Non-perception lightweight three-dimensional data processing method and device and storage medium

Abstract

The invention discloses a non-perceptual lightweight three-dimensional data processing method and device and a storage medium, relates to the technical field of three-dimensional data processing, and discloses the non-perceptual lightweight three-dimensional data processing method, which comprises the following steps: acquiring key feature information of preprocessed three-dimensional model data, carrying out importance evaluation processing on the key feature information based on a graph theory algorithm to obtain an importance score of the key feature information; after removing the key feature information of which the importance score is smaller than a preset score in the key feature information, performing data simplification processing on the key feature information based on a preset simplification algorithm; and determining a compression strategy according to the data type of the three-dimensional model data, and after encoding the three-dimensional model data, performing compression processing on the encoded data based on the compression strategy to obtain compressed data. And dimension reduction and simplification processing is performed on the three-dimensional data based on the importance score of the key feature information, so that the data processing efficiency is improved.

Description

Technical Field

[0001] The present application relates to the field of three-dimensional data processing technology, and in particular to a method, device and storage medium for imperceptible lightweight three-dimensional data processing. Background Art

[0002] When processing 3D model data, the processing algorithms selected usually include geometric simplification and texture compression algorithms. Geometric simplification refers to the use of algorithms to analyze the geometric features of the 3D model and remove details that have little impact on the overall visual effect, so that the simplified model remains visually similar to the original model, while the number of vertices, faces and other geometric information are significantly reduced, thereby reducing storage requirements and rendering complexity. Texture compression uses a texture compression algorithm to compress the texture image of the 3D model, so that the compressed texture image can maintain a certain visual quality while significantly reducing storage requirements and speeding up texture loading and rendering.

[0003] However, existing geometric simplification and texture compression methods often require manual setting of parameters, such as error thresholds, compression rates, etc. Manual parameter setting relies on subjective judgment. If the simplification degree is too high, the shape of the 3D model will be deformed and important detail features will be lost. If the compression rate is not set properly, texture details will be lost. Therefore, the current 3D model data processing method has a low degree of automation and low processing efficiency.

[0004] The above contents are only used to assist in understanding the technical solution of the present application and do not constitute an admission that the above contents are prior art. Summary of the invention

[0005] The main purpose of this application is to provide a non-perceptual lightweight three-dimensional data processing method, device and storage medium, aiming to solve the current technical problem of low efficiency in three-dimensional model data processing.

[0006] To achieve the above objectives, the present application proposes a non-perceptual lightweight three-dimensional data processing method, the method comprising: Acquire key feature information of the preprocessed three-dimensional model data, and perform importance evaluation processing on the key feature information based on a graph theory algorithm to obtain an importance score of the key feature information; After removing the key feature information whose importance score is less than a preset score from the key feature information, performing data simplification processing on the key feature information based on a preset simplification algorithm; A compression strategy is determined according to the data type of the three-dimensional model data, and after the three-dimensional model data is encoded, the encoded data is compressed based on the compression strategy to obtain compressed data.

[0007] In one embodiment, after the step of determining a compression strategy according to the data type of the three-dimensional model data, and compressing the encoded data after encoding the three-dimensional model data based on the compression strategy to obtain compressed data, the step further includes: Respond to the decompression information sent by the client; Read and decompress the data to be decompressed based on a preset decompression algorithm to obtain decompressed target three-dimensional model data; The target three-dimensional model data is output to the client.

[0008] In one embodiment, the step of reading and decompressing the data to be decompressed based on a preset decompression algorithm to obtain the decompressed target three-dimensional model data includes: Reading the characteristic information, metadata and the data to be decompressed during the compression processing of the three-dimensional model data based on the preset decompression algorithm; The data to be decompressed is decompressed, and the decompressed data is restored based on the feature information and the metadata to obtain the target three-dimensional model data.

[0009] In one embodiment, after removing the key feature information whose importance score is less than a preset score from the key feature information, the step of performing data simplification processing on the key feature information based on a preset simplification algorithm includes: Determine, in the key feature information, feature information to be reduced in dimension whose importance score feature value is less than a preset score; Performing dimensionality reduction processing on the feature information to be reduced in dimension based on a preset dimensionality reduction algorithm to remove the key feature information whose importance score is less than a preset score in the key feature information; The key feature information is subjected to data simplification processing based on the preset simplification algorithm.

[0010] In one embodiment, the step of determining a compression strategy according to the data type of the three-dimensional model data, and compressing the encoded data after encoding the three-dimensional model data based on the compression strategy to obtain compressed data includes: Acquire a data type of the three-dimensional model data, and determine the compression strategy associated with the data type; Performing encoding processing on the three-dimensional model data based on a preset encoding algorithm to obtain the encoded data; The encoded data is compressed according to a compression algorithm corresponding to the compression strategy to obtain the compressed data.

[0011] In one embodiment, the step of obtaining key feature information of the preprocessed three-dimensional model data, and performing importance evaluation processing on the key feature information based on a graph theory algorithm to obtain an importance score of the key feature information includes: Based on a geometric feature extraction method, obtaining the key feature information of the preprocessed three-dimensional model data; or based on a deep learning model, obtaining the key feature information of the preprocessed three-dimensional model data; The key feature information is evaluated for importance based on a graph theory algorithm to obtain the importance evaluation information.

