Virtual model disassembling method and device and related equipment

Abstract

The invention provides a virtual model disassembling method and device and related equipment, and belongs to the technical field of image processing. The virtual model disassembling method provided by the embodiment of the invention comprises the following steps: respectively determining a semantic importance score, a geometric complexity score, a scene correlation score and an equipment suitability score of a plurality of components in a to-be-disassembled virtual model; according to the semantic importance score, the geometric complexity score, the scene correlation score and the equipment suitability score, respectively determining comprehensive scores of the plurality of components; and generating a disassembling strategy according to the comprehensive score, and disassembling the to-be-disassembled virtual model according to the disassembling strategy. According to the virtual model disassembling method, the disassembling effect of the virtual model can be improved.

Description

Technical Field

[0001] This application relates to the field of image processing technology, and in particular to a virtual model disassembly method, apparatus and related equipment. Background Technology

[0002] As the application scenarios of various 3D virtual models become increasingly complex, model decomposition technology has been widely used in many fields such as Virtual Reality (VR), industrial design, urban architecture, and medicine. However, existing model decomposition methods simplify the virtual model to be decomposed into pure geometric entities and decompose the components according to the topological structure, resulting in poor decomposition effects of the virtual model. Summary of the Invention

[0003] This application provides a virtual model disassembly method, apparatus, and related equipment, which can solve the technical problem of poor virtual model disassembly effect.

[0004] In a first aspect, embodiments of this application provide a virtual model decomposition method, the method comprising:

[0005] Semantic importance score, geometric complexity score, scene relevance score, and device adaptability score of multiple components in the virtual model to be disassembled are determined respectively.

[0006] The comprehensive scores of the multiple components are determined based on the semantic importance score, the geometric complexity score, the scene relevance score, and the device adaptability score.

[0007] A disassembly strategy is generated based on the comprehensive score, and the virtual model to be disassembled is disassembled according to the disassembly strategy.

[0008] Optionally, determining the semantic importance scores of multiple components in the virtual model to be disassembled includes:

[0009] Semantic tag sets for the plurality of components are determined respectively, and tag importance scores for the plurality of components are determined respectively based on the semantic tag sets and the application scenario of the virtual model to be disassembled.

[0010] Obtain the hierarchical relationship of the multiple components, and determine the hierarchical importance score of each of the multiple components based on the hierarchical relationship;

[0011] Each of the multiple components is assigned a role importance score in the virtual model to be disassembled;

[0012] The semantic importance scores of the multiple components are obtained by weighted summation of the tag importance score, the hierarchy importance score, and the role importance score.

[0013] Optionally, generating a disassembly strategy based on the comprehensive score includes:

[0014] Based on the comprehensive score and the preset equipment complexity threshold, the first disassembly level of each of the multiple components is determined.

[0015] The value density of each of the plurality of components is determined based on the product of the first disassembly level and the comprehensive score.

[0016] Based on the first disassembly level and the value density, a second disassembly level is determined for each of the plurality of components, and a disassembly strategy is generated based on the second disassembly level.

[0017] Optionally, determining the second disassembly level of the plurality of components based on the first disassembly level and the value density includes:

[0018] Based on the ranking result obtained by sorting the value density of the multiple components, the first disassembly level of the components located within the preset sequence range is adjusted to the second disassembly level.

[0019] Among them, components that are outside the preset sequence range in the sorting results are configured to keep their component models unchanged.

[0020] Optionally, generating a disassembly strategy based on the second disassembly level includes:

[0021] The adjusted disassembly level of the target component is determined based on the target component, a preset first threshold, and a preset second threshold. The first threshold is a first threshold representing the difference between the second disassembly levels of the target component and those of components at the next higher level. The second threshold is a second threshold representing the difference between the second disassembly levels of the target component and those of components at the next lower level. Based on the adjusted disassembly level of the target component and the adjusted disassembly levels of adjacent components within a preset spatial range, a weighted sum is performed according to preset weights to obtain the third disassembly level of the target component. The target component is any one of the plurality of components.

[0022] A disassembly strategy is generated based on the third disassembly level of the plurality of components, and the virtual model to be disassembled is disassembled according to the disassembly strategy.

[0023] Optionally, determining the comprehensive score of the multiple components based on the semantic importance score, the geometric complexity score, the scene relevance score, and the device adaptability score includes:

[0024] Based on the application scenario of the virtual model to be disassembled, the corresponding score weights of the semantic importance score, geometric complexity score, scenario relevance score and device adaptability score of the multiple components in the application scenario are determined respectively.

[0025] The semantic importance score, set complexity score, scene relevance score, and device adaptability score are weighted and summed according to the scoring weights to obtain the comprehensive score of the multiple components.

[0026] Secondly, embodiments of this application provide a virtual model disassembly device, the device comprising:

[0027] The first processing module is used to determine the semantic importance score, geometric complexity score, scene relevance score and device adaptability score of multiple components in the virtual model to be disassembled.

[0028] The second processing module is used to determine the comprehensive score of the multiple components based on the semantic importance score, the geometric complexity score, the scene relevance score, and the device adaptability score.

[0029] The third processing module is used to generate a dismantling strategy based on the comprehensive score, and to dismantle the virtual model to be dismantled according to the dismantling strategy.

[0030] Thirdly, embodiments of this application provide an electronic device, including: a processor, a memory, and a program stored in the memory and executable on the processor, wherein when the program is executed by the processor, it implements the steps of the virtual model disassembly method as described in the first aspect.

[0031] Fourthly, embodiments of this application provide a computer-readable storage medium storing a computer program, which, when executed by a processor, implements the steps of the virtual model disassembly method as described in the first aspect.

[0032] Fifthly, embodiments of this application provide a computer program product, including computer instructions that, when executed by a processor, implement the steps of the virtual model disassembly method as described in the first aspect.

[0033] In this embodiment, a comprehensive score is generated based on semantic importance, geometric complexity, scene relevance, and device adaptability, and a disassembly strategy is generated. The virtual model to be disassembled is then disassembled according to the generated disassembly strategy. This disassembly method avoids the limitations of relying solely on topological structural elements for disassembly. It disassembles the virtual model to be disassembled based on a comprehensive multi-dimensional score, thereby improving the disassembly effect of the virtual model. Attached Figure Description

[0034] To more clearly illustrate the technical solutions of the embodiments of this application, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0035] Figure 1 This is one of the flowcharts of a virtual model disassembly method provided in the embodiments of this application;

[0036] Figure 2 This is the second flowchart of a virtual model disassembly method provided in the embodiments of this application;

[0037] Figure 3 This is a schematic diagram of the structure of a virtual model disassembly device provided in an embodiment of this application;

[0038] Figure 4 This is a schematic diagram of the structure of an electronic device provided in an embodiment of this application. Detailed Implementation

[0039] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0040] The terms "first," "second," etc., used in this application are used to distinguish similar objects and not to describe a specific order or sequence. It should be understood that such terms can be used interchangeably where appropriate so that embodiments of this application can be implemented in orders other than those illustrated or described herein, and the objects distinguished by "first" and "second" are generally of the same class, not limited in number; for example, the first object can be one or more. Furthermore, "and / or" in this application indicates at least one of the connected objects. For example, the scope of protection of "A and / or B" covers at least three scenarios: Scenario 1: including A but not B; Scenario 2: including B but not A; Scenario 3: including both A and B. Additionally, the character " / " generally indicates that the preceding and following objects are in an "or" relationship.

