Virtual visual 3D modeling mode based on character parameter description

Abstract

The invention relates to a virtual visualization 3D modeling mode based on character parameter description, comprising the following steps: S1, text is sent to a 3D generator, and AI automatically generates a corresponding 3D model; s2, based on the 3D modeling software of the text, a Shape-E model of OpenAI is adopted, so that the design idea can be converted into a visual 3D model more quickly; s3, quickly generating 3D assets, adopting 3D Gen deduced by Meta, and supporting complex text prompt and high texture quality, so that a user can realize quick iteration and accelerate design and development processes; and S4, language embedding radiation field LERF: a mode of embedding a natural language into a NeRF neural radiation field model allows a wide concept to be queried through the natural language, and interactively extracts a three-dimensional correlation map in real time. The method has the advantages that the 3D model can be quickly and accurately constructed, the operation is simple and convenient, and the process is simplified.

Description

Technical Field

[0001] The present invention relates to the technical field of 3D modeling, and in particular to a virtual visualization 3D modeling method based on text parameter description. Background Art

[0002] 3D modeling is a technology that uses professional software to create three-dimensional models, which converts real-world objects or fictional objects into data forms that computers can understand and process. This technology is widely used in movies, games, architecture, engineering, and manufacturing, creating many amazing visual effects and innovative designs.

[0003] 3D modeling through the description of file parameters can greatly reduce the time spent on modeling. However, the current modeling method of text parameter description cannot quickly, accurately and realistically build a 3D model based on the text parameter description, and has certain limitations. Summary of the invention

[0004] 1. Technical Problems Solved

[0005] The technical problem to be solved by the present invention is that the existing text parameter description 3D modeling method cannot quickly and accurately build a 3D model.

[0006] 2. Technical Solution

[0007] In order to solve the above technical problems, a virtual visualization 3D modeling method based on text parameter description includes the following steps:

[0008] S1. Text to 3D Generator, using Spline AI and Masterpiece Studio. These tools allow users to create 3D models and animations through simple text descriptions. Users only need to provide text-based descriptions, and AI will automatically generate the corresponding 3D models.

[0009] S2. Text-based 3D modeling software, using OpenAI's Shape-E model, which can generate realistic 3D assets through basic text descriptions, helping users save time in designing and visualizing complex design ideas. The introduction of Shape-E greatly simplifies the complexity of the design process, allowing design ideas to be converted into visual 3D models more quickly;

[0010] S3, fast 3D asset generation, using 3D Gen launched by Meta, which can generate high-quality 3D assets based on text prompts within 1 minute. It supports complex text prompts and high texture quality, allowing users to achieve rapid iteration and accelerate the design and development process;

[0011] S4. Language Embedded Radiance Field (LERF), which embeds natural language into the NeRF neural radiance field model, allows querying a wide range of concepts through natural language and extracting three-dimensional correlation maps interactively in real time.

[0012] Furthermore, the AI ​​in the S1 automatically generates corresponding 3D models for seamless textures, supports team collaboration, and makes real-time sharing and editing and creation simple and easy.

[0013] Furthermore, the specific method of using the Shape-E model in S2 includes:

[0014] S21, generating a text prompt of the 3D model;

[0015] S22, generating a synthetic view image of the 3D model;

[0016] S23, setting the ratio of guidance;

[0017] S24, specify the number of outputs;

[0018] S25, select rendering mode;

[0019] S26. Select the size of the renderer.

[0020] Furthermore, in said S22, when synthesizing the view image, the background is deleted from the image.

[0021] Furthermore, the 3DGen in S3 can further edit and customize the generated object texture to facilitate subsequent model processing.

[0022] Furthermore, the LERF in S4 supports users to browse three-dimensional scenes through natural language descriptions.

[0023] 3. Beneficial Effects

[0024] The advantages of the present invention compared with the prior art are:

[0025] By combining multiple software, realistic 3D resources are constructed through text, parameters, and instructions, simplifying the design steps and streamlining the complexity of the design. The resulting 3D models can be seamlessly textured, making real-time sharing and editing and creation simple and easy, while effectively ensuring the realism of the modeling. DETAILED DESCRIPTION

[0026] The technical solutions in the embodiments of the present invention are described clearly and completely below. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0028] In addition, in the present invention, descriptions such as "first", "second", etc. are only used for descriptive purposes and cannot be understood as indicating or implying their relative importance or implicitly indicating the number of the indicated technical features. Therefore, the features defined as "first" or "second" may explicitly or implicitly include at least one of the features. In the description of the present invention, the meaning of "plurality" is at least two, such as two, three, etc., unless otherwise clearly and specifically defined.

