Accurate lightweight method, equipment and program product for three-dimensional model

Abstract

The present application proposes a method, device and program product for accurate lightweighting of three-dimensional models. The method includes obtaining incremental data generated during the editing process of the three-dimensional model, and comparing the original model nodes with the edited model nodes to perform spatial proximity calculations to obtain spatial proximity calculation values. According to the calculated values, the original model nodes are simplified to obtain a lightweight three-dimensional model. The spatial proximity calculation is obtained based on the dimensions of spatial semantic relevance, structural relevance, visual coverage relevance, and volume relevance. This solution reduces redundant nodes, reduces the data of the three-dimensional model, the storage space of the three-dimensional model and the computational complexity, and improves the efficiency of lightweighting the three-dimensional model.

Description

Technical Field

[0001] The present application relates to the technical field of constructing digital twin three-dimensional models, and specifically to a precise lightweight method, equipment and program product for three-dimensional models. Background Art

[0002] In the field of digital twin technology, it involves the mapping and simulation of physical entities in the real world, among which building a virtual 3D model corresponding to the physical model is one of the key links. Due to the complexity and diversity of 3D models, massive amounts of data will inevitably be generated during this construction process.

[0003] Therefore, lightweight processing of 3D models is one of the key technologies in the field of digital twin 3D modeling. Traditional 3D model lightweighting generally adopts a "one-size-fits-all" approach, which simply performs a unified standard lightweight operation on the 3D model without fully considering the specific characteristics of different 3D models and the diverse needs of users.

[0004] Based on this, lightweighting of three-dimensional models is a technical difficulty in this field, and precise lightweighting is one of the key directions for the development of lightweight technology. Summary of the invention

[0005] In view of this, in order to solve the problem of low accuracy of lightweighting of three-dimensional models, a method, equipment and program product for accurate lightweighting of three-dimensional models are provided.

[0006] According to one aspect of an embodiment of the present application, a method for accurately lightweighting a three-dimensional model is disclosed, comprising:

[0007] Acquiring incremental data formed by editing the three-dimensional model, the incremental data corresponding to the edited model node of the three-dimensional model;

[0008] An original model node corresponding to the original data of the three-dimensional model distribution is subjected to spatial proximity calculation on the original model node and the edited model node to obtain a spatial proximity calculation value;

[0009] The original model nodes are simplified according to the spatial proximity calculation values ​​to obtain a lightweight three-dimensional model.

[0010] According to one aspect of an embodiment of the present application, the spatial proximity calculation value is obtained by performing proximity correlation calculation based on one or more dimensions of spatial semantic correlation, structural relationship, visual coverage correlation, and volume correlation between the original model node and the edited model node.

[0011] According to one aspect of the embodiment of the present application, the algorithm of spatial semantic relevance includes:

[0012] Constructing a spatial semantic hierarchical relationship between the original model node and the edited model node, wherein the relationship includes a spatial hierarchy, and the root node of the spatial hierarchy is 0;

[0013] Calculate the spatial correlation degree L1=1 / (1+R) between the original model node and the edited model node according to the spatial level;

[0014] Wherein, R is the depth value of the spatial level between the original model node and the edited model node.

[0015] According to one aspect of the embodiment of the present application, the relationship further includes an object level and / or a region level; the algorithm of the spatial semantic relevance further includes:

[0016] Acquire the object impact factor d1 of the object level based on prior data, and / or acquire the region impact factor q1 of the region level;

[0017] Calculate and obtain the value S of the spatial semantic correlation semantic =L1*d1*q1, S semantic The value range is [0, 1].

[0018] According to one aspect of the embodiment of the present application, the algorithm for the structural relationship includes:

[0019] Acquire a structural relationship between the original model node and the edited model node, wherein the structural relationship includes the structure of the spatial hierarchy, the topological relationship, and predefined rules; the predefined rules include: consistent family type of the model, consistent spatial belonging of the model, and consistent material of the model;

[0020] According to the structure of the spatial hierarchy, determine whether the original model node and the edited model node have the same structure of the spatial hierarchy; if not, they do not belong to the same structure, and output the value S of the structural relationship struc The result is 0;

[0021] If yes, then determine whether the topological relationship of the original model node and the edited model node is consistent at the same time according to the topological relationship; if not, they do not belong to the same structure, and output the value S of the structural relationship. struc The result is 0;

[0022] If yes, then determine whether the predefined rules of the original model node and the edited model node are consistent with each other according to the predefined rules; if not, they do not belong to the same structure, and output the value S of the structure relationship. struc The result is 0; if so, they belong to the same structure, and the value S of the relationship between the structures is output. struc The result is 1.

