user for a target three-dimensional scene construction task, and converting the natural language editing instruction into a step instruction set including a plurality of step instructions;

[0059] S200: Based on each step instruction in the step instruction set, calling a pre-stored interactive operation set to determine an operation instruction corresponding to each step instruction;

[0060] S300: extracting parameter indicator representation information associated with each step instruction in the natural language editing instruction, and using the parameter indicator representation information to generate operation parameters corresponding to the operation instructions under the step instructions with the associated relationship based on the operation parameter standard model of the target editing user;

[0061] S400: Based on the operation instructions and operation parameters corresponding to each step instruction in the step instruction set, a three-dimensional scene editing operation instruction set is generated and a three-dimensional scene is constructed, the natural language editing instructions supplemented by the target editing user are obtained, and the constructed three-dimensional scene is corrected.

[0062] It should be noted that the existing interactive methods and tools for 3D design and creation have the following objective defects in practical applications: the scene editor-based method has high requirements for technical personnel and a high learning cost. Practitioners need to undergo professional learning and training to master the scene editing function. When using the existing scene editor for scene editing, it is often necessary to adjust more parameters to achieve the desired effect. This also requires practitioners to have strong professional knowledge and proficiency to master the role of all parameters and set more appropriate parameter properties. Different 3D platform software has different editors and control methods, and different platform terminals of the same software also have different interactive methods. Practitioners need to learn and use new interactive forms and understand new parameters when switching between different software or platforms.

[0063] In order to solve the above problems, this embodiment obtains the natural language editing instructions input by the target editing user, converts the natural language editing instructions into step instructions and operation instructions, and extracts parameter index characterization information. Using the parameter index characterization information, based on the target editing user's operation parameter standard model, the operation parameters corresponding to the operation instructions under the step instructions with an associated relationship are generated, and finally based on the operation instructions and operation parameters corresponding to each step instruction in the step instruction set, a three-dimensional scene editing operation instruction set is generated and a three-dimensional scene is constructed. The present invention uses a large language model to replace the three-dimensional editor with a three-dimensional scene editing framework based on natural language control, converts the natural language editing instructions input by the user into the operation parameters of the corresponding operation instructions under each step instruction, introduces an operation parameter standard model to record the parameter index characterization habit characteristics of each user, and improves the adaptability of the parameter index characterization habit characteristics to the user by periodically optimizing the operation parameter standard model, improves the efficiency and flexibility of three-dimensional scene editing, and reduces the threshold for three-dimensional scene editing operations.

[0064] In a preferred embodiment, a natural language editing instruction input by a target editing user for a target three-dimensional scene construction task is obtained, and the natural language editing instruction is converted into a step instruction set including a plurality of step instructions, specifically including:

[0065] S110: upon receiving a 3D scene construction request for a target 3D scene construction task transmitted by a target editing user, extracting a user identifier of the target editing user in the 3D scene construction request and building a 3D scene construction environment, and after the 3D scene construction environment is built, feeding back input guidance information to the target editing user;

[0066] S120: Obtaining the natural language editing instructions input by the target editing user in response to the input guidance information, and converting the natural language editing instructions into a step instruction set including a plurality of step instructions for the three-dimensional scene construction environment.

[0067] Furthermore, obtaining the natural language editing instruction input by the target editing user for the input guide information, and converting the natural language editing instruction into a step instruction set including a plurality of step instructions for the three-dimensional scene construction environment, specifically includes:

[0068] S121: obtaining a natural language editing instruction input by a target editing user in response to the input guidance information, and performing word segmentation, part-of-speech analysis, and operation-related vocabulary extraction on the natural language editing instruction using a large language model based on natural language processing to obtain an operation-related vocabulary set;

[0069] S122: Based on the pre-stored mapping relationship table between operation-related words and three-dimensional scene construction steps, the three-dimensional scene construction steps corresponding to each operation-related word in the operation-related vocabulary set are matched to construct a step instruction set including several step instructions for the three-dimensional scene construction environment.

[0070] In this embodiment, the natural language editing instructions input by the target editing user for the target three-dimensional scene construction task are obtained, the operation-related vocabulary set is extracted using a large language model, and then a step instruction set is constructed through matching.

