Automatic editing method and device for three-dimensional model chartlet, equipment and medium

Abstract

The invention discloses an automatic editing method and device for a three-dimensional model map, equipment and a medium, and belongs to the technical field of Internet. The method comprises the following steps: acquiring an initial three-dimensional model; acquiring mapping guide information of the user; wherein the mapping guide information comprises a demand description text and / or a reference image; inputting the map guidance information into a pre-constructed map generation model, and generating a target map based on the map generation model; and performing mapping processing on the initial three-dimensional model according to the target map to obtain a target three-dimensional model. According to the technical scheme, the map guide information of the user is analyzed and understood by adopting the large language model, the target map is automatically generated by adopting the map generation model, and map processing is performed, so that complex and tedious manual operation can be avoided, and the editing efficiency of the three-dimensional model map is improved.

Description

Technical Field

[0001] The present application belongs to the field of Internet technology, and specifically relates to a method, device, equipment and medium for automatically editing three-dimensional model mapping. Background Art

[0002] 3D model texture editing is an indispensable key step in the 3D model design process. Reasonable texture design can give the 3D model a realistic appearance and rich details, and lay a good foundation for subsequent rendering and animation production.

[0003] Ordinary users have limited ability to express their needs and ideas, which forces modelers to spend a lot of time communicating and exchanging ideas with users to ensure that the user's design needs are accurately captured, which is very inefficient. Modelers rely on professional skills and rich experience to edit textures so that the texture effects of 3D models can accurately meet the user's design needs. The texture editing process involves a lot of manual operations, such as manually editing and adjusting the position, size and color of the texture, which is very time-consuming and laborious.

[0004] Therefore, how to automatically, quickly and accurately understand the user's design requirements for stickers and perform sticker processing is an urgent problem that people in this field need to solve. Summary of the invention

[0005] The embodiments of the present application provide a method, device, equipment and medium for automatically editing three-dimensional model textures, the purpose of which is to avoid complex and tedious manual operations and improve the editing efficiency of three-dimensional model textures.

[0006] In a first aspect, an embodiment of the present application provides a method for automatically editing a three-dimensional model texture, the method comprising:

[0007] Obtaining an initial three-dimensional model;

[0008] Acquire the user's sticker guidance information; wherein the sticker guidance information includes a requirement description text and / or a reference image;

[0009] Inputting the mapping guidance information into a pre-built mapping generation model, and generating a target mapping based on the mapping generation model;

[0010] The initial three-dimensional model is subjected to mapping processing according to the target mapping to obtain a target three-dimensional model.

[0011] In a second aspect, an embodiment of the present application provides an automatic editing device for a three-dimensional model texture, the device comprising:

[0012] A three-dimensional model acquisition module, used to acquire an initial three-dimensional model;

[0013] A guidance information acquisition module, used to acquire the user's sticker guidance information; wherein the sticker guidance information includes a requirement description text and / or a reference image;

[0014] A target texture generation module, used for inputting the texture guidance information into a pre-built texture generation model, and generating a target texture based on the texture generation model;

[0015] The mapping processing module is used to perform mapping processing on the initial three-dimensional model according to the target mapping to obtain a target three-dimensional model.

[0016] In a third aspect, an embodiment of the present application provides an electronic device, which includes a processor, a memory, and a program or instruction stored in the memory and executable on the processor, wherein the program or instruction, when executed by the processor, implements the steps of the method described in the first aspect.

[0017] In a fourth aspect, an embodiment of the present application provides a readable storage medium, on which a program or instruction is stored, and when the program or instruction is executed by a processor, the steps of the method described in the first aspect are implemented.

[0018] In a fifth aspect, an embodiment of the present application provides a chip, comprising a processor and a communication interface, wherein the communication interface is coupled to the processor, and the processor is used to run a program or instruction to implement the method described in the first aspect.

[0019] In an embodiment of the present application, an initial three-dimensional model is obtained; user's mapping guidance information is obtained; wherein the mapping guidance information includes a requirement description text and / or a reference image; the mapping guidance information is input into a pre-built mapping generation model, and a target mapping is generated based on the mapping generation model; the initial three-dimensional model is mapped according to the target mapping to obtain a target three-dimensional model. The above-mentioned automatic editing method of three-dimensional model mapping can avoid complicated and tedious manual operations and improve the editing efficiency of three-dimensional model mapping by using a large language model to parse and understand the user's mapping guidance information, and using a mapping generation model to automatically generate a target mapping and perform mapping processing. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] Figure 1 It is a flowchart of the automatic editing method of the three-dimensional model texture provided in the first embodiment of the present application;

