Instant light color rendering system for game scene

Abstract

The invention belongs to the technical field of game rendering, and discloses an instant light color rendering system for a game scene, which comprises the following steps: collecting scene illumination information based on a game engine, and generating initial illumination data; carrying out heterogeneous light source data fusion processing to generate integrated illumination data; performing multi-dimensional illumination attribute clustering analysis and feature extraction to generate illumination feature data; performing scene region division processing, and constructing region illumination feature data; performing regional material-illumination interaction feature analysis to obtain interaction feature data; performing illumination propagation characteristic analysis to generate propagation characteristic data; designing a region rendering priority prediction engine; transmitting the feature data in real time to perform rendering demand intelligent analysis; performing global instant light color rendering adaptive decision analysis to form a rendering decision; adjusting the initial illumination data according to the decision, and executing instant light color rendering of the game scene; and the visual expressive force and the rendering performance of the game are greatly improved.

Description

Technical Field

[0001] The present invention relates to the field of game rendering technology, and more specifically, to a real-time light color rendering system for game scenes. Background Art

[0002] Existing real-time light and color rendering systems for game scenes have significant deficiencies in dealing with multiple dynamic light source changes and accurate rendering of visual focus areas in open world environments. In particular, in dealing with scenes of sunrise and sunset transitions, extreme weather events, or complex interactions between light sources and media, traditional rendering methods cannot effectively capture the subtle gradual changes in light intensity and color. Existing technologies usually use global unified rendering parameters and cannot perform differentiated processing based on the importance and particularity of scene areas, resulting in irrational resource allocation, or missing details in key visual areas, or excessive consumption of computing resources in non-critical areas. Existing technologies lack the ability to intelligently correlate the characteristics of light propagation and material properties, and cannot make adaptive rendering decisions based on the unique response characteristics of different materials to light, resulting in difficulty in achieving a good balance between rendering quality and performance when dealing with complex scenes with highly heterogeneous material distributions. In particular, in high-end games that require accurate simulation of the effects of light propagation in multiple media (such as translucent fabrics, multi-layered water bodies, and volumetric smoke), traditional rendering systems cannot meet the dual requirements of physical accuracy and visual expression.

[0003] In view of this, the present invention proposes a real-time light color rendering system for game scenes to solve the above problems. Summary of the invention

[0004] In order to overcome the above-mentioned defects of the prior art and to achieve the above-mentioned purpose, the present invention provides the following technical solution: a real-time light color rendering system for a game scene, comprising: Acquisition and processing module: collect scene lighting information based on the game engine to generate initial lighting data; perform heterogeneous light source data fusion processing based on the initial lighting data to generate integrated lighting data; Data extraction module: perform multi-dimensional illumination attribute clustering analysis based on the integrated illumination data to generate multi-dimensional illumination attribute clustering data; perform multi-dimensional illumination feature extraction based on the multi-dimensional illumination attribute clustering data to generate multi-dimensional illumination feature data; Analysis module: Perform scene area division processing based on multi-dimensional illumination feature data to generate regional multi-dimensional illumination feature data; perform regional material-illumination interaction feature analysis based on regional multi-dimensional illumination feature data to generate regional material-illumination interaction feature data; Immediate rendering module: perform lighting propagation feature analysis based on regional multi-dimensional lighting feature data to generate lighting propagation feature data; design an intelligent judgment engine for predicting regional rendering priorities based on regional material-light interaction feature data to obtain a regional rendering priority prediction engine; transmit the lighting propagation feature data to the regional rendering priority prediction engine in real time for intelligent analysis of regional dynamic rendering requirements to generate regional dynamic rendering requirement data; perform adaptive decision-making analysis on global immediate light color rendering based on the regional dynamic rendering requirement data to generate a global immediate light color rendering decision; perform immediate light color rendering adjustment processing on the initial lighting data according to the global immediate light color rendering decision to generate immediate light color rendering data, and perform immediate light color rendering of the game scene through the immediate light color rendering data.

[0005] Further, collect scene lighting information based on the game engine to generate initial lighting data; perform heterogeneous light source data fusion processing on the initial lighting data to generate integrated lighting data, including: Collect scene lighting information based on the game engine to generate initial lighting data; Perform light source type recognition processing on the initial lighting data to generate light source type feature data; Design a heterogeneous light source representation model based on the light source type feature data to generate a heterogeneous light source representation model; Perform heterogeneous light source data fusion processing based on the heterogeneous light source representation model to generate integrated lighting data.

[0006] Further, perform multi-dimensional lighting attribute clustering analysis on the integrated lighting data to generate multi-dimensional lighting attribute clustering data; perform multi-dimensional lighting feature extraction on the multi-dimensional lighting attribute clustering data to generate multi-dimensional lighting feature data, including: Extract lighting intensity feature data from the integrated lighting data; Extract spectral distribution feature data from the integrated lighting data; Use a preset multi-dimensional clustering algorithm to perform initial multi-dimensional lighting attribute clustering analysis on the lighting intensity feature data and the spectral distribution feature data to generate initial multi-dimensional lighting attribute clustering data; Perform visual-unrelated lighting data screening processing based on the initial multi-dimensional lighting attribute clustering data to generate multi-dimensional lighting attribute clustering data; Perform multi-dimensional lighting feature extraction on the multi-dimensional lighting attribute clustering data to generate multi-dimensional lighting feature data.

[0007] Further, perform visual-unrelated lighting data screening processing based on the initial multi-dimensional lighting attribute clustering data to generate multi-dimensional lighting attribute clustering data, including: Performing visual perception threshold analysis on the initial multi-dimensional lighting attribute clustering data to generate visual perception threshold data, where the visual perception threshold data includes lighting intensity threshold data and spectral sensitivity data; Performing low-impact light source aggregation processing based on the lighting intensity threshold data to generate aggregated low-impact light source data; Performing visual redundant lighting recognition processing based on the aggregated low-impact light source data and the spectral sensitivity data to generate visual redundant lighting data; Performing visual redundant lighting multi-dimensional lighting attribute clustering data screening processing on the initial multi-dimensional lighting attribute clustering data according to the visual redundant lighting data to generate multi-dimensional lighting attribute clustering data, and transmitting the visual redundant lighting data to the rendering engine to execute a simplified rendering job.

[0008] Further, performing scene area division processing based on the multi-dimensional lighting feature data to generate regional multi-dimensional lighting feature data; performing regional material-light interaction feature analysis based on the regional multi-dimensional lighting feature data to generate regional material-light interaction feature data, including: Performing scene area division processing based on the multi-dimensional lighting feature data to generate regional multi-dimensional lighting feature data; Performing regional material property mapping analysis based on the regional multi-dimensional lighting feature data to generate regional material property data; Performing regional material-light interaction feature analysis based on the regional material property data to generate regional material-light interaction feature data.