[0023] According to one aspect of an embodiment of the present application, the algorithm for visual coverage correlation includes:

[0024] Calculating the minimum bounding boxes of the original model node and the edited model node respectively;

[0025] Determine whether the minimum bounding box of the original model node intersects with the minimum bounding box of the edited model node;

[0026] If yes, then output the value S of the visual coverage correlation bbox The result is 1;

[0027] If not, then output the value S of the visual coverage correlation bbox The result is 0.

[0028] According to one aspect of the embodiment of the present application, the algorithm of the volume correlation constructs an algorithm model of the volume correlation:

[0029] ,

[0030] like ,but ;like ,but ;

[0031] in, The value of volume correlation is 0 or 1; The total volume of the current 3D model; The maximum volume threshold of the 3D model; The minimum volume threshold for the 3D model.

[0032] According to one aspect of the embodiment of the present application, the algorithm model of the spatial proximity calculation is:

[0033] ,

[0034] Among them, S proximity : spatial proximity value, ranging from [0,1]; S semantic : The value of spatial semantic correlation, ranging from [0,1]; S bbox : The value of visual coverage correlation, ranging from [0,1]; S struct : The value of the structural relationship, ranging from [0,1]; The value of volume correlation is 0 or 1;

[0035] α=0.75, the basic weight of structural correlation; β=0.7, the basic weight of visual coverage correlation;

[0036] w s=0.45, the weight of spatial semantic relevance; w b =0.65, weight of visual coverage correlation; w c =0.35, weight of structural relationship.

[0037] A computer device comprises a memory, a processor and a computer program stored in the memory, wherein the processor executes the computer program to implement the steps of the above method.

[0038] A computer program product comprises a computer program, wherein the computer program implements the steps of the above method when executed by a processor.

[0039] The solution of accurate lightweight 3D model is adopted. The model nodes of the 3D model are accurately classified according to the editing results, the complete information of the edited model nodes is retained, and the original model nodes are simplified through the spatial proximity calculation value, which accurately reduces the data volume of the 3D model. The lightweight 3D model reduces the consumption of storage space and computing resources while ensuring the model accuracy, and improves the efficiency of lightweight 3D model.

[0040] Furthermore, through spatial proximity calculation, accurate simplification can be achieved based on multi-dimensional information such as semantics, structure, and vision between model nodes, ensuring that the simplified model retains important features and simplifies redundant data, making the lightweight three-dimensional model of this solution cost-effective. BRIEF DESCRIPTION OF THE DRAWINGS

[0041] The above and other objects, features and advantages of the present application will become more apparent by describing in detail exemplary embodiments thereof with reference to the attached drawings.

[0042] Figure 1 It is a flow chart of a precise lightweight method for a three-dimensional model provided according to an embodiment of the present application.

[0043] Figure 2 It is a flow chart of a method for assigning a simplified weight to each original model node according to a spatial proximity calculation value according to an embodiment.

[0044] Figure 3 is a flowchart of an algorithm for spatial semantic relevance according to an embodiment.

[0045] Figure 4 is a schematic diagram showing the relationship between spatial semantic levels according to an embodiment.

[0046] Figure 5 is a flowchart of an algorithm for spatial semantic correlation according to another embodiment.

[0047] Figure 6 is an algorithm flow chart showing the structural relationship according to an embodiment.

[0048] Figure 7 is a flowchart of an algorithm for visual coverage correlation according to an embodiment.

[0049] Figure 8 is a comparison diagram of performance parameters of a three-dimensional model after optimization according to an embodiment. DETAILED DESCRIPTION

[0050] Example embodiments will now be described more fully with reference to the accompanying drawings. However, example embodiments can be implemented in a variety of forms and should not be construed as limited to the examples set forth herein; rather, these example embodiments are provided so that the description of the present application will be more comprehensive and complete and the concepts of the example embodiments will be fully conveyed to those skilled in the art. The accompanying drawings are only schematic illustrations of the present application and are not necessarily drawn to scale. The same reference numerals in the figures represent the same or similar parts, and thus their repeated description will be omitted.

