A three-dimensional visual simulation construction method and system

Abstract

The present invention provides a three-dimensional visual scene simulation construction method and system, which relates to the field of image simulation. The method includes: obtaining the geometric parameters of the scene model, and constructing a material-free scene model according to a parameterized model generation component; setting the material of the material-free scene model according to a dynamic material component to obtain a dynamic material scene model, wherein the dynamic material component includes a basic material layer and an environmental response layer; setting the animation of the dynamic material scene model according to a modular animation component to obtain a scene model with animation and dynamic materials, so as to construct simulation resources according to the scene model with animation and dynamic materials. The present invention modularizes animation and decouples resources, thus having high reusability and flexibility, replacing the traditional manual modeling that requires manual adjustment of each character's joints, resulting in transitions between animations that are often stiff and unnatural; the process parameter automation in the technical solution of the present invention greatly improves the efficiency of model production.

Description

Technical Field

[0001] The present invention relates to the field of image simulation technology, and in particular to a three-dimensional visual scene simulation construction method and system. Background Art

[0002] In urban planning, medical training, game development and other fields, three-dimensional visual simulation technology is crucial. A large number of model resources are often required for display in simulation. The model is mainly composed of materials, textures, animations and other parts.

[0003] Currently, 3D modeling in existing technologies relies on manual design (such as using 3DMax, Maya, and Blender), and the reuse rate of models is less than 15%, resulting in low efficiency in model resource production. For example, in visual simulation software, the same type of tank needs to be repeatedly modeled in different scenes such as deserts and snowy areas (the tank's texture should be yellowish in desert scenes, but white in snowy areas). This operation will greatly increase the redundancy of the model resource library. On the other hand, when simulation software uses high-precision models (with more than 1 million faces) for real-time rendering, the frame rate is often difficult to exceed 30fps, which cannot meet the smoothness requirement of XR-type simulation scenes that must be greater than 90fps (users will feel dizzy or vomit if the running speed is lower than 90fps);

[0004] Secondly, the materials and textures of traditional models are statically bound, and the textures use Physically Based Rendering (PBR) materials, which require pre-generated static texture sets (such as Albedo, Normal, Roughness, etc.). Generating static texture sets is time-consuming and cannot dynamically adapt to multiple lighting environments. For example, when the exterior glass of a building changes from day to night in the training environment, multiple sets of material maps need to be switched. In this scenario, the material map storage memory will increase by about 1.5GB (assuming 4K textures are used). Real-time material adjustment using dynamic lighting (in the same scene) relies on shader programming, which is cumbersome and time-consuming, and will greatly extend the project development cycle;

[0005] Furthermore, traditional models suffer from excessive coupling between animation resources. Traditional character animation data is strongly bound to the skeletal model, requiring frame-by-frame adjustment of skeletal weights to reuse animations from different character models. For example, when migrating character A's "walking" animation to character B of a different body type, the probability of body model interpenetration often exceeds 60% due to differences in height, limb length, and body weight. Manual correction and readjustment of the skeleton and weights for this animation clip takes approximately two hours.

[0006] Therefore, the three-dimensional visual simulation model in the prior art has technical problems such as low efficiency in model resource production, static binding of materials and textures that cannot be dynamically adapted, and excessive coupling of animation resources. Summary of the Invention

[0007] Based on this, the purpose of the present invention is to provide a three-dimensional visual simulation construction method and system to solve the technical problems of low model resource production efficiency, inability to dynamically adapt static binding of materials and textures, and excessive coupling of animation resources in the existing three-dimensional visual simulation model.

[0008] In one aspect, the present invention provides a method for constructing a three-dimensional visual scene simulation, comprising:

[0009] Acquire geometric parameters of the scene model and construct a material-free scene model based on a parametric model generation component, wherein the geometric parameters include basic size, number of segments, and curvature radius; the parametric model generation component includes a rigid template library and a flexible template library; and the scene model includes a rescue vehicle model, a landslide mountain model, and a landslide rock model;

[0010] Setting the material of the material-free scene model according to the dynamic material component to obtain a dynamic material scene model, wherein the dynamic material component includes a basic material layer and an environment response layer;

[0011] The dynamic material scene model is animated according to the modular animation component to obtain a scene model with animation and dynamic materials, so as to construct a simulation resource according to the scene model with animation and dynamic materials.