[0009] Further, the performing regional material property mapping analysis based on the regional multi-dimensional lighting feature data to generate regional material property data includes: Performing surface material reflection property analysis based on the regional multi-dimensional lighting feature data to generate surface material reflection property data; Performing surface material scattering property analysis based on the regional multi-dimensional lighting feature data to generate surface material scattering property data; Performing material-light interaction response matrix calculation based on the surface material reflection property data and the surface material scattering property data to generate a material-light interaction response matrix; Performing regional material property mapping analysis based on the material-light interaction response matrix to generate regional material property data.

[0010] Further, the performing regional material property mapping analysis based on the material-light interaction response matrix to generate regional material property data includes: Performing physically based rendering model feature analysis on the regional material property data to generate physically based rendering model feature data; Perform regional material-light interaction feature analysis based on the feature data of the physically based rendering model to generate regional material-light interaction feature data.

[0011] Furthermore, perform light propagation feature analysis based on the regional multi-dimensional light feature data to generate light propagation feature data; design an intelligent judgment engine for predicting the regional rendering priority based on the regional material-light interaction feature data to obtain a regional rendering priority prediction engine; transmit the light propagation feature data to the regional rendering priority prediction engine in real time for intelligent analysis of the regional dynamic rendering requirements to generate regional dynamic rendering requirement data; perform adaptive decision-making analysis for global real-time light color rendering based on the regional dynamic rendering requirement data to generate a global real-time light color rendering decision; perform real-time light color rendering adjustment processing on the initial light data according to the global real-time light color rendering decision to generate real-time light color rendering data, and perform real-time light color rendering of the game scene through the real-time light color rendering data, including: Perform light propagation feature analysis based on the regional multi-dimensional light feature data to generate light propagation feature data; Perform visual importance scoring node analysis based on the regional material-light interaction feature data to generate visual importance scoring node data, and design a decision tree priority node through the visual importance scoring node data; Design a neural network rendering priority tree for light effect and rendering resource allocation based on the preset deep neural network model for the decision tree priority node and the regional material-light interaction feature data to obtain a rendering priority neural network model; Design an intelligent judgment engine for predicting the regional rendering priority according to the rendering priority neural network model to obtain a regional rendering priority prediction engine; Transmit the light propagation feature data to the regional rendering priority prediction engine in real time for intelligent analysis of the regional dynamic rendering requirements to generate regional dynamic rendering requirement data; Perform adaptive decision-making analysis for global real-time light color rendering based on the regional dynamic rendering requirement data to generate a global real-time light color rendering decision; Perform real-time light color rendering adjustment processing on the initial light data according to the global real-time light color rendering decision to generate real-time light color rendering data, and perform real-time light color rendering of the game scene through the real-time light color rendering data.

[0012] Furthermore, the design of the neural network rendering priority tree for light effect and rendering resource allocation based on the preset deep neural network model for the decision tree priority node and the regional material-light interaction feature data to obtain a rendering priority neural network model includes: Design of a neural network rendering priority tree architecture for allocating lighting effects and rendering resources to decision tree priority nodes based on a preset deep neural network model, and generating a rendering priority neural network model architecture; Perform user visual attention prediction analysis based on regional material-light interaction feature data to generate user visual attention prediction data; Perform priority feature weight learning processing on the lighting effects and rendering resources of the rendering priority neural network model architecture according to the user visual attention prediction data to generate a rendering priority neural network model.

[0013] An instant color rendering device for a game scene, characterized in that the instant color rendering device for a game scene includes: a memory and at least one processor, and instructions are stored in the memory; the at least one processor calls the instructions in the memory so that the instant color rendering device for a game scene executes an instant color rendering method for a game scene.

[0014] The technical effects and advantages of an instant color rendering system for a game scene according to the present invention: Through lighting data collection and light source type identification, the present invention establishes a heterogeneous light source representation model to achieve unified management and coordinated rendering of multiple types of light sources. The system uses multi-dimensional lighting attribute clustering analysis and feature extraction to accurately capture changes in lighting intensity and spectral distribution in the scene, and through a visual-irrelevant lighting data screening mechanism, effectively eliminates redundant lighting information that has a weak impact on the visual effect, significantly improving the rendering efficiency. Based on scene area division and material-light interaction feature analysis, the system realizes accurate lighting rendering for different regional material characteristics, and through physical-based rendering model analysis, presents a more realistic lighting effect. The regional rendering priority prediction engine analyzes the lighting propagation characteristics in real time, and intelligently allocates rendering resources in combination with a deep neural network model to ensure high-efficiency processing capabilities when the scene dynamically changes. The system generates an adaptive rendering decision according to the user visual attention prediction data and regional dynamic rendering requirement analysis, preferentially processes visually important regions, and effectively reduces the consumption of computing resources while ensuring high-quality visual performance. This adaptive instant color rendering technology for heterogeneous light source data and complex rendering requirements realizes a more efficient and realistic lighting effect in the game scene. Brief Description of the Drawings

[0015] Figure 1 It is a schematic diagram of an instant color rendering system for a game scene according to the present invention; Figure 2 It is a schematic diagram of an instant color rendering method for a game scene according to the present invention. Detailed Embodiments

[0016] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0017] Embodiment 1; Please refer to Figure 1 As shown, a real-time light color rendering system for a game scene in this embodiment includes: an acquisition and processing module: collecting scene lighting information based on a game engine to generate initial lighting data; performing heterogeneous light source data fusion processing according to the initial lighting data to generate integrated lighting data; A data extraction module: performing multi-dimensional lighting attribute clustering analysis according to the integrated lighting data to generate multi-dimensional lighting attribute clustering data; performing multi-dimensional lighting feature extraction according to the multi-dimensional lighting attribute clustering data to generate multi-dimensional lighting feature data; A feature analysis module: performing scene area division processing according to the multi-dimensional lighting feature data to generate area multi-dimensional lighting feature data; performing area material-light interaction feature analysis based on the area multi-dimensional lighting feature data to generate area material-light interaction feature data; A real-time rendering module: performing lighting propagation feature analysis according to the area multi-dimensional lighting feature data to generate lighting propagation feature data; designing an intelligent judgment engine for area rendering priority prediction based on the area material-light interaction feature data to obtain an area rendering priority prediction engine; transmitting the lighting propagation feature data to the area rendering priority prediction engine in real time for intelligent analysis of area dynamic rendering requirements to generate area dynamic rendering requirement data; performing adaptive decision analysis of global real-time light color rendering according to the area dynamic rendering requirement data to generate a global real-time light color rendering decision; performing real-time light color rendering adjustment processing on the initial lighting data according to the global real-time light color rendering decision to generate real-time light color rendering data, and performing real-time light color rendering of the game scene through the real-time light color rendering data.