[0012] In one embodiment, before the step of obtaining key feature information of the preprocessed three-dimensional model data and performing importance evaluation processing on the key feature information based on a graph theory algorithm to obtain an importance score of the key feature information, the step further includes: Receiving original three-dimensional model data, and performing noise removal processing on the original three-dimensional data based on Gaussian filtering, median filtering, bilateral filtering or wavelet transform to obtain denoised data; Data reconstruction and simplification processing are performed based on the denoised data to obtain the pre-processed three-dimensional model data.

[0013] In one embodiment, after the step of receiving the original three-dimensional model data and performing noise removal processing on the original three-dimensional data based on Gaussian filtering, median filtering, bilateral filtering or wavelet transform to obtain denoised data, the method further includes: The denoised data is simplified based on a voxel grid simplification method to obtain the pre-processed three-dimensional model data.

[0014] In addition, to achieve the above-mentioned purpose, the present application also proposes a three-dimensional data processing device, which includes: a memory, a processor, and a computer program stored on the memory and executable on the processor, wherein the computer program is configured to implement the steps of the imperceptible lightweight three-dimensional data processing method as described above.

[0015] In addition, to achieve the above-mentioned purpose, the present application also proposes a storage medium, which is a computer-readable storage medium, and a computer program is stored on the storage medium. When the computer program is executed by the processor, the steps of the imperceptible lightweight three-dimensional data processing method as described above are implemented.

[0016] One or more technical solutions proposed in this application have at least the following technical effects: When processing the three-dimensional model data, the key feature information of the three-dimensional model data is first obtained, including key feature points, edges, and surfaces used for calculating and processing the three-dimensional model data. At the same time, the importance of these key feature information is evaluated through a graph theory algorithm to obtain a corresponding importance score, so as to remove the key feature information whose importance score is less than the preset score. Then, other key feature information is simplified based on a preset simplification algorithm. No manual operation is required in this process. Finally, the compression strategy corresponding to the three-dimensional model data is determined, and the three-dimensional model data after encoding is compressed based on the compression strategy to obtain compressed data of the three-dimensional model data. Based on this, the redundancy and importance of the three-dimensional data can be automatically analyzed, and the compression parameters and strategies can be dynamically adjusted to ensure the best balance between data compression rate and fidelity. There is no need for manual intervention or parameter setting by the user, so as to improve the efficiency of three-dimensional data processing and compression and achieve imperceptible processing. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the present application and, together with the description, serve to explain the principles of the present application.

[0018] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, for ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative labor.

[0019] Figure 1 A schematic diagram of a flow chart provided for the first embodiment of the non-perceptual lightweight three-dimensional data processing method of the present application; Figure 2 A schematic diagram of the process of data dimensionality reduction and simplification processing for the non-perceptual lightweight three-dimensional data processing method of this application; Figure 3 A schematic diagram of the encoding and compression process of the imperceptible lightweight three-dimensional data processing method of this application; Figure 4 A flowchart diagram of the second embodiment of the non-perceptual lightweight three-dimensional data processing method of the present application; Figure 5 A schematic diagram of an exemplary implementation flow provided for the second embodiment of the present application; Figure 6 A schematic diagram of a flow chart provided for the third embodiment of the method for imperceptible lightweight three-dimensional data processing of the present application; Figure 7 This is a schematic diagram of a process for extracting key features based on a deep learning model in the third embodiment of the present application; Figure 8This is a schematic diagram of a process for preprocessing original three-dimensional data in the fourth embodiment of the present application; Fig. 9 This is a schematic diagram of the device structure of the hardware operating environment involved in the imperceptible lightweight three-dimensional data processing method in the embodiment of the present application.

[0020] The purpose, features and advantages of this application will be further described in conjunction with the embodiments and with reference to the accompanying drawings. DETAILED DESCRIPTION

[0021] It should be understood that the specific embodiments described herein are only used to explain the technical solutions of the present application and are not used to limit the present application.

[0022] The main solution of the embodiment of the present application is: obtaining key feature information of the preprocessed three-dimensional model data, and performing importance evaluation processing on the key feature information based on a graph theory algorithm to obtain an importance score of the key feature information; After removing the key feature information whose importance score is less than a preset score from the key feature information, performing data simplification processing on the key feature information based on a preset simplification algorithm; A compression strategy is determined according to the data type of the three-dimensional model data, and after the three-dimensional model data is encoded, the encoded data is compressed based on the compression strategy to obtain compressed data.

[0023] In the prior art, geometric simplification and texture compression methods often require manual setting of parameters, such as error threshold, compression rate and other parameters. Manual setting of parameters relies on subjective judgment. If the simplification degree is too high, the shape of the 3D model will be deformed and important detail features will be lost. If the compression rate is not set properly, texture details will be lost. Therefore, the current 3D model data processing method has a low degree of automation and has the defect of poor processing effect.

