Lightweight transmission and loading optimization method for high-precision three-dimensional model cloud rendering
By employing cloud-based preprocessing and client-side hierarchical loading, the transmission and loading process of high-precision 3D models has been optimized, solving the bandwidth limitations and latency issues inherent in traditional methods, and achieving efficient rendering data transmission and a smooth user experience.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- HUANENG YINGKOU THERMAL POWER CO LTD
- Filing Date
- 2025-06-20
- Publication Date
- 2026-06-26
AI Technical Summary
In the process of cloud rendering of high-precision 3D models, existing technologies face bandwidth limitations and network latency issues in traditional transmission methods, making it difficult to achieve the best balance between high precision and low latency. In particular, they lack refined processing strategies for different important parts of the rendering data, resulting in insufficient transmission efficiency and loading speed.
The rendering scene data of the high-precision 3D model is preprocessed through a cloud rendering server to construct a hierarchical detail structure and spatial partitioning structure, and is initially compressed. The importance of the rendered primitives is evaluated based on the client's viewpoint data, an importance score is generated, important data points are transmitted first, and non-important data points are compressed or delayed. The client performs importance reassessment and hierarchical loading, decoding high-priority data first and loading low-priority data gradually.
It significantly improves transmission efficiency and loading speed, ensuring that critical information is delivered quickly, optimizing the user's 3D model browsing experience, and providing smooth, real-time rendering effects.
Smart Images
Figure CN120689483B_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of network transmission, and in particular to a lightweight transmission and loading optimization method for high-precision 3D model cloud rendering. Background Technology
[0002] With the rapid development of 3D modeling technology and cloud computing, the application of cloud rendering for high-precision 3D models is becoming increasingly widespread, such as in online design, virtual reality, augmented reality, and digital twins. However, high-precision 3D models typically contain massive amounts of geometric, texture, and material data. After rendering in the cloud, the rendering results need to be transmitted to the client via the network. Traditional transmission methods face challenges of bandwidth limitations and network latency, which may lead to client-side data reception lag and slow loading, severely impacting user experience. While some compression and streaming methods exist in existing technologies, it is often difficult to achieve the optimal balance between high precision and low latency. In particular, there is a lack of refined processing strategies for different important parts of the rendered data, leaving room for improvement in transmission efficiency and loading speed. Summary of the Invention
[0003] To address the aforementioned problems in existing technologies, the present invention aims to provide a lightweight transmission and loading optimization method for high-precision 3D model cloud rendering, including a lightweight transmission method for cloud rendering, executed by a cloud rendering server, comprising:
[0004] Step S1: Obtain the original rendering scene data of the high-precision 3D model, and preprocess the original rendering scene data. The preprocessing includes constructing the hierarchical detail structure of the rendering scene data and generating the spatial partitioning structure of the rendering scene data, and performing preliminary compression on the general data in the rendering scene data.
[0005] Step S2: Based on the viewpoint data transmitted by the client in real time, the importance of the rendering primitives in the rendering scene data is evaluated to generate an importance score for the rendering primitives. The importance evaluation is based on at least one of the following: the distance of the rendering primitive to the client's viewpoint, the projected area on the client's screen, the geometric curvature, the visibility, and user interaction and attention factors.
[0006] Step S3: Construct the feature matrix and weight vector of the rendered primitive. Calculate the importance score vector of the rendered primitive by performing matrix multiplication on the feature matrix and the weight vector. Determine the important data points in the rendered primitive that need to be transmitted first based on the importance score vector.
[0007] Step S4: Lightweight encoding is performed on the important data points and they are transmitted with priority, while high-intensity compression encoding or delayed transmission is performed on the unimportant data points.
[0008] Furthermore, the client-side loading step, performed by the client, includes:
[0009] Step Y1: Receive the rendering data stream after the lightweight transmission.
[0010] Step Y2: Based on at least one of the client's current rendering state, display resolution, and user interaction, re-evaluate the client importance of the rendering data blocks in the rendering data stream.