[0018] Preferably, collecting scene lighting information based on a game engine to generate initial lighting data; performing heterogeneous light source data fusion processing according to the initial lighting data, including: (1) Collecting scene lighting information based on a game engine to generate initial lighting data; (2) Performing light source type recognition processing according to the initial lighting data to generate light source type feature data; (3) Designing a heterogeneous light source representation model according to the light source type feature data to generate a heterogeneous light source representation model; (4)Perform heterogeneous light source data fusion processing according to the heterogeneous light source representation model to generate integrated illumination data.

[0019] In the embodiment of the present invention, a light collection framework is deployed in the game engine, supporting six basic light source types: directional light, point light, spotlight, area light, ambient light, and volume light. The collection framework captures light parameters every frame, including intensity (lumen value), RGB color value, attenuation coefficient, illumination range (and shadow resolution). The system collects data through the UpdateLightData() function of the engine before each frame rendering, with a sampling frequency of 60Hz to ensure data real-time. The collected data is stored in a 512KB circular buffer indexed by light source ID and timestamp, allowing retrospective analysis of light changes in the most recent 240 frames to generate initial illumination data containing a complete parameter set. The light source type recognition processing uses a feature matching algorithm, setting up a recognition rule library: directional light source (no attenuation, unidirectional), point light source (omnidirectional attenuation, point emission), spotlight (conical attenuation, angle limit), etc. The system executes a three-step recognition process: parameter pattern matching, outlier detection, and type confirmation. For hybrid light sources, the dominant characteristic determination method is used to determine their main type. The system sets parameter rationality thresholds: light intensity range from 0 to 8000 lumens, color temperature range from 1500K to 15000K, and parameters outside the range are clipped. The recognition accuracy reaches 98.5%, and standardized light source type feature data is output. The heterogeneous light source representation model adopts a hierarchical design: basic illumination layer, spatial distribution layer, and time-varying characteristic layer. The basic layer uses unified photometric parameters (luminous flux, luminous intensity, illuminance) to describe all light sources; the spatial layer uses spherical harmonic functions (3rd order, 16 coefficients) to represent complex illumination distributions; the time-varying layer uses spline interpolation curves to record the change of light parameters over time. The model supports four attenuation functions: linear, inverse square, exponential, and custom curves, controlled parametrically. The model verification test is carried out in four scenarios (outdoor sunlight, indoor mixed light, dynamic weather, night city), with the spatial accuracy deviation less than 3% and the time consistency score reaching 4.8, generating a standardized heterogeneous light source representation model. The heterogeneous light source data fusion adopts a partition processing strategy, dividing the scene into a voxel grid of 8×8×4. The individual contributions of six light source types are calculated within each voxel, and the physical light superposition principle (linear superposition + saturation control) is applied for synthesis. The fusion process includes direct illumination synthesis, indirect reflection calculation, and volume light scattering processing, with a calculation accuracy of 16-bit floating point. The system sets a light importance scoring index, ignoring secondary light sources with contributions less than 2% to improve performance. The fused illumination data is stored in the form of a light probe grid with a resolution of 32³, covering the entire scene space, generating globally consistent integrated illumination data.

[0020] Preferably, multi-dimensional light attribute clustering analysis is performed based on integrated light data to generate multi-dimensional light attribute clustering data; multi-dimensional light feature extraction is performed based on the multi-dimensional light attribute clustering data to generate multi-dimensional light feature data, including: (1) Extract light intensity features from the integrated light data to generate light intensity feature data; (2) Extract spectral distribution features from the integrated light data to generate spectral distribution feature data; (3) Use a preset multi-dimensional clustering algorithm to perform initial multi-dimensional light attribute clustering analysis on the light intensity feature data and the spectral distribution feature data to generate initial multi-dimensional light attribute clustering data; (4) Based on the initial multi-dimensional light attribute clustering data, perform screening processing on visually irrelevant light data to generate multi-dimensional light attribute clustering data; (5) Extract multi-dimensional light features from the multi-dimensional light attribute clustering data to generate multi-dimensional light feature data.

[0021] In the embodiments of the present invention, first, the light intensity features are extracted. The system adopts a three-stage processing flow: parameter identification, numerical normalization, and distribution calculation. Four types of key intensity parameters are extracted from the integrated light data: peak illuminance (unit: lux), average luminance (unit: cd / m²), attenuation coefficients (three parameters: α, β, γ), and spatial distribution function. The system assigns standardized identifiers (DL_01, PL_03, and SL_02) to each light source and sets the valid range of parameters (illuminance: 0 - 120000 lux, luminance: 0 - 10000 cd / m²). The spatial light distribution is recorded through a sampling grid (32×32×16 resolution) to generate light intensity feature data containing a complete set of intensity parameters. For the spectral distribution feature extraction, color space conversion technology is used to convert the RGB light data to the CIE-Lab color space for processing. The system extracts three groups of key color features: color temperature parameter (1500K to 15000K, accuracy ±50K), hue angle (0° to 360°, accuracy ±0.5°), and color saturation (0 to 100%, accuracy ±1%). For high dynamic range light sources, the gamut coverage rate and color transition curve are additionally recorded. The system uses the four-point sampling method to analyze the color space distribution. Each sampling point records a complete color vector, and a continuous color distribution map is generated through cubic spline interpolation to output standardized spectral distribution feature data. For the initial clustering analysis, an improved K-means++ algorithm is used to construct an 8-dimensional feature vector by combining the light intensity and spectral distribution features. The system presets the number of clusters K = 7, and the initial cluster centers are selected through the maximum distance optimization strategy. The maximum number of steps for the clustering iteration is set to 100, and the convergence threshold is 0.001. For mixed light sources, a fuzzy clustering coefficient (0 - 1.0) is introduced to represent the membership degree. The system evaluates the clustering results using the Davies-Bouldin index to ensure the clustering quality. After clustering, a unique ID (LC_01 to LC_07) is assigned to each light group to generate initial multi-dimensional light attribute clustering data containing complete clustering information. For the screening of visually irrelevant light data, a triple screening mechanism is adopted: brightness contribution degree evaluation (threshold 1.5%), field of view coverage analysis (threshold 2%), and visual saliency calculation (threshold 0.15). The system performs ray tracing detection on occluded light sources to calculate the occlusion rate. The actual contribution of each light source to the final image is evaluated through a visual perception model, and the light sources with a contribution degree lower than the threshold are marked as "low priority". The screening process retains the light source data but reduces its sampling rate in rendering to generate refined multi-dimensional light attribute clustering data. In the multi-dimensional light feature extraction stage, the system extracts five types of key features from the optimized clustering: light gradient map (16-bit precision), color transition matrix (4×4×4 resolution), shadow boundary feature map, specular light distribution map, and ambient occlusion map. The feature extraction uses spherical harmonics (3rd order, 16 coefficients) to represent the complex light distribution, generating complete multi-dimensional light feature data, which provides a basis for subsequent scene area division and rendering optimization.