[0024] The present application provides a solution, which automatically obtains the key feature information of the pre-processed 3D model data and scores the importance of these key feature information through a graph theory algorithm, performs dimensionality reduction processing on the key features with lower importance based on the importance evaluation information, i.e. removes the key feature information with lower importance scores in the key feature information, so as to reduce the amount of data for subsequent calculations, and then simplifies and compresses the remaining data after the dimensionality reduction processing. This process does not require user participation, so that the 3D model data processing is performed without perception. The 3D model data is processed based on the non-perceptual intelligent optimization processing method, which effectively improves the processing effect of the 3D model data.

[0025] It should be noted that the execution subject of this embodiment may be a computing service device with data processing, network communication and program running functions, such as a tablet computer, a personal computer, a mobile phone, etc., or an electronic device capable of realizing the above functions, a three-dimensional data processing device, etc. The following takes a three-dimensional data processing device as an example to illustrate this embodiment and the following embodiments.

[0026] In order to better understand the technical solution of the present application, a detailed description will be given below in conjunction with the accompanying drawings and specific implementation methods.

[0027] The present application embodiment provides a non-perceptual lightweight three-dimensional data processing method, referring to Figure 1 , Figure 1 This is a flow chart of the first embodiment of the imperceptible lightweight three-dimensional data processing method of the present application.

[0028] In this embodiment, the imperceptible lightweight three-dimensional data processing method includes steps S10 to S30: Step S10, obtaining key feature information of the preprocessed three-dimensional model data, and performing importance evaluation processing on the key feature information based on a graph theory algorithm to obtain an importance score of the key feature information.

[0029] It should be noted that key feature data refers to the model's feature points, edges (lines) and faces, etc., which are details and attribute information that are crucial to model recognition, analysis, rendering or subsequent applications. For example, in a cube three-dimensional model, its key feature information contains at least eight feature points, lines between faces, and other content that can keep the shape of the cube unchanged. The importance of feature data evaluated based on its contribution to the overall shape and details of the three-dimensional model is an importance score, which is usually expressed in numerical form such as 0-100 points, or in a graded form such as AD. Among them, before extracting the key feature information of the three-dimensional model data, it is usually necessary to preprocess the three-dimensional model data to remove the noise of the three-dimensional model data while reducing the repeated or highly similar parts of the data, that is, reducing redundant data.

[0030] In this embodiment, the key feature information of the preprocessed three-dimensional model data can be extracted through the built-in algorithm of the three-dimensional data processing device, or the key features such as edges, faces and other information can be extracted through the built-in deep learning model of the device such as the convolutional neural network (CNN).

[0031] Furthermore, after extracting the key feature information of the three-dimensional model data, there is data whose key feature information has little impact on the subsequent calculation of the three-dimensional model. For example, when the three-dimensional model data is a building, the key feature information obtained includes hundreds of vertex data of the building, while in the actual calculation process, at least four key feature points (representing the shape of the top of the model) can be used for spatial calculation of the three-dimensional model. The importance of the remaining data is relatively low, which usually affects the storage and query efficiency of subsequent data. Therefore, it is necessary to reduce the number of vertices and faces in the model through data dimensionality reduction and model simplification. As for the key feature information such as the top surface, edge and number of faces that need to be reduced, they are selected from the importance evaluation information of these key feature information.

[0032] Specifically, after obtaining the key feature information, it is necessary to evaluate the importance of each feature based on its contribution to the overall shape and details. During the evaluation process, graph-based algorithms, such as the PageRank algorithm (page ranking algorithm) or the HITS (Hyperlink-Induced Topic Search) algorithm, are used to evaluate the importance of features. For example, node importance evaluation: use graph-based algorithms such as PageRank, HITS, and Betweenness Centrality to evaluate the importance of each node. Similarly, when evaluating edge importance, algorithms such as Edge Betweenness Centrality can be used to evaluate the importance of each edge. For face features, the importance of the entire region can be evaluated based on the importance of the nodes and edges it contains. For example, the average or weighted sum of the importance of all nodes in the region is calculated, and the calculation result is used as the importance score of the region.

[0033] It should be noted that after acquiring the key feature information, the three-dimensional data processing device automatically calculates the importance evaluation information of these features based on an algorithm, so as to perform subsequent data dimensionality reduction processing based on the importance evaluation information.

[0034] It is understandable that each key feature information corresponds to importance evaluation information. For example, if the extracted key feature information contains 100 vertices, then the 100 vertices correspond to 100 importance scores.

[0035] Step S20: After removing the key feature information whose importance score is less than a preset score from the key feature information, data simplification processing is performed on the key feature information based on a preset simplification algorithm.

[0036] In this embodiment, after obtaining the key feature information and the importance score, it is necessary to perform lightweight processing on the key feature information of the three-dimensional model data to reduce the redundancy of the three-dimensional model data. Lightweight processing refers to the dimensionality reduction of the data and the simplification of the model data. The dimensionality reduction process is to remove the key feature information with a low importance score, and the simplification process is to further reduce the amount of data that needs to be stored after the dimensionality reduction process. Therefore, the dimensionality reduction of the key feature information is completed by removing the key feature information whose importance score is less than the preset score. Among them, if the importance score is a numerical system, the feature information to be reduced in dimension with an importance score less than the preset score is selected from the key feature data, wherein the larger the numerical value in the numerical system, the higher the degree of importance.