[0041] See Figure 1 , Figure 1 This is a flowchart of a virtual model disassembly method provided in an embodiment of this application, such as... Figure 1 As shown, the method includes the following steps:

[0042] Step 101: Determine the semantic importance score, geometric complexity score, scene relevance score, and device adaptability score of multiple components in the virtual model to be disassembled.

[0043] The virtual model can be a virtual model constructed by 3D modeling software such as Unity Engine, Unreal Engine, and Cityengine, or a virtual model constructed by other 3D modeling software or methods (such as Solidworks software); in this application, the virtual model can also be simply referred to as the model, and the virtual model to be disassembled can also be simply referred to as the virtual model to be disassembled.

[0044] The step of determining the semantic importance scores of multiple components in the virtual model to be disassembled can be to construct a set of semantic labels for each of the multiple components, determine the score corresponding to each semantic label according to the set of semantic labels, and then obtain the semantic importance score of each component.

[0045] Taking component c as an example (component c can be any of the multiple components), the determination of the geometric complexity score of multiple components in the virtual model to be disassembled can refer to the following first formula:

[0046]

[0047] In the first formula, This can represent the geometric complexity score of component c; , , and These can represent the weights corresponding to volume ratio, surface area ratio, curvature complexity, and detail density, respectively. , , and It can satisfy ; It can represent the volume of component c; This can represent the total volume of the virtual model to be disassembled; It can represent the surface area of ​​component c; This can represent the total surface area of ​​the virtual model to be disassembled; This can represent the curvature complexity of component c; The calculation can be referenced from the following second formula:

[0048]

[0049] In the second formula, It can represent the set of faces of component c; It can represent a surface The set of edges; It can represent an edge The dihedral angles on both sides; It can represent the detail density of component c; The calculation can be referenced from the following third formula:

[0050] .

[0051] In the third formula; It can represent the detail density of component c; It can represent the set of faces of component c; It can represent the volume of component c;

[0052] The multiple components in the virtual model to be disassembled can also have their respective geometric complexity scores determined in the same way as component c described above, which will not be elaborated here;

[0053] Understandably, the geometric complexity score of component c above can also be used. express;

[0054] For example, the scenario relevance score can be used to measure the importance of a component to a specific application scenario. For instance, in an educational scenario, a scenario relevance score with pedagogical significance would be increased. Again, taking component c as an example, the determination of the scenario relevance scores of multiple components in the virtual model to be decomposed can refer to the following fourth formula:

[0055]

[0056] In the fourth formula, It can represent a component In application scenarios Relevance score in; application scenarios The application scenario of the virtual model to be disassembled can be represented. The application scenario can be specified by those skilled in the art as needed. In this application, the application scenario can also be simply referred to as the scenario. It can represent application scenarios The set of key features (taking application scenario s as an example of an educational scenario, the learning focus of the educational scenario can be the set of key features; taking application scenario s as an example of an educational scenario). Taking industrial design as an example, the core function of an industrial design scenario can be a set of key features. express Key features in; It can represent a component With key features The degree of matching, The range of values ​​can be selected as follows: ; It can represent key features In application scenarios Importance weights in; This can represent component c in the scene. Importance of its position in the middle; This can represent the interactive value of component c in the scene s;

[0057] in, The calculation can be referenced from the following fifth formula:

[0058]

[0059] In the fifth formula, This can represent component c in the scene. Importance of its position in the middle; It can represent a measure of the visibility of component c; This can represent the centrality measure of component c in the virtual model to be disassembled; and They represent and The corresponding weights;

[0060] in, The calculation can be referenced in the sixth formula below:

[0061]

[0062] In the sixth formula, This can represent the interactive value of component c in the scene s; It can represent application scenarios The collection of interaction types in; It can represent the interaction type The corresponding weights; It can represent a component Supported interaction types The degree of.

[0063] It is understood that the multiple components in the virtual model to be disassembled can also determine their respective scene relevance scores in the same way as component c described above, which will not be elaborated here;

[0064] For example, device compatibility scores can be used to measure the rendering and processing adaptability of the multiple components on a specific computing device (e.g., a server or computer and device d described below).

[0065] Continuing with component c as an example, the determination of the device adaptability scores for multiple components in the virtual model to be disassembled can be referenced to the following formula (number seven):

[0066] ,

[0067] In the seventh formula, It can be used to represent the device adaptability score of component c on device d; It can be used to represent the rendering adaptability score of component c on device d; It can be used to represent the memory compatibility score of component c on device d; This can be used to represent the compatibility score of component c on device d; where, , ,and The weights for rendering adaptability score, memory adaptability score, and processing adaptability score can be represented separately, and can be set as needed by those skilled in the art, and can meet the requirements. ;

[0068] In the seventh formula It can represent a component In the equipment The rendering compatibility score on the screen. The calculation can be referenced in the following eighth formula:

[0069]

[0070] In the eighth formula, It can represent a component In the equipment Rendering compatibility score; This can represent the rendering complexity of component c. This can be determined by the number of polygons and / or material complexity of component c, etc. Can represent devices The rendering performance threshold; It can represent the rendering penalty factor. It can be configured as needed by those skilled in the art;

[0071] In the seventh formula It can represent a component In the equipment Memory compatibility rating on the screen The calculation can be referenced in the following ninth formula:

[0072]

[0073] In the ninth formula, It can represent a component In the equipment Memory compatibility rating; This can represent the memory requirements of component c; Can represent devices The memory threshold; It can represent a memory penalty factor; It can be configured as needed by those skilled in the art;

[0074] In the seventh formula It can represent a component In the equipment On-processing adaptability score, The calculation can be referenced in the following formula number ten:

[0075]

[0076] In the tenth formula, It can represent a component In the equipment On-processing adaptability score; It can represent a component In the equipment The processing complexity (such as the complexity of physics simulation and / or the computational complexity of animation, etc.); For equipment The processing performance threshold; This can represent the processing penalty factor. It can be configured as needed by those skilled in the art.

[0077] In this step, the semantic importance score, geometric complexity score, scene relevance score and device adaptability score of multiple components in the virtual model to be disassembled are determined respectively. This can characterize the features of the virtual model to be disassembled from multiple different dimensions and provide a data basis for determining the comprehensive score.

[0078] Step 102: Determine the comprehensive score of the multiple components based on the semantic importance score, the geometric complexity score, the scene relevance score, and the device adaptability score.

[0079] The determination of the comprehensive score of the multiple components can be achieved by weighted summation of the semantic importance score, the geometric complexity score, the scene relevance score, and the device adaptability score, or by nonlinear fusion of the semantic importance score, the geometric complexity score, the scene relevance score, and the device adaptability score.

[0080] In this step, the comprehensive scores of the multiple components are determined based on the semantic importance score, the geometric complexity score, the scene relevance score, and the device adaptability score, so that the comprehensive score can integrate the feature information of the multiple components and provide a data foundation for generating the disassembly strategy.

[0081] Step 103: Generate a dismantling strategy based on the comprehensive score, and dismantle the virtual model to be dismantled according to the dismantling strategy.