[0021] Figure 2 It is a flowchart of the automatic editing method of the three-dimensional model texture provided in the second embodiment of the present application;

[0022] Figure 3 This is a scene example diagram of the automatic editing method of the three-dimensional model texture provided in the second embodiment of the present application;

[0023] Figure 4 This is a flowchart of an automatic editing method for a three-dimensional model texture provided in the second embodiment of the present application;

[0024] Figure 5 It is a flowchart of the automatic editing method of the three-dimensional model texture provided in the third embodiment of the present application;

[0025] Figure 6 It is a flowchart of the automatic editing method of the three-dimensional model texture provided in the fourth embodiment of the present application;

[0026] Figure 7 It is a structural schematic diagram of an automatic editing device for a three-dimensional model texture provided in Embodiment 5 of the present application;

[0027] Figure 8 It is a structural diagram of an electronic device provided in Example 6 of the present application. DETAILED DESCRIPTION

[0028] In order to make the purpose, technical scheme and advantages of the present application clearer, the specific embodiments of the present application are further described in detail below in conjunction with the accompanying drawings. It is understood that the specific embodiments described herein are only used to explain the present application, rather than to limit the present application. It should also be noted that, for the convenience of description, only the part related to the present application but not all the contents are shown in the accompanying drawings. Before discussing the exemplary embodiments in more detail, it should be mentioned that some exemplary embodiments are described as processes or methods depicted as flow charts. Although the flow chart describes each operation (or step) as a sequential process, many of the operations therein can be implemented in parallel, concurrently or simultaneously. In addition, the order of each operation can be rearranged. The process can be terminated when its operation is completed, but it can also have additional steps not included in the accompanying drawings. The process can correspond to a method, a function, a procedure, a subroutine, a subprogram, etc.

[0029] The following will be combined with the drawings in the embodiments of the present application to clearly describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are part of the embodiments of the present application, rather than all the embodiments. All other embodiments obtained by ordinary technicians in this field based on the embodiments in the present application belong to the scope of protection of this application.

[0030] The terms "first", "second", etc. in the specification and claims of the present application are used to distinguish similar objects, and are not used to describe a specific order or sequence. It should be understood that the data used in this way can be interchangeable under appropriate circumstances, so that the embodiments of the present application can be implemented in an order other than those illustrated or described here, and the objects distinguished by "first", "second", etc. are generally of one type, and the number of objects is not limited. For example, the first object can be one or more. In addition, "and / or" in the specification and claims represents at least one of the connected objects, and the character " / " generally indicates that the objects associated with each other are in an "or" relationship.

[0031] The following, in conjunction with the accompanying drawings, describes in detail the automatic editing method, device, equipment and medium for three-dimensional model mapping provided by the embodiments of the present application through specific embodiments and their application scenarios.

[0032] Embodiment 1

[0033] Figure 1 1 is a flow chart of the automatic editing method of the three-dimensional model texture provided in the first embodiment of the present application. Figure 1 As shown, the specific steps include:

[0034] S101, obtaining an initial three-dimensional model;

[0035] First of all, this application is applicable to scenarios where users want to edit the texture of a three-dimensional model. Based on the above usage scenarios, it can be understood that the execution subject of this application can be a three-dimensional model editing system or its plug-in. Specifically, the acquisition of texture guidance information, the generation of target textures, and the texture processing of the three-dimensional model can be performed by the three-dimensional model editing system or its plug-in, and the target three-dimensional model obtained after the texture processing meets the user's texture design requirements for the three-dimensional model.

[0036] A 3D model is a data structure that digitally represents the 3D shape and appearance of an object. A 3D model editing system can be a web page or software application for creating, modifying, and managing 3D models. A plug-in is a small software component that can extend or enhance functionality. It is understandable that in this technical solution, the plug-in of the 3D model editing system can support the automatic editing function of the 3D model texture.

[0037] The initial 3D model may refer to a 3D model that needs to be mapped. The initial 3D model may be obtained by having the user import or upload the initial 3D model, or by automatically reading the 3D model as the initial 3D model after the user creates a new 3D model.

[0038] S102, obtaining the user's sticker guidance information; wherein the sticker guidance information includes a requirement description text and / or a reference image;

[0039] The texture guide information may refer to information used to guide the texture generation model to generate a target texture. It can be understood that the content of the texture guide information is description information of the design requirements for the texture.

[0040] The sticker guide information includes a requirement description text and / or a reference image, wherein the requirement description text may be text data of the design requirement for the sticker, and the reference image may be image data including the design requirement pattern for the sticker. The sticker guide information is input and / or uploaded by the user.