[0071] In a preferred embodiment, based on each step instruction in the step instruction set, calling a pre-stored interactive operation set to determine the operation instruction step corresponding to each step instruction specifically includes:

[0072] S210: calling the interactive operation set pre-stored in the three-dimensional scene construction environment, and extracting a plurality of interactive operations recorded in the interactive operation set and the interactive effect of each interactive operation;

[0073] S220: Using a large language model based on causal reasoning, causal relationship judgment is performed on each step instruction in the step instruction set and the interaction effect of each interactive operation, and the interactive operation with causal relationship is used as the operation instruction of the corresponding step instruction.

[0074] Furthermore, extracting parameter indicator representation information associated with each step instruction in the natural language editing instruction, using the parameter indicator representation information, based on the target editing user's operation parameter standard model, generating an operation parameter step corresponding to the operation instruction under the step instruction with the associated relationship, specifically including:

[0075] S310: Using a large language model based on semantic analysis, performing word segmentation processing, semantic analysis, parameter indicator representation vocabulary extraction, and operation-related word matching on the natural language editing instruction, and constructing parameter indicator representation information; wherein the parameter indicator representation information includes a plurality of parameter indicator representation words that have a matching relationship with the operation-related words;

[0076] S360: Input each parameter indicator representation word together with the operation-related words with matching relationship into the operation parameter standard model of the target editing user, and generate operation parameters corresponding to the operation instructions under the operation instructions determined by the operation-related words with matching relationship.

[0077] In this embodiment, after obtaining a step instruction set containing several step instructions, a large language model is used to extract the operation instructions corresponding to the step instructions and the parameter indicator representation information of each operation instruction, so as to convert the natural language editing instructions input by the user into step instructions for representing the editing steps, operation instructions for representing the operation method and operation parameters for representing the operation indicators, and replace the three-dimensional editor with a three-dimensional scene editing framework based on natural language control.

[0078] In a preferred embodiment, based on the operation instructions and operation parameters corresponding to each step instruction in the step instruction set, a 3D scene editing operation instruction set is generated and a 3D scene is constructed, the natural language editing instruction supplemented by the target editing user is obtained, and the constructed 3D scene is corrected, specifically including:

[0079] S410: generating a 3D scene editing operation instruction set based on the operation instruction and operation parameters corresponding to each step instruction in the step instruction set, inputting the 3D scene editing operation instruction into a 3D scene construction environment, and constructing a 3D scene for obtaining a target 3D scene construction task;

[0080] S420: Obtain the natural language editing instructions supplementally input by the target editing user for the constructed three-dimensional scene, and convert the supplementally input natural language editing instructions into operation instructions and operation parameters corresponding to each step instruction to modify the constructed three-dimensional scene.

[0081] On this basis, using the parameter index characterization information, based on the target editing user's operation parameter standard model, before generating the operation parameter step corresponding to the operation instruction under the step instruction with the associated relationship, it also includes:

[0082] S320: constructing an initial standard model of operating parameters; wherein the initial standard model of operating parameters stores standard operating parameters corresponding to operating instructions of each parameter indicator representation vocabulary;

[0083] S330: obtaining the natural language editing instructions input and supplemented by the target editing user for each historical three-dimensional scene construction task, extracting the operation parameters and corresponding parameter indicator representation vocabulary with the same operation instructions in the natural language editing instructions input and supplemented in the same historical three-dimensional scene construction task, and constructing them into an operation parameter standard model correction sample;

[0084] S340: correcting the operation parameters of the same operation instructions and the corresponding parameter index representation vocabulary in the sample according to the several operation parameter standard models in each correction cycle, and using the operation parameters of the operation instructions corresponding to the parameter index representation vocabulary in the supplemented input natural language editing instruction as correction values ​​to correct the operation parameters of the operation instructions corresponding to the parameter index representation vocabulary in the input natural language editing instruction;

[0085] S350: updating the operation parameter standard model using the operation parameter of the operation instruction corresponding to the vocabulary represented by each parameter indicator after correction, and storing the updated operation parameter standard model in association with the user identification of the target editing user;