[0082] The process of generating a disassembly strategy based on the comprehensive score can be described later, which involves determining the disassembly level of the multiple components based on the comprehensive score (which can be the initial disassembly level, the first disassembly level, the second disassembly level, or the third disassembly level). Then, a disassembly strategy is generated based on the disassembly level, and the virtual model to be disassembled is disassembled according to the disassembly strategy.

[0083] The disassembly strategy may include: determining the disassembly operation type for the virtual model to be disassembled;

[0084] For example, by comparing the disassembly level (described later) with the hierarchical level of the current component in the virtual model to be disassembled, the type of operation that each component should perform includes at least one of the following:

[0085] Keep operation: When the component's decomposition level is equal to the current hierarchy level, the component remains unchanged (i.e., it is not further decomposed) and requires no further processing. These components are typically in an optimal state of detail, balancing visual quality and computational resource requirements.

[0086] Simplify: When a component's decomposition level is lower than the current hierarchy level, the component needs to be simplified. This can occur when the computing resources of the specific device are limited or the component is located in a visually non-critical area. By reducing the level of detail in the decomposition, computing resources can be saved.

[0087] Decompose: When a component's decomposition level is higher than the current hierarchy level, the component's granularity needs to be increased. This is typically applied to visually critical areas or functionally important components, enhancing user experience or ensuring functional integrity through increased granularity.

[0088] The decomposition strategy may further include determining the degree of simplification corresponding to the simplification operation; for example, determining the specific degree of simplification based on the level difference between the component's decomposition level and the current hierarchical level; specifically, the degree of simplification is positively correlated or proportional to the level difference, and can also be adjusted through a preset geometric simplification coefficient. This mechanism ensures that the simplification process is gradual and controllable, avoiding a sharp decline in visual quality due to over-simplification; the degree of simplification determines the parameter configuration when performing subsequent geometric simplification operations, directly affecting the final visual quality.

[0089] Furthermore, the decomposition strategy may also include determining the decomposition processing method; for example, based on the characteristics of the components, the decomposition level, and the operation type, the system selects the best processing method from the available method library: for geometric simplification, possible methods include edge collapse, quadrilateral simplification, voxelization, detail texture replacement, etc.; for component decomposition, possible methods include semantic segmentation, geometric segmentation, application of preset templates, etc.; the selection process is based on the predicted quality score, comprehensively considering visual preservation, semantic integrity, and computational efficiency; and can further calculate the expected quality score for each method, selecting the method with the highest score as the final execution strategy.

[0090] Furthermore, the disassembly strategy may also include: generating processing instructions for each component based on the disassembly operation type, the disassembly level, the simplification degree, and the disassembly processing method, and then disassembling the virtual model to be disassembled according to the processing instructions.

[0091] For example, the method is applied to both Virtual Reality (VR) and a regular Personal Computer (PC). Existing decomposition methods either cause lag on the PC or insufficient feature representation on the VR side. The method disclosed in this application can be adjusted according to the characteristics of the device. When applied to a PC, virtual model decomposition can simplify non-critical components to ensure smooth operation. When applied to a VR side, virtual model decomposition can retain more details required for immersion, ultimately achieving the best balance of experience on both PC and VR sides.

[0092] In this step, a disassembly strategy is generated based on the comprehensive score, and the virtual model to be disassembled is disassembled according to the disassembly strategy. This can determine the disassembly strategy based on the geometric complexity of the virtual model itself, different scene requirements and device capabilities, based on understanding the semantic structure and other features of the virtual model to be disassembled. While maintaining key semantic features, it can achieve the best balance between performance and visual quality, and realize the efficient application of the virtual model to be disassembled in multi-scene and multi-device application environments.

[0093] In this embodiment, a comprehensive score is generated based on semantic importance, geometric complexity, scene relevance, and device adaptability, and a disassembly strategy is generated. The virtual model to be disassembled is then disassembled according to the generated disassembly strategy. This disassembly method avoids the limitations of relying solely on topological structural elements for disassembly. It disassembles the virtual model to be disassembled based on a comprehensive multi-dimensional score, thereby improving the disassembly effect of the virtual model.

[0094] In some implementations, determining the semantic importance scores of multiple components in the virtual model to be disassembled includes:

[0095] Semantic tag sets for the plurality of components are determined respectively, and tag importance scores for the plurality of components are determined respectively based on the semantic tag sets and the application scenario of the virtual model to be disassembled.

[0096] Obtain the hierarchical relationship of the multiple components, and determine the hierarchical importance score of each of the multiple components based on the hierarchical relationship;

[0097] Each of the multiple components is assigned a role importance score in the virtual model to be disassembled;

[0098] The semantic importance scores of the multiple components are obtained by weighted summation of the tag importance score, the hierarchy importance score, and the role importance score.

[0099] The determination of the semantic tag set of the multiple components can be made through dimensions such as the function, structure, interactivity and aesthetics of each component, so that the virtual model to be decomposed can be expressed and processed in a structured way.

[0100] It is easy to understand that the semantic tag set of the above components may include one or more semantic tags; the semantic tag set can be determined by one or more methods such as manual annotation, machine vision recognition, artificial intelligence recognition, and historical data import.

[0101] For example, a steering wheel component in a virtual car model might be labeled with tags like "interactivity" and "necessary," thus determining its semantic tag set, and the importance score for these tags can be relatively high. Conversely, a decorative wheel hub component might be labeled with "aesthetics" and "decorative," thus determining its semantic tag set, and the importance score for this component can be relatively low. These semantic tags not only describe the functional significance of the components but also provide crucial information for subsequent generation and disassembly strategies.

[0102] The step involves determining the label importance scores of the multiple components based on the semantic label set and the application scenario of the virtual model to be decomposed. This can be achieved by determining the numerical weight of each label in the semantic label set of each component according to a preset semantic label weight mapping function under the application scenario; and by performing a weighted summation according to the preset intrinsic importance value of each label and the numerical weight to determine the label importance score of each component, thereby determining the label importance scores of the multiple components respectively.

[0103] Specifically, the preset semantic tag weight mapping function for the aforementioned application scenario can be found in the following eleventh formula:

[0104]

[0105] In the eleventh formula, tag is used to represent a semantic label; This is used to represent the preset semantic tag weight mapping function in the application scenario; "core", "essential", "functional", "structural", "educational", "decorative", "aesthetic", and "hidden" respectively represent core semantic tags, essential semantic tags, functional semantic tags, structural semantic tags, educational semantic tags, decorative semantic tags, aesthetic semantic tags, and hidden semantic tags; " is used to indicate that the semantic tag is a certain semantic tag or a certain type of semantic tag, for example: The `else` option is used to indicate the case where the semantic label is the core semantic label; `else` is used to indicate other cases of the semantic label, which are not listed here; the above-mentioned preset semantic label weight mapping function can map the semantic label to a numerical weight of [0,1].

[0106] Specifically, obtaining the hierarchical relationship of the multiple components and determining the hierarchical importance score of each of the multiple components based on the hierarchical relationship may include the following steps:

[0107] A hierarchical tree structure is constructed based on the parent-child relationship between components;

[0108] Starting from the root node, the initial decomposition level of each component is set recursively; optionally, the root node can be the simulation model to be decomposed, or any of the multiple components.