[0051] In addition, the described features, structures or characteristics may be combined in one or more example embodiments in any suitable manner. In the following description, many specific details are provided to provide a full understanding of the example embodiments of the present application. However, those skilled in the art will appreciate that the technical solution of the present application may be practiced while omitting one or more of the specific details, or other methods, components, steps, etc. may be adopted. In other cases, known structures, methods, implementations or operations are not shown or described in detail to avoid obscuring the present application and making the various aspects of the present application obscure.

[0052] In the context of the widespread application of digital twin technology, the core foundation is to model physical scenes and build virtual 3D model scenes covering a variety of 3D model scenes from cities, buildings to indoor spaces. For example, a city is a 3D model scene, a building is a 3D model scene, and a room is also a 3D model scene. From the perspective of application scenarios, some 3D model scenes tend to be fire emergency, water conservancy or smart factories, etc., and the construction requirements of the corresponding 3D models vary according to different application scenarios.

[0053] On the other hand, in the 3D model scene, there are many different types of model nodes, such as 3D models of buildings, such as doors, windows, stairs, ceilings, etc.; 3D models of fire emergency, such as fire extinguishers, fire hydrants, fire trucks, etc.; 3D models of water conservancy, such as dams, water flows, pipelines, etc.; 3D models of smart factories, such as gantry cranes, conveyor lines, transfer lines, etc. are model nodes. 3D models in different fields define and arrange different model nodes to ensure that each model node can accurately reflect the specific needs of its application scenario, and also indirectly reflect the spatial proximity between model nodes.

[0054] For different 3D model scenes, all kinds of model nodes in the scene are edited. The editing content includes adding, deleting, modifying and checking the model nodes. Specifically, it is the data editing of the geometric structure of the points, lines, surfaces and bodies of the model nodes. In the process of 3D model editing, it is often necessary to edit a part of the model node, which will form new incremental data (deletion or replacement is a kind of "increment"). The corresponding model node mapped by the incremental data is the core content (that is, the edited model node), and the content will not be simplified.

[0055] On the contrary, the original data that has not been edited and its corresponding model nodes usually need to be simplified. The simplification method is based on the spatial proximity of the original model nodes and the edited nodes, and different weights are constructed to achieve differentiated and accurate 3D model simplification, thereby reducing the overall data volume of the 3D model, making it lighter, while retaining the core and important data of the edited model nodes.

[0056] Based on this, combined with the Figure 1 , a precise lightweight method for three-dimensional models, comprising:

[0057] Step 100, obtaining incremental data formed by editing the three-dimensional model, the incremental data corresponding to the edited model node of the three-dimensional model;

[0058] Step 200, performing spatial proximity calculation on the original model nodes corresponding to the original data of the three-dimensional model distribution and the edited model nodes to obtain spatial proximity calculation values;

[0059] Step 300: simplify the original model nodes according to the calculated spatial proximity values, and obtain a lightweight three-dimensional model.

[0060] Specifically, during the 3D model editing process, the user performs operations such as adding, deleting, modifying, and checking, such as editing points, lines, surfaces, and bodies, or materials, textures, animations, etc. The generated change data is incremental data, and the 3D model node corresponding to the incremental data is the edited model node. Therefore, the identification of the edited model node is determined by comparing the model node status before and after editing.

[0061] The original model data and incremental data are relative and different. In the original data of the 3D model, there are many unedited original data, and the nodes corresponding to the original data are the original model nodes. It should be noted that for the edited model nodes, there will also be unedited original data, but the model nodes are not the original model nodes. Therefore, the core difference is: taking the model node as the smallest identifiable object, all the data of the model node has not been edited, then the model is the original model node; correspondingly, if the model node contains edited data, then the model node is the edited model node.

[0062] For the determined original model node, the spatial proximity between the original model node and the edited model node is calculated. The algorithm of the spatial proximity includes any one or more combinations of spatial semantic correlation, structural phase relationship, visual coverage correlation, and volume correlation, thereby obtaining the corresponding spatial proximity calculation value.