[0012] The above-mentioned three-dimensional visual scene simulation construction method uses dynamic material generation technology. Compared with traditional model mapping, dynamic materials are generated using adjustable parameters without static binding to maps, and dynamic materials can achieve dynamic effects according to the parameters of the environmental response layer, thereby shortening the material generation time while improving the model appearance effect; secondly, the present invention modularizes animation and decouples resources, with high reusability and flexibility, replacing the traditional manual modeling that requires manual adjustment of each character joint, resulting in the transition between animations being often stiff and unnatural; furthermore, the process parameters in the technical solution of the present invention are automated, greatly improving the model production efficiency; and solving the technical problems in the prior art of low model resource production efficiency, static binding of materials and maps, and excessive coupling of animation resources in three-dimensional visual scene simulation models.

[0013] In addition, the above-mentioned three-dimensional visual scene simulation construction method according to the present invention may also have the following additional technical features:

[0014] Furthermore, the modular animation components include action atomization, physical constraints, and real-time animation synthesis, wherein the real-time animation synthesis represents real-time transitions between different actions;

[0015] The steps of animating the dynamic material scene model according to the modular animation component to obtain a scene model with animation and dynamic materials include:

[0016] Use linear interpolation in a one-dimensional blend space to transition the animation. The character speed parameter controls the speed of the blended animation. The one-dimensional blend space is an atom that changes posture only in two axes. The transition action includes transitioning from walking forward to running forward.

[0017] Complex motion fusion is used in a two-dimensional hybrid space to adjust the skeleton posture through dual-parameter interpolation. Complex motion fusion includes turning during jumping; the dual parameters are horizontal speed and vertical speed.

[0018] Animation synthesis is performed using an inverse dynamics algorithm in a specific scenario; wherein the specific scenario includes a scenario in which the end position and posture of an atom are known; the inverse dynamics algorithm is used to automatically calculate the rotation of intermediate atoms by specifying the position and rotation of the end atoms, wherein the end atoms include hands and feet, and the intermediate atoms include elbows and knees.

[0019] Furthermore, the step of animating the dynamic material scene model according to the modular animation component to obtain a scene model with animation and dynamic materials also includes:

[0020] In the dynamic material mountain model and the dynamic material rock model, real-time calculations are performed based on the physical engine, and a physical simulation module is added to the simulation engine so that each physical simulation module cooperates with each other to complete the landslide animation playback. Among them, the physical simulation module includes a gravity module, a collision module and a rolling module.

[0021] Furthermore, in the step of obtaining the geometric parameters of the scene model and constructing the material-free scene model based on the parameterized model generation component:

[0022] A material-free scene model is obtained according to the parameter constraint algorithm and the instantiation model generation algorithm.

[0023] Furthermore, in the step of setting the material of the material-free scene model according to the dynamic material component to obtain the dynamic material scene model:

[0024] A dynamic material scene model is obtained according to the basic material layer parameters, the environmental response layer parameters and the dynamic material mixing algorithm, wherein the basic material layer parameters include basic color, metalness and roughness; the environmental response layer parameters include light intensity, ambient humidity and air temperature.

[0025] Furthermore, in the environmental response layer parameters, the calculation formula for the specular reflection intensity is:

[0026] R(θ)=R+(1-R)(1-cosθ) 5 ;

[0027] Where: R(θ) represents the intensity of specular reflection; R is the base reflectivity of the material, and θ is the angle of incidence.

[0028] Furthermore, in the base material layer parameters, the calculation formula for metalness is:

[0029] Metallic=0.5+0.5×sin(2π×time / 24);

[0030] Where Metallic represents metallicity; time represents the current time, and its value range is [1, 24].