[0109] Based on the initial decomposition level, the connection relationships, inclusion relationships, and dependency relationships among the multiple components, the hierarchical relationship of the multiple components is obtained.

[0110] The hierarchical relationship of the multiple components can be a hierarchical tree structure, which can record the inclusion relationship between components and can also serve as the basis for the decomposition level in the above decomposition strategy.

[0111] The process of obtaining the hierarchical relationship of the multiple components can be achieved by recursively setting the initial decomposition level of each component starting from the root node, ensuring that the model decomposition has top-down hierarchical coherence.

[0112] The recursive setting of the initial decomposition level for each component can be found in the following twelfth formula:

[0113]

[0114] In the twelfth formula, This can represent recursively setting components. The initial decomposition level; This can represent the component The hierarchy is set to ; It can represent a component The collection of all child components; Indicates the component Each subcomponent Perform the operation; This can indicate that the function is called recursively to child components, incrementing the level by 1.

[0115] The step of determining the hierarchical importance scores of the multiple components according to the hierarchical relationship can be: determining the hierarchical importance scores of the multiple components according to the hierarchical relationship using a preset model or a preset evaluation method;

[0116] The step of determining the role importance score of the plurality of components in the virtual model to be disassembled can be that each of the plurality of components determines its role importance score in the virtual model to be disassembled based on its corresponding structural features, functional features and / or interactive features.

[0117] Continuing with component c as an example (component c can be any of the multiple components), the weighted summation of the tag importance score, the hierarchical importance score, and the role importance score yields the semantic importance scores of the multiple components, as shown in the following formula thirteen:

[0118]

[0119] In the thirteenth formula, This can represent the semantic importance score of component c; , and The weights for tag importance score, role importance score, and hierarchy importance score can be represented separately, and can be set as needed by those skilled in the art, and can meet the requirements. ; It can represent a component A collection of semantic tags; It can represent semantic tags The intrinsic importance value; It can represent semantic tags Weights in the current application scenario (e.g.: The label can be determined based on the semantic label weight mapping function of the current application scenario. (weights) Rate the importance of the functional role of the components; This can represent the hierarchical importance score of component c in the hierarchical relationship;

[0120] Among them, in the thirteenth formula The calculation can be referenced in the following fourteenth formula:

[0121] ;

[0122] In the fourteenth formula, This can represent the depth of component c in the hierarchical relationship; It can be used to represent the number of child components of component c; It can represent the number of components of the virtual model to be disassembled (the number of components of the virtual model to be disassembled can be the number of components of the multiple components).

[0123] In the fourteenth formula, and They can be represented separately and The corresponding weights and The specific value can be set by those skilled in the art as needed.

[0124] In this embodiment, the label importance score, the hierarchy importance score, and the role importance score are weighted and summed to obtain the semantic importance score of the multiple components. The semantic importance score can measure the importance of the multiple components at the functional and semantic levels, ensuring that the core semantic information can be completely preserved after the virtual model is decomposed, improving the feature expression ability of the virtual model after decomposition, and thus improving the decomposition effect of the virtual model.

[0125] In some implementations, determining the comprehensive score of the multiple components based on the semantic importance score, the geometric complexity score, the scene relevance score, and the device adaptability score includes:

[0126] Based on the application scenario of the virtual model to be disassembled, the corresponding score weights of the semantic importance score, geometric complexity score, scenario relevance score and device adaptability score of the multiple components in the application scenario are determined respectively.

[0127] The semantic importance score, set complexity score, scene relevance score, and device adaptability score are weighted and summed according to the scoring weights to obtain the comprehensive score of the multiple components.

[0128] Specifically, continuing with component c as an example, the semantic importance score, set complexity score, scene relevance score, and device adaptability score are weighted and summed according to the scoring weights to obtain the comprehensive score of the multiple components. This can be referred to in the following fifteenth formula:

[0129]

[0130] In the fifteenth formula, It can represent a component In application scenarios and equipment The overall score; , , and These can be used to represent semantic importance score, geometric complexity score, scene relevance score, and device compatibility score in the application scenario, respectively. The scoring weights in the ratings; , , and All of these can be set as needed by those skilled in the art or determined according to a preset application scenario scoring weight table, and can meet the requirements. ; , , and These can be represented by semantic importance score, geometric complexity score, scene relevance score, and device compatibility score, respectively. The geometric complexity score can be used... The higher the geometric complexity, the lower the tendency should be to retain the original level of detail.

[0131] For example, the scoring weights can be determined based on a preset application scenario scoring weight table. Taking industrial design scenario, education and training scenario, and virtual reality scenario as examples, the scenario scoring weight table can be referenced in Table 1 below:

[0132] Table 1:

[0133]

[0134] In Table 1 above, , , and These can represent the semantic importance score, geometric complexity score, scene relevance score, and device compatibility score in the application scenario, respectively. The scoring weights in the system.

[0135] In this embodiment, based on the application scenario of the virtual model to be disassembled, the semantic importance score, geometric complexity score, scenario relevance score, and device adaptability score of the multiple components are determined according to the application scenario of the application scenario; and the comprehensive score of the multiple components is obtained by weighted summation; this can comprehensively consider the characteristics of the multiple components in various aspects, generate disassembly strategies according to the application scenario and device capabilities, improve the feature expression ability of the virtual model after disassembly, and thus improve the disassembly effect of the virtual model.

[0136] In some implementations, generating a disassembly strategy based on the comprehensive score includes:

[0137] Based on the comprehensive score and the preset equipment complexity threshold, the first disassembly level of each of the multiple components is determined.

[0138] The value density of each of the plurality of components is determined based on the product of the first disassembly level and the comprehensive score.

[0139] Based on the first disassembly level and the value density, a second disassembly level is determined for each of the plurality of components, and a disassembly strategy is generated based on the second disassembly level.

[0140] The step of determining the first disassembly level of the plurality of components based on the comprehensive score and the preset device complexity threshold may include:

[0141] Based on the overall score, the preset maximum disassembly level, and the preset disassembly level range, the initial disassembly level of each of the multiple components is determined.