[0041] S103, inputting the mapping guidance information into a pre-built mapping generation model, and generating a target mapping based on the mapping generation model;

[0042] The map generation model can be an artificial intelligence model that can generate images based on input data. Among them, the artificial intelligence model is a mathematical and computational model used to realize artificial intelligence functions. The artificial intelligence model learns and extracts features from a large amount of data through a machine learning algorithm, thereby realizing the understanding and solution of complex problems. Accordingly, the way to pre-build a map generation model can use a guide map generation model to learn and extract features from a large amount of image data and its corresponding guidance information, thereby realizing the function of generating images based on input data. The map guidance information is used as an input parameter to call the map generation model, and the map generation model returns an image after calculation, which is the target map.

[0043] S104, performing mapping processing on the initial three-dimensional model according to the target mapping to obtain a target three-dimensional model.

[0044] The method of mapping the initial three-dimensional model according to the target map can select a suitable UV unfolding method according to the initial three-dimensional model to unfold the surface of the initial three-dimensional model into a two-dimensional plane, apply the target map to the UV unfolding plane of the initial three-dimensional model, and use a rendering engine to render the three-dimensional model with the target map after mapping to obtain the target three-dimensional model.

[0045] In the technical solution, optionally, after mapping the initial three-dimensional model according to the target mapping to obtain the target three-dimensional model, the method further includes:

[0046] When receiving feedback information from the user, the feedback information and the mapping guidance information are input into a pre-built mapping generation model, a new target mapping is generated based on the mapping generation model, and the initial three-dimensional model is mapped according to the new target mapping to obtain a new target three-dimensional model.

[0047] After receiving the target 3D model, if the user believes that the target 3D model does not meet the design requirements for the texture, the user can also input feedback information on the target 3D model to further optimize the texture of the target 3D model. The feedback information can be supplementary information and / or modification information for the texture guidance information.

[0048] The mapping generation model integrates the feedback information and the mapping guidance information to generate a new image, which is the new target map. The method of mapping the initial three-dimensional model according to the new target map is consistent with the above-mentioned method of mapping the initial three-dimensional model according to the target map.

[0049] The advantage of this arrangement of the present invention is that, upon receiving feedback information from the user, the initial three-dimensional model is mapped based on the feedback information and mapping guidance information to obtain a new target three-dimensional model. This allows the target three-dimensional model to fit the user's design requirements for mapping, thereby improving the user experience.

[0050] In an embodiment of the present application, an initial three-dimensional model is obtained; user's mapping guidance information is obtained; wherein the mapping guidance information includes a requirement description text and / or a reference image; the mapping guidance information is input into a pre-built mapping generation model, and a target mapping is generated based on the mapping generation model; the initial three-dimensional model is mapped according to the target mapping to obtain a target three-dimensional model. The above-mentioned automatic editing method of three-dimensional model mapping can avoid complicated and tedious manual operations and improve the editing efficiency of three-dimensional model mapping by using a large language model to parse and understand the user's mapping guidance information, and using a mapping generation model to automatically generate a target mapping and perform mapping processing.

[0051] Embodiment 2

[0052] Figure 2 It is a flow chart of the automatic editing method of three-dimensional model mapping provided in the second embodiment of the present application. This scheme makes a better improvement on the above embodiment, and the specific improvement is: before the mapping guidance information is input into the pre-built mapping generation model, and the target mapping is generated based on the mapping generation model, the method also includes: inputting the mapping guidance information into the pre-built large language model to obtain the user's mapping intention data; if the mapping intention data does not meet the execution condition, generating a dialogue message according to the mapping intention data, so as to obtain the mapping intention data that meets the execution condition according to the user's reply content to the dialogue message; accordingly, inputting the mapping guidance information into the pre-built mapping generation model, and generating the target mapping based on the mapping generation model, including: inputting the mapping intention data into the pre-built mapping generation model, and generating the target mapping based on the mapping generation model.

[0053] like Figure 2 As shown, the specific steps include:

[0054] S201, obtaining an initial three-dimensional model;

[0055] S202, obtaining the user's sticker guidance information; wherein the sticker guidance information includes a requirement description text and / or a reference image;

[0056] S203, inputting the sticker guidance information into a pre-built large language model to obtain sticker intention data of the user;

[0057] The sticker intention data may refer to information extracted from the sticker guidance information, which is more directional and easier for the sticker generation model to understand. The sticker intention data may include sticker prompt words and corresponding sticker parameters.