[0086] The operation parameters of the operation instructions corresponding to the parameter indicator representation vocabulary in the input natural language editing instruction are the operation parameters of the operation instructions corresponding to each parameter indicator representation vocabulary in the operation parameter standard model updated and obtained in the previous correction cycle;

[0087] Among them, the operating parameters of the operating instructions corresponding to the parameter indicator representation vocabulary in the natural language editing instructions input in the first correction cycle are the standard operating parameters of the operating instructions corresponding to each parameter indicator representation vocabulary in the initial operating parameter standard model.

[0088] In this embodiment, by obtaining the natural language editing instruction input by the target editing user, the natural language editing instruction is converted into a step instruction and an operation instruction, and the parameter index characterization information is extracted. The parameter index characterization information is used to generate the operation parameters corresponding to the operation instructions under the step instructions with an associated relationship based on the standard model of the operation parameters of the target editing user. Finally, based on the operation instructions and operation parameters corresponding to each step instruction in the step instruction set, a three-dimensional scene editing operation instruction set is generated and a three-dimensional scene is constructed. For example, when the natural language editing instruction "adjust the brightness of the environment a little higher" input by the user is obtained, the natural language editing instruction is converted into a step instruction "adjust the brightness value of the main light source" and an operation instruction "SetMainLight{}", and the operation parameter "OldInedsity*1.5" is determined in the standard model of the operation parameters according to the parameter index characterization vocabulary "some". Therefore, this embodiment uses a large language model to replace the 3D editor with a 3D scene editing framework based on natural language control, converts the natural language editing instructions input by the user into operation parameters of the corresponding operation instructions under each step instruction, introduces an operation parameter standard model to record the parameter indicator characterization habit characteristics of each user, and improves the adaptability of the parameter indicator characterization habit characteristics to the user by periodically optimizing the operation parameter standard model, thereby improving the efficiency and flexibility of 3D scene editing and lowering the threshold for 3D scene editing operations.

[0089] In practical applications, this embodiment can greatly reduce the threshold for editing 3D scenes, so that people without 3D software production experience can also use this method to produce and edit 3D scenes. It can greatly improve the efficiency of 3D scene editing. Operations that originally required multiple steps to complete can now be completed with just one command, which can greatly reduce the requirements for personnel capabilities. Users no longer need to understand and learn the functions of various parameters, and only need to give a vague modification direction to complete the production and modification of the scene.

[0090] Reference Figure 3 , Figure 3 It is a structural block diagram of an embodiment of a three-dimensional scene editing device of the present invention.

[0091] like Figure 3 As shown, the three-dimensional scene editing device proposed in the embodiment of the present invention includes:

[0092] An acquisition module 10 is used to acquire a natural language editing instruction input by a target editing user for a target three-dimensional scene construction task, and convert the natural language editing instruction into a step instruction set including a plurality of step instructions;

[0093] A determination module 20, configured to call a pre-stored interactive operation set based on each step instruction in the step instruction set, and determine an operation instruction corresponding to each step instruction;

[0094] The extraction module 30 is used to extract parameter index representation information associated with each step instruction in the natural language editing instruction, and generate operation parameters corresponding to the operation instructions under the step instructions with the associated relationship based on the operation parameter standard model of the target editing user by using the parameter index representation information;

[0095] The generation module 40 is used to generate a three-dimensional scene editing operation instruction set and construct a three-dimensional scene based on the operation instructions and operation parameters corresponding to each step instruction in the step instruction set, obtain the natural language editing instructions supplemented by the target editing user, and modify the constructed three-dimensional scene.

[0096] Other embodiments or specific implementations of the three-dimensional scene editing device of the present invention can refer to the above-mentioned method embodiments, which will not be described in detail here.

[0097] In addition, the present invention also proposes a three-dimensional scene editing device, which includes: a memory, a processor, and a three-dimensional scene editing program stored in the memory and executable on the processor, wherein the three-dimensional scene editing program implements the steps of the three-dimensional scene editing method described above when executed by the processor.