[0142] Based on the preset equipment complexity threshold and the initial disassembly level, the first disassembly level of each of the plurality of components is determined respectively;

[0143] Specifically, taking component c in application scenario s and device d as examples, the initial disassembly levels of the multiple components are determined based on the comprehensive score, the preset maximum disassembly level (which can also be the maximum disassembly level that the specific computing device can support), and the preset disassembly level range (which can also be the disassembly level range that the specific computing device can support). This can be referenced in the following sixteenth formula:

[0144]

[0145] In the sixteenth formula, It can represent a component In application scenarios and equipment The initial disassembly level in; Can represent devices Maximum supported disassembly level; It can represent a component The overall score can represent the component's score. Importance or priority; The range of values ​​can be ; Can represent devices Equipment-related decomposition coefficients can be used to adjust the impact of equipment on the decomposition level; This indicates that a preset disassembly level range can be represented; This represents the floor function, which can be used to ensure that the initial floor level is an integer;

[0146] Continuing with the example of component c in application scenario s and device d, the determination of the first disassembly level of the plurality of components based on the preset device complexity threshold and the initial disassembly level can be referenced in the following formula seventeen:

[0147]

[0148] In the seventeenth formula, It can be used to represent components In application scenarios and equipment The first level of disassembly in the process; and The devices can be represented separately. The corresponding first preset device complexity threshold and second preset device complexity threshold, and can satisfy ; It can represent a component The overall score; It can be used to represent components In application scenarios and equipment The initial disassembly level in;

[0149] In the seventeenth formula, in the component The overall score is lower than In the case of components The first disassembly level is set to 0 (i.e., no components are disassembled). Any further disassembly, retain components (the original model); in the components The overall score is greater than or equal to and components The overall score is less than or equal to In this case, the component The initial decomposition level as a component The first level of disassembly; in the components The overall score is greater than In this case, the component The initial decomposition level is raised one level as a component. The first disassembly level, and components The first disassembly level shall not exceed the maximum disassembly level mentioned above. ;

[0150] Constrained by the aforementioned preset device complexity threshold, components with relatively high overall scores and less complex designs will retain more details; components with relatively low overall scores or overly complex designs will be significantly simplified.

[0151] Continuing with the example of component c in application scenario s and device d, the value density of the multiple components is determined based on the product of the first disassembly level and the comprehensive score, which can be referenced in the following eighteenth formula:

[0152]

[0153] In the eighteenth formula, It can represent a component The value density is used to represent the value density level of each of the plurality of components; It can represent a component The overall score; It can be used to represent components In application scenarios and equipment The first level of disassembly in the process;

[0154] In this embodiment, a second disassembly level is determined for each of the plurality of components based on the first disassembly level and the value density, and a disassembly strategy is generated based on the second disassembly level. The disassembly strategy is optimized by the device capabilities of a specific computing device, the inherent complexity of each component, and the value density of each component, thereby improving the feature representation capability of the virtual model after disassembly and thus improving the disassembly effect of the virtual model.

[0155] In some implementations, determining the second disassembly level of the plurality of components based on the first disassembly level and the value density includes:

[0156] Based on the ranking result obtained by sorting the value density of the multiple components, the first disassembly level of the components located within the preset sequence range is adjusted to the second disassembly level.

[0157] Among them, components that are outside the preset sequence range in the sorting results are configured to keep their component models unchanged.

[0158] The preset sequence range can be a manually set range, or it can be limited by the specific computing device (e.g., device d) to support only a maximum of [specific range]. The scope is determined by decomposing each component;

[0159] For example, the plurality of components can be arranged according to value density. Sort the components from highest to lowest value to obtain the ranking results; prioritize components with high value density, as they contribute more to the overall optimization.

[0160] A greedy strategy can be applied to allocate the dismantling level; taking a specific computing device as device d as an example, since the device... It can only support a maximum of The components are decomposed, and based on the sorting results, the first... Each component retains its first level of disassembly. Unchanged, the first disassembly level will remain unchanged. As the second disassembly level For sorting in For subsequent components, set their second decomposition level to 0 and do not decompose them (i.e., configure them to keep the component model unchanged).

[0161] Under the constraints of equipment capacity and / or preset sequence range, the component with the highest value density is selected for decomposition to achieve optimization. The second decomposition level for each component is determined by calculating value density, sorting by value density, and applying a greedy strategy. This allows for optimization of the number of component splits.

[0162] In this embodiment, based on the ranking result obtained by sorting the value density of the multiple components, the first disassembly level of the components located within the preset sequence range is adjusted to the second disassembly level, thereby optimizing the number of component splits among the multiple components, thereby optimizing the consumption of computing resources and the feature retention capability after the virtual model is disassembled, improving the feature expression capability after the virtual model is disassembled, and thus improving the disassembly effect of the virtual model.

[0163] In some implementations, generating a disassembly strategy based on the second disassembly level includes:

[0164] The adjusted disassembly level of the target component is determined based on the target component, a preset first threshold, and a preset second threshold. The first threshold is a first threshold representing the difference between the second disassembly levels of the target component and those of components at the next higher level. The second threshold is a second threshold representing the difference between the second disassembly levels of the target component and those of components at the next lower level. Based on the adjusted disassembly level of the target component and the adjusted disassembly levels of adjacent components within a preset spatial range, a weighted sum is performed according to preset weights to obtain the third disassembly level of the target component. The target component is any one of the plurality of components.

[0165] A disassembly strategy is generated based on the third disassembly level of the plurality of components, and the virtual model to be disassembled is disassembled according to the disassembly strategy.

[0166] Wherein, the first threshold and the second threshold can be thresholds that can be set as needed by those skilled in the art to maintain the relative smoothness of component disassembly levels between different layers;

[0167] The target component is the aforementioned component. For example, if the component If there is a parent component (i.e., the component at the next higher level), then its decomposition level cannot exceed the parent component's decomposition level plus the first threshold (e.g., a preset first upper limit difference). For example, in components There is a parent component, and the component The disassembly level exceeds the disassembly level of the parent component plus In this case, add the parent component's decomposition level. As a component Adjusted disassembly level of (target component);

[0168] If the component Without a parent component, the component can be... The corresponding second disassembly level Adjusted disassembly level as the target component;

[0169] If the component If there are child components (components at the next level), then its decomposition level cannot be lower than the child component's decomposition level plus the second threshold (e.g., a preset second upper limit difference). ), indicating component The disassembly level can be as low as that of the sub-component. In other words, the decomposition level of a parent component cannot be lower than the difference between the decomposition levels of all its child components minus the lower limit. The minimum value. For example, in the component It has child components, and the component The disassembly level is lower than the disassembly level of the sub-component plus... In this case, reduce the decomposition level of the sub-component. As a component Adjusted disassembly level of (target component);

[0170] It is understandable that by determining the adjustment and disassembly level of the target component, the adjustment and disassembly level of the adjacent components of the target component within a preset spatial range can be determined, as well as the adjustment and disassembly level of each of the multiple components; thereby achieving hierarchical coordinated adjustment of the multiple components.

[0171] Specifically, continue with components For example, the weighted summation according to preset weights to obtain the third disassembly level of the target component can be referred to in the following nineteenth formula:

[0172]

[0173] In the nineteenth formula, It can represent a component The third level of disassembly; This can represent the adjusted disassembly level of component c; It can represent adjacent components The adjusted disassembly level;

[0174] It can represent a component The set of adjacent components; express The first in Each component, i.e., component Adjacent components ; It can represent a component With adjacent components The association weights between components are used to represent the degree of influence that adjacent components have on the current component; It can represent a spatial consistency smoothing coefficient, used to control the degree of influence of adjacent components on the adjustment of the current component's decomposition level. It can be configured as needed by those skilled in the art; The function is used to represent a rounding function (such as a rounding function or a banker's rounding function; the specific rounding rules can be set as needed by those skilled in the art).

[0175] It is understandable that in the nineteenth formula above, one can... Understand as The weight, Understand as The weights;

[0176] In this embodiment, since a disassembly strategy is generated based on the third target disassembly level of the plurality of components, and the virtual model to be disassembled is disassembled according to the disassembly strategy, the disassembly levels of the upper and lower level components of the plurality of components and adjacent components within a preset space range can remain relatively consistent during the virtual model disassembly process, avoiding drastic changes, improving the visual coordination and feature expression ability of the virtual model after disassembly, and improving the disassembly effect of the virtual model.