[0022] Preferably, visual-irrelevant light data is screened out based on the initial multi-dimensional light attribute clustering data to generate multi-dimensional light attribute clustering data, including: (1) Performing visual perception threshold analysis on the initial multi-dimensional light attribute clustering data to generate visual perception threshold data, where the visual perception threshold data includes light intensity threshold data and spectral sensitivity data; (2) Performing low-impact light source aggregation processing based on the light intensity threshold data to generate aggregated low-impact light source data; (3) Performing visual redundant light identification processing based on the aggregated low-impact light source data and the spectral sensitivity data to generate visual redundant light data; (4) Performing screening processing of multi-dimensional light attribute clustering data of visual redundant light on the initial multi-dimensional light attribute clustering data according to the visual redundant light data to generate multi-dimensional light attribute clustering data, and transmitting the visual redundant light data to the rendering engine to perform a simplified rendering operation.

[0023] In the embodiments of the present invention, visual perception threshold analysis is first performed. Based on the initial multi-dimensional light attribute clustering data, the system calculates the light perception threshold by applying an improved Weber-Fechner model. The processing flow includes two parts: brightness contribution evaluation and spectral sensitivity analysis. For brightness contribution evaluation, the six-point sampling method is adopted. A 32×32 sampling grid is set within the main view angle of the scene, and the illuminance contribution of each light source to the grid points is calculated. The system sets the brightness contribution threshold to 1.2%, and the light sources with a contribution lower than this value are marked as "low priority". For spectral sensitivity analysis, the CIE1931 standard color observer function is used to map the sensitivity of the human eye to different wavelengths, and the perception weight is calculated according to the color temperature and color gamut distribution of the light source. The system generates visual perception threshold data including brightness threshold mapping and spectral sensitivity curves. For the aggregation processing of low-impact light sources, a spatial partitioning and feature grouping strategy is adopted. The system divides the scene into a voxel grid of 8×8×4, and performs clustering analysis on the low-impact light sources within each voxel. The aggregation conditions are set as follows: the spatial distance is less than 15% of the unit length of the scene and the similarity of light characteristics is greater than 85%. The system applies a hierarchical clustering algorithm to iteratively merge the light sources that meet the conditions. Each aggregated light source group is assigned a unique identifier (ALCG_001~ALCG_n), and the original parameters of the constituent light sources are recorded. The aggregation processing replaces multiple low-impact light sources with a simplified light source model with equivalent parameters, generating aggregated low-impact light source data. For visual redundant light recognition, an image quality assessment method is used. The system sets three key viewpoints, renders the scene from different angles, and compares the full lighting and the lighting after removing the aggregated light source group respectively. The structural similarity index (SSIM) is used to evaluate the image difference, and the redundancy determination threshold is set to 0.96 (full score 1.0). The aggregated light source groups with an SSIM higher than the threshold are marked as "visually redundant". The system records their IDs, positions, influence ranges, and visual contribution degrees, generating visually redundant light data. In the screening process, the system removes the light source groups marked as visually redundant from the initial multi-dimensional light attribute clustering data and retains their metadata for possible subsequent recovery. The system designs an adaptive screening strategy for different rendering scenes and viewpoints: the weak light sources within the main view are processed with a downsampling rate (down to 25%); the light sources outside the main view are completely replaced with a simplified lighting model; the distant light sources are replaced with precomputed light maps. The processed data and the visually redundant light data are respectively transmitted to the rendering engine. The former is used for the standard rendering channel, and the latter is used for simplified rendering. The system achieves the goal of significantly improving the rendering efficiency while maintaining the visual quality.

[0024] Preferably, the scene area is divided according to the multi-dimensional light feature data to generate area multi-dimensional light feature data; based on the area multi-dimensional light feature data, area material-light interaction feature analysis is performed to generate area material-light interaction feature data, including: (1)Perform scene area division processing based on multi-dimensional illumination feature data to generate area multi-dimensional illumination feature data; (2)Perform area material property mapping analysis based on the area multi-dimensional illumination feature data to generate area material property data; (3)Perform area material-illumination interaction feature analysis based on the area material property data to generate area material-illumination interaction feature data.

[0025] In the embodiments of the present invention, scene area division processing is first performed. Based on multi-dimensional light feature data, the system hierarchically divides the scene using an adaptive quadtree segmentation algorithm. Three main evaluation metrics are set during the division process: the light gradient change rate (threshold 20% / m), the material complexity (threshold 0.65, full score 1.0), and the visual importance (based on the center-surround attenuation model). The system initially divides the scene into four basic regions: the core interaction area, the environmental transition area, the background area, and the special effect area. For areas with drastic light changes (gradient > 30% / m), fine-grained division (minimum grid size 2×2 m) is used to ensure accurate capture of light boundaries; for areas with uniform light, coarse-grained division (maximum grid size 10×10 m) is used to improve efficiency. Each region is assigned a unique identifier (such as RA_001 to RA_128), and its boundary data, light feature vector, and spatial position are recorded to generate region multi-dimensional light feature data containing complete spatial partition information. The regional material property mapping analysis uses a physically based rendering parameter extraction technique. The system parameterizes the material properties within each divided region and extracts four groups of key parameters: base color / albedo (RGB value accuracy 16 bits), metallicity (0 to 1.0, accuracy 0.01), roughness (0 to 1.0, accuracy 0.01), and subsurface scattering coefficient (0 to 1.0, accuracy 0.05). The system sets the sampling density within each region to 4 points per square meter and increases it to 16 points per square meter for high-complexity regions (such as the crystal chandelier area). The material property clustering uses the K-Means++ algorithm (K = 8) to group similar materials (similarity > 92%) within the region into the same category. The system separately establishes a high-precision parameter model for special materials (such as translucent glass, metal reflective surfaces) to generate region material property data containing material microscopic properties and classification information. The material-light interaction feature analysis uses a simplified ray tracing and BRDF evaluation method. The system sets the interaction sampling rate for each material category to 1024 samples per square meter and records the key interaction parameters: direct reflection intensity (0 to 1.0), specular concentration (0 to 1.0), and diffuse reflection coefficient (0 to 1.0). Three typical light conditions are considered during the analysis process: high-intensity directional light (greater than 7500 lumens), medium-intensity ambient light (2000 to 5000 lumens), and low-intensity point light source combination (less than 1500 lumens). The system calculates the light response curves of each material at different incident angles (0°, 45°, 75°) and evaluates the view dependence. The interaction features are recorded with 32-bit floating-point precision, including a complete angle-response mapping table, forming region material-light interaction feature data, providing a scientific basis for subsequent rendering optimization.