[0011] Step Y3: Construct the client feature matrix and client weight vector of the rendering data block, and determine the high-priority loading data in the rendering data block that needs to be loaded and rendered first through matrix multiplication.
[0012] Step Y4: Prioritize decoding, decompressing, and loading the high-priority loaded data for rendering, and gradually decode, decompress, and load the low-priority loaded data for detailed rendering, thereby gradually improving the refinement of the rendered image.
[0013] Furthermore, the hierarchical detail structure includes multiple geometric representations of the 3D model at different levels of geometric detail. These geometric representations are represented by the number of polygons or texture resolution, ranging from coarse to fine, so as to dynamically select and adjust the display complexity of the 3D model according to the client's viewpoint distance or the importance of the rendered primitives. The geometric data, texture data, and material data in the rendered scene data are subjected to preliminary lossless or lightly lossy compression to reduce the data volume and prepare for subsequent transmission.
[0014] Furthermore, in step S2, the formula for calculating the importance score of the rendered primitive is: ,in This represents the distance from the rendered primitive to the client's viewpoint. This represents the projected area of the rendered primitive on the client screen. This represents the geometric curvature of the region where the rendered primitive is located. Represents the visibility factor of rendered primitives. This represents the attention factor of the rendered primitive. This represents the corresponding weighting coefficient.
[0015] Furthermore, in step S3, the construction of the feature matrix and weight vector of the rendered primitive is achieved by performing matrix multiplication on the feature matrix and the weight vector: standardizing the distance, projected area, geometric curvature, visibility, and attention factor contained in the rendered primitive to ensure the dimensionality consistency of each feature in the feature matrix; the feature matrix for The matrix, where The total number of rendered primitives. The number of features corresponding to each rendered primitive, each row of the feature matrix represents a feature vector of the rendered primitive, and each component of the feature vector corresponds to at least one of the following: distance, projected area, geometric curvature, visibility, and attention factor of the rendered primitive; the column vector of the weight vector, wherein each element corresponds to the weighting coefficient of the corresponding feature in the feature matrix.
[0016] Furthermore, in step S3, the important data points in the rendered primitives that need to be transmitted first are determined according to the importance score vector, a transmission importance threshold is set, and the rendered primitives with scores greater than or equal to the transmission importance threshold in the importance score vector are marked as the important data points. The transmission importance threshold is dynamically adjusted according to the current network bandwidth, client device performance, or user tolerance for transmission latency.
[0017] Furthermore, in step S4, the transmission method of the non-critical data points includes high-intensity compression coding or delayed transmission. The high-intensity compression coding adopts a lossy compression algorithm, including low-resolution compression of the texture of non-critical areas. The delayed transmission refers to transmitting the non-critical data points after the network bandwidth allows or after the transmission of the critical data points is completed.
[0018] Furthermore, the client loading step also includes, after step Y1, the client dynamically adjusts the buffering strategy or loading order of the rendering data blocks according to the priority of the data blocks in the rendering data stream, the client's system resource load, and the current display performance, so as to ensure that the high-priority loaded data can occupy and utilize the client resources for fast decoding, decompression, or uploading to the graphics processor, while the low-priority loaded data is delayed or dynamically unloaded, thereby optimizing the client's resource utilization efficiency.
[0019] Furthermore, in step Y3, high-priority data in the rendering data blocks that need to be loaded and rendered first are determined, along with the client feature matrix. for The matrix, where The total number of the rendered data blocks that have been received or are yet to be loaded. The number of client-side features corresponding to each rendered data block; each row of the client feature matrix represents a feature vector of the rendered data block; each component of the feature vector corresponds to at least one of the geometric detail level, rendering readiness, or user interaction area correlation of the rendered data block; the client weight vector for The column vector contains elements, each corresponding to a weighted coefficient of the corresponding feature in the client feature matrix; a loading importance threshold is set, and rendering data blocks whose scores in the client importance score vector are greater than or equal to the loading importance threshold are marked as high-priority loading data.