[0177] In some implementations, please refer to Figure 2 , Figure 2 This is a second flowchart of a virtual model disassembly method provided in an embodiment of this application, the method comprising:

[0178] The model semantic information acquisition and preprocessing can be used to determine the semantic label set of the multiple components respectively;

[0179] The component hierarchy structure analysis can be used to determine the hierarchical relationship and / or disassembly level of the multiple components;

[0180] The comprehensive score of the multiple components can be determined by using a multi-dimensional scoring mechanism. The semantic importance score, geometric complexity score, scene relevance score and device adaptability score of the multiple components can be determined by the multi-dimensional scoring mechanism, and then the comprehensive score of the multiple components can be determined.

[0181] Generating a dismantling strategy; wherein, generating the dismantling strategy may include: determining the initial dismantling level of the plurality of components respectively; optimizing the number of components that need to be further dismantled (for example, optimizing the number of component splits by using value density ranking and applying a greedy strategy as described above); performing hierarchical coordination and adjustment of the dismantling levels of the plurality of components; and determining the final dismantling strategy (i.e., final decision generation).

[0182] In some implementations, please refer to Figure 2The method also includes the implementation and application of the disassembly results;

[0183] The method may specifically include: performing geometric operations on the virtual model to be disassembled using a preset geometric manipulator; monitoring the real-time performance of the specific computing device to obtain real-time performance indicators; dynamically adjusting the disassembly strategy based on the real-time performance indicators; and evaluating and providing feedback on the disassembly results, wherein the disassembly results may be obtained by disassembling the virtual model to be disassembled according to the disassembly strategy.

[0184] Specifically, the step of performing geometric operations on the virtual model to be disassembled using a preset geometric manipulator may include: disassembling, merging, and / or subdividing the multiple components;

[0185] More specifically, the step of performing geometric operations on the virtual model to be disassembled using a preset geometric manipulator may include: performing simplified operations on multiple components of the virtual model to be disassembled, performing disassembly operations, preserving geometric features, and processing boundary connections; thereby maintaining the visual continuity and functional integrity of the virtual model after disassembly.

[0186] More specifically, continuing with component c as an example, the simplification operation performed on multiple components in the virtual model to be disassembled can be found in the following formula 20:

[0187]

[0188] In the twentieth formula, Used to represent a simplified operation execution function for component c; This can represent the original geometric model of component c; It can represent the simplification ratio, which in turn represents the degree to which simplification is required; The calculation method can be the difference between the current level and the final decomposition level of component c, multiplied by a preset coefficient. To confirm; It can indicate the use of The algorithm applies a simplification ratio to the original geometric model of component c. Perform simplified operations; It can represent a preset simplification method;

[0189] More specifically, the simplification method may include one or more of the following algorithms: edge collapse algorithm, quadrilateral merging simplification algorithm, voxelization simplification algorithm, detail texture mapping algorithm, and contour preservation simplification algorithm;

[0190] The applicable scenarios and popular descriptions of the simplification method described in the twentieth formula can be found in Table 2:

[0191] Table 2:

[0192]

[0193] More specifically, continuing with component c as an example, the disassembly operation of multiple components in the virtual model to be disassembled can be seen in the following formula:

[0194]

[0195] In the twenty-first formula, This can represent the function to execute the disassembly operation of component c; This can represent the original geometric model of component c; This can represent the number of sub-components after component c is decomposed; It can represent the original geometric model of component c, using a preset disassembly method. Disassembled into Number of sub-components; It can represent a preset decomposition method; the decomposition method may include one or more of the following: semantic segmentation decomposition algorithm, convex decomposition algorithm, hierarchical space partitioning algorithm, skeleton-guided decomposition algorithm, and preset template decomposition algorithm;

[0196] In the twenty-first formula, the applicable scenarios and common descriptions of the disassembly method can be found in Table 3:

[0197] Table 3:

[0198]

[0199] More specifically, continuing with component c as an example, the geometric feature preservation of multiple components in the virtual model to be disassembled can be found in the following formula twenty-two:

[0200]

[0201] In the twenty-second formula, This can represent the geometric feature preservation function of component c; It can represent a temporary model of component c (e.g., a simplified or decomposed model). It can represent a set of key features; It can represent the features that need to be retained; It can represent the preset feature retention weights; It can be used to control the intensity of feature preservation; for example, during enhancement, these features are prioritized for preservation. The higher the weight of the key features, the more obvious the feature is preserved. It can be used to represent a pair use and The parameters are used to preserve geometric features.

[0202] More specifically, continuing with component c as an example, the boundary connection processing of multiple components in the virtual model to be disassembled can be found in the following formula twenty-three:

[0203]

[0204] In the twenty-third formula, This can represent the boundary connection handling function of component c; It can represent a temporary model of the adjacent components of component c; It can represent a shared boundary (that is, a common boundary between two components). It can represent the preset boundary matching tolerance. This indicates the allowable error range. Functions can represent boundary connection processing operations.

[0205] Specifically, the real-time performance monitoring of the specific computing device to obtain real-time performance indicators may include one or more of the following: frame rate monitoring, rendering load monitoring, performance early warning monitoring, and device resource utilization monitoring, to obtain performance indicators such as frame rate, rendering load (which can be used to characterize or determine rendering complexity), memory usage, and performance anomalies.

[0206] More specifically, the frame rate monitoring can be calculated by taking the reciprocal of the time required to take one frame, as shown in the following formula (number twenty-four):

[0207]

[0208] In the twenty-fourth formula, It can be used to represent the frame rate at time t. It can represent the time of a single frame.

[0209] More specifically, the rendering load can be obtained by weighted summing of the polygon count and material complexity of each visible component, as shown in Formula Twenty-Five below:

[0210]

[0211] In the twenty-fifth formula, This can represent the rendering load at time t; It can represent the set of visible components at time t; component c can represent Visible components in; This can represent the number of polygons in component c; It can represent a component The complexity of the materials; and The weights for the number of polygons and the material complexity can be represented separately. and All of these can be configured as needed by those skilled in the art.

[0212] More specifically, performance warning monitoring may include triggering a warning when the frame rate is lower than the target frame rate minus the fault tolerance range, or when the rendering load exceeds a threshold. The warning is used to alert the user.

[0213] Specifically, the step of dynamically adapting the disassembly strategy based on the real-time performance indicators may include: adjusting the disassembly strategy in real time based on changes in the performance indicators fed back by the specific computing device during operation, to ensure that the best balance between performance and quality is always achieved.

[0214] Specifically, the evaluation and feedback of the disassembly results can be based on the visual quality and / or semantic preservation of the disassembly results (i.e., whether the model can effectively convey the original design intent after disassembly and / or simplification) to optimize various disassembly parameters in the disassembly strategy.

[0215] In summary, the four core mechanisms described above enable the disassembly strategy to be efficiently transformed into specific geometric operations, monitor performance indicators in real time, dynamically adapt to changes in the operating environment, and continuously optimize parameter configuration through quality assessment, thereby achieving the best presentation of the virtual model in different environments.