[0098] The specific implementation of the 3D scene editing device of the present application is basically the same as the above-mentioned embodiments of the 3D scene editing method, and will not be repeated here.

[0099] In addition, the present invention also provides a readable storage medium, the readable storage medium includes a computer readable storage medium, on which a three-dimensional scene editing program is stored. The readable storage medium may be Figure 1 The memory 1005 in the terminal may also be at least one of a ROM (Read-Only Memory) / RAM (Random Access Memory), a magnetic disk, and an optical disk. The readable storage medium includes a number of instructions for enabling a three-dimensional scene editing device having a processor to execute the three-dimensional scene editing method described in each embodiment of the present invention.

[0100] The specific implementation in the readable storage medium of the present application is basically the same as the above-mentioned embodiments of the three-dimensional scene editing method, and will not be repeated here.

[0101] It is understood that, in the description of this specification, the description with reference to the terms "one embodiment", "another embodiment", "other embodiments", or "first to Nth embodiments" etc. means that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present invention. In this specification, the schematic representation of the above terms does not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described may be combined in any one or more embodiments or examples in a suitable manner.

[0102] It should be noted that, in this article, the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, article or system including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or system. In the absence of further restrictions, an element defined by the sentence "comprises a ..." does not exclude the existence of other identical elements in the process, method, article or system including the element.

[0103] The serial numbers of the above embodiments of the present invention are only for description and do not represent the advantages or disadvantages of the embodiments.

[0104] Through the description of the above implementation methods, those skilled in the art can clearly understand that the above-mentioned embodiment methods can be implemented by means of software plus a necessary general hardware platform, and of course by hardware, but in many cases the former is a better implementation method. Based on such an understanding, the technical solution of the present invention is essentially or the part that contributes to the prior art can be embodied in the form of a software product, which is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disk) as described above, and includes a number of instructions for a terminal device (which can be a mobile phone, a computer, a server, or a network device, etc.) to execute the methods described in each embodiment of the present invention.

[0105] The above are only preferred embodiments of the present invention, and are not intended to limit the patent scope of the present invention. Any equivalent structure or equivalent process transformation made using the contents of the present invention specification and drawings, or directly or indirectly applied in other related technical fields, are also included in the patent protection scope of the present invention.

Claims

1. A three-dimensional scene editing method, characterized in that: The following steps are involved: Obtaining a natural language editing instruction input by a target editing user for a target three-dimensional scene construction task, and converting the natural language editing instruction into a step instruction set including a plurality of step instructions; Based on each step instruction in the step instruction set, calling a pre-stored interactive operation set to determine an operation instruction corresponding to each step instruction; Extract parameter indicator representation information associated with each step instruction in the natural language editing instruction, and use the parameter indicator representation information to generate operation parameters corresponding to the operation instructions under the step instructions with the associated relationship based on the operation parameter standard model of the target editing user; specifically comprising: using a large language model based on semantic analysis to perform word segmentation, semantic analysis, parameter indicator representation vocabulary extraction and operation-related word matching on the natural language editing instruction to construct parameter indicator representation information; wherein the parameter indicator representation information includes a plurality of parameter indicator representation words that have a matching relationship with the operation-related words; each parameter indicator representation word is input into the operation parameter standard model of the target editing user together with the operation-related words that have a matching relationship, to generate operation parameters corresponding to the operation instructions under the operation instructions determined by the operation-related words that have a matching relationship; Based on the operation instructions and operation parameters corresponding to each step instruction in the step instruction set, a 3D scene editing operation instruction set is generated and a 3D scene is constructed, and the natural language editing instructions supplemented by the target editing user are obtained to correct the constructed 3D scene; specifically comprising: based on the operation instructions and operation parameters corresponding to each step instruction in the step instruction set, a 3D scene editing operation instruction set is generated, and the 3D scene editing operation instructions are input into a 3D scene construction environment to construct a 3D scene for obtaining a target 3D scene construction task; the natural language editing instructions supplemented by the target editing user for the constructed 3D scene are obtained, and the supplemented natural language editing instructions are converted into the operation instructions and operation parameters corresponding to each step instruction to correct the constructed 3D scene; Wherein, before the step of generating the operation parameters corresponding to the operation instructions under the step instructions with the associated relationship based on the operation parameter standard model of the target editing user by using the parameter indicator representation information, it also includes: constructing an initial operation parameter standard model; wherein the initial operation parameter standard model stores the standard operation parameters of the operation instructions corresponding to each parameter indicator representation vocabulary; obtaining the natural language editing instructions input and supplemented by the target editing user for each historical three-dimensional scene construction task, extracting the operation parameters and corresponding parameter indicator representation vocabulary with the same operation instructions in the natural language editing instructions input and supplemented in the same historical three-dimensional scene construction task, and constructing them as operation parameter standard model correction samples; according to the operation parameters and corresponding parameter indicator representation vocabulary with the same operation instructions in several operation parameter standard model correction samples in each correction cycle, the operation parameters of the operation instructions corresponding to the parameter indicator representation vocabulary in the supplemented natural language editing instructions are used as correction values ​​to correct the operation parameters of the operation instructions corresponding to the parameter indicator representation vocabulary in the input natural language editing instructions; updating the operation parameter standard model using the operation parameters of the operation instructions corresponding to each parameter indicator representation vocabulary after correction, and storing the updated operation parameter standard model in association with the user identification of the target editing user; Among them, the operating parameters of the operating instructions corresponding to the parameter indicator representation vocabulary in the input natural language editing instructions are the operating parameters of the operating instructions corresponding to each parameter indicator representation vocabulary in the operating parameter standard model updated in the previous correction cycle; among them, the operating parameters of the operating instructions corresponding to the parameter indicator representation vocabulary in the natural language editing instructions input in the first correction cycle are the standard operating parameters of the operating instructions corresponding to each parameter indicator representation vocabulary in the initial operating parameter standard model.