[0063] Through the calculation value of spatial proximity, the complete data of the edited model node is retained, and only the original model node is simplified. Specifically, according to the calculation value of the spatial proximity of the original model node, the closer the value is to 1, the closer it is to the edited model node, indicating that the original model node and the edited model node have the same importance, and the more complete data needs to be retained. Conversely, the closer the calculation value of the spatial proximity of the original model node is to 0, the farther it is from the edited model node, indicating that the original model node needs to be simplified, including lightweight processing such as merging and deleting model nodes, so as to reduce the amount of data of the original model node. Finally, the simplified original model nodes and the edited model nodes are updated and saved, and a lightweight three-dimensional model is output.

[0064] In other embodiments, the Figure 2 In step 200, a simplified weight is assigned to each original model node according to the spatial proximity calculation value, and the algorithm strategy includes:

[0065] Step 210, obtaining the type of the original model node, and counting the number of original model nodes of each type;

[0066] Step 220, dividing the basic levels according to the ratio of the number of original model nodes of the corresponding type to the total number of original model nodes;

[0067] Step 230, obtaining the mean value of the spatial proximity calculation of the original model nodes of the same type, and dividing the calculation value level according to the ratio of the spatial proximity calculation value of the original model nodes of each type to the mean value;

[0068] Step 240, determining the simplification weight according to the product of the basic level and the calculated value level of each original model node.

[0069]

[0070] Combined with the above table, the algorithm strategy of simplifying weights is illustrated. For a 3D model, the number of different types of model nodes determines its position in the data volume of the 3D model, and is also an important reference basis for lightweight and accurate objects of the model.

[0071] On this basis, each model node of the corresponding type is combined and the value (e.g. S an , S bn , S dn ) and the mean of the spatial proximity of the model nodes of the same type (e.g. S av , S bv , S dv ), and the calculated value level is divided according to the ratio of the two to reflect the deviation of the model node in the model node of its type.

[0072] Based on this, the simplified weight Q of each original model node is obtained by multiplying the base level of each original model node by the calculated value level (for example, Q a1 , Q a2... Q an ), according to the simplified weight Q, the corresponding threshold is set, and the original model nodes are accurately classified according to the threshold, and lightweight processing such as merging and deleting is performed. In this scheme, a simplified weight Q is assigned to each original model node, and the original model nodes can be pre-processed, and different original model nodes can be preliminarily classified to input high-quality data for subsequent spatial proximity calculations.

[0073] The solution of accurate lightweighting of 3D models is adopted. The model nodes of the 3D model are accurately classified according to the editing results, the complete information of the edited model nodes is retained, and the original model nodes are accurately simplified through the spatial proximity calculation value, and the data volume of the 3D model is accurately reduced. The lightweight 3D model reduces the consumption of storage space and computing resources while ensuring the model accuracy, and improves the efficiency of lightweighting of 3D models.

[0074] In one embodiment, in combination with Figure 3 , the algorithms for spatial semantic relevance include:

[0075] Step 410, constructing a spatial semantic hierarchical relationship between the original model node and the edited model node, the relationship including the spatial hierarchy, the root node of the spatial hierarchy being 0;

[0076] Step 420, calculating the spatial correlation degree L1=1 / (1+R) between the original model node and the edited model node according to the spatial level;

[0077] Wherein, R is the depth value of the spatial level between the original model node and the edited model node.

[0078] Specifically, combined with Figure 4 , establish the spatial semantic hierarchy relationship between the original model node and the edited model node. Combined with specific application examples, for example, in the building information scenario, according to the floor 1 / 2 / 3 is the same spatial semantic hierarchy, the corresponding floor structure 1 / 2 / 3 is the same spatial semantic hierarchy, but belongs to the next level of the floor. The nodes of the corresponding structure of the corresponding floor include unedited nodes / edited nodes (corresponding to the original model node / edited model node), where the arrows express the spatial proximity of the unedited nodes and the edited nodes. Combined with the levels between the nodes, the unedited node A (original model node A) is selected. Since the root node of the spatial level is 0, the depth value between the unedited node A (original model node A) and the edited node A (edited model node A) is: unedited node A-["Structure 1-Floor 1-Floor 2-Structure 2"]-edited node A, that is, from the unedited node A to the edited node A, there are 4 spatial semantic levels in between, so the depth value R from the unedited node A to the edited node A is 4, and its corresponding spatial correlation degree L1=1 / (1+R), that is, L1=1 / 5. The spatial correlation degrees of other original model nodes can be obtained by analogy, and the corresponding spatial correlation degree L1 can be obtained. In other embodiments, the spatial correlation degree L1 can be directly regarded as a spatial proximity calculation value, which is used to simplify the original model nodes. This method can reduce the complexity of calculation and improve lightweight efficiency.