[0216] It should be noted that the virtual model disassembly method described above can be executed by an electronic device, that is, all steps included in the above method are executed by the electronic device, which can be a server, computer or mobile terminal (such as a mobile phone) or other electronic devices.

[0217] See Figure 3 , Figure 3 This is a schematic diagram of the structure of a virtual model disassembly device provided in an embodiment of this application, as shown below. Figure 3 As shown, the virtual model disassembly device 300 includes:

[0218] The first processing module 301 is used to determine the semantic importance score, geometric complexity score, scene relevance score and device adaptability score of multiple components in the virtual model to be disassembled.

[0219] The second processing module 302 is used to determine the comprehensive score of the multiple components based on the semantic importance score, the geometric complexity score, the scene relevance score, and the device adaptability score.

[0220] The third processing module 303 is used to generate a dismantling strategy based on the comprehensive score, and to dismantle the virtual model to be dismantled according to the dismantling strategy.

[0221] Optionally, determining the semantic importance scores of multiple components in the virtual model to be disassembled includes:

[0222] Semantic tag sets for the plurality of components are determined respectively, and tag importance scores for the plurality of components are determined respectively based on the semantic tag sets and the application scenario of the virtual model to be disassembled.

[0223] Obtain the hierarchical relationship of the multiple components, and determine the hierarchical importance score of each of the multiple components based on the hierarchical relationship;

[0224] Each of the multiple components is assigned a role importance score in the virtual model to be disassembled;

[0225] The semantic importance scores of the multiple components are obtained by weighted summation of the tag importance score, the hierarchy importance score, and the role importance score.

[0226] Optionally, generating a disassembly strategy based on the comprehensive score includes:

[0227] Based on the comprehensive score and the preset equipment complexity threshold, the first disassembly level of each of the multiple components is determined.

[0228] The value density of each of the plurality of components is determined based on the product of the first disassembly level and the comprehensive score.

[0229] Based on the first disassembly level and the value density, a second disassembly level is determined for each of the plurality of components, and a disassembly strategy is generated based on the second disassembly level.

[0230] Optionally, determining the second disassembly level of the plurality of components based on the first disassembly level and the value density includes:

[0231] Based on the ranking result obtained by sorting the value density of the multiple components, the first disassembly level of the components located within the preset sequence range is adjusted to the second disassembly level.

[0232] Among them, components that are outside the preset sequence range in the sorting results are configured to keep their component models unchanged.

[0233] Optionally, generating a disassembly strategy based on the second disassembly level includes:

[0234] The adjusted disassembly level of the target component is determined based on the target component, a preset first threshold, and a preset second threshold. The first threshold is a first threshold representing the difference between the second disassembly levels of the target component and those of components at the next higher level. The second threshold is a second threshold representing the difference between the second disassembly levels of the target component and those of components at the next lower level. Based on the adjusted disassembly level of the target component and the adjusted disassembly levels of adjacent components within a preset spatial range, a weighted sum is performed according to preset weights to obtain the third disassembly level of the target component. The target component is any one of the plurality of components.

[0235] A disassembly strategy is generated based on the third disassembly level of the plurality of components, and the virtual model to be disassembled is disassembled according to the disassembly strategy.

[0236] Optionally, determining the comprehensive score of the multiple components based on the semantic importance score, the geometric complexity score, the scene relevance score, and the device adaptability score includes:

[0237] Based on the application scenario of the virtual model to be disassembled, the corresponding score weights of the semantic importance score, geometric complexity score, scenario relevance score and device adaptability score of the multiple components in the application scenario are determined respectively.

[0238] The semantic importance score, set complexity score, scene relevance score, and device adaptability score are weighted and summed according to the scoring weights to obtain the comprehensive score of the multiple components.

[0239] The virtual model disassembly device 300 is capable of implementing each process of the above-described embodiments of the virtual model disassembly method. The technical features correspond one-to-one and can achieve the same technical effect. To avoid repetition, it will not be described again here.

[0240] This application also provides an electronic device, including: a processor, a memory, and a program stored in the memory and executable on the processor. When the program is executed by the processor, it implements the various processes of the above-described virtual model decomposition method embodiments and achieves the same technical effect. To avoid repetition, it will not be described again here.

[0241] For details, see Figure 4 This application also provides an electronic device, including a bus 401, a transceiver 402, an antenna 403, a bus interface 404, a processor 405, and a memory 406;

[0242] The processor 405 is used to determine the semantic importance score, geometric complexity score, scene relevance score and device adaptability score of multiple components in the virtual model to be disassembled.

[0243] The comprehensive scores of the multiple components are determined based on the semantic importance score, the geometric complexity score, the scene relevance score, and the device adaptability score.

[0244] A disassembly strategy is generated based on the comprehensive score, and the virtual model to be disassembled is disassembled according to the disassembly strategy.

[0245] Optionally, determining the semantic importance scores of multiple components in the virtual model to be disassembled includes:

[0246] Semantic tag sets for the plurality of components are determined respectively, and tag importance scores for the plurality of components are determined respectively based on the semantic tag sets and the application scenario of the virtual model to be disassembled.

[0247] Obtain the hierarchical relationship of the multiple components, and determine the hierarchical importance score of each of the multiple components based on the hierarchical relationship;

[0248] Each of the multiple components is assigned a role importance score in the virtual model to be disassembled;

[0249] The semantic importance scores of the multiple components are obtained by weighted summation of the tag importance score, the hierarchy importance score, and the role importance score.

[0250] Optionally, generating a disassembly strategy based on the comprehensive score includes:

[0251] Based on the comprehensive score and the preset equipment complexity threshold, the first disassembly level of each of the multiple components is determined.

[0252] The value density of each of the plurality of components is determined based on the product of the first disassembly level and the comprehensive score.

[0253] Based on the first disassembly level and the value density, a second disassembly level is determined for each of the plurality of components, and a disassembly strategy is generated based on the second disassembly level.

[0254] Optionally, determining the second disassembly level of the plurality of components based on the first disassembly level and the value density includes:

[0255] Based on the ranking result obtained by sorting the value density of the multiple components, the first disassembly level of the components located within the preset sequence range is adjusted to the second disassembly level.

[0256] Among them, components that are outside the preset sequence range in the sorting results are configured to keep their component models unchanged.

[0257] Optionally, generating a disassembly strategy based on the second disassembly level includes:

[0258] The adjusted disassembly level of the target component is determined based on the target component, a preset first threshold, and a preset second threshold. The first threshold is a first threshold representing the difference between the second disassembly levels of the target component and those of components at the next higher level. The second threshold is a second threshold representing the difference between the second disassembly levels of the target component and those of components at the next lower level. Based on the adjusted disassembly level of the target component and the adjusted disassembly levels of adjacent components within a preset spatial range, a weighted sum is performed according to preset weights to obtain the third disassembly level of the target component. The target component is any one of the plurality of components.

[0259] A disassembly strategy is generated based on the third disassembly level of the plurality of components, and the virtual model to be disassembled is disassembled according to the disassembly strategy.

[0260] Optionally, determining the comprehensive score of the multiple components based on the semantic importance score, the geometric complexity score, the scene relevance score, and the device adaptability score includes:

[0261] Based on the application scenario of the virtual model to be disassembled, the corresponding score weights of the semantic importance score, geometric complexity score, scenario relevance score and device adaptability score of the multiple components in the application scenario are determined respectively.