[0020] Furthermore, in step Y4, during the initial rendering, the client quickly generates and displays a rough skeleton of the 3D model using high-priority loaded data. After the high-priority loaded data rendering is completed, the client progressively decodes, decompresses, and loads the low-priority loaded data to perform detailed rendering. The detailed rendering includes at least one of the following: loading higher-level detail models of the 3D model to improve geometric accuracy; loading higher-resolution texture data of the 3D model to improve texture clarity; applying more complex materials or lighting effects to improve visual realism. The client applies the detailed rendering to the displayed 3D model in real time by dynamically updating the rendering pipeline or per-pixel shader, thereby progressively improving the refinement of the rendered image.
[0021] Compared with the prior art, the beneficial effects of the present invention are as follows:
[0022] This invention introduces a viewpoint-aware importance assessment model to achieve refined grading and differentiated transmission of rendering data, significantly improving transmission efficiency. Based on factors such as the client's real-time viewpoint, the model's projected area on the screen, and the complexity of geometric details, this invention accurately identifies the most critical data points for the user's visual experience. By prioritizing and efficiently transmitting important data points while intensively compressing or delaying the transmission of less important data points, it avoids the bandwidth waste and transmission bottlenecks caused by treating all data equally in traditional methods. This ensures the rapid delivery of key information, thereby achieving rapid response and display of high-precision 3D models within limited bandwidth.
[0023] This invention implements a matrix-based importance reassessment and hierarchical loading mechanism on the client side, significantly optimizing the local loading and rendering experience of high-precision 3D models. The client can dynamically reassess the loading priority of received data based on its device performance, rendering status, and user interaction. By utilizing matrix calculations for efficient batch processing of data, high-priority data is decoded, loaded, and rendered first, allowing users to quickly obtain a rough overview of the model and interact with it. Subsequently, low-priority data is gradually loaded, progressively improving the image detail. This effectively alleviates the problems of long loading times for large models and initial rendering stuttering, providing users with a smoother, real-time 3D model browsing experience. Attached Figure Description
[0024] Figure 1 This is an exemplary flowchart of the lightweight transmission method of the present invention.
[0025] Figure 2 This is an exemplary flowchart of the loading optimization step of the present invention. Detailed Implementation
[0026] The present invention will be further described below with reference to specific embodiments.
[0027] This application discloses a lightweight transmission and loading optimization method for high-precision 3D model cloud rendering. It aims to significantly improve the rendering efficiency, response speed, and user experience of complex 3D models under different device and network conditions by intelligently managing and optimizing the transmission and client loading process of 3D rendering data. This method is particularly suitable for cloud rendering scenarios. Through server-side data preprocessing, importance assessment, and encoded transmission, as well as client-side receiving, reassessment, and hierarchical loading rendering, it achieves a balance between high-precision visual effects and a smooth interactive experience.
[0028] In one embodiment, the method comprises two main parts: a cloud-based lightweight rendering transport method executed by a cloud rendering server, and a client-side loading step executed by the client. These two parts work together to optimize the rendering process of the 3D model.
[0029] like Figure 1 The diagram illustrates the lightweight cloud rendering transmission method of this embodiment, executed by the cloud rendering server, and includes:
[0030] Step S1: Obtain the original rendering scene data of the high-precision 3D model and preprocess the original rendering scene data. The preprocessing includes constructing the hierarchical detail structure of the rendering scene data and generating the spatial partitioning structure of the rendering scene data, and performing preliminary compression on the general data in the rendering scene data.
[0031] In one embodiment, step S1 is a crucial step in server-side data preparation.