[0037] Optionally, if the importance score is a letter grade, the data to be reduced in dimension with a feature grade of a preset grade is selected and deleted from the key feature data, wherein the preset grade is usually the grade that needs to be reduced in dimension.

[0038] Therefore, as an optional implementation, step S20 includes steps S21 to S23: Step S21, determining the feature information to be reduced in dimension whose importance score feature value is less than a preset score in the key feature information.

[0039] Step S22, performing dimensionality reduction processing on the feature information to be reduced in dimension based on a preset dimensionality reduction algorithm, so as to remove the key feature information whose importance score is less than a preset score in the key feature information.

[0040] In this embodiment, according to the importance score of the features, a preset dimensionality reduction algorithm such as principal component analysis (PCA) and multidimensional scaling (MDS) may be used to perform dimensionality reduction processing on the three-dimensional data.

[0041] For example, please refer to Figure 2 When the feature data to be reduced in dimension is selected and processed, the data can be reduced in dimension by using preset dimensionality reduction algorithms such as principal component analysis (PCA) / multi-dimensional scaling (MDS) and other algorithm technologies.

[0042] Optionally, nonlinear dimensionality reduction techniques such as ISOMAP (isometric mapping) or LLE (local linear embedding) can be used to reduce the dimensionality of key feature information, thereby better preserving the local and global structural information of the data.

[0043] Step S23: performing data simplification processing on the key feature information based on the preset simplification algorithm.

[0044] In this embodiment, after the data is processed for dimensionality reduction, it is necessary to further simplify the reduced data, such as using a preset simplification algorithm such as a quadratic error metric (QEM) algorithm to remove unnecessary vertices and faces while maintaining the basic shape and key features of the model, thereby obtaining the target key feature information after the data simplification processing.

[0045] For example, please continue to refer to Figure 2 After the key feature information is reduced in dimension, other key feature information after dimension reduction is simplified based on the preset simplification algorithm QEM algorithm, thereby obtaining a simplified three-dimensional model.

[0046] Optionally, in addition to the QEM algorithm, a progressive mesh simplification algorithm, such as the Edge Collapse algorithm, can be used to simplify the model by gradually removing edges and vertices that have the least impact on the model shape.

[0047] In this embodiment, the dimensionality reduction and simplification of the three-dimensional model data are performed by scoring the importance of key feature information. In this process, the importance of each feature information is automatically assessed and a simplification algorithm that meets the requirements is automatically selected for processing. There is no need for the user to manually set the simplification parameters, thereby improving the efficiency of automated processing.

[0048] Step S30, determining a compression strategy according to the data type of the three-dimensional model data, and after encoding the three-dimensional model data, compressing the encoded data based on the compression strategy to obtain compressed data.

[0049] The compression strategy refers to the compression algorithm selected when compressing the 3D model data after dimensionality reduction and simplification. There are many different compression algorithms preset in the 3D data processing equipment.

[0050] Therefore, in this embodiment, when compressing the 3D model data, a compression algorithm that meets the requirements can be selected based on the data type of the 3D model data, thereby eliminating the need for the user to manually set compression parameters and select a corresponding compression algorithm, thereby improving compression efficiency. When compressing the 3D model data, encoding processing is usually required before data compression, thereby improving subsequent compression efficiency.

[0051] Specifically, step S30 includes steps S31 to S33: Step S31, obtaining the data type of the three-dimensional model data, and determining the compression strategy associated with the data type.

[0052] In this embodiment, the 3D model data includes point cloud, mesh, voxel and other types, and each data type corresponds to different data features, so the associated compression strategy can be determined based on the data type of the 3D model data. For example, the current compression strategies include A, B and C. When the data type is point cloud data, the corresponding strategy is A, and when the data type is mesh data, the corresponding compression strategy is B, and so on. The specific compression strategy information is not limited in this application, and the corresponding association relationship between the strategy and the type is pre-set by the user in the device.

[0053] Step S32: encoding the three-dimensional model data based on a preset encoding algorithm to obtain the encoded data.

[0054] For example, please refer to Figure 3 After specifying the corresponding compression strategy in multiple efficient compression algorithms, the dimensionality-reduced and lightweight three-dimensional model data is encoded through technologies such as preset encoding algorithms such as predictive coding or transform coding, so as to obtain encoded data for subsequent compression processing.

[0055] Step S33: compress the encoded data according to the compression algorithm corresponding to the compression strategy to obtain the compressed data.

[0056] Please continue to refer to Figure 3 The compression algorithm corresponding to the compression strategy may include a Huffman coding algorithm or a run-length coding algorithm. Based on this, when the compression algorithm is applied based on the compression strategy, the encoded data is compressed by the Huffman coding algorithm or the run-length coding to obtain the compressed data, thus completing the compressed data storage.

[0057] Optionally, in addition to Huffman coding and run-length coding, more efficient compression algorithms can be introduced, such as dictionary compression algorithms such as LZ77 and LZ78, or prediction-based coding techniques such as the EBCOT (Embedded Block Coding Optimized Truncation) algorithm in JPEG2000, to improve the compression ratio and compression efficiency.

[0058] This embodiment selects a corresponding compression strategy according to the type of three-dimensional model data. After encoding the three-dimensional model data based on a preset encoding algorithm, the encoded data is automatically compressed using the compression algorithm of the automatically selected compression strategy, without the need for manual operation, thereby improving data processing efficiency.