2. The three-dimensional scene editing method according to claim 1, characterized in that: Obtain the natural language editing instruction input by the target editing user for the target three-dimensional scene construction task, and convert the natural language editing instruction into a step instruction set including a plurality of step instructions, specifically including: Upon receiving a 3D scene construction request for a target 3D scene construction task transmitted by a target editing user, extracting a user identifier of the target editing user in the 3D scene construction request and building a 3D scene construction environment, and after the 3D scene construction environment is built, feeding back input guidance information to the target editing user; The natural language editing instructions input by the target editing user in response to the input guide information are obtained, and the natural language editing instructions are converted into a step instruction set including a plurality of step instructions for the three-dimensional scene construction environment.

3. The three-dimensional scene editing method according to claim 2, characterized in that: The steps of obtaining a natural language editing instruction input by a target editing user for the input guide information and converting the natural language editing instruction into a step instruction set including a plurality of step instructions for the three-dimensional scene construction environment include: Acquire a natural language editing instruction input by a target editing user in response to the input guide information, and use a large language model based on natural language processing to perform word segmentation, part-of-speech analysis, and operation-related vocabulary extraction on the natural language editing instruction to obtain an operation-related vocabulary set; Based on the pre-stored mapping relationship table between operation-related words and three-dimensional scene construction steps, the three-dimensional scene construction steps corresponding to each operation-related word in the operation-related vocabulary set are matched to construct a step instruction set containing several step instructions for the three-dimensional scene construction environment.

4. The three-dimensional scene editing method according to claim 3, characterized in that: Based on each step instruction in the step instruction set, calling the pre-stored interactive operation set to determine the operation instruction step corresponding to each step instruction specifically includes: Calling the interactive operation set pre-stored in the three-dimensional scene construction environment, extracting a plurality of interactive operations recorded in the interactive operation set and the interactive effect of each interactive operation; By using a large language model based on causal reasoning, a causal relationship is judged between each step instruction in the step instruction set and the interaction effect of each interactive operation, and the interactive operation with a causal relationship is used as the operation instruction of the corresponding step instruction.