[0079] In one embodiment, in combination with Figure 5 , the relationship of the spatial semantic level also includes the object level and / or the region level; the algorithm of spatial semantic relevance also includes:

[0080] Step 430, obtaining an object impact factor d1 at an object level and / or obtaining a region impact factor q1 at a region level based on prior data;

[0081] Step 440, calculate and obtain the value S of spatial semantic correlation semantic =L1*d1*q1,S semantic The value range is [0, 1].

[0082] Specifically, the object level is divided according to the spatial semantic level, and the object influence factor d1 is set according to the prior data and different types of object levels as follows: door = 1.2, wall = 1.0, decoration body = 0.8, equipment object = 1.0.

[0083] Furthermore, the regional level is divided according to the spatial semantic level, and the regional impact factor q1 is set according to the prior data and different types of regional levels as follows: fire partition = 0.6, equipment area = 1.2, public area = 1.0, outdoor = 0.5, indoor = 1.0, road = 0.3.

[0084] Based on the spatial association degree L1 and two spatial semantic levels of object influence factor d1 and regional influence factor q1, the spatial semantic relevance score S is obtained. semantic The calculation formula is:

[0085] S semantic =L1*d1*q1; that is, S semantic = spatial correlation degree L1*object impact factor d1*regional impact factor q1; where S semantic >1, S semantic The results are unified as 1.

[0086] In other embodiments, the corresponding object level and regional level will change adaptively according to different application scenarios. For example, in the fire protection scenario, the factor of the fire partition in the regional impact factor q1 will increase to 1.5, and the outdoor factor will decrease to 0.3. Similarly, for the road and bridge application scenario, the factors of doors and walls will decrease, and the equipment object factor (for example, road surface) will increase.

[0087] Based on this, the strategy for adaptively changing the object influence factor d1 and the regional influence factor q1 is: obtaining the scene theme of the three-dimensional model; obtaining objects and regions related to the scene theme through AI big data statistics; and assigning factors to objects and regions based on prior data.

[0088] In one embodiment, in combination with Figure 6 , the algorithms for structural relationships include:

[0089] Step 510, obtaining the structural relationship between the original model node and the edited model node, the structural relationship including the spatial hierarchical structure, topological relationship and predefined rules; the predefined rules include: the family type of the model is consistent, the spatial belonging of the model is consistent, and the material of the model is consistent;

[0090] Step 520, judging whether the original model node and the edited model node have a common spatial hierarchical structure according to the spatial hierarchical structure;

[0091] Step 530: If not, they do not belong to the same structure, and the value S of the structure correlation is output. struc The result is 0;

[0092] Step 540: If yes, then determine whether the topological relationship of the original model node and the edited model node is consistent based on the topological relationship; if not, they do not belong to the same structure, then go to step 530, that is, output the value S of the structural relationship. struc The result is 0;

[0093] Step 550: If yes, then determine whether the predefined rules of the original model node and the edited model node are consistent with each other according to the predefined rules; if not, they do not belong to the same structure, then go to step 530, that is, output the value S of the structure relationship. struc The result is 0;

[0094] Step 560: If yes, they belong to the same structure, and the value S of the structure relationship is output. struc The result is 1.

[0095] Specifically, the structure of the spatial hierarchy is consistent with the structure of the spatial hierarchy of spatial semantic relevance in the above embodiment. Topological relationships are the relationships between various spatial data that satisfy the principles of topological geometry, such as the topological relationship of multiple BIM model elements being close to each other in space and sharing boundaries, the inclusion relationship of one element completely containing another or more elements, the connection relationship between model elements being interconnected through specific geometric or physical connection methods, and the intersection relationship of model elements partially overlapping in space.

[0096] Considering personalized customization, predefined rules are set, including: consistent family types of models, such as building component families including structural components such as columns, beams, and slabs, or enclosure components such as walls, doors, and windows, electromechanical equipment families including water supply and drainage equipment, electrical equipment, and HVAC equipment, and site component families including topography, landforms, and landscape components; consistent spatial affiliation of models, such as room affiliation, system affiliation, and regional affiliation; consistent model materials, such as color, texture, glossiness, transparency, etc.