[0058] A large language model (LLM) may refer to a large-scale language model based on a deep learning algorithm, which has very powerful natural language understanding and generation capabilities. By pre-building a large language model, a large language model may be guided to learn and extract features from a large amount of sticker guidance information and its corresponding sticker intent data, thereby realizing a function of obtaining sticker intent data according to the input sticker guidance information.

[0059] The sticker guidance information is used as an input parameter to call the large language model, which returns sticker intent data after calculation. For example, if the sticker guidance information includes the requirement description text "change the character's skirt to a Japanese-style white skirt", the sticker intent data may include sticker prompt words such as "skirt", "Japanese style" and "white".

[0060] In this technical solution, optionally, the sticker guidance information is input into a pre-built large language model to obtain the sticker intention data of the user, including:

[0061] Inputting the mapping guidance information into at least two pre-built large language models to obtain output results of the large language models;

[0062] According to the pre-configured weight relationship of each language model, the output results of each language model are weighted and comprehensively calculated to obtain the user's mapping intention data.

[0063] The sticker guidance information is used as an input parameter to call at least two pre-built large language models (such as GPT, BERT, and T5, etc.). After calculation, each language model returns the corresponding sticker intent data. Among them, GPT (Generative Pre-trained Transformer) is based on the unsupervised pre-trained Transformer architecture and is good at generating coherent and fluent language; BERT (Bidirectional Encoder Representations from Transformers) adopts a bidirectional Transformer encoder architecture and is good at understanding context and complex semantics; T5 (Text-to-Text Transfer Transformer) adopts a unified Transformer architecture and is good at tasks such as question answering, summarization, and translation.

[0064] The weight relationship of each language model may refer to the weight coefficients of the weighted comprehensive calculation of each language model, wherein the weight coefficient of a large language model may refer to the degree of influence of the output result of the large language model on the final determined mapping intent data. The weight relationship of each language model may be pre-configured according to the model performance, model complexity and task relevance of each language model.

[0065] The output results of each language model are weighted and comprehensively calculated to obtain the user's mapping intention data. The output results of each language model are converted into vector representations, and the target vector is obtained by weighted average calculation of each vector representation according to the weight relationship. The target vector is reversely converted to a vector to obtain the mapping intention data.

[0066] The benefit of this arrangement of the present solution is that by inputting the sticker guidance information into at least two large language models and performing weighted comprehensive calculation on the output results of the large language models according to the weight relationship between the large language models, the user's sticker intention data can be obtained, which can reduce the limitations of a single model and focus on improving the influence of a particular model on a specific task according to the actual application scenario, thereby obtaining more accurate sticker intention data.

[0067] S204, when the sticker intent data does not satisfy the execution condition, generating a dialogue message according to the sticker intent data, so as to obtain sticker intent data satisfying the execution condition according to the user's reply content to the dialogue message;

[0068] The sticker intent data not satisfying the execution condition may refer to a situation where the sticker intent data cannot enable the sticker generation model to generate a sticker. The dialogue information may be information generated by the large language model to guide the user to provide more data required by the sticker generation model, such as "Can you provide further drawing styles?" The user inputs a reply content according to the dialogue information, and the large language model determines the user's sticker intent data according to the sticker guidance information and the reply content.

[0069] Figure 3 : is a scene example diagram of the automatic editing method of the three-dimensional model texture provided in the second embodiment of the present application. Figure 3 As shown, the dialog box displays the conversation information generated by the large language model and the user's reply content. The user can enter the reply content in the input box, and the three-dimensional model is rendered on the right.

[0070] In the technical solution, optionally, the dialogue information includes at least one of dialogue information for confirming style, dialogue information for confirming color, and dialogue information for confirming material;

[0071] Correspondingly, when the sticker intent data does not meet the execution condition, a dialog message is generated according to the sticker intent data, so as to obtain sticker intent data that meets the execution condition according to the user's reply content to the dialog message, including:

[0072] In the case where the sticker intent data does not satisfy the style execution condition, generating a style confirmation dialogue message according to the sticker intent data, so as to obtain a sticker style satisfying the execution condition according to the user's reply content to the style confirmation dialogue message;

[0073] and / or,

[0074] In the case where the sticker intent data does not satisfy the color execution condition, generating a confirmation color dialogue message according to the sticker intent data, so as to obtain a sticker color satisfying the execution condition according to the user's reply content to the confirmation color dialogue message;

[0075] and / or,

[0076] In the case that the texture intent data does not satisfy the material execution condition, a confirmation material dialogue message is generated according to the texture intent data, so as to obtain a texture material satisfying the execution condition according to the user's reply content to the confirmation material dialogue message.