[0059] It should be noted that during the data processing and compression process, the processing equipment will also use technical means such as encrypted transmission, access control, and data desensitization to ensure the security of three-dimensional data and the protection of user privacy during transmission, storage, and processing.

[0060] This embodiment provides an imperceptible lightweight three-dimensional data processing method, which automatically obtains the key feature information of the pre-processed three-dimensional model data, and then performs dimensionality reduction processing on the key feature information based on the importance score of the key feature information to remove irrelevant data with low importance that affects the calculation of the three-dimensional data, and then performs data simplification processing on other key feature information after the dimensionality reduction processing to reduce the number of points and faces of the three-dimensional model data, and further reduce the complexity of the processed data. Finally, the corresponding compression strategy is selected according to the type of three-dimensional model data, and after the three-dimensional model data after the lightweight processing is completed, the compression algorithm corresponding to the compression strategy is used to compress the three-dimensional data to obtain the compressed data for subsequent spatial calculation. Based on this, the user is unaware of the simplification and compression process of the three-dimensional model data, and the user does not need to manually set the simplification parameters and compression parameters of the three-dimensional model data, thereby improving the processing efficiency of the three-dimensional model data.

[0061] Based on the first embodiment of the present application, in the second embodiment of the present application, the same or similar contents as those in the first embodiment can refer to the above introduction, and will not be repeated in the following. Figure 4 After step S30, the imperceptible lightweight three-dimensional data processing method further includes steps S40 to S60: Step S40, responding to the decompression information sent by the client; In this embodiment, the decompression information sent by the user at the client can be responded to, based on an efficient decompression algorithm, to ensure that the decompression process is completed in a short time, reducing the user's waiting time, and at the same time improving the perception between the user and the processing device after the data compression process.

[0062] It is understandable that after the 3D model data is compressed, the user's perception needs are often not fully considered, resulting in the processed model deviating from the user's expectations in some aspects. Therefore, in this embodiment, the data is quickly displayed in response to the user's actual request to improve the user's perception effect after the 3D model data is compressed.

[0063] Step S50, reading and decompressing the data to be decompressed based on a preset decompression algorithm to obtain decompressed target three-dimensional model data.

[0064] In this embodiment, the data to be decompressed can be directly read and decompressed by the decompression algorithm to obtain the target three-dimensional model data. In the decompression process, the feature information before decompression and the metadata retained during compression can be used to perform quality recovery processing on the decompressed three-dimensional data, such as detail enhancement, texture repair and decoding processing, to improve the decompression effect.

[0065] Therefore, as an optional implementation, the feature information, metadata and the data to be decompressed during the compression process of the three-dimensional model data can be read based on a preset decompression algorithm, and then the data to be decompressed can be decompressed, and the decompressed data can be restored based on the feature information and metadata to obtain the target three-dimensional model data. It can be understood that the feature information and metadata are stored in a database table, and the data details can be enhanced through the metadata, and the data can be decoded through the feature information. In the decompression process, in order to improve the data decompression efficiency, the decompression algorithm can be a decompression algorithm based on GPU (Graphics Processing Unit) acceleration, which uses the parallel processing capabilities of the GPU to accelerate the decompression process and reduce user waiting time.

[0066] Please refer to Figure 5 After the user initiates a decompression request, the device's server receives the request and starts the decompression algorithm to improve the decompression efficiency through GPU acceleration and parallel processing. It then loads the compressed data from the storage medium, i.e., reads the compressed data, and decompresses the compressed data, including data decoding, data recovery, and detail enhancement processing. After the decompression is completed, the decompressed data is generated.

[0067] Step S60: output the target three-dimensional model data to the client.

[0068] Please continue to refer to Figure 5 After obtaining the decompressed data, the decompressed data is sent to the user device, thereby displaying the decompressed three-dimensional model in the client.

[0069] This embodiment provides a non-perceptual lightweight three-dimensional data processing method. After responding to the data decompression request of the user terminal, it performs non-perceptual decompression processing on the three-dimensional model data. By setting an efficient decompression algorithm, it ensures that the decompression process is completed in a short time, reducing the user waiting time. During the decompression process, known information is used to decompress, decode and enhance the data, thereby improving the processing efficiency of subsequent spatial data.

[0070] Based on the first embodiment of the present application, in the third embodiment of the present application, the same or similar contents as those in the first embodiment can refer to the above introduction, and will not be repeated later. Figure 6 , step S10 also includes steps S11-S12: Step S11, determining a feature extraction method for the preprocessed three-dimensional model data, and acquiring the key feature information based on the feature extraction method.

[0071] In this embodiment, the feature extraction method includes a deep learning model feature extraction method and a geometric feature extraction method. Therefore, if the feature extraction method is a deep feature extraction method, the key feature information of the pre-processed three-dimensional model data is obtained based on the geometric feature extraction method, wherein the geometric feature extraction method includes SIFT (Scale Invariant Feature Transform), SURF (Speeded Robust Feature), etc., combined with morphological analysis to identify the key feature information in the three-dimensional model data.