[0032] Acquiring Raw Rendering Scene Data: The server first retrieves complete, high-precision 3D model data from a data storage system or asset repository. This may include the model's geometric information, topological information, texture data, material properties, animation data, lighting information, and other entities and attributes in the scene. This raw data is typically very large and unsuitable for direct transmission. Preprocessing: Preprocessing optimizes the data structure and size for efficient subsequent transmission and rendering. Constructing Hierarchical Detail Structure: This structure contains multiple geometric representations of the 3D model at different levels of geometric detail. For example, the original model can be polygonally simplified to generate a series of model versions with different numbers of vertices and faces, which can dynamically select and adjust their display complexity based on the client's viewpoint distance. Additionally, texture images of different resolutions can be generated for the texture data. Generating Spatial Partition Structure: To efficiently perform view frustum clipping and construct hierarchical detail structure selection, the server constructs a spatial partition structure, such as an octree, quadtree, or bounding box hierarchy. These structures divide the entire 3D scene into several sub-regions, each containing a portion of rendering primitives, so that the client only loads and renders data within the viewpoint's visible area. Preliminary compression: Before transmission, general data in the rendered scene data is preliminarily compressed. For example, lossless compression algorithms or lightweight lossy compression algorithms can be used to reduce the data volume while maintaining good data quality, preparing it for subsequent transmission.
[0033] Step S2: Based on the viewpoint data transmitted by the client in real time, the importance of the rendering primitives in the rendering scene data is evaluated to generate an importance score for the rendering primitives. The importance evaluation is based on at least one of the following: the distance of the rendering primitive to the client's viewpoint, the projected area on the client's screen, the geometric curvature, the visibility, and user interaction and attention factors.
[0034] In step S2, the formula for calculating the importance score of rendered primitives is: ,in This represents the distance from the rendered primitive to the client's viewpoint. This represents the projected area of the rendered primitive on the client's screen. This represents the geometric curvature of the region where the rendered primitive is located. Represents the visibility factor of rendered primitives. This represents the attention factor of the rendered primitive. This represents the corresponding weighting coefficient.
[0035] In one embodiment, the parameters and weights of the formula are defined as follows:
[0036] This represents the overall importance score of a single rendered primitive. The higher the score, the more critical the primitive is to the user's visual experience or interaction behavior from the current viewpoint, and the more it should be prioritized for transmission and rendering.
[0037] This represents the Euclidean distance from the rendered primitive to the client's viewpoint. 1 / The term implies that the closer the distance, the greater the contribution of the term, which aligns with the visual perception that "near objects are more important".
[0038] This represents the projected area of the rendered primitive on the client's screen. This value is calculated based on the primitive's actual size and distance from the viewpoint, reflecting the visual proportion the primitive occupies on the screen. The larger the projected area, the greater this contribution.
[0039] This represents the geometric curvature of the region containing the rendered primitives. For example, for a mesh model, it could be the average rate of change of vertex normals or Gaussian curvature. The greater the curvature, the greater the contribution of this term.
[0040] This represents the visibility factor of the rendered primitive. This factor is typically 0 or 1, indicating whether the primitive is within the current view frustum and not completely occluded by other objects. A value of 1 indicates visibility, and 0 indicates invisibility, ensuring that only visible data is transmitted. In more complex implementations, it can be a value between 0 and 1, representing partial occlusion.
[0041] This represents the attention factor of rendered primitives. It's a metric that reflects user intent. For example, if a user is pointing or clicking on an object with their mouse, or if eye tracking indicates the user is gazing at an area, then the primitives within that area... The value will increase significantly.
[0042] This represents the corresponding weighting coefficients. These weighting coefficients are configurable and are used to adjust the relative contribution of each importance assessment factor to the total score. In one embodiment, these weights can be adjusted according to the application type. For example, for applications requiring high interaction accuracy, the weights can be appropriately increased. , The weights of and can be increased for applications that need to emphasize visual details. These weights are typically optimized through expert experience, user research data, or machine learning algorithms.
[0043] Step S3: Construct the feature matrix and weight vector of the rendering primitives. Calculate the importance score vector of the rendering primitives by performing matrix multiplication on the feature matrix and weight vector, and determine the important data points in the rendering primitives that need to be transmitted first based on the importance score vector.