[0029] 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 embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not used to limit the invention.

[0030] The present invention will now be further described.

[0031] The present invention provides a virtual visualization 3D modeling method based on text parameter description, comprising the following steps:

[0032] S1. Text to 3D generator, using Spline AI and Masterpiece Studio. These tools allow users to create 3D models and animations through simple text descriptions. Users only need to provide text-based descriptions, and AI will automatically generate the corresponding 3D model. AI will seamlessly texture the automatically generated corresponding 3D model, support team collaboration, and make real-time sharing and editing creation simple and easy;

[0033] S2. Text-based 3D modeling software, using OpenAI's Shape-E model, can generate realistic 3D resources through basic text descriptions, helping users save time in designing and visualizing complex design ideas. The introduction of Shape-E greatly simplifies the complexity of the design process, allowing design ideas to be converted into visual 3D models more quickly. The specific usage of the Shape-E model includes:

[0034] S21, generating a text prompt of the 3D model;

[0035] S22, generating a synthetic view image of the 3D model, removing the background from the image;

[0036] S23, setting the ratio of guidance;

[0037] S24, specify the number of outputs;

[0038] S25, select rendering mode;

[0039] S26. Select the size of the renderer;

[0040] S3, fast 3D asset generation, using 3D Gen launched by Meta. 3DGen can further edit and customize the generated object textures to facilitate subsequent model processing. It can generate high-quality 3D assets based on text prompts within 1 minute. It supports complex text prompts and high texture quality, allowing users to achieve rapid iteration and accelerate the design and development process.

[0041] S4. Language Embedded Radiance Field (LERF). LERF supports users to browse three-dimensional scenes through natural language descriptions. It embeds natural language into the NeRF neural radiation field model, allowing a wide range of concepts to be queried through natural language and extract three-dimensional correlation maps interactively in real time.

[0042] The invention and its implementation methods are described, and such description is not restrictive. In short, if ordinary technicians in the field are inspired by it and design structural methods and embodiments similar to the technical solution without creativity without departing from the purpose of the invention, they should all fall within the protection scope of the invention.

Claims

1. A virtual visualization 3D modeling method based on text parameter description, characterized in that: The following steps are involved: S1. Text to 3D Generator, using Spline AI and Masterpiece Studio. These tools allow users to create 3D models and animations through simple text descriptions. Users only need to provide text-based descriptions, and AI will automatically generate the corresponding 3D models. S2. Text-based 3D modeling software, using OpenAI's Shape-E model, which can generate realistic 3D assets through basic text descriptions, helping users save time in designing and visualizing complex design ideas. The introduction of Shape-E greatly simplifies the complexity of the design process, allowing design ideas to be converted into visual 3D models more quickly; S3, fast 3D asset generation, using 3D Gen launched by Meta, which can generate high-quality 3D assets based on text prompts within 1 minute. It supports complex text prompts and high texture quality, allowing users to achieve rapid iteration and accelerate the design and development process; S4. Language Embedded Radiance Field (LERF), which embeds natural language into the NeRF neural radiance field model, allows querying a wide range of concepts through natural language and extracting three-dimensional correlation maps interactively in real time.

2. The virtual visualization 3D modeling method based on text parameter description according to claim 1, characterized in that: The AI ​​in the S1 automatically generates corresponding 3D models for seamless textures, supports team collaboration, and makes real-time sharing and editing creation simple and easy.

3. The virtual visualization 3D modeling method based on text parameter description according to claim 1, characterized in that: The specific usage of the Shape-E model in S2 includes: S21, generating a text prompt of the 3D model; S22, generating a synthetic view image of the 3D model; S23, setting the ratio of guidance; S24, specify the number of outputs; S25, select rendering mode; S26. Select the size of the renderer.

4. The virtual visualization 3D modeling method based on text parameter description according to claim 2 is characterized in that: In the S22, when synthesizing the view image, the background is deleted from the image.

5. The virtual visualization 3D modeling method based on text parameter description according to claim 1, characterized in that: The 3DGen in S3 can also further edit and customize the generated object texture to facilitate subsequent model processing.

6. The virtual visualization 3D modeling method based on text parameter description according to claim 1, characterized in that: The LERF in S4 supports users to browse three-dimensional scenes through natural language descriptions.