[0026] Preferably, regional material property mapping analysis is performed according to the regional multi-dimensional light feature data to generate regional material property data, including: (1)Analyze the reflection characteristics of the surface material based on the regional multi-dimensional light characteristic data to generate the surface material reflection characteristic data; (2)Analyze the scattering characteristics of the surface material based on the regional multi-dimensional light characteristic data to generate the surface material scattering characteristic data; (3)Calculate the material-light interaction response matrix based on the surface material reflection characteristic data and the surface material scattering characteristic data to generate the material-light interaction response matrix; (4)Conduct regional material characteristic mapping analysis based on the material-light interaction response matrix to generate the regional material characteristic data.

[0027] In the embodiment of the present invention, first, extract the light parameters from the regional multi-dimensional light characteristic data for the analysis of the surface material reflection characteristics. The system adopts a four-parameter reflection model, including the incident angle (0° to 90°, accuracy 0.5°), light intensity (0 to 10000 lumens, accuracy 5 lumens), color temperature (1500K to 10000K, accuracy 50K), and light source size (point light source to area light source, radius 0 to 5 meters). The improved Cook-Torrance micro-surface model is used in the analysis process, and four types of reflection type identifiers are set: high specular reflection (SR>0.75), medium specular reflection (0.4<SR≤0.75), mixed reflection (0.2<SR≤0.4), and high diffuse reflection (SR≤0.2), where SR is the specular reflection ratio. The system constructs an incident-exit angle mapping table (resolution 18×18), records the BRDF response curves of various materials, sets the material analysis grid density to 9 sampling points per square meter, and enhances the key surfaces to 25 points / square meter to generate the standardized surface material reflection characteristic data. The surface material scattering characteristic analysis uses the volume light transport equation, and the system sets the scattering analysis parameters: scattering coefficient σs (0 to 20, accuracy 0.1), absorption coefficient σa (0 to 10, accuracy 0.1), and phase function g (-0.9 to 0.9, accuracy 0.05). The analysis process includes the measurement of the light penetration depth and the calculation of the scattering direction distribution. The system applies a 16-direction scattering detector to translucent materials (such as wax, skin, cloud), measures the propagation characteristics of light in the material, and sets the scattering sampling rate to 2048 samples / material type. The material classification uses the scattering threshold criterion: high scattering materials (σs>8), medium scattering materials (3<σs≤8), and low scattering materials (σs≤3). The system records the scattering curves and depth distribution maps of each material to generate the surface material scattering characteristic data. The calculation of the material-light interaction response matrix integrates the reflection and scattering characteristic data. The system constructs an M×L-dimensional response matrix, where M represents the number of material types (set to 12 types), and L represents the number of light condition types (set to 8 types). The matrix calculation formula is , where α and β are weight coefficients, and ref and scat are the normalized reflection and scattering response values respectively. The system applies additional weight adjustment factors for special visual effects (such as metal highlights and glass refractions). The response threshold is set to 0.65 (out of 1.0), and material-lighting combinations exceeding the threshold are marked as "high-priority rendering objects". Matrix calculations are accelerated using the GPU, and the processing time is controlled within 5 ms to generate a complete material-lighting interaction response matrix. The regional material property mapping analysis is based on the interaction response matrix for material classification and optimization. The system adopts a four-dimensional optical property scoring system: visual saliency (0 to 1.0), computational complexity (1 to 10), surface detail fidelity (0 to 1.0), and light sensitivity (0 to 1.0). Hierarchical clustering algorithm is used for material classification, with the maximum number of categories set to 8 and the minimum intra-class similarity set to 0.8. The system assigns rendering processing strategies to each material category: high-priority materials use high-precision shaders; medium-priority materials use the standard PBR pipeline; low-priority materials adopt a simplified lighting model. Rendering parameter adjustment is based on the results of property mapping. For example, the reflection sampling rate of a high-specular metal surface is increased to 16x, while the standard diffuse surface maintains a 4x sampling rate, generating optimized regional material property data to provide a scientific basis for resource allocation in the rendering system.

[0028] Preferably, based on the regional material property data, regional material-lighting interaction feature analysis is performed to generate regional material-lighting interaction feature data, including: (1) Performing physical-based rendering model feature analysis on the regional material property data to generate physical-based rendering model feature data; (2) Performing regional material-lighting interaction feature analysis based on the physical-based rendering model feature data to generate regional material-lighting interaction feature data.

[0029] In the embodiments of the present invention, the regional material property data is applied to a physical-based rendering model (PBR) to analyze parameters such as the BRDF, refractive index, metallicity, and roughness of the material. The system uses the Cook-Torrance microfacet model or Disney BRDF to physically model the material properties, identify the differences from the ideal model, and generate physical-based rendering model feature data. Subsequently, ray tracing algorithms and radiometric techniques are used to simulate the interaction process between the material and light, and finally, regional material-lighting interaction feature data is generated to accurately express the visual performance of the material under different lighting conditions.

[0030] Preferably, the real-time rendering module includes the following steps: (1) Performing light propagation feature analysis based on the regional multi-dimensional lighting feature data to generate light propagation feature data; (2) Analyze the visual importance scoring node based on the regional material-light interaction feature data, generate the visual importance scoring node data, and design the decision tree priority node through the visual importance scoring node data; (3) Design the neural network rendering priority tree for light effect and rendering resource allocation based on the preset deep neural network model for the decision tree priority node and the regional material-light interaction feature data, so as to obtain the rendering priority neural network model; (4) Design the intelligent judgment engine for regional rendering priority prediction according to the rendering priority neural network model, so as to obtain the regional rendering priority prediction engine; (5) Transmit the light propagation feature data to the regional rendering priority prediction engine in real time for intelligent analysis of the regional dynamic rendering requirements, and generate the regional dynamic rendering requirement data; (6) Conduct adaptive decision analysis for global real-time light color rendering according to the regional dynamic rendering requirement data, and generate the global real-time light color rendering decision; (7) Perform real-time light color rendering adjustment processing on the initial light data according to the global real-time light color rendering decision, generate the real-time light color rendering data, and execute the real-time light color rendering of the game scene through the real-time light color rendering data.