[0077] The fact that the sticker intent data does not satisfy the style execution condition may refer to the situation that the sticker intent data does not include sticker style data, and the sticker generation model cannot generate stickers. Among them, sticker style data may refer to a unique artistic feature or expression method shown by the sticker, which may include painting style, artistic style, theme style, emotional style, and element style. Confirmation style dialogue information may refer to dialogue information used to guide the user to further provide information related to the sticker style data, such as "Can you further provide the painting style?" The user enters the reply content according to the confirmation style dialogue information, and the large language model determines the sticker style data according to the reply content.

[0078] The fact that the sticker intent data does not satisfy the color execution condition may refer to a situation where the sticker intent data does not include sticker color data, and the sticker generation model cannot generate a sticker. Among them, sticker color data may refer to the color characteristics of the sticker, which may include color style, color mood, color theme, and color texture. The color confirmation dialogue information may refer to dialogue information used to guide the user to further provide information related to the sticker color data, such as "Can you further provide a color theme?" The user enters a reply content based on the color confirmation dialogue information, and the large language model determines the sticker color data based on the reply content.

[0079] The fact that the texture intent data does not satisfy the material execution condition may refer to the situation that the texture intent data does not include the texture material data, and the texture generation model cannot generate the texture. The texture material data may refer to the material characteristics of the texture in terms of vision and touch, and may include surface texture, reflection properties, and texture characteristics. The material confirmation dialogue information may refer to dialogue information for guiding the user to further provide information related to the texture material data, such as "Can you further provide the surface texture?" The user enters the reply content according to the material confirmation dialogue information, and the large language model determines the texture material data according to the reply content.

[0080] The advantage of this arrangement of the present solution is that, by generating dialogue information based on the sticker intent data when the sticker intent data does not meet the execution conditions, and obtaining sticker intent data that meets the execution conditions based on the user's reply to the dialogue information, the user can be better guided to input the required information of the sticker generation model, thereby improving the success rate of target sticker generation and user experience.

[0081] S205, inputting the texture intent data into a pre-built texture generation model, and generating a target texture based on the texture generation model;

[0082] The texture generation model is called with the texture intent data as an input parameter. After calculation, the texture generation model returns an image, which is the target texture.

[0083] S206, performing mapping processing on the initial three-dimensional model according to the target mapping to obtain a target three-dimensional model.

[0084] Figure 4 2 is a flowchart of an automatic editing method for a three-dimensional model texture provided in the second embodiment of the present application. Figure 4 As shown, the specific steps include: S401, the user uploads the 3D model and obtains the rendering; S402, the user inputs the map modification description and / or reference picture; S403, the large language model parses the user needs, the participant map and the rendering; S404, dynamically generates the dialogue content; S405, generates prompt words and drawing parameters according to the dialogue content; S406, calls the drawing model to modify the map; S407, displays the modification result and obtains the user feedback; S408, determines whether the user is satisfied with the modification result; if satisfied, ends the process; if not satisfied, executes S404.

[0085] The benefit of this setting is that by inputting the sticker guidance information into the pre-built large language model to obtain the user's sticker intention data, the powerful ability of the large language model in natural language understanding and generation can be brought into play, helping the sticker generation model to accurately capture the user's design requirements for stickers.

[0086] Embodiment 3

[0087] Figure 5 It is a flowchart of the automatic editing method of three-dimensional model mapping provided in Example 3 of the present application. This solution makes a better improvement on the above-mentioned embodiment, and the specific improvement is: before the mapping guidance information is input into the pre-built mapping generation model, and the target mapping is generated based on the mapping generation model, the method also includes: obtaining the user's mapping selection label and / or mapping template editing result; wherein, the preset mapping information includes the mapping selection label, and the preset mapping information includes preset prompt words and preset parameters; accordingly, the mapping guidance information is input into the pre-built mapping generation model, and the target mapping is generated based on the mapping generation model, including: inputting the mapping selection label and / or mapping template editing result into the pre-built mapping generation model, and generating the target mapping based on the mapping generation model.

[0088] like Figure 5 As shown, the specific steps include:

[0089] S501, obtaining an initial three-dimensional model;

[0090] S502, obtaining a sticker selection label and / or sticker template editing result of a user; wherein the preset sticker information includes the sticker selection label, and the preset sticker information includes preset prompt words and preset parameters;

[0091] The preset sticker information may be pre-set directional information that is easy for the sticker generation model to understand, and may include preset prompt words and preset parameters. The preset prompt words are pre-set sticker prompt words, and the preset parameters are pre-set sticker parameters. The preset sticker information selected by the user is the sticker selection label. In response to the user's sticker information selection request, the user's sticker selection label may be obtained.