[0044] In step S3, the feature matrix and weight vector of the rendered primitive are constructed by performing matrix multiplication on the feature matrix and the weight vector: distance, projected area, geometric curvature, visibility, and attention factor included in the rendered primitive are standardized to ensure the consistency of the dimensions of each feature in the feature matrix; the feature matrix... for The matrix, where The total number of rendered primitives. The number of features corresponding to each rendered primitive, each row of the feature matrix represents a feature vector of the rendered primitive, and each component of the feature vector corresponds to at least one of the following: distance, projected area, geometric curvature, visibility, and attention factor of the rendered primitive; the column vector of the weight vector, wherein each element corresponds to the weighting coefficient of the corresponding feature in the feature matrix.
[0045] In step S3, important data points in the rendered primitives that need to be transmitted first are determined based on the importance score vector. A transmission importance threshold is set, and rendered primitives with scores greater than or equal to the transmission importance threshold in the importance score vector are marked as important data points. The transmission importance threshold is dynamically adjusted based on the current network bandwidth, client device performance, or user tolerance for transmission latency.
[0046] Step S4: Lightweight encoding is performed on important data points and they are transmitted with priority, while high-intensity compression encoding or delayed transmission is performed on non-important data points.
[0047] In one embodiment, step S4 is the execution phase of the server-side transmission strategy, which embodies the essence of differentiated transmission.
[0048] Important Data Point Processing: Lightweight encoding is applied to the important data points. This means using efficient, fast decoding encoding algorithms while maintaining high data quality as much as possible. For example, progressive mesh encoding, efficient geometric data compression algorithms, or high-quality texture compression formats can be used. This encoded data is prioritized for transmission to the client over the network. Non-Important Data Point Processing: For the non-important data points, the system adopts a more aggressive transmission strategy to maximize bandwidth and server resource savings. High-Intensity Compression Encoding: Lossy compression algorithms are used to achieve a higher compression ratio within acceptable visual loss limits. For example, low-resolution compression can be applied to textures in non-critical areas; more aggressive simplification can be applied to geometric details. Delayed Transmission: Delayed transmission refers to transmitting the non-important data points only after network bandwidth allows or after the important data points have been transmitted. This means that non-important data is only sent when the network is idle or after the client has finished rendering the main content, thus avoiding blocking the transmission channel of critical data and ensuring the rapid presentation of core visual content.
[0049] In step S4, the transmission methods for non-critical data points include high-intensity compression coding or delayed transmission. High-intensity compression coding uses a lossy compression algorithm, including low-resolution compression of textures in non-critical areas. Delayed transmission refers to transmitting non-critical data points only after network bandwidth allows or after the transmission of critical data points has been completed.
[0050] like Figure 2 The diagram shows the client loading steps in this embodiment, which are executed by the client and include:
[0051] Step Y1 involves receiving the rendered data stream after lightweight transmission; in one embodiment, step Y1 is the client receiving data sent by the server. The client continuously listens for the rendered data stream from the server and caches it for processing. Because the server has performed differentiated transmission and encoding, the data streams received by the client will exhibit different priorities, meaning that important data will arrive first.
[0052] Step Y2: Based on at least one of the client's current rendering state, display resolution, and user interaction, re-evaluate the client importance of the rendering data blocks in the rendering data stream.
[0053] In one embodiment, step Y2 is a secondary intelligent assessment performed by the client based on its own real-time situation. Although the server has already performed an importance assessment, the client may have its own unique real-time state, and these factors require the client to adjust the importance itself.
[0054] Client-side rendering status: This includes factors such as the current frame rate (e.g., frame rate smoothness), VRAM usage, and CPU load. If client resources are limited, it may be necessary to reduce data precision requirements and prioritize loading coarse models. Display resolution: If the client's screen resolution is low, high-resolution textures or extremely fine geometric details may not need to be loaded immediately. User interaction: Similar to the server-side, if the user performs new zoom, pan, or rotate operations on the client, or clicks on an object, the client will re-evaluate the importance of received or being received data blocks based on these new interactions, prioritizing the parsing and rendering of the area or object the user is currently focusing on. For example, if the user zooms in on an area, the data blocks in that area will immediately have a higher priority, even if they were not initially prioritized on the server. Client-side feature matrix. and weight vector The client feature matrix for The matrix, where The total number of the rendered data blocks that have been received or are yet to be loaded. The number of client-side features corresponding to each rendered data block. Each row of the feature matrix represents a feature vector of the rendered data block, and each component corresponds to at least one of the following: geometric detail level, rendering readiness, or correlation of the user interaction area of the rendered data block. The client-side weight vector... for The vector consists of columns, where each element corresponds to a weighted coefficient of a feature in the client feature matrix. These coefficients can be dynamically adjusted based on client performance, network conditions, and user preferences. Loading importance threshold: A loading importance threshold is set, and rendering data blocks in the client importance score vector with scores greater than or equal to the loading importance threshold are marked as high-priority loading data. This threshold can also be dynamically adjusted based on the client's real-time resource status and the user's requirements for rendering smoothness.