[0097] Determine whether the original model node and the edited model node belong to the same structure, and the spatial hierarchical structure, topological relationship and predefined rules are established at the same time, and output the value S of the structural relationship struc The result is 1 if yes, 0 otherwise.

[0098] Furthermore, in order to control the influence of structural phase relationship on adjacency, a weight control factor α is introduced, and its constant value range is (0, 1), and the final structural phase relationship value (N)S is obtained. struct , the calculation formula is:

[0099] (N)S struct =α*S struct，

[0100] Through this solution, the structural consistency of the model nodes can be accurately evaluated based on the spatial hierarchical structure, topological relationship and predefined rules. In addition, the multi-dimensional judgment method can deeply analyze the similarity of the model at different levels, improve accuracy, and reduce the risks or errors caused by relying only on a single dimension.

[0101] In one embodiment, in combination with Figure 7 , the algorithms for visual coverage correlation include:

[0102] Step 610, respectively calculating the minimum bounding boxes of the original model node and the edited model node;

[0103] Step 620, determining whether the minimum bounding box of the original model node intersects with the minimum bounding box of the edited model node;

[0104] Step 630: If yes, then output the value S of the visual coverage correlation bbox The result is 1;

[0105] Step 640: If not, output the value S of the visual coverage correlation bbox The result is 0.

[0106] Specifically, all geometric elements of the original model node are traversed, and for the point set data, its coordinate range in the three-dimensional space is determined, the minimum and maximum values ​​of the x, y, and z coordinates in the point set are found, and the minimum bounding box of the original model node is constructed according to the obtained coordinate range. The coordinate range comparison method is adopted, that is, for the minimum bounding box of the original model node and the minimum bounding box of the edited model node, their coordinate ranges in the three coordinate axis directions are compared; or the space segmentation method, that is, the method of judging the intersection is to segment the three-dimensional space, for example, using an octree structure, and mapping the minimum bounding box of the original model node and the minimum bounding box of the edited model node to the octree space respectively, and judging the intersection between the two. The above intersection judgment method is used to determine the intersection between the minimum bounding box of the original model node and the minimum bounding box of the edited model node. If the two intersect, the value S of the visual coverage correlation. bbox is 1, if the two do not intersect, the visual coverage correlation value S bbox is 0.

[0107] Furthermore, in order to control the impact of visual coverage correlation on adjacency, a weight control factor β is introduced, and the constant value range is (0, 1), and the final visual coverage correlation value (N) S is obtained. bbox , the calculation formula is:

[0108] (N)S bbox =β*S′ bbox，

[0109] This solution determines the visual coverage relationship in a concise and efficient way by calculating and comparing the minimum bounding boxes of the two. In the overlapping parts of visual coverage, duplicate data can be deleted to reduce duplicate data and improve the lightweight effect. In addition, in the subsequent 3D model visualization process, the rendering order can be determined or optimized according to the parts where the model nodes may occlude or cover each other, thereby improving the rendering efficiency.

[0110] In one embodiment, the algorithm of volume correlation constructs an algorithm model of volume correlation:

[0111] ,

[0112] like ,but ;like ,but ;

[0113] in, The value of volume correlation is 0 or 1; The total volume of the current 3D model; The maximum volume threshold of the 3D model; The minimum volume threshold for the 3D model.

[0114] Specifically, in the project practice, the volume of the 3D model involved is huge, and the total volume of the 3D model Often spans multiple orders of magnitude (e.g., from 10 3 Cubic meters to city level 10 9 When the volume of a 3D model scene is used to influence spatial proximity, the linear use of volume values ​​in a single dimension may cause the calculation to lose sensitivity. The logarithmic function can compress this order of magnitude change into a more manageable range and maintain a smooth response. Based on this, the solution of this embodiment is designed:

[0115] ,

[0116] like ,but ;like ,but ;

[0117] In the case of large-scale 3D models, , ; In the small volume scene of 3D model , . After logarithmic scaling, even if Across multiple orders of magnitude, the formula still produces smooth volume correlation values The logarithmic normalization is used to adjust the volume impact factor, which reasonably balances the fault tolerance of large-volume scenes and the sensitivity of small-volume scenes.