[0092] The sticker template editing result may be a requirement description text and / or a reference picture edited by the user based on a preset template. In response to the sticker template editing request of the user, the sticker template editing result of the user may be obtained.

[0093] S503, inputting the texture selection label and / or texture template editing result into a pre-built texture generation model, and generating a target texture based on the texture generation model;

[0094] The texture selection label and / or the texture template editing result is used as an input parameter to call the texture generation model, and the texture generation model returns an image after calculation, and the image is the target texture.

[0095] S504: Perform mapping processing on the initial three-dimensional model according to the target mapping to obtain a target three-dimensional model.

[0096] The advantage of this arrangement of the present scheme is that by obtaining the user's texture selection label and / or texture template editing result, and inputting the texture selection label and / or texture template editing result into the texture generation model to obtain the target texture, the user's operation can be simplified, and it is ensured that the data input into the texture generation model meets the data requirements of the texture generation model, thereby improving the robustness of the texture generation model.

[0097] Embodiment 4

[0098] Figure 6 It is a flow chart of the automatic editing method of three-dimensional model mapping provided by the fourth embodiment of the present application. This solution makes a better improvement on the above embodiment, and the specific improvement is: the drawing model adopts a stable diffusion model; accordingly, the mapping guidance information is input into a pre-built mapping generation model, and a target mapping is generated based on the mapping generation model, including: the mapping guidance information is input into a pre-built stable diffusion model, and a target mapping is generated based on the stable diffusion model.

[0099] like Figure 6 As shown, the specific steps include:

[0100] S601, obtaining an initial three-dimensional model;

[0101] S602, obtaining the user's sticker guidance information; wherein the sticker guidance information includes a requirement description text and / or a reference image;

[0102] S603, inputting the mapping guidance information into a pre-built stable diffusion model, and generating a target mapping based on the stable diffusion model;

[0103] The stable diffusion model is a text-based AI technology based on the diffusion model. The stable diffusion model establishes the association between images and texts by learning massive image-text pairing data, so that it can generate corresponding images based on the input text description.

[0104] The code and model parameters of the stable diffusion model are open source, and developers can customize and fine-tune them according to their own needs; compared with the image generation model based on the generative adversarial network, the stable diffusion model has higher generation efficiency and faster generation speed; the stable diffusion model can generate high-resolution, realistic and natural images, and performs well in terms of image details, color, and composition; the stable diffusion model can generate images of various styles and themes based on the input text, and has strong creativity.

[0105] The stable diffusion model is called with the map guide information as an input parameter. The stable diffusion model returns an image after calculation, and the image is the target map.

[0106] S604: Perform mapping processing on the initial three-dimensional model according to the target mapping to obtain a target three-dimensional model.

[0107] The advantage of this arrangement of the present invention is that by adopting a stable diffusion model to generate a target map, the generation speed of the target map can be increased, and the image quality of the generated target map can be improved.

[0108] Embodiment 5

[0109] Figure 7 Schematic diagram of the structure of the automatic editing device for three-dimensional model mapping provided in Example 5 of the present application. Figure 7 As shown, the device comprises:

[0110] A three-dimensional model acquisition module 710 is used to acquire an initial three-dimensional model;

[0111] The guidance information acquisition module 720 is used to obtain the user's sticker guidance information; wherein the sticker guidance information includes a requirement description text and / or a reference image;

[0112] A target texture generation module 730, configured to input the texture guidance information into a pre-built texture generation model, and generate a target texture based on the texture generation model;

[0113] The texture processing module 740 is used to perform texture processing on the initial three-dimensional model according to the target texture to obtain a target three-dimensional model.

[0114] In an embodiment of the present application, a three-dimensional model acquisition module is used to acquire an initial three-dimensional model; a guidance information acquisition module is used to acquire a user's mapping guidance information; wherein the mapping guidance information includes a requirement description text and / or a reference image; a target mapping generation module is used to input the mapping guidance information into a pre-built mapping generation model, and generate a target mapping based on the mapping generation model; and a mapping processing module is used to perform mapping processing on the initial three-dimensional model according to the target mapping to obtain a target three-dimensional model. The above-mentioned automatic editing device for three-dimensional model mapping can avoid complex and tedious manual operations and improve the editing efficiency of three-dimensional model mapping by using a large language model to parse and understand the user's mapping guidance information, and using a mapping generation model to automatically generate a target mapping and perform mapping processing.