[0072] Optionally, if the feature extraction method is a deep learning model extraction method, the key feature information of the preprocessed three-dimensional model data is obtained based on the deep learning model. It is understandable that after the deep learning model training is completed, the key feature information of the three-dimensional model data can be automatically identified.

[0073] For example, please refer to Figure 7 In the process of feature extraction based on deep learning network, the three-dimensional model data is extracted and analyzed through deep learning models such as convolutional neural network CNN, so as to extract key features such as points, edges and faces, and obtain the key feature set, that is, key feature information.

[0074] Step S12: performing importance evaluation processing on the key feature information based on a graph theory algorithm to obtain the importance evaluation information.

[0075] In this embodiment, when evaluating the importance of features, the centrality or influence of the features in the network is calculated based on a graph theory algorithm to determine their importance, thereby effectively knowing which feature information needs to be processed for dimensionality reduction without the need for the user to manually select parameters.

[0076] Based on the first embodiment of the present application, in the fourth embodiment of the present application, the same or similar contents as those in the first embodiment can be referred to the above description, and will not be repeated in the following. On this basis, before step S10, the non-perceptual lightweight three-dimensional data processing method further includes steps S01~S02: Step S01, receiving original three-dimensional model data, and performing noise removal processing on the original three-dimensional data based on Gaussian filtering, median filtering, bilateral filtering or wavelet transform to obtain denoised data.

[0077] Step S02: reconstruct and simplify the data based on the denoised data to obtain the pre-processed three-dimensional model data.

[0078] In this embodiment, 3D model data from a scanning device, modeling software or a database can be received, and then when noise removal is performed, a Gaussian filter or a median filter algorithm is used to remove noise points in the data to improve data quality. Alternatively, a bilateral filter or a wavelet transform technique can also be used to remove noise, thereby having a better effect when processing noises of different types and complexity.

[0079] After removing the noise, the repeated or highly similar parts in the three-dimensional data are analyzed and the redundancy is reduced through data reconstruction and simplification.

[0080] Optionally, after step S01, the denoised data may be simplified based on a voxel grid simplification method, such as octree partitioning, by recursively partitioning the space to reduce redundant data while maintaining the geometric and topological structure of the data, thereby obtaining preprocessed three-dimensional model data.

[0081] For example, please refer to Figure 8 After obtaining the original three-dimensional data, the noise is removed by Gaussian filtering and median filtering, and then the redundant data is reduced by data reconstruction and simplification, and finally the preprocessed three-dimensional model data is obtained.

[0082] This embodiment removes noise and redundant data generated during the scanning process of the three-dimensional model file through the preprocessing stage, thereby improving the quality of the three-dimensional model data.

[0083] In order to help understand the contents obtained after combining the above-mentioned embodiments, taking the three-dimensional model of a car as an example, the present application first removes the noise and redundant data generated during the scanning process of the three-dimensional model file through the preprocessing stage; then, in the feature extraction and analysis stage, the key features such as the body lines, lights, and wheels of the car are automatically identified; then, in the lightweight processing stage, the key features with low importance are removed through data dimension reduction and model simplification technology, and the number of vertices and faces in the model is reduced, while the basic shape and key features of the car are maintained for subsequent analysis and processing; finally, in the efficient compression algorithm stage, the lightweight model is encoded and compressed to obtain a compact data file. When the user needs to view or edit the car model, the high-quality three-dimensional model can be quickly restored through the imperceptible decompression process.

[0084] It should be noted that before and after the three-dimensional model data of the car is processed, the model does not change at the macro level for the user, that is, there is basically no difference between the content of the model before and after processing that the user sees.

[0085] The present application provides a three-dimensional data processing device, which includes: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor can execute the imperceptible lightweight three-dimensional data processing method in the above-mentioned first embodiment.

[0086] Reference below Fig. 9 , which shows a schematic diagram of the structure of a three-dimensional data processing device suitable for implementing the embodiment of the present application. The three-dimensional data processing device in the embodiment of the present application may include but is not limited to mobile terminals such as mobile phones, laptop computers, digital broadcast receivers, personal digital assistants (PDA, Personal Digital Assistant), tablet computers (PAD, Portable Application Description), portable multimedia players (PMP, Portable Media Player), vehicle terminals (such as vehicle navigation terminals), etc., and fixed terminals such as digital TVs, desktop computers, etc. Fig. 9 The three-dimensional data processing device shown is only an example and should not bring any limitation to the functions and scope of use of the embodiments of the present application.

[0087] like Fig. 9As shown, the three-dimensional data processing device may include a processing device 1001 (such as a central processing unit, a graphics processor, etc.), which can perform various appropriate actions and processes according to a program stored in a read-only memory (ROM) 1002 or a program loaded from a storage device 1003 to a random access memory (RAM) 1004. Various programs and data required for the operation of the three-dimensional data processing device are also stored in the random access memory 1004. The processing device 1001, the read-only memory 1002, and the random access memory 1004 are connected to each other via a bus 1005. An input / output (I / O) interface 1006 is also connected to the bus. Typically, the following systems can be connected to the input / output interface 1006: input devices 1007 including, for example, a touch screen, a touchpad, a keyboard, a mouse, an image sensor, a microphone, an accelerometer, a gyroscope, etc.; output devices 1008 including, for example, a liquid crystal display (LCD), a speaker, a vibrator, etc.; storage devices 1003 including, for example, a magnetic tape, a hard disk, etc.; and communication devices 1009. The communication device 1009 can allow the three-dimensional data processing device to communicate with other devices wirelessly or by wire to exchange data. Although the figure shows a three-dimensional data processing device with various systems, it should be understood that it is not required to implement or have all the systems shown. More or fewer systems may be implemented or have alternatively.