[0031] Another aspect of the present invention provides a three-dimensional visual simulation construction system, the system comprising:

[0032] An acquisition module is used to obtain geometric parameters of a scene model and construct a material-free scene model based on a parametric model generation component, wherein the geometric parameters include basic size, number of segments, and curvature radius; the parametric model generation component includes a rigid template library and a flexible template library; and the scene model includes a rescue vehicle model, a landslide mountain model, and a landslide rock model;

[0033] A setting module, configured to perform material setting on the material-free scene model according to a dynamic material component to obtain a dynamic material scene model, wherein the dynamic material component includes a basic material layer and an environment response layer;

[0034] A construction module is used to animate the dynamic material scene model according to the modular animation component to obtain a scene model with animation and dynamic materials, so as to construct simulation resources according to the scene model with animation and dynamic materials.

[0035] Another aspect of the present invention provides a computer-readable storage medium having a computer program stored thereon, which implements the above-mentioned three-dimensional visual scene simulation construction method when executed by a processor.

[0036] Another aspect of the present invention provides a data processing device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the above-mentioned three-dimensional visual scene simulation construction method when executing the program. BRIEF DESCRIPTION OF THE DRAWINGS

[0037] Figure 1 Flowchart of a method for constructing a three-dimensional visual scene simulation according to a first embodiment of the present invention;

[0038] The following specific embodiments will further illustrate the present invention in conjunction with the above-mentioned drawings. DETAILED DESCRIPTION

[0039] To facilitate understanding of the present invention, the present invention will be described more fully below with reference to the accompanying drawings. The drawings illustrate several embodiments of the present invention. However, the present invention may be implemented in many different forms and is not limited to the embodiments described herein. Rather, these embodiments are provided to provide a more thorough and comprehensive understanding of the present invention.

[0040] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one skilled in the art to which this invention pertains. The terms used in this specification of the present invention are for the purpose of describing specific embodiments only and are not intended to limit the present invention. The term "and / or" as used herein includes any and all combinations of one or more of the associated listed items.

[0041] In order to solve the technical problems in the existing technology of low model resource production efficiency of three-dimensional visual simulation models, the inability to dynamically adapt static binding of materials and textures, and the excessive coupling of animation resources, the present application provides a three-dimensional visual simulation construction method and system. Through dynamic material generation technology, compared with traditional model mapping, dynamic materials are generated using adjustable parameters without the need for static binding to textures, and dynamic materials can achieve dynamic effects according to the parameters of the environmental response layer, thereby shortening the material generation time while improving the model appearance effect; secondly, the present invention modularizes animation and decouples resources, with high reusability and flexibility, replacing the traditional manual modeling that requires manual adjustment of each character joint, resulting in the transition between animations being often stiff and unnatural; furthermore, the process parameters in the technical solution of the present invention are automated, greatly improving the model production efficiency; it solves the technical problems in the existing technology of low model resource production efficiency, the inability to dynamically adapt static binding of materials and textures, and the excessive coupling of animation resources.

[0042] To facilitate understanding of the present invention, several embodiments of the present invention are provided below. However, the present invention can be implemented in many different forms and is not limited to the embodiments described herein. Rather, these embodiments are provided to provide a more thorough and comprehensive disclosure of the present invention.

[0043] Example 1

[0044] See also Figure 1 , which shows a method for constructing a three-dimensional visual scene simulation in a first embodiment of the present invention, and includes steps S101 to S103:

[0045] S101: Acquire geometric parameters of a scene model, and construct a material-free scene model based on component generation of the parameterized model.

[0046] In this embodiment, the geometric parameters include basic size, number of segments and curvature radius; the parameterized model generation component includes a rigid template library and a flexible template library; the scene model includes a rescue vehicle model, a landslide mountain model and a landslide rock model.