[0055] Step Y3: Construct the client feature matrix and client weight vector of the rendering data block. Through matrix multiplication, determine the high-priority data to be loaded and rendered in the rendering data block.
[0056] Step Y4 prioritizes decoding, decompressing, and loading high-priority data for rendering, and then gradually decodes, decompresses, and loads low-priority data for detailed rendering, gradually improving the refinement of the rendered image.
[0057] In one embodiment, step Y4 is the final step in the client performing data loading and rendering, achieving progressive rendering.
[0058] Rough Skeleton Fast Rendering: During the initial rendering, the client utilizes high-priority data defined in Y3, prioritizing decoding and decompression before uploading it to the graphics processor to quickly generate and display a rough skeleton or low-resolution view of the 3D model. This ensures users see an interactive initial view in the shortest possible time, even if details are not yet complete. This improves the user experience and avoids long white screen waits. Detailed Rendering: After the high-priority data rendering is complete, the client progressively decodes, decompresses, and loads the low-priority data. This low-priority data is used to add detail to the displayed rough model, gradually improving the rendering's precision. Higher-Level Detail Model: Loads higher-level detail models of the 3D model to improve geometric accuracy, making the model surface smoother and more realistic. Higher-Resolution Texture Data: Loads higher-resolution texture data of the 3D model to improve texture clarity, enriching surface details such as scratches and wear. Complex Materials or Lighting Effects: Applyes more complex materials or lighting effects to enhance visual realism and create a more immersive experience. Dynamically updated rendering pipeline: The client dynamically updates the rendering pipeline or per-pixel shaders to apply these detailed renderings to the displayed 3D model in real time. This means that the client does not need to re-render the entire scene, but gradually adds details on the existing basis, thereby efficiently improving the refinement of the rendered image and achieving a smooth transition from blurry to clear, from coarse to fine.
[0059] In this embodiment, the hierarchical detail structure includes multiple geometric representations of the 3D model at different geometric detail levels. The geometric representation is the number of polygons or texture resolution from coarse to fine, so as to dynamically select and adjust the display complexity of the 3D model according to the client's viewpoint distance or the importance of the rendered primitives. The geometric data, texture data and material data in the rendering scene data are initially compressed without loss or with light loss to reduce the data volume and prepare for subsequent transmission.
[0060] The client loading step in this embodiment further includes, after step Y1, the client dynamically adjusts the buffering strategy or loading order of the rendering data blocks according to the priority of the data blocks in the rendering data stream, the client's system resource load, and the current display performance, so as to ensure that high-priority loaded data can occupy and utilize client resources first for fast decoding, decompression, or uploading to the graphics processor, while low-priority loaded data is delayed or dynamically unloaded, thereby optimizing the client's resource utilization efficiency.
[0061] In step Y3, the high-priority data in the rendering data blocks that need to be loaded and rendered first is determined, and the client feature matrix is used. for The matrix, where The total number of the rendered data blocks that have been received or are yet to be loaded. The number of client-side features corresponding to each rendered data block; each row of the client feature matrix represents a feature vector of the rendered data block; each component of the feature vector corresponds to at least one of the geometric detail level, rendering readiness, or user interaction area correlation of the rendered data block; the client weight vector for The column vector contains elements, each corresponding to a weighted coefficient of the corresponding feature in the client feature matrix; a loading importance threshold is set, and rendering data blocks whose scores in the client importance score vector are greater than or equal to the loading importance threshold are marked as high-priority loading data.