[0115] The automatic editing device of the three-dimensional model mapping in the embodiment of the present application can be a device, or a component, integrated circuit, or chip in a terminal. The device can be a mobile electronic device or a non-mobile electronic device. Exemplarily, the mobile electronic device can be a mobile phone, a tablet computer, a laptop computer, a PDA, an in-vehicle electronic device, a wearable device, an ultra-mobile personal computer (ultra-mobile personal computer, UMPC), a netbook or a personal digital assistant (personal digital assistant, PDA), etc., and the non-mobile electronic device can be a server, a network attached storage (Network Attached Storage, NAS), a personal computer (personal computer, PC), a television (television, TV), a teller machine or a self-service machine, etc., which is not specifically limited in the embodiment of the present application.

[0116] The automatic editing device for three-dimensional model mapping in the embodiment of the present application may be a device having an operating system. The operating system may be an Android operating system, an IOS operating system, or other possible operating systems, which are not specifically limited in the embodiment of the present application.

[0117] The automatic editing device for three-dimensional model mapping provided in the embodiment of the present application can implement each process implemented in the above-mentioned embodiments one to four, and will not be described again here to avoid repetition.

[0118] Embodiment 6

[0119] like Figure 8As shown, the embodiment of the present application also provides an electronic device 800, including a processor 801, a memory 802, and a program or instruction stored in the memory 802 and executable on the processor 801. When the program or instruction is executed by the processor 801, each process of the embodiment of the automatic editing method of the above-mentioned three-dimensional model mapping is implemented, and the same technical effect can be achieved. To avoid repetition, it will not be repeated here.

[0120] It should be noted that the electronic devices in the embodiments of the present application include the mobile electronic devices and non-mobile electronic devices mentioned above.

[0121] Embodiment 8

[0122] An embodiment of the present application also provides a readable storage medium, on which a program or instruction is stored. When the program or instruction is executed by a processor, each process of the above-mentioned automatic editing method for three-dimensional model mapping is implemented, and the same technical effect can be achieved. To avoid repetition, it will not be repeated here.

[0123] The processor is a processor in the electronic device described in the above embodiment. The readable storage medium includes a computer readable storage medium, such as a computer read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk.

[0124] Embodiment 9

[0125] An embodiment of the present application further provides a chip, which includes a processor and a communication interface, wherein the communication interface is coupled to the processor, and the processor is used to run programs or instructions to implement the various processes of the above-mentioned automatic editing method embodiment of the three-dimensional model mapping, and can achieve the same technical effect. To avoid repetition, it will not be repeated here.

[0126] It should be understood that the chip mentioned in the embodiments of the present application can also be called a system-level chip, a system chip, a chip system or a system-on-chip chip, etc.

[0127] It should be noted that, in this article, the terms "comprise", "include" or any other variant thereof are intended to cover non-exclusive inclusion, so that the process, method, article or device including a series of elements includes not only those elements, but also includes other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, the elements defined by the sentence "comprise one..." do not exclude the presence of other identical elements in the process, method, article or device including the element. In addition, it should be pointed out that the scope of the methods and devices in the embodiments of the present application is not limited to performing functions in the order shown or discussed, and may also include performing functions in a substantially simultaneous manner or in reverse order according to the functions involved, for example, the described method may be performed in an order different from that described, and various steps may also be added, omitted, or combined. In addition, the features described with reference to certain examples may be combined in other examples.

[0128] 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 application, or the part that contributes to the prior art, can be embodied in the form of a computer software product, which is stored in a storage medium (such as ROM / RAM, a disk, or an optical disk), and includes a number of instructions for a terminal (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 application.

[0129] The embodiments of the present application are described above in conjunction with the accompanying drawings, but the present application is not limited to the above-mentioned specific implementation methods. The above-mentioned specific implementation methods are merely illustrative and not restrictive. Under the guidance of the present application, ordinary technicians in this field can also make many forms without departing from the purpose of the present application and the scope of protection of the claims, all of which are within the protection of the present application.

[0130] The above are only preferred embodiments of the present application and the technical principles used. The present application is not limited to the specific embodiments described herein, and various obvious changes, readjustments and substitutions that can be made by those skilled in the art will not deviate from the scope of protection of the present application. Therefore, although the present application is described in more detail through the above embodiments, the present application is not limited to the above embodiments, and may include more other equivalent embodiments without departing from the concept of the present application, and the scope of the present application is determined by the scope of the claims.

Claims

1. A method for automatically editing a three-dimensional model map, characterized in that: The method comprises: Obtaining an initial three-dimensional model; Acquire the user's sticker guidance information; wherein the sticker guidance information includes a requirement description text and / or a reference image; Inputting the mapping guidance information into a pre-built mapping generation model, and generating a target mapping based on the mapping generation model; The initial three-dimensional model is subjected to mapping processing according to the target mapping to obtain a target three-dimensional model.