[0088] In particular, according to the embodiments disclosed in the present application, the process described above with reference to the flowchart can be implemented as a computer software program. For example, the embodiments disclosed in the present application include a computer program product, which includes a computer program carried on a computer-readable medium, and the computer program includes a program code for executing the method shown in the flowchart. In such an embodiment, the computer program can be downloaded and installed from a network through a communication device, or installed from a storage device 1003, or installed from a read-only memory 1002. When the computer program is executed by the processing device 1001, the above-mentioned functions defined in the method of the embodiment disclosed in the present application are executed.

[0089] The three-dimensional data processing device provided by the present application adopts the imperceptible lightweight three-dimensional data processing method in the above embodiment, which can solve the technical problem of low efficiency of current three-dimensional model data processing. Compared with the prior art, the beneficial effects of the three-dimensional data processing device provided by the present application are the same as the beneficial effects of the imperceptible lightweight three-dimensional data processing method provided by the above embodiment, and the other technical features in the three-dimensional data processing device are the same as the features disclosed in the method of the previous embodiment, which will not be repeated here.

[0090] It should be understood that the various parts disclosed in this application can be implemented by hardware, software, firmware or a combination thereof. In the description of the above embodiments, specific features, structures, materials or characteristics can be combined in any one or more embodiments or examples in a suitable manner.

[0091] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art who is familiar with the present technical field can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.

[0092] The present application provides a computer-readable storage medium having computer-readable program instructions (ie, computer programs) stored thereon, and the computer-readable program instructions are used to execute the imperceptible lightweight three-dimensional data processing method in the above-mentioned embodiment.

[0093] The computer-readable storage medium provided in the present application may be, for example, a USB flash drive, but is not limited to electrical, magnetic, optical, electromagnetic, infrared or semiconductor systems, systems or devices, or any combination of the above. More specific examples of computer-readable storage media may include, but are not limited to: an electrical connection with one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM, Erasable Programmable Read Only Memory or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM, CD-Read Only Memory), an optical storage device, a magnetic storage device, or any suitable combination of the above. In this embodiment, the computer-readable storage medium may be any tangible medium containing or storing a program, which may be used by or in combination with an instruction execution system, system or device. The program code contained on the computer-readable storage medium may be transmitted using any appropriate medium, including but not limited to: wires, optical cables, radio frequencies (RF, Radio Frequency), etc., or any suitable combination of the above.

[0094] The computer-readable storage medium may be included in the three-dimensional data processing device; or may exist independently without being assembled into the three-dimensional data processing device.

[0095] The computer-readable storage medium carries one or more programs. When the one or more programs are executed by the three-dimensional data processing device, the three-dimensional data processing device: Acquire key feature information of the preprocessed three-dimensional model data, and perform importance evaluation processing on the key feature information based on a graph theory algorithm to obtain an importance score of the key feature information; After removing the key feature information whose importance score is less than a preset score from the key feature information, performing data simplification processing on the key feature information based on a preset simplification algorithm; A compression strategy is determined according to the data type of the three-dimensional model data, and after the three-dimensional model data is encoded, the encoded data is compressed based on the compression strategy to obtain compressed data.

[0096] Computer program code for performing the operations of the present application may be written in one or more programming languages ​​or a combination thereof, including object-oriented programming languages ​​such as Java, Smalltalk, C++, and conventional procedural programming languages ​​such as "C" or similar programming languages. The program code may be executed entirely on the user's computer, partially on the user's computer, as a separate software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In the case of a remote computer, the remote computer may be connected to the user's computer via any type of network, including a local area network (LAN) or a wide area network (WAN), or may be connected to an external computer (e.g., via the Internet using an Internet service provider).

[0097] The flowcharts and block diagrams in the accompanying drawings illustrate the possible architecture, functions and operations of the systems, methods and computer program products according to various embodiments of the present application. In this regard, each box in the flowchart or block diagram can represent a module, a program segment or a part of a code, and the module, a program segment or a part of the code contains one or more executable instructions for realizing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the box can also occur in a different order from the order marked in the accompanying drawings. For example, two boxes represented in succession can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each box in the block diagram and / or flowchart, and the combination of boxes in the block diagram and / or flowchart can be implemented with a dedicated hardware-based system that performs a specified function or operation, or can be implemented with a combination of dedicated hardware and computer instructions.

[0098] The modules involved in the embodiments described in this application may be implemented by software or hardware, wherein the name of the module does not constitute a limitation on the unit itself in some cases.

[0099] The readable storage medium provided in this application is a computer-readable storage medium, which stores computer-readable program instructions (i.e., computer programs) for executing the above-mentioned imperceptible lightweight three-dimensional data processing method, and can solve the technical problem of low efficiency in current three-dimensional model data processing. Compared with the prior art, the beneficial effects of the computer-readable storage medium provided in this application are the same as the beneficial effects of the imperceptible lightweight three-dimensional data processing method provided in the above-mentioned embodiment, and will not be elaborated here.