[0062] In step Y4, during the initial rendering, the client uses high-priority loaded data to quickly generate and display a rough skeleton of the 3D model. After the high-priority loaded data rendering is complete, the client gradually decodes, decompresses, and loads low-priority loaded data to perform detailed rendering. Detailed rendering includes at least one of the following: loading higher-level detail models of the 3D model to improve geometric accuracy; loading higher-resolution texture data of the 3D model to improve texture clarity; applying more complex materials or lighting effects to improve visual realism. The client applies detailed rendering to the displayed 3D model in real time by dynamically updating the rendering pipeline or per-pixel shader, thereby gradually improving the refinement of the rendered image.
[0063] Those skilled in the art will understand that the above embodiments are merely exemplary, and various modifications and equivalent substitutions can be made without departing from the spirit and scope of the invention. For example, specific feature point algorithms, optimizer selection, distortion model details, etc., can be adjusted according to actual needs. Therefore, the scope of protection of the present invention should be defined by the appended claims.
Claims
1. A lightweight transmission and loading optimization method for high-precision 3D model cloud rendering, characterized by: This includes a lightweight delivery method for cloud rendering, executed by the cloud rendering server, including: Step S1: Obtain the original rendering scene data of the high-precision 3D model, and preprocess the original rendering scene data. The preprocessing includes constructing the hierarchical detail structure of the rendering scene data and generating the spatial partitioning structure of the rendering scene data, and performing preliminary compression on the general data in the rendering scene data. Step S2: Based on the viewpoint data transmitted by the client in real time, the importance of the rendering primitives in the rendering scene data is evaluated to generate an importance score for the rendering primitives. The importance evaluation is based on at least one of the following: the distance of the rendering primitive to the client's viewpoint, the projected area on the client's screen, the geometric curvature, the visibility, and user interaction and attention factors. Step S3: Construct the feature matrix and weight vector of the rendered primitive. Calculate the importance score vector of the rendered primitive by performing matrix multiplication on the feature matrix and the weight vector. Determine the important data points in the rendered primitive that need to be transmitted first based on the importance score vector. Step S4: Lightweight encoding is performed on the important data points and they are transmitted with priority, while high-intensity compression encoding or delayed transmission is performed on the non-important data points.
2. The lightweight transmission and loading optimization method for high-precision 3D model cloud rendering according to claim 1, characterized in that: This includes the client-side loading step, which is executed by the client and includes: Step Y1: Receive the rendering data stream after the lightweight transmission; Step Y2: Based on at least one of the client's current rendering state, display resolution, and user interaction, re-evaluate the client importance of the rendering data blocks in the rendering data stream; Step Y3: Construct the client feature matrix and client weight vector of the rendering data block, and determine the high-priority loading data in the rendering data block that needs to be loaded and rendered first through matrix multiplication. Step Y4: Prioritize decoding, decompressing, and loading the high-priority loaded data for rendering, and gradually decode, decompress, and load the low-priority loaded data for detailed rendering, thereby gradually improving the refinement of the rendered image.
3. The lightweight transmission and loading optimization method for high-precision 3D model cloud rendering according to claim 1, characterized in that: The hierarchical detail structure includes multiple geometric representations of the 3D model at different levels of geometric detail. The geometric representation is a number of polygons or texture resolution from coarse to fine, so as to dynamically select and adjust the display complexity of the 3D model according to the client viewpoint distance or the importance of the rendered primitives. The geometric data, texture data, and material data in the rendered scene data are subjected to preliminary lossless or lightly lossy compression to reduce the data volume and prepare for subsequent transmission.
4. The lightweight transmission and loading optimization method for high-precision 3D model cloud rendering according to claim 1, characterized in that: In step S2, the formula for calculating the importance score of the rendered primitive is: ,in This represents the distance from the rendered primitive to the client's viewpoint. This represents the projected area of the rendered primitive on the client screen. This represents the geometric curvature of the region where the rendered primitive is located. Represents the visibility factor of rendered primitives. This represents the attention factor of the rendered primitive. This represents the corresponding weighting coefficient.