[0031] In the embodiments of the present invention, first, light propagation feature analysis is performed on the regional multi-dimensional light feature data. The system extracts key parameters such as light angle, intensity, and color temperature from the original data, and uses an optimized bidirectional path tracing algorithm to analyze the propagation behavior of light in the scene. The light sampling density is set to 32 sampling points per square meter, the maximum depth of path tracing is 5 bounces, and the light propagation accuracy threshold is 4%. The system records the reflection, scattering, and absorption coefficients, and generates a light density distribution map, a shadow projection map, and an indirect illumination contribution map with 16-bit precision respectively, forming a complete light propagation feature data matrix. The visual importance scoring node analysis is based on the regional material-light interaction feature data. The system uses an improved visual saliency algorithm, combined with four key indicators: brightness contrast (weight 0.35), specular quality (weight 0.25), texture complexity (weight 0.2), and line-of-sight attention (weight 0.2) for comprehensive scoring. The scoring threshold is set to 0.7 (full score 1.0), and the regions above this threshold are marked as high-priority rendering regions. The system constructs a four-layer decision tree structure, with the number of leaf nodes set to 32, the minimum number of split samples to 10, and the information gain ratio is used as the splitting criterion, generating a decision tree priority node containing the priority status of each region in the scene. The neural network rendering priority tree design uses a preset deep convolutional neural network model. The network structure includes four convolutional layers (number of filters: 48, 96, 192, and 256) and three fully connected layers (number of nodes: 512, 256, and 64), using the LeakyReLU activation function (α = 0.15) and Batch normalization layers. The training process uses the Adam optimizer (β1 = 0.9, β2 = 0.999), the initial learning rate is 0.008, and the batch size is 96. The model is trained for 150 epochs on a 200,000 sample set, and the early stopping strategy (patience = 12) is used to avoid overfitting. The validation uses SSIM (threshold ≥ 0.85) and performance evaluation (target frame rate ≥ 60fps), generating a rendering priority neural network model. The intelligent judgment engine design uses a three-stage pipeline architecture, including a feature analyzer, a priority predictor, and a resource scheduler. The engine has a built-in cache (capacity 64MB), and the real-time processing delay is controlled within 8ms. The engine is optimized using SIMD instructions, and the single-frame processing ability reaches 120 scene regions. Its stability is tested and verified in four typical scenarios (daytime outdoor, nighttime indoor, underwater, and special effects intensive areas), generating a regional rendering priority prediction engine. The light propagation feature data is sent into the prediction engine in real time through an asynchronous transmission channel, using an incremental update strategy, only transmitting the changed data, and the compression ratio reaches 4:1. The prediction engine applies the pre-trained model to the input data, assigns the rendering resource percentage and processing priority to each scene region, generating regional dynamic rendering demand data. The global real-time light color rendering decision analysis uses a hierarchical optimization algorithm, integrating the demands of each region and considering the performance budget (target frame rate, video memory limit).The system defines three levels of rendering quality, dynamically balancing visual quality and performance. The decision-making process is updated 15 times per second to generate global real-time light and color rendering decisions. During the rendering adjustment process, the system dynamically adjusts light source parameters, material response characteristics, and sampling rate allocation based on the decision data. High-quality processing (sampling rate 8x) is applied to visually important areas, standard processing (4x) to secondary areas, and simplified processing (2x) to background areas. Finally, the rendering is executed through an efficient shader pipeline to achieve a balance between high-quality visual effects and high performance.

[0032] Preferably, based on a preset deep neural network model, a neural network rendering priority tree design for lighting effects and rendering resource allocation is carried out on the decision tree priority nodes and regional material-light interaction feature data to obtain a rendering priority neural network model, including: (1) Designing the architecture of the neural network rendering priority tree for lighting effects and rendering resource allocation based on the preset deep neural network model for the decision tree priority nodes to generate the architecture of the rendering priority neural network model; (2) Conducting user visual attention prediction analysis based on the regional material-light interaction feature data to generate user visual attention prediction data; (3) Performing priority feature weight learning processing on the architecture of the rendering priority neural network model according to the user visual attention prediction data to generate the rendering priority neural network model.

[0033] In the embodiments of the present invention, a deep neural network model is first preset. It is designed with a VGG variant structure, including 4 convolutional layers and 3 fully connected layers. The number of filters in the convolutional layers is 32 - 64 - 128 - 256 in sequence, and the number of nodes in the fully connected layers is 512 - 256 - 128. LeakyReLU is selected as the activation function (α = 0.2) to avoid the problem of neuron death; the Softmax function is used in the output layer to achieve the multi-region priority probability distribution. The learning rate is initially set to 0.003, and the Adam optimizer (β1 = 0.9, β2 = 0.999) is used to improve the training stability, and the batch size is set to 48. Specialized design is carried out for the decision tree priority nodes, and different network parameters are configured for regions with different lighting conditions: the nodes in the strong light region use more convolutional kernels (increased to 96) to capture the high-light details; the nodes in the shadow region are configured with a deeper receptive field (9×9 pixels) to enhance the edge detection ability; an attention mechanism module (number of attention heads = 4) is added in the dynamic light source region to process the time-varying light. The residual connection structure is adopted between the nodes to alleviate the problem of gradient disappearance. The neural networks of each priority node are integrated into a unified tree-like architecture, and a three-level rendering priority hierarchy is designed: the high-priority layer (core visual region), the medium-priority layer (secondary visual region), and the basic layer (background region). Each level controls the resource allocation ratio through a gating unit (gating threshold = 0.65) to ensure that the key regions obtain sufficient rendering resources. In the tree-like architecture, the decision results of the upper-layer nodes will affect the weight parameters of the lower-layer nodes to achieve global collaborative optimization and generate a complete rendering priority neural network model architecture. The key elements of visual attention are extracted from the region material-light interaction feature data, including brightness contrast (weight 0.3), color saturation (weight 0.25), texture complexity (weight 0.25), and motion saliency (weight 0.2). The improved Itti-Koch model is used to analyze visual saliency, and a deep learning enhancement module is introduced to improve the prediction accuracy. A regional sampling grid (16×16) is set, and the saliency score is calculated for each region. The regions with scores exceeding 0.7 are marked as high-attention regions. The prediction model is calibrated through eye movement tracking verification (sample size of 50 person-times), and the verification accuracy reaches 85% to generate the user visual attention prediction data matrix. Based on the generated visual attention prediction data, weight learning is carried out on the rendering priority neural network model. Cross-validation (K = 5) is adopted to ensure the generalization of the model, and the ratio of the training set to the validation set is 7:3. The loss function design combines the mean square error (MSE) and the structural similarity (SSIM) loss, taking into account both numerical accuracy and visual effect consistency. A dynamic learning rate adjustment strategy is set: if the validation loss decreases for 3 consecutive rounds, the learning rate is increased by 10%; if the validation loss increases, the learning rate is decreased by 15%.Set resource allocation coefficients for different priority regions: The high-attention region (coefficient = 1.8) gets the most rendering sampling points; the medium-attention region (coefficient = 1.0) gets the standard sampling rate; the low-attention region (coefficient = 0.6) reduces the sampling rate to save resources. The early stopping mechanism (patience = 10) is set in the weight learning process, and training stops when the validation loss stabilizes to avoid overfitting. The effectiveness of the model is verified through A / B testing (user satisfaction increased by 20%), and finally an optimized rendering priority neural network model is generated to achieve intelligent allocation of lighting effects and rendering resources.