[0047] Specifically, this solution includes a parametric model generation component, a dynamic material component, and a modular animation component. The parametric model generation component primarily consists of a structural template library and model instantiation. The structural template library categorizes and stores existing basic models to form a template library. Non-existent model categories must be added in advance. As the number of templates in the structural template library increases, the speed of model parameterization generation also increases. The structural template library contains a variety of basic model templates, including rigid and flexible structure templates. The parameters of the basic model templates have a constraint mechanism, specifically implemented using a parameter constraint algorithm to determine the rationality of the parameter settings. Rigid structures include buildings and vehicles, while flexible structures include vegetation and fluids. The instantiated model is generated using the instantiated model generation algorithm.

[0048] Specifically, regarding rigid structure templates: by setting the model geometric parameters of the rigid structure, users can obtain the desired rigid template, where the geometric parameters include basic dimensions (including length, width, and height), number of segments (controlling surface smoothness), and curvature radius (affecting the degree of curvature of the arc structure). For example, when using a building template, users can adjust the number of floors (1 to 50 floors), the density of windows on each floor (0.1 / m2), and the number of windows per floor (0.1 / m2). 2 ~1.0 piece / m 2 ) and building roof types (including: flat roof, spire, dome).

[0049] Furthermore, regarding the flexible structure template: the flexible structure is generated by a physical simulation algorithm. For example, the vegetation template uses a fractal algorithm to generate the branch distribution. The parameters that can be set include the branch bifurcation angle (15°~45°), the layer depth between each branch (3~7 levels), the branch leaf density (100 pieces / m 2 ~500 pieces / m 2 ) and the health of the leaves (color, leaf integrity).

[0050] Furthermore, the parameter constraint mechanism dynamically verifies physical plausibility. For example, the width-to-height ratio of a real-life vehicle chassis should be limited to 1:1.5 to 1:3. Models generated outside this ratio will experience physics simulation failure during runtime due to imbalanced center of gravity during the engine's physics calculations. Furthermore, parameter constraints can eliminate interpolation between component models due to interpolation caused by component size. This eliminates interpolation by using a mesh retopology algorithm, offsetting vertices and reconstructing the meshes.

[0051] After setting the parameters, an instanced model needs to be generated. The model generated by the instanced model generation algorithm requires control over the model's refinement level (e.g., number of vertices and triangles) and UV mapping. In simulation training, it is often necessary to use models of varying refinement based on their distance from the camera. This technique is known as multi-level detail (LOD). When the model is farther from the camera, details are unclear, so the model is displayed as a low-poly model. This reduces the computational and rendering pressure on the graphics card, thereby improving program performance. LOD (Level of Detail) chains are created during model generation. Each level is dynamically switched based on specific rules, such as from high-poly to low-poly. The face count is calculated using a differential edge collapse algorithm. For low-poly models, only key feature points, such as building outlines, are retained. High-poly models are defined as LOD0, with more than 100,000 faces; low-poly models are defined as LOD2, with fewer than 5,000 faces. The materials also change with the level. The reflectivity of the high-poly material switches from PBR (Physically Based Rendering) to the low-poly Phong model, and the texture resolution is gradually reduced from 4K to 256×256.

[0052] S102: Setting the material of the non-material scene model according to the dynamic material component to obtain a dynamic material scene model.

[0053] Dynamic material components process materials in layers, allowing for real-time composite rendering during program execution. This invention divides dynamic material components into a base material layer and an environment response layer. The rendering process utilizes a cache reuse mechanism and multi-threaded material compilation technology. The cache reuse mechanism caches static materials (such as building exterior walls) between frames, significantly reducing the GPU's computational load while increasing reuse rates. Multi-threaded material compilation technology significantly improves material synthesis and output efficiency while ensuring thread consistency. The base material layer and environment response layer are described below:

[0054] Specifically, the base material layer stores non-dynamic properties, including base color (Albedo), metallic, and roughness. It uses HDR compression to reduce video memory usage. The approximate formula for calculating metallic is: metallic = 0.5 + 0.5 × sin(2π × time / 24), where metallic represents metallic; time represents the current time, with a value range of [1, 24]. Controlling the time input dynamically changes the material of a metal model, simulating the effect of a metal model's surface changing over time.