5. A three-dimensional scene editing device, characterized in that: include: An acquisition module is used to acquire a natural language editing instruction input by a target editing user for a target three-dimensional scene construction task, and convert the natural language editing instruction into a step instruction set including a plurality of step instructions; A determination module, configured to call a pre-stored interactive operation set based on each step instruction in the step instruction set, and determine an operation instruction corresponding to each step instruction; The extraction module is used to extract parameter indicator representation information associated with each step instruction in the natural language editing instruction, and generate operation parameters corresponding to the operation instructions under the step instructions with the associated relationship based on the operation parameter standard model of the target editing user by using the parameter indicator representation information; specifically comprising: using a large language model based on semantic analysis to perform word segmentation, semantic analysis, parameter indicator representation vocabulary extraction and operation-related word matching on the natural language editing instruction to construct parameter indicator representation information; wherein the parameter indicator representation information includes a plurality of parameter indicator representation words that have a matching relationship with the operation-related words; each parameter indicator representation word is input together with the operation-related words that have a matching relationship into the operation parameter standard model of the target editing user to generate operation parameters corresponding to the operation instructions under the operation instructions determined by the operation-related words that have a matching relationship; A generation module is used to generate a 3D scene editing operation instruction set and construct a 3D scene based on the operation instructions and operation parameters corresponding to each step instruction in the step instruction set, obtain the natural language editing instructions supplemented by the target editing user, and modify the constructed 3D scene; specifically comprising: generating a 3D scene editing operation instruction set based on the operation instructions and operation parameters corresponding to each step instruction in the step instruction set, inputting the 3D scene editing operation instructions into a 3D scene construction environment, and constructing a 3D scene for obtaining a target 3D scene construction task; obtaining the natural language editing instructions supplemented by the target editing user for the constructed 3D scene, converting the supplemented natural language editing instructions into the operation instructions and operation parameters corresponding to each step instruction, so as to modify the constructed 3D scene; Wherein, before generating the operation parameters corresponding to the operation instructions under the step instructions with the associated relationship based on the operation parameter standard model of the target editing user by using the parameter indicator representation information, it also includes: constructing an initial operation parameter standard model; wherein the initial operation parameter standard model stores the standard operation parameters of the operation instructions corresponding to each parameter indicator representation vocabulary; obtaining the natural language editing instructions input and supplemented by the target editing user for each historical three-dimensional scene construction task, extracting the operation parameters and corresponding parameter indicator representation vocabulary with the same operation instructions in the natural language editing instructions input and supplemented in the same historical three-dimensional scene construction task, and constructing them as operation parameter standard model correction samples; according to the operation parameters and corresponding parameter indicator representation vocabulary with the same operation instructions in several operation parameter standard model correction samples in each correction cycle, the operation parameters of the operation instructions corresponding to the parameter indicator representation vocabulary in the supplemented natural language editing instructions are used as correction values ​​to correct the operation parameters of the operation instructions corresponding to the parameter indicator representation vocabulary in the input natural language editing instructions; updating the operation parameter standard model using the operation parameters of the operation instructions corresponding to each parameter indicator representation vocabulary after correction, and storing the updated operation parameter standard model in association with the user identification of the target editing user; Among them, the operating parameters of the operating instructions corresponding to the parameter indicator representation vocabulary in the input natural language editing instructions are the operating parameters of the operating instructions corresponding to each parameter indicator representation vocabulary in the operating parameter standard model updated in the previous correction cycle; among them, the operating parameters of the operating instructions corresponding to the parameter indicator representation vocabulary in the natural language editing instructions input in the first correction cycle are the standard operating parameters of the operating instructions corresponding to each parameter indicator representation vocabulary in the initial operating parameter standard model.

6. A three-dimensional scene editing device, characterized in that: The three-dimensional scene editing device includes: a memory, a processor, and a three-dimensional scene editing program stored in the memory and executable on the processor. When the three-dimensional scene editing program is executed by the processor, the steps of the three-dimensional scene editing method according to any one of claims 1 to 4 are implemented.

7. A storage medium, characterized in that: The storage medium stores a three-dimensional scene editing program, and when the three-dimensional scene editing program is executed by the processor, the steps of the three-dimensional scene editing method according to any one of claims 1 to 4 are implemented.

Images