[0034] Through the collection of lighting data and the identification of light source types, the present invention establishes a heterogeneous light source representation model to achieve unified management and coordinated rendering of multiple types of light sources. The system uses multi-dimensional lighting attribute clustering analysis and feature extraction to accurately capture the changes in lighting intensity and spectral distribution in the scene, and through a visually irrelevant lighting data screening mechanism, effectively eliminates redundant lighting information that has a weak impact on the visual effect, significantly improving the rendering efficiency. Based on scene area division and material-light interaction feature analysis, the system realizes accurate lighting rendering for different regional material characteristics, and through the analysis of the physically based rendering model, presents a more realistic lighting effect. The regional rendering priority prediction engine analyzes the lighting propagation characteristics in real time, and combines with the deep neural network model to intelligently allocate rendering resources to ensure high-efficiency processing ability when the scene changes dynamically. The system generates an adaptive rendering decision based on the user's visual attention prediction data and regional dynamic rendering requirement analysis, preferentially processes visually important regions, and effectively reduces the consumption of computing resources while ensuring high-quality visual performance. This adaptive real-time color rendering technology for heterogeneous light source data and complex rendering requirements realizes a more efficient and realistic lighting effect in game scenes.

[0035] Embodiment 2; Please refer to Figure 2 As shown, for the parts not described in detail in this embodiment, refer to the description content of Embodiment 1. A real-time color rendering method for game scenes is provided, including: S1. Collect scene lighting information based on the game engine to generate initial lighting data; perform heterogeneous light source data fusion processing on the initial lighting data to generate integrated lighting data; S2. Perform multi-dimensional lighting attribute clustering analysis on the integrated lighting data to generate multi-dimensional lighting attribute clustering data; perform multi-dimensional lighting feature extraction on the multi-dimensional lighting attribute clustering data to generate multi-dimensional lighting feature data; S3. Perform scene area division processing on the multi-dimensional lighting feature data to generate regional multi-dimensional lighting feature data; perform regional material-light interaction feature analysis based on the regional multi-dimensional lighting feature data to generate regional material-light interaction feature data; S4. Analyze the light propagation characteristics based on the regional multi-dimensional light characteristic data to generate light propagation characteristic data; design an intelligent judgment engine for predicting the regional rendering priority based on the regional material-light interaction characteristic data to obtain a regional rendering priority prediction engine; transmit the light propagation characteristic data to the regional rendering priority prediction engine in real time for intelligent analysis of the regional dynamic rendering requirements to generate regional dynamic rendering requirement data; conduct adaptive decision-making analysis for global real-time light color rendering based on the regional dynamic rendering requirement data to generate a global real-time light color rendering decision; perform real-time light color rendering adjustment processing on the initial light data according to the global real-time light color rendering decision to generate real-time light color rendering data, and execute the real-time light color rendering of the game scene through the real-time light color rendering data.

[0036] This embodiment also provides a real-time light color rendering device for a game scene. The real-time light color rendering device for a game scene includes a memory and a processor. Computer-readable instructions are stored in the memory. When the computer-readable instructions are executed by the processor, the processor executes the steps of the real-time light color rendering method for a game scene in the above embodiments.

[0037] The above formulas are all dimensionless and take their numerical calculations. The formulas are obtained by collecting a large amount of data for software simulation to get a formula closest to the real situation. The preset parameters and threshold selection in the formulas are set by those skilled in the art according to the actual situation.

[0038] The above is only the preferred embodiment of the present invention. The protection scope of the present invention is not limited to the above embodiments. All technical solutions within the idea of the present invention belong to the protection scope of the present invention. It should be pointed out that for ordinary technical users in the technical field, several improvements and refinements made without departing from the principle of the present invention should also be regarded as the protection scope of the present invention.

Claims

1. An instant light color rendering system for a game scene, characterized in that, Including: Acquisition and processing module: Collecting scene lighting information based on the game engine to generate initial lighting data; Performing heterogeneous light source data fusion processing according to the initial lighting data to generate integrated lighting data; Data extraction module: Performing multi-dimensional lighting attribute clustering analysis according to the integrated lighting data to generate multi-dimensional lighting attribute clustering data; Performing multi-dimensional lighting feature extraction according to the multi-dimensional lighting attribute clustering data to generate multi-dimensional lighting feature data; Analysis module: Performing scene area division processing according to the multi-dimensional lighting feature data to generate area multi-dimensional lighting feature data; Performing area material-lighting interaction feature analysis based on the area multi-dimensional lighting feature data to generate area material-lighting interaction feature data; Real-time rendering module: Performing lighting propagation feature analysis according to the area multi-dimensional lighting feature data to generate lighting propagation feature data; Designing an intelligent judgment engine for predicting the area rendering priority based on the area material-lighting interaction feature data to obtain an area rendering priority prediction engine; Transmitting the lighting propagation feature data to the area rendering priority prediction engine in real time for intelligent analysis of the area dynamic rendering requirements to generate area dynamic rendering requirement data; Performing adaptive decision analysis for global real-time light color rendering according to the area dynamic rendering requirement data to generate a global real-time light color rendering decision; Performing real-time light color rendering adjustment processing on the initial lighting data according to the global real-time light color rendering decision to generate real-time light color rendering data, and performing real-time light color rendering of the game scene through the real-time light color rendering data.

2. The real-time light color rendering system for game scenes according to claim 1, wherein Collecting scene lighting information based on the game engine to generate initial lighting data; Performing heterogeneous light source data fusion processing according to the initial lighting data to generate integrated lighting data, including: Collecting scene lighting information based on the game engine to generate initial lighting data; Performing light source type recognition processing according to the initial lighting data to generate light source type feature data; Designing a heterogeneous light source representation model according to the light source type feature data to generate a heterogeneous light source representation model; Performing heterogeneous light source data fusion processing according to the heterogeneous light source representation model to generate integrated lighting data.