[0055] Secondly, the environmental response layer is implemented based on real-time environmental information. For example, the interaction between the model surface and the lighting is calculated. The calculation of the intensity of the highlight reflection is based on the angle between the forward vector of the rendering camera and the normal. The approximate calculation formula is: R(θ)=R+(1-R)(1-cosθ) 5 , where R is the base reflectivity of the material and θ is the angle of incidence. Ambient Occlusion (AO) is achieved through a blend of precomputed baking and real-time screen-space AO. Furthermore, in rainy and snowy environments, the material blends normal map details based on the scene's wetness parameter (0-1), reducing surface roughness to better reflect the wetness of rain and snow.

[0056] S103 , animating the dynamic material scene model according to the modular animation component to obtain a scene model with animation and dynamic materials, and constructing simulation resources according to the scene model with animation and dynamic materials.

[0057] The modular animation component mainly consists of three functions: action atomization, physical constraints, and real-time animation synthesis. For static models, the parametric model generation component and the dynamic material component are sufficient for construction, while models with skeletal animation require the modular animation component.

[0058] Specifically, action atomization involves breaking down a complex animation into its most basic atoms, such as a character's arm swing, lower limb walking, or neck rotation. This involves using each character's joints as fundamental atoms and adjusting the movements of each atom to complete a complex action. Physical constraints limit the range of motion of each atom. For example, a normal adult's neck rotation angle is typically 60° to 80° left and right. Physical constraints control the left and right rotation angles within this range in real time. Real-time animation synthesis refers to the real-time transition between different actions. Linear interpolation is used to transition actions in a one-dimensional blend space, and the character speed parameter controls the blending synthesis speed of the animation. Among them, the one-dimensional blend space is for atoms to change posture only in two-axis space; the transition action includes the transition from walking forward to running forward; complex action fusion is used in the two-dimensional blend space, and the skeleton posture is adjusted by dual-parameter interpolation, among which complex action fusion includes turning during jumping; the dual parameters are horizontal speed and vertical speed; the inverse dynamics algorithm is used for animation synthesis in specific scenarios; among which, the specific scenarios include scenarios where the end position and posture of an atom are known; the inverse dynamics algorithm is used to automatically calculate the rotation of the intermediate atoms by specifying the position and rotation of the terminal atoms, among which the terminal atoms include hands and feet, and the intermediate atoms include elbows and knees.

[0059] As a specific example, in a simulation training system for landslide rescue, multiple vehicles are required to jointly transport materials and manpower for rescue work. When the required vehicle models are lacking, the technical solution of the present invention is used to generate all vehicles according to the following process, specifically:

[0060] First, call the vehicle template library in the rigid structure template and input parameters such as length, width, and height to generate a new vehicle model; if the vehicle template does not exist in the vehicle template library, use other models to modify or remodel a simple vehicle template and import it into the library.

[0061] Next, we set the base color, metalness, and roughness parameters to generate a "green camouflage" base material layer. The environmental material in the environmental response layer uses a mud pollution layer with parameters such as low light, high humidity, and moderate air temperature. The two materials are dynamically fused to generate a car body material with mud stains.

[0062] Furthermore, vehicle movement requires animation support. Through modular animation components, atomic animations of wheels, suspension, exhaust, etc. are synthesized, and physical constraints are applied to each atom to play animations while the vehicle is moving. Among them, physical constraints include the steering range of wheel atoms.

[0063] Repeat the above steps to generate all vehicles required for simulation rescue training.

[0064] In addition, the simulation needs to simulate landslide scenarios under a variety of different situations. Using the technical solution in the present invention, there is no need to model multiple times. The required model can be generated according to the following steps to perform landslide simulation.