5. The lightweight transmission and loading optimization method for high-precision 3D model cloud rendering according to claim 1, characterized in that: In step S3, the feature matrix and weight vector of the rendered primitive are constructed by performing matrix multiplication on the feature matrix and the weight vector: distance, projected area, geometric curvature, visibility, and attention factor included in the rendered primitive are standardized to ensure the consistency of the dimensions of each feature in the feature matrix; the feature matrix... for The matrix, where The total number of rendered primitives. The number of features corresponding to each rendered primitive, each row of the feature matrix represents a feature vector of the rendered primitive, and each component of the feature vector corresponds to at least one of the following: distance, projected area, geometric curvature, visibility, and attention factor of the rendered primitive; the column vector of the weight vector, wherein each element corresponds to the weighting coefficient of the corresponding feature in the feature matrix.
6. The lightweight transmission and loading optimization method for high-precision 3D model cloud rendering according to claim 1, characterized in that: In step S3, the important data points in the rendered primitives that need to be transmitted first are determined according to the importance score vector. A transmission importance threshold is set, and the rendered primitives with scores greater than or equal to the transmission importance threshold in the importance score vector are marked as the important data points. The transmission importance threshold is dynamically adjusted according to the current network bandwidth, client device performance, or user tolerance for transmission latency.
7. The lightweight transmission and loading optimization method for high-precision 3D model cloud rendering according to claim 1, characterized in that: In step S4, the transmission method of the non-critical data points includes high-intensity compression coding or delayed transmission. The high-intensity compression coding adopts a lossy compression algorithm, including low-resolution compression of the texture of non-critical areas. The delayed transmission refers to transmitting the non-critical data points after the network bandwidth allows or after the transmission of the critical data points is completed.
8. The lightweight transmission and loading optimization method for high-precision 3D model cloud rendering according to claim 2, characterized in that: The client loading step further includes, after step Y1, the client dynamically adjusts the buffering strategy or loading order of the rendering data blocks according to the priority of the data blocks in the rendering data stream, the client's system resource load, and the current display performance, so as to ensure that the high-priority loaded data can occupy and utilize the client resources for fast decoding, decompression, or uploading to the graphics processor, while the low-priority loaded data is processed with delay or dynamically unloaded.
9. The lightweight transmission and loading optimization method for high-precision 3D model cloud rendering according to claim 2, characterized in that: In step Y3, the high-priority data in the rendering data blocks that need to be loaded and rendered first is determined, and the client feature matrix is used. for The matrix, where The total number of the rendered data blocks that have been received or are yet to be loaded. The number of client-side features corresponding to each rendered data block; each row of the client feature matrix represents a feature vector of the rendered data block; each component of the feature vector corresponds to at least one of the geometric detail level, rendering readiness, or user interaction area correlation of the rendered data block; the client weight vector for The column vector contains elements, each corresponding to a weighted coefficient of the corresponding feature in the client feature matrix; a loading importance threshold is set, and rendering data blocks whose scores in the client importance score vector are greater than or equal to the loading importance threshold are marked as high-priority loading data.
10. The lightweight transmission and loading optimization method for high-precision 3D model cloud rendering according to claim 2, characterized in that: In step Y4, during the initial rendering, the client uses high-priority loaded data to quickly generate and display a rough skeleton of the 3D model; after the high-priority loaded data rendering is completed, the client gradually decodes, decompresses and loads the low-priority loaded data to perform detailed rendering, which includes at least one of the following: loading higher-level detailed models of the 3D model to improve geometric accuracy. Load higher resolution texture data from 3D models to improve texture clarity; apply more complex materials or lighting effects to enhance visual realism; The client dynamically updates the rendering pipeline or per-pixel shader to apply detailed rendering to the displayed 3D model in real time, gradually improving the precision of the rendered image.