3. The real-time light color rendering system for game scenes according to claim 1, characterized in that Performing multi-dimensional lighting attribute clustering analysis according to the integrated lighting data to generate multi-dimensional lighting attribute clustering data; Performing multi-dimensional lighting feature extraction according to the multi-dimensional lighting attribute clustering data to generate multi-dimensional lighting feature data, including: Performing light intensity feature extraction according to the integrated lighting data to generate light intensity feature data; Performing spectral distribution feature extraction according to the integrated lighting data to generate spectral distribution feature data; Performing initial multi-dimensional lighting attribute clustering analysis on the light intensity feature data and the spectral distribution feature data using a preset multi-dimensional clustering algorithm to generate initial multi-dimensional lighting attribute clustering data; Performing visual unrelated lighting data screening processing based on the initial multi-dimensional lighting attribute clustering data to generate multi-dimensional lighting attribute clustering data; Performing multi-dimensional lighting feature extraction according to the multi-dimensional lighting attribute clustering data to generate multi-dimensional lighting feature data.

4. The real-time light color rendering system for game scenes according to claim 3, wherein Performing visual-irrelevant illumination data screening processing based on the initial multi-dimensional illumination attribute clustering data to generate multi-dimensional illumination attribute clustering data, including: Performing visual perception threshold analysis based on the initial multi-dimensional illumination attribute clustering data to generate visual perception threshold data, where the visual perception threshold data includes illumination intensity threshold data and spectral sensitivity data; Performing low-impact light source aggregation processing based on the illumination intensity threshold data to generate aggregated low-impact light source data; Performing visual redundant illumination recognition processing based on the aggregated low-impact light source data and the spectral sensitivity data to generate visual redundant illumination data; Performing multi-dimensional illumination attribute clustering data screening processing of visual redundant illumination on the initial multi-dimensional illumination attribute clustering data according to the visual redundant illumination data to generate multi-dimensional illumination attribute clustering data, and transmitting the visual redundant illumination data to the rendering engine to execute a simplified rendering job.

5. The real-time light color rendering system for a game scene according to claim 1, wherein Performing scene area division processing according to the multi-dimensional illumination feature data to generate area multi-dimensional illumination feature data; Performing area material-illumination interaction feature analysis based on the area multi-dimensional illumination feature data to generate area material-illumination interaction feature data, including: Performing scene area division processing according to the multi-dimensional illumination feature data to generate area multi-dimensional illumination feature data; Performing area material property mapping analysis according to the area multi-dimensional illumination feature data to generate area material property data; Performing area material-illumination interaction feature analysis according to the area material property data to generate area material-illumination interaction feature data.

6. The real-time light color rendering system for a game scene according to claim 5, characterized in that, Performing area material property mapping analysis according to the area multi-dimensional illumination feature data to generate area material property data, including: Performing surface material reflection property analysis according to the area multi-dimensional illumination feature data to generate surface material reflection property data; Performing surface material scattering property analysis according to the area multi-dimensional illumination feature data to generate surface material scattering property data; Calculating a material-illumination interaction response matrix according to the surface material reflection property data and the surface material scattering property data to generate a material-illumination interaction response matrix; Performing area material property mapping analysis according to the material-illumination interaction response matrix to generate area material property data.

7. The real-time light color rendering system for game scenes according to claim 5, characterized in that Performing area material property mapping analysis according to the material-illumination interaction response matrix to generate area material property data, including: Performing physically-based rendering model feature analysis on the area material property data to generate physically-based rendering model feature data; Performing area material-illumination interaction feature analysis according to the physically-based rendering model feature data to generate area material-illumination interaction feature data.

8. The real-time light color rendering system for game scenes according to claim 1, characterized in that, Performing illumination propagation feature analysis according to the area multi-dimensional illumination feature data to generate illumination propagation feature data; Designing an intelligent judgment engine for area rendering priority prediction based on the area material-illumination interaction feature data to obtain an area rendering priority prediction engine; Transmitting the illumination propagation feature data to the area rendering priority prediction engine in real time for intelligent analysis of area dynamic rendering requirements to generate area dynamic rendering requirement data; Adaptive decision analysis for global real-time light color rendering based on region dynamic rendering requirement data to generate global real-time light color rendering decisions; Perform real-time light color rendering adjustment processing on the initial light data according to the global real-time light color rendering decision to generate real-time light color rendering data, and perform real-time light color rendering of the game scene through the real-time light color rendering data, including: Perform light propagation feature analysis based on region multi-dimensional light feature data to generate light propagation feature data; Perform visual importance scoring node analysis based on region material-light interaction feature data to generate visual importance scoring node data, and design decision tree priority nodes through the visual importance scoring node data; Design a neural network rendering priority tree for light effect and rendering resource allocation based on the preset deep neural network model for the decision tree priority nodes and region material-light interaction feature data to obtain a rendering priority neural network model; Design an intelligent judgment engine for region rendering priority prediction according to the rendering priority neural network model to obtain a region rendering priority prediction engine; Transmit the light propagation feature data to the region rendering priority prediction engine in real time for intelligent analysis of region dynamic rendering requirements to generate region dynamic rendering requirement data; Adaptive decision analysis for global real-time light color rendering based on region dynamic rendering requirement data to generate global real-time light color rendering decisions; Perform real-time light color rendering adjustment processing on the initial light data according to the global real-time light color rendering decision to generate real-time light color rendering data, and perform real-time light color rendering of the game scene through the real-time light color rendering data.

9. The real-time light color rendering system for game scenes according to claim 8, wherein, The design of the neural network rendering priority tree for light effect and rendering resource allocation based on the preset deep neural network model for the decision tree priority nodes and region material-light interaction feature data to obtain a rendering priority neural network model includes: Design the architecture of the neural network rendering priority tree for light effect and rendering resource allocation based on the preset deep neural network model for the decision tree priority nodes to generate the architecture of the rendering priority neural network model; Perform user visual attention prediction analysis based on region material-light interaction feature data to generate user visual attention prediction data; Perform priority feature weight learning processing of light effect and rendering resources on the architecture of the rendering priority neural network model according to the user visual attention prediction data to generate the rendering priority neural network model.

10. A real-time light color rendering device for a game scene, characterized in that, The real-time light color rendering device for game scenes includes: a memory and at least one processor, and instructions are stored in the memory; the at least one processor calls the instructions in the memory so that the real-time light color rendering device for game scenes executes a real-time light color rendering method for game scenes.

Images