[0262] The semantic importance score, set complexity score, scene relevance score, and device adaptability score are weighted and summed according to the scoring weights to obtain the comprehensive score of the multiple components.

[0263] exist Figure 4In this context, a bus architecture (represented by bus 401) is used. Bus 401 can include any number of interconnected buses and bridges, linking various circuits including one or more processors represented by processor 405 and memory represented by memory 406. Bus 401 can also link various other circuits such as peripheral devices, voltage regulators, and power management circuits, which are well known in the art and therefore will not be described further herein. Bus interface 404 provides an interface between bus 401 and transceiver 402. Transceiver 402 can be a single element or multiple elements, such as multiple receivers and transmitters, providing a unit for communicating with various other devices over a transmission medium. Data processed by processor 405 is transmitted over a wireless medium via antenna 403, which further receives data and transmits data to processor 405.

[0264] Processor 405 is responsible for managing bus 401 and general processing, and can also provide various functions, including timing, peripheral interface, voltage regulation, power management, and other control functions. Memory 406 can be used to store data used by processor 405 during operation.

[0265] Optionally, the processor 405 may be a central processing unit (CPU), an application-specific integrated circuit (ASIC), a field-programmable gate array (FPGA), or a complex programmable logic device (CPLD).

[0266] This application also provides a computer-readable storage medium storing a computer program. When executed by a processor, this computer program implements the various processes of the above-described virtual model decomposition method embodiments and achieves the same technical effects. To avoid repetition, it will not be described again here. The computer-readable storage medium may be a read-only memory (ROM), a random access memory (RAM), a magnetic disk, or an optical disk, etc.

[0267] This application also provides a computer program product, including computer instructions. When executed by a processor, these computer instructions implement the various processes of the above-described virtual model decomposition method embodiments and achieve the same technical effects. To avoid repetition, they will not be described again here.

[0268] It should be noted that, in this document, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes that element. Furthermore, it should be noted that the scope of the methods and apparatuses in the embodiments of this application is not limited to performing functions in the order discussed, but may also include performing functions substantially simultaneously or in the reverse order, depending on the functions involved. For example, the described methods may be performed in a different order than described, and various steps may be added, omitted, or combined. Additionally, features described with reference to certain examples may be combined in other examples.

[0269] Through the above description of the embodiments, those skilled in the art can clearly understand that the methods of the above embodiments can be implemented by means of software plus necessary general-purpose hardware platforms. Of course, they can also be implemented by hardware, but in many cases the former is a better implementation method. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disk) and includes several instructions to cause a terminal (which may be a mobile phone, computer, server, air conditioner, or network device, etc.) to execute the methods described in the various embodiments of this application.

[0270] The embodiments of this application have been described above with reference to the accompanying drawings. However, this application is not limited to the specific embodiments described above. The specific embodiments described above are merely illustrative and not restrictive. Those skilled in the art can make many other forms under the guidance of this application without departing from the spirit and scope of the claims, and all of these forms are within the protection scope of this application.

Claims

1. A method for disassembling a virtual model, characterized in that, The method includes: Semantic importance score, geometric complexity score, scene relevance score, and device adaptability score of multiple components in the virtual model to be disassembled are determined respectively. The comprehensive scores of the multiple components are determined based on the semantic importance score, the geometric complexity score, the scene relevance score, and the device adaptability score. A disassembly strategy is generated based on the comprehensive score, and the virtual model to be disassembled is disassembled according to the disassembly strategy.

2. The method according to claim 1, characterized in that, The process of determining the semantic importance scores of multiple components in the virtual model to be disassembled includes: Semantic tag sets for the plurality of components are determined respectively, and tag importance scores for the plurality of components are determined respectively based on the semantic tag sets and the application scenario of the virtual model to be disassembled. Obtain the hierarchical relationship of the multiple components, and determine the hierarchical importance score of each of the multiple components based on the hierarchical relationship; Each of the multiple components is assigned a role importance score in the virtual model to be disassembled; The semantic importance scores of the multiple components are obtained by weighted summation of the tag importance score, the hierarchy importance score, and the role importance score.

3. The method according to claim 2, characterized in that, The step of generating a disassembly strategy based on the comprehensive score includes: Based on the comprehensive score and the preset equipment complexity threshold, the first disassembly level of each of the multiple components is determined. The value density of each of the plurality of components is determined based on the product of the first disassembly level and the comprehensive score. Based on the first disassembly level and the value density, a second disassembly level is determined for each of the plurality of components, and a disassembly strategy is generated based on the second disassembly level.

4. The method according to claim 3, characterized in that, The step of determining the second disassembly level of the plurality of components based on the first disassembly level and the value density includes: Based on the ranking result obtained by sorting the value density of the multiple components, the first disassembly level of the components located within the preset sequence range is adjusted to the second disassembly level. Among them, components that are outside the preset sequence range in the sorting results are configured to keep their component models unchanged.

5. The method according to claim 3 or 4, characterized in that, The step of generating a disassembly strategy based on the second disassembly level includes: The adjusted disassembly level of the target component is determined based on the target component, a preset first threshold, and a preset second threshold. The first threshold is a first threshold representing the difference between the second disassembly levels of the target component and those of components at the next higher level. The second threshold is a second threshold representing the difference between the second disassembly levels of the target component and those of components at the next lower level. Based on the adjusted disassembly level of the target component and the adjusted disassembly levels of adjacent components within a preset spatial range, a weighted sum is performed according to preset weights to obtain the third disassembly level of the target component. The target component is any one of the plurality of components. A disassembly strategy is generated based on the third disassembly level of the plurality of components, and the virtual model to be disassembled is disassembled according to the disassembly strategy.

6. The method according to any one of claims 1 to 4, characterized in that, The step of determining the comprehensive score of the multiple components based on the semantic importance score, the geometric complexity score, the scene relevance score, and the device adaptability score includes: Based on the application scenario of the virtual model to be disassembled, the corresponding score weights of the semantic importance score, geometric complexity score, scenario relevance score and device adaptability score of the multiple components in the application scenario are determined respectively. The semantic importance score, set complexity score, scene relevance score, and device adaptability score are weighted and summed according to the scoring weights to obtain the comprehensive score of the multiple components.

7. A virtual model disassembly device, characterized in that, The device includes: The first processing module is used to determine the semantic importance score, geometric complexity score, scene relevance score and device adaptability score of multiple components in the virtual model to be disassembled. The second processing module is used to determine the comprehensive score of the multiple components based on the semantic importance score, the geometric complexity score, the scene relevance score, and the device adaptability score. The third processing module is used to generate a dismantling strategy based on the comprehensive score, and to dismantle the virtual model to be dismantled according to the dismantling strategy.

8. An electronic device, characterized in that, include: A processor, a memory, and a program stored in the memory and executable on the processor, wherein the program, when executed by the processor, implements the steps of the method as described in any one of claims 1 to 6.

9. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program that, when executed by a processor, implements the steps of the method as described in any one of claims 1 to 6.

10. A computer program product, characterized in that, Includes computer instructions that, when executed by a processor, implement the steps of the method as described in any one of claims 1 to 6.

Images