[0118] In one embodiment, based on the precise lightweight method of the above three-dimensional model, the algorithm model for calculating spatial proximity is:

[0119] ,

[0120] Among them, S proximity : spatial proximity value, ranging from [0,1]; S semantic : The value of spatial semantic correlation, ranging from [0,1]; S bbox : The value of visual coverage correlation, ranging from [0,1]; S struct : The value of the structural relationship, ranging from [0,1]; The value of volume correlation is 0 or 1;

[0121] α=0.75, the basic weight of structural correlation; β=0.7, the basic weight of visual coverage correlation;

[0122] w s =0.45, the weight of spatial semantic relevance; w b =0.65, weight of visual coverage correlation; w c =0.35, weight of structural relationship.

[0123] Specifically, the algorithm model for calculating spatial proximity involves weight control factors α, β, including the weights of structural phase relationship and visual coverage correlation, and adjusts the impact of volume correlation.

[0124] In addition, s is the weight of spatial semantic relevance, ranging from [0,1]; w b is the weight of visual coverage relevance; w c is the weight of the structural relationship.

[0125] In other examples, in order to obtain the best weight value, the following method can be used to determine it, as follows:

[0126] By manually annotating the proximity value S true , optimize the weights to minimize S proximity With S true To evaluate the error, the mean square error (MSE) is introduced to optimize the objective function:

[0127] ,

[0128] For a data point {i}, its error Deviation (i) is defined as:

[0129] ,

[0130] Where N is the number of data points; 1 / N is the scaling factor for taking the mean to avoid unreasonable optimization caused by error accumulation.

[0131] The proximity value calculated by the formula; The true value obtained by manual annotation.

[0132] The gradient descent method is used to optimize the weight parameters, specifically:

[0133] Set the initial weight parameter, that is:

[0134] w s = 0.5, the initial weight of the weight of spatial semantic relevance; w b =0.6, weight of visual coverage correlation; w c =0.4, weight of structural relationship.

[0135] The initial weights of α and β are set to α=0.7 and β=0.8 respectively.

[0136] Based on the above weight combination, S is calculated according to the algorithm model of spatial proximity calculation. proximity .

[0137] Then the weight parameters are iteratively optimized by the gradient descent method. After several iterations, when the value of the objective function no longer decreases significantly, the final weight is output, that is:

[0138] w s ≈0.45; w b ≈0.65; w c ≈0.35; α≈0.75, β≈0.7.

[0139] For ease of calculation, the unified weight value is: w s =0.45; w b =0.65; w c =0.35;α=0.75,β=0.7.

[0140] Combination width Figure 8, now combined with the results of lightweight 3D models in a specific application example, specifically, the performance parameter comparison values ​​of a 3D scene of a shopping mall after optimization by this solution, the 3D model optimization while retaining the key features of the 3D model, in which the number of objects, the number of geometric models, and the number of triangular facets are greatly reduced, and the data volume of some indicators is reduced by more than 70%, achieving lightweight 3D models. In addition, the efficiency of the frame rendering time of the 3D model is also greatly improved by 4 times, with almost no impact on the user's visualization experience, achieving accurate lightweight 3D models and comprehensively improving the user experience.

[0141] In one embodiment, a computer device is also provided, including a memory, a processor, and a computer program stored in the memory, and the processor executes the computer program to implement the steps of the above-mentioned fusion-driven three-dimensional model rapid construction method.

[0142] In one embodiment, a computer program product includes a computer program, and when the computer program is executed by a processor, the steps of the above-mentioned fusion-driven three-dimensional model rapid construction method are implemented.

[0143] It should be noted that, although several modules or units of the equipment for action execution are mentioned in the above detailed description, this division is not mandatory. In fact, according to the embodiments of the present application, the features and functions of two or more modules or units described above can be embodied in one module or unit. On the contrary, the features and functions of one module or unit described above can be further divided into being embodied by multiple modules or units.

[0144] In addition, although the steps of the method in the present application are described in a specific order in the drawings, this does not require or imply that the steps must be performed in this specific order, or that all the steps shown must be performed to achieve the desired results. Additionally or alternatively, some steps may be omitted, multiple steps may be combined into one step, and / or one step may be decomposed into multiple steps, etc.

[0145] Those skilled in the art will readily appreciate other embodiments of the present application after considering the specification and practicing the invention disclosed herein. The present application is intended to cover any modification, use or adaptation of the present application, which follows the general principles of the present application and includes common knowledge or customary techniques in the art that are not disclosed in the present application. The specification and examples are intended to be exemplary only, and the true scope and spirit of the present application are indicated by the appended claims.