[0065] Specifically:

[0066] First, the landslide is a fluid. The fluid in the flexible template is called and the slope, soil type and sliding area in the mountain model parameters are set to 35°, clay and 200m respectively. 2 ;

[0067] Secondly, there will be rock mixing during the landslide process. The stone template library in the rigid structure template is called, and the rock model parameters are set to the rock size range and quantity range to dynamically generate rock models with random size, random quantity, and irregular shape according to the input parameters;

[0068] Secondly, dynamic material parameter settings: input base color, metalness, roughness and reflectivity to generate the base material layer, input air humidity, soil moisture and light intensity to generate the environment material layer in the environment response layer, and the two material layers are mixed in real time;

[0069] Furthermore, landslides require landslide animations, while rock and soil landslides require real-time calculations based on the gravity and friction of the physics engine. A physics simulation module is added to the simulation engine so that the modules can collaborate with each other to complete the landslide animation playback when the program is running. The physics simulation module includes a gravity module, a collision module, and a rolling module.

[0070] By adjusting the above parameters, landslide simulation can be performed under different conditions.

[0071] Furthermore, the technical solution of the present invention offers the advantage of high cross-platform compatibility. Through parametric modeling, the same resource is guaranteed to perform consistently across different rendering engines and hardware platforms. Furthermore, model material parameters use standardized ranges, such as a metalness range of 0 to 1, to avoid visual distortion caused by engine differences. Traditional manual modeling requires separate export for different platforms, and various parameter adjustments must be made before export to adapt to the target engine, which is time-consuming and labor-intensive.

[0072] The technical solution of the present invention is used to compare the expansion of the simulation model with the traditional method. The data are shown in Table 1:

[0073] Table 1:

[0074]

[0075] In summary, the three-dimensional visual scene simulation construction method in the above-mentioned embodiment of the present invention, through dynamic material generation technology, is compared with traditional model mapping. The dynamic material is generated in a manner with adjustable parameters without static binding to the map, and the dynamic material can achieve dynamic effects according to the parameters of the environmental response layer, thereby shortening the material generation time while improving the model appearance effect; secondly, the present invention modularizes the animation and decouples the resources, which has high reusability and flexibility, replacing the traditional manual modeling that requires manual adjustment of each character joint, resulting in the transition between animations being often stiff and unnatural; furthermore, the process parameters in the technical solution of the present invention are automated, which greatly improves the model production efficiency; it solves the technical problems in the prior art of low model resource production efficiency of the three-dimensional visual scene simulation model, the inability to dynamically adapt the static binding of materials and maps, and the excessive coupling of animation resources.

[0076] Example 2

[0077] A second embodiment of the present invention provides a three-dimensional visual simulation construction system, comprising:

[0078] An acquisition module is used to obtain geometric parameters of a scene model and construct a material-free scene model based on a parametric model generation component, wherein the geometric parameters include basic size, number of segments, and curvature radius; the parametric model generation component includes a rigid template library and a flexible template library; and the scene model includes a rescue vehicle model, a landslide mountain model, and a landslide rock model;

[0079] A setting module, configured to perform material setting on the material-free scene model according to a dynamic material component to obtain a dynamic material scene model, wherein the dynamic material component includes a basic material layer and an environment response layer;

[0080] A construction module is used to animate the dynamic material scene model according to the modular animation component to obtain a scene model with animation and dynamic materials, so as to construct simulation resources according to the scene model with animation and dynamic materials.

[0081] In summary, the three-dimensional visual simulation construction system in the above-mentioned embodiment of the present invention, through dynamic material generation technology, is compared with traditional model mapping. The dynamic material is generated in a manner with adjustable parameters without static binding to the map, and the dynamic material can achieve dynamic effects according to the parameters of the environmental response layer, thereby shortening the material generation time while improving the model appearance effect; secondly, the present invention modularizes the animation and decouples the resources, which has high reusability and flexibility, replacing the traditional manual modeling that requires manual adjustment of each character joint, resulting in the transition between animations being often stiff and unnatural; furthermore, the process parameters in the technical solution of the present invention are automated, which greatly improves the model production efficiency; it solves the technical problems in the prior art of low model resource production efficiency of the three-dimensional visual simulation model, the inability to dynamically adapt the static binding of materials and maps, and the excessive coupling of animation resources.

[0082] In addition, an embodiment of the present invention further provides a computer-readable storage medium having a computer program stored thereon, which implements the steps of the method in the above embodiment when the program is executed by a processor.

