Method and system for generating object mesh based on three-dimensional model

Abstract

The application discloses a kind of object grid generation method and system based on three-dimensional model, the method includes: obtaining the three-dimensional model of object;The segmentation boundary is determined by carrying out segmentation operation to three-dimensional model;Based on the segmentation boundary, cut three-dimensional model into multiple segmentation regions;Boundary zone is generated between adjacent segmentation regions;The grid region of the boundary zone is executed to realize boundary reconstruction by remeshing operation, and the reconstructed segmentation boundary is obtained;Based on the reconstructed segmentation boundary, generate the update grid of object;And between detachable component and fixed component separated by segmentation boundary, build multiple connection structures.The application realizes structure decomposition and boundary reconstruction optimization based on the three-dimensional model of object, improves the structure compatibility, assembly stability and connection transition smoothness in additive manufacturing process, improves the rationality and manufacturing adaptation ability of reconstructed morphology.

Description

Technical Field

[0001] This invention relates to the field of 3D printing technology, specifically to meshes for generating objects based on 3D models, and more specifically, to meshes for segmented regions, boundary bands, segmented boundaries, and connection structures of objects generated based on 3D models. Background Technology

[0002] In the field of 3D printing (or additive manufacturing), the challenge has always been how to transform a digitally designed 3D model (usually represented by a triangular mesh, quadrilateral mesh, or other polygonal surface) into a physically manufactureable entity.

[0003] Models that are mathematically valid in a digital environment are often unsuitable for direct manufacturing due to inherent physical, mechanical, and manufacturing limitations. Such models may be geometrically too complex to be constructed as single components; they may also lead to low manufacturing efficiency due to excessive time consumption or complex processing paths; they may result in material waste due to a lack of supporting structures; or they may be incompatible with subsequent assembly processes. Furthermore, digital models often contain inherent defects such as holes, self-intersections, non-manifold geometries, and irregular mesh partitioning, all of which can hinder successful printing.

[0004] Therefore, a significant gap remains between digital design and physical manufacturability, and traditional preprocessing techniques, such as simple slicing, scaling, or mesh repair, are insufficient to fully bridge these gaps. In conclusion, there is an urgent need to develop a 3D printing method that can intelligently adjust and optimize the relationship between digital design and physical manufacturability for the fabrication of 3D models. Summary of the Invention

[0005] This invention overcomes the shortcomings of the prior art and provides a method and system for generating object meshes based on a three-dimensional model. It generates object meshes based on the three-dimensional model of the object, realizes the structural decomposition and reconstruction optimization of the object in additive manufacturing, improves the production compatibility of additive manufacturing, the stability and smoothness of structural connection and assembly, and the rationality of the reconstruction form setting.

[0006] To achieve the above objectives, the technical solution adopted by this invention is: a method for generating object meshes based on a three-dimensional model, comprising the following steps: Obtain a 3D model of the object; perform a segmentation operation on the 3D model to determine at least one segmentation boundary; based on the segmentation operation, determine the segmentation boundary and divide the 3D model into multiple segmented regions; construct boundary bands between adjacent segmented regions; perform a remeshing operation on the mesh region of the boundary band to reconstruct the segmentation boundary; generate a new mesh of the object based on the reconstructed segmentation boundary; and generate a connection structure between adjacent segmented regions based on the reconstructed segmentation boundary.

[0007] In a preferred embodiment of the present invention, determining the segmentation boundary based on the segmentation operation includes: At least one candidate segmentation boundary is generated based on the three-dimensional model; a segmentation boundary is selected and determined from the candidate segmentation boundaries according to the preset segmentation boundary determination conditions; the segmentation boundary determination conditions include: manufacturing constraints of the segmentation boundary, assembly compatibility of the segmentation boundary, and connection compatibility.

[0008] In a preferred embodiment of the present invention, the creation of the segmented region includes: The surface of the 3D model is divided into multiple segments based on the defined segmentation boundaries, creating segmented regions; wherein, the segmented regions include: a first type of segmented region for forming parts that can be detached from the main body of the object and manufactured independently; and a second type of segmented region for forming parts that are manufactured as a whole in a single 3D printing process.

[0009] In a preferred embodiment of the present invention, obtaining the candidate segmentation boundary includes: Receive a 3D model containing geometric elements that define the shape and topology of the object; Candidate segmentation boundaries are obtained based on preset manufacturing requirements and morphological factors of the 3D model. The manufacturing requirements include manufacturing constraints, such as gravity direction, preferred build orientation for printing, preferred orientation for separating or inserting and assembling separated parts, necessary support for 3D printing or other manufacturing processes, structural stability, and manufacturability. The morphological factors include geometric discontinuities, surface features, curvature variations, appearance transitions, and seed region-based transitions. The candidate segmentation boundaries are represented as independent mesh structures.

[0010] In a preferred embodiment of the present invention, a multidimensional evaluation is performed during the step of determining the segmentation boundary: Manufacturing constraints on the segmentation boundary include: the orientational relationship of the generated connection surface relative to the direction of gravity or the orientation of the structure; the predicted stability of the segmented region after segmentation; and the expected requirements for manufacturing support structures. Assembly and connection compatibility of segmentation boundaries include: analyzing the compatibility of candidate segmentation boundaries with the assembly and reconnection process of several segmented areas in the object after individual manufacturing; the examination of some candidate segmentation boundaries includes: the feasibility of having sufficient material thickness along the boundary, the feasibility of generating connection structures, connectors or interlocking structures along the boundary, and the compatibility of candidate segmentation boundaries with chain, layered or multi-component assembly structures. The performance of candidate segmentation boundaries is characterized by the results of candidate segmentation boundaries after mesh reconstruction or remeshing, including: expected boundary smoothness based on curvature or other relevant geometric factors, robustness of candidate boundaries after remeshing operations, and ability to maintain appearance properties.

[0011] In a preferred embodiment of the present invention, the remeshing operation involves remeshing the mesh region of the segmented boundary surrounded by the boundary band using a mesh finer than the original mesh of the 3D model, including: Remeshing is performed on the surface and inside the boundary zone respectively. When the size of the existing triangle or longer side inside the boundary zone is greater than a threshold, the existing triangle or longer side inside the boundary zone is subdivided. The subdivision method includes: inserting additional vertices along the edge adjacent to the dividing boundary, and reducing or not subdividing the mesh area outside the boundary zone as needed to achieve local remeshing. Within the boundary band after local remeshing, additional vertices are located using a finer mesh than the original mesh of the 3D model, including: aligning vertex positions to the spatial mesh of the finer mesh; and aligning the fine mesh by averaging small geometric perturbations to complete the remeshing operation.

[0012] In a preferred embodiment of the present invention, after the remeshing operation is completed, the segmentation boundary is re-derived based on a finer mesh representation with higher geometric resolution to obtain the reconstructed segmentation boundary. The quality of the reconstructed segmentation boundary is then evaluated to determine whether it meets a preset quality threshold. If it does not meet the threshold, it is either eliminated or further refined. The quality evaluation indicators include: boundary smoothness, curvature continuity, and compatibility with subsequent downstream processing.

[0013] In a preferred embodiment of the present invention, after