2. The automatic editing method of three-dimensional model mapping according to claim 1, characterized in that: Before inputting the mapping guidance information into a pre-built mapping generation model and generating a target mapping based on the mapping generation model, the method further includes: Inputting the sticker guidance information into a pre-built large language model to obtain sticker intention data of the user; In the case where the sticker intent data does not satisfy the execution condition, a dialogue message is generated according to the sticker intent data, so as to obtain sticker intent data satisfying the execution condition according to the user's reply content to the dialogue message; Accordingly, the mapping guidance information is input into a pre-built mapping generation model, and a target mapping is generated based on the mapping generation model, including: The texture intent data is input into a pre-built texture generation model, and a target texture is generated based on the texture generation model.

3. The automatic editing method of three-dimensional model mapping according to claim 2, characterized in that: The dialogue information includes at least one of a style confirmation dialogue information, a color confirmation dialogue information, and a material confirmation dialogue information; Correspondingly, when the sticker intent data does not meet the execution condition, a dialog message is generated according to the sticker intent data, so as to obtain sticker intent data that meets the execution condition according to the user's reply content to the dialog message, including: In the case where the sticker intent data does not satisfy the style execution condition, generating a style confirmation dialogue message according to the sticker intent data, so as to obtain a sticker style satisfying the execution condition according to the user's reply content to the style confirmation dialogue message; and / or, In the case where the sticker intent data does not satisfy the color execution condition, generating a confirmation color dialogue message according to the sticker intent data, so as to obtain a sticker color satisfying the execution condition according to the user's reply content to the confirmation color dialogue message; and / or, In the case that the texture intent data does not satisfy the material execution condition, a confirmation material dialogue message is generated according to the texture intent data, so as to obtain a texture material satisfying the execution condition according to the user's reply content to the confirmation material dialogue message.

4. The automatic editing method of three-dimensional model mapping according to claim 2, characterized in that: The sticker guidance information is input into a pre-built large language model to obtain the sticker intention data of the user, including: Inputting the mapping guidance information into at least two pre-built large language models to obtain output results of the large language models; According to the pre-configured weight relationship of each language model, the output results of each language model are weighted and comprehensively calculated to obtain the user's mapping intention data.

5. The automatic editing method of three-dimensional model mapping according to claim 1, characterized in that: After the initial three-dimensional model is subjected to mapping processing according to the target mapping to obtain the target three-dimensional model, the method further includes: When receiving feedback information from the user, the feedback information and the mapping guidance information are input into a pre-built mapping generation model, a new target mapping is generated based on the mapping generation model, and the initial three-dimensional model is mapped according to the new target mapping to obtain a new target three-dimensional model.

6. The automatic editing method of three-dimensional model mapping according to claim 1, characterized in that: Before inputting the mapping guidance information into a pre-built mapping generation model and generating a target mapping based on the mapping generation model, the method further includes: Obtaining a sticker selection label and / or sticker template editing result of a user; wherein the preset sticker information includes the sticker selection label, and the preset sticker information includes preset prompt words and preset parameters; Accordingly, the mapping guidance information is input into a pre-built mapping generation model, and a target mapping is generated based on the mapping generation model, including: The texture selection label and / or texture template editing result is input into a pre-built texture generation model, and a target texture is generated based on the texture generation model.

7. The automatic editing method of three-dimensional model mapping according to claim 1, characterized in that: The drawing model adopts a stable diffusion model; Accordingly, the mapping guidance information is input into a pre-built mapping generation model, and a target mapping is generated based on the mapping generation model, including: The mapping guidance information is input into a pre-built stable diffusion model, and a target mapping is generated based on the stable diffusion model.

8. A method for automatically editing a three-dimensional model map, characterized in that: The method comprises: A three-dimensional model acquisition module, used to acquire an initial three-dimensional model; A guidance information acquisition module, used to acquire the user's sticker guidance information; wherein the sticker guidance information includes a requirement description text and / or a reference image; A target texture generation module, used for inputting the texture guidance information into a pre-built texture generation model, and generating a target texture based on the texture generation model; The mapping processing module is used to perform mapping processing on the initial three-dimensional model according to the target mapping to obtain a target three-dimensional model.

9. An electronic device, characterized in that: The method comprises a processor, a memory and a program or instruction stored in the memory and executable on the processor, wherein the program or instruction, when executed by the processor, implements the steps of the automatic editing method of the three-dimensional model mapping as described in any one of claims 1 to 7.

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

Images