[0100] The above descriptions are only some embodiments of the present application, and are not intended to limit the patent scope of the present application. All equivalent structural changes made using the contents of the present application specification and drawings under the technical concept of the present application, or direct / indirect applications in other related technical fields are included in the patent protection scope of the present application.

Claims

1. A non-perceptual lightweight three-dimensional data processing method, characterized in that: The imperceptible lightweight three-dimensional data processing method comprises: Acquire key feature information of the preprocessed three-dimensional model data, and perform importance evaluation processing on the key feature information based on a graph theory algorithm to obtain an importance score of the key feature information; After removing the key feature information whose importance score is less than a preset score from the key feature information, performing data simplification processing on the key feature information based on a preset simplification algorithm; A compression strategy is determined according to the data type of the three-dimensional model data, and after the three-dimensional model data is encoded, the encoded data is compressed based on the compression strategy to obtain compressed data.

2. The imperceptible lightweight three-dimensional data processing method according to claim 1, characterized in that: After the step of determining a compression strategy according to the data type of the three-dimensional model data, and after encoding the three-dimensional model data, compressing the encoded data after encoding based on the compression strategy to obtain compressed data, the step further includes: Respond to the decompression information sent by the client; Read and decompress the data to be decompressed based on a preset decompression algorithm to obtain decompressed target three-dimensional model data; The target three-dimensional model data is output to the client.

3. The imperceptible lightweight three-dimensional data processing method according to claim 2, characterized in that: The step of reading and decompressing the data to be decompressed based on a preset decompression algorithm to obtain the decompressed target three-dimensional model data comprises: Reading the characteristic information, metadata and the data to be decompressed during the compression processing of the three-dimensional model data based on the preset decompression algorithm; The data to be decompressed is decompressed, and the decompressed data is restored based on the feature information and the metadata to obtain the target three-dimensional model data.

4. The imperceptible lightweight three-dimensional data processing method according to claim 1, characterized in that: After removing the key feature information whose importance score is less than the preset score in the key feature information, the step of performing data simplification processing on the key feature information based on a preset simplification algorithm includes: Determine, in the key feature information, feature information to be reduced in dimension whose importance score feature value is less than a preset score; Performing dimensionality reduction processing on the feature information to be reduced in dimension based on a preset dimensionality reduction algorithm to remove the key feature information whose importance score is less than a preset score in the key feature information; The key feature information is subjected to data simplification processing based on the preset simplification algorithm.

5. The imperceptible lightweight three-dimensional data processing method according to claim 1, characterized in that: The step of determining a compression strategy according to the data type of the three-dimensional model data, and compressing the encoded data after encoding the three-dimensional model data based on the compression strategy to obtain compressed data comprises: Acquire a data type of the three-dimensional model data, and determine the compression strategy associated with the data type; Performing encoding processing on the three-dimensional model data based on a preset encoding algorithm to obtain the encoded data; The encoded data is compressed according to a compression algorithm corresponding to the compression strategy to obtain the compressed data.

6. The imperceptible lightweight three-dimensional data processing method according to claim 1, characterized in that: The step of obtaining the key feature information of the preprocessed three-dimensional model data and performing importance evaluation processing on the key feature information based on a graph theory algorithm to obtain the importance score of the key feature information includes: Based on a geometric feature extraction method, obtaining the key feature information of the preprocessed three-dimensional model data; or based on a deep learning model, obtaining the key feature information of the preprocessed three-dimensional model data; The key feature information is evaluated for importance based on a graph theory algorithm to obtain the importance evaluation information.

7. The imperceptible lightweight three-dimensional data processing method according to claim 1, characterized in that: Before the step of obtaining the key feature information of the preprocessed three-dimensional model data and performing importance evaluation processing on the key feature information based on a graph theory algorithm to obtain the importance score of the key feature information, the step further includes: Receiving original three-dimensional model data, and performing noise removal processing on the original three-dimensional data based on Gaussian filtering, median filtering, bilateral filtering or wavelet transform to obtain denoised data; Data reconstruction and simplification processing are performed based on the denoised data to obtain the pre-processed three-dimensional model data.

8. The imperceptible lightweight three-dimensional data processing method according to claim 7, characterized in that: After the step of receiving the original three-dimensional model data and performing noise removal processing on the original three-dimensional data based on Gaussian filtering, median filtering, bilateral filtering or wavelet transform to obtain denoised data, the method further includes: The denoised data is simplified based on a voxel grid simplification method to obtain the pre-processed three-dimensional model data.

9. A three-dimensional data processing device, characterized in that: The three-dimensional data processing device includes: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the computer program is configured to implement the steps of the imperceptible lightweight three-dimensional data processing method as described in any one of claims 1 to 8.

10. A storage medium, characterized in that: The storage medium is a computer-readable storage medium, and a computer program is stored on the storage medium. When the computer program is executed by a processor, the steps of the imperceptible lightweight three-dimensional data processing method as described in any one of claims 1 to 8 are implemented.

Images