[0083] In addition, an embodiment of the present invention further provides a data processing device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the steps of the method in the above embodiment when executing the program.

[0084] The logic and / or steps represented in the flowcharts or otherwise described herein, for example, can be considered as a sequenced list of executable instructions for implementing the logical functions, and can be embodied in any computer-readable medium for use by, or in conjunction with, an instruction execution system, apparatus, or device (e.g., a computer-based system, a system including a processor, or other system that can fetch and execute instructions from an instruction execution system, apparatus, or device). For purposes of this specification, a "computer-readable medium" can be any device that can contain, store, communicate, propagate, or transport a program for use by, or in conjunction with, an instruction execution system, apparatus, or device.

[0085] More specific examples (a non-exhaustive list) of computer-readable media include the following: an electrical connection with one or more wires (electronic devices), a portable computer disk cartridge (magnetic devices), random access memory (RAM), read-only memory (ROM), erasable and programmable read-only memory (EPROM or flash memory), fiber optic devices, and portable compact disk read-only memory (CDROM). In addition, the computer-readable medium may even be paper or other suitable medium on which the program is printed, since the program may be obtained electronically, for example, by optically scanning the paper or other medium and then editing, interpreting, or processing it in another suitable manner as necessary, and then storing it in a computer memory.

[0086] It should be understood that various components of the present invention may be implemented using hardware, software, firmware, or a combination thereof. In the above-described embodiments, multiple steps or methods may be implemented using software or firmware stored in a memory and executed by a suitable instruction execution system. For example, if implemented using hardware, as in another embodiment, any one of the following technologies known in the art or a combination thereof may be used: a discrete logic circuit having logic gate circuits for implementing logic functions on data signals, an application-specific integrated circuit having suitable combinational logic gate circuits, a programmable gate array (PGA), a field-programmable gate array (FPGA), etc.

[0087] Throughout this specification, reference to terms such as "one embodiment," "some embodiments," "examples," "specific examples," or "some examples" means that a specific feature, structure, material, or characteristic described in connection with that embodiment or example is included in at least one embodiment or example of the present invention. In this specification, schematic representations of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in any one or more embodiments or examples.

[0088] While embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions, and variations may be made to the embodiments without departing from the principles and spirit of the invention, and that the scope of the invention is defined by the claims and their equivalents.

Claims

1. A three-dimensional visual simulation construction method, characterized in that: include: Acquire geometric parameters of the scene model and construct a material-free scene model based on a parametric model generation component, wherein the geometric parameters include basic size, number of segments, and curvature radius; the parametric model generation component includes a rigid template library and a flexible template library; and the scene model includes a rescue vehicle model, a landslide mountain model, and a landslide rock model; Setting the material of the material-free scene model according to the dynamic material component to obtain a dynamic material scene model, wherein the dynamic material component includes a basic material layer and an environment response layer; Animating the dynamic material scene model according to the modular animation component to obtain a scene model with animation and dynamic materials, and constructing a simulation resource according to the scene model with animation and dynamic materials; The modular animation components include action atomization, physical constraints, and real-time animation synthesis, wherein the real-time animation synthesis represents the real-time transition between different actions; The steps of animating the dynamic material scene model according to the modular animation component to obtain a scene model with animation and dynamic materials include: Use linear interpolation in a one-dimensional blend space to transition the animation. The character speed parameter controls the speed of the blended animation. The one-dimensional blend space is an atom that changes posture only in two axes. The transition action includes transitioning from walking forward to running forward. Complex motion fusion is used in a two-dimensional hybrid space to adjust the skeleton posture through dual-parameter interpolation. Complex motion fusion includes turning during jumping; the dual parameters are horizontal speed and vertical speed. Animation synthesis is performed using an inverse dynamics algorithm in a specific scenario; wherein the specific scenario includes a scenario in which the end position and posture of an atom are known; the inverse dynamics algorithm is used to automatically calculate the rotation of intermediate atoms by specifying the position and rotation of the end atoms, wherein the end atoms include hands and feet, and the intermediate atoms include elbows and knees.