Claims

1. A precise lightweight method for a three-dimensional model, characterized in that: include: Acquiring incremental data formed by editing the three-dimensional model, the incremental data corresponding to the edited model node of the three-dimensional model; The original model node corresponding to the original data of the three-dimensional model distribution is subjected to spatial proximity calculation to obtain a spatial proximity calculation value; the spatial proximity calculation value is obtained by performing proximity correlation calculation based on the spatial semantic correlation between the original model node and the edited model node; The original model nodes are simplified according to the spatial proximity calculation value to obtain a lightweight three-dimensional model. The algorithm of spatial semantic relevance includes: Constructing a spatial semantic hierarchical relationship between the original model node and the edited model node, wherein the relationship includes a spatial hierarchy, and the root node of the spatial hierarchy is 0; Calculate the spatial correlation degree L1=1 / (1+R) between the original model node and the edited model node according to the spatial level; Wherein, R is the depth value of the spatial level between the original model node and the edited model node; The relationship also includes object level and / or region level; the algorithm of spatial semantic relevance also includes: Acquire the object impact factor d1 of the object level based on prior data, and / or acquire the region impact factor q1 of the region level; Calculate and obtain the value S of the spatial semantic correlation semantic =L1*d1*q1, S semantic The value range is [0, 1].

2. The precise lightweight method of a three-dimensional model according to claim 1, characterized in that: The spatial proximity calculation value also includes: obtaining the value by performing proximity correlation calculation according to one or more dimensions of structural relationship, visual coverage correlation, and volume correlation between the original model node and the edited model node.

3. The precise lightweight method of a three-dimensional model according to claim 2, characterized in that: The algorithm of the structural relationship includes: Acquire a structural relationship between the original model node and the edited model node, wherein the structural relationship includes the structure of the spatial hierarchy, the topological relationship, and predefined rules; the predefined rules include: consistent family type of the model, consistent spatial belonging of the model, and consistent material of the model; According to the structure of the spatial hierarchy, determine whether the original model node and the edited model node have the same structure of the spatial hierarchy; if not, they do not belong to the same structure, and output the value S of the structural relationship struc The result is 0; If yes, then determine whether the topological relationship of the original model node and the edited model node is consistent at the same time according to the topological relationship; if not, they do not belong to the same structure, and output the value S of the structural relationship. struc The result is 0; If yes, then according to the predefined rules, determine whether the predefined rules of the original model node and the edited model node are consistent at the same time; if not, they do not belong to the same structure, and output the value S of the structure relationship struc The result is 0; if so, they belong to the same structure, and the value S of the relationship between the structures is output. struc The result is 1.

4. The precise lightweight method for a three-dimensional model according to claim 3, characterized in that: The algorithm of visual coverage correlation includes: Calculating the minimum bounding boxes of the original model node and the edited model node respectively; Determine whether the minimum bounding box of the original model node intersects with the minimum bounding box of the edited model node; If yes, then output the value S of the visual coverage correlation bbox The result is 1; If not, then output the value S of the visual coverage correlation bbox The result is 0.

5. The precise lightweight method for a three-dimensional model according to claim 4, characterized in that: The algorithm of the volume correlation constructs the algorithm model of the volume correlation: ， like < ,but = =0; if > ,but = =1; in, : The value of volume correlation is 0 or 1; : The total volume of the current 3D model; : Maximum volume threshold of the 3D model; : The minimum volume threshold of the 3D model.

6. The precise lightweight method for a three-dimensional model according to claim 5, characterized in that: The algorithm model of the spatial proximity calculation is: ， Among them, S proximity : spatial proximity value, ranging from [0,1]; S semantic : The value of spatial semantic correlation, ranging from [0,1]; S bbox : The value of visual coverage correlation, ranging from [0,1]; S struct : The value of the structural relationship, ranging from [0,1]; : The value of volume correlation is 0 or 1; α=0.75, the basic weight of structural correlation; β=0.7, the basic weight of visual coverage correlation; w s =0.45, the weight of spatial semantic relevance; w b =0.65, weight of visual coverage correlation; w c =0.35, weight of structural relationship.

7. A computer device comprising a memory, a processor and a computer program stored in the memory, characterized in that: The processor executes the computer program to implement the steps of the method according to any one of claims 1 to 6.

8. A computer program product, comprising a computer program, characterized in that When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 6 are implemented.

Images