2. The method for constructing a three-dimensional visual scene simulation according to claim 1, wherein: The step of animating the dynamic material scene model according to the modular animation component to obtain a scene model with animation and dynamic material also includes: In the dynamic material mountain model and the dynamic material rock model, real-time calculations are performed based on the physical engine, and a physical simulation module is added to the simulation engine so that each physical simulation module cooperates with each other to complete the landslide animation playback. Among them, the physical simulation module includes a gravity module, a collision module and a rolling module.

3. The method for constructing a three-dimensional visual scene simulation according to claim 1, wherein: In the steps of obtaining the geometric parameters of the scene model and generating components based on the parameterized model to construct a material-free scene model: A material-free scene model is obtained according to the parameter constraint algorithm and the instantiation model generation algorithm.

4. The method for constructing a three-dimensional visual simulation according to claim 1, wherein: In the step of setting the material of the material-free scene model according to the dynamic material component to obtain the dynamic material scene model: A dynamic material scene model is obtained according to the basic material layer parameters, the environmental response layer parameters and the dynamic material mixing algorithm, wherein the basic material layer parameters include basic color, metalness and roughness; the environmental response layer parameters include light intensity, ambient humidity and air temperature.

5. The method for constructing a three-dimensional visual simulation according to claim 4, wherein: In the environmental response layer parameters, the calculation formula for the high light reflection intensity is: R(θ)=R+(1-R)(1-cosθ) 5 ; Where: R(θ) represents the intensity of specular reflection; R is the base reflectivity of the material, and θ is the angle of incidence.

6. The method for constructing a three-dimensional visual scene simulation according to claim 4, wherein: In the base material layer parameters, the calculation formula for metalness is: Metallic=0.5+0.5×sin(2π×time / 24); Where Metallic represents metallicity; time represents the current time, and its value range is [1, 24].

7. A three-dimensional visual simulation construction system, characterized in that: The system comprises: An acquisition module is used to obtain geometric parameters of a scene model and construct a material-free scene model based on a parametric model generation component, wherein the geometric parameters include basic size, number of segments, and curvature radius; the parametric model generation component includes a rigid template library and a flexible template library; and the scene model includes a rescue vehicle model, a landslide mountain model, and a landslide rock model; A setting module, configured to perform material setting on the material-free scene model according to a dynamic material component to obtain a dynamic material scene model, wherein the dynamic material component includes a basic material layer and an environment response layer; A construction module, configured to animate the dynamic material scene model according to the modular animation component to obtain a scene model with animation and dynamic materials, and to construct simulation resources according to the scene model with animation and dynamic materials; The modular animation components include action atomization, physical constraints, and real-time animation synthesis, wherein the real-time animation synthesis represents the real-time transition between different actions; The steps of animating the dynamic material scene model according to the modular animation component to obtain a scene model with animation and dynamic materials include: Use linear interpolation in a one-dimensional blend space to transition the animation. The character speed parameter controls the speed of the blended animation. The one-dimensional blend space is an atom that changes posture only in two axes. The transition action includes transitioning from walking forward to running forward. Complex motion fusion is used in a two-dimensional hybrid space to adjust the skeleton posture through dual-parameter interpolation. Complex motion fusion includes turning during jumping; the dual parameters are horizontal speed and vertical speed. Animation synthesis is performed using an inverse dynamics algorithm in a specific scenario; wherein the specific scenario includes a scenario in which the end position and posture of an atom are known; the inverse dynamics algorithm is used to automatically calculate the rotation of intermediate atoms by specifying the position and rotation of the end atoms, wherein the end atoms include hands and feet, and the intermediate atoms include elbows and knees.

8. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the program is executed by a processor, the three-dimensional visual scene simulation construction method according to any one of claims 1 to 6 is implemented.

9. A data processing device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein: When the processor executes the program, the three-dimensional visual scene simulation construction method according to any one of claims 1 to 6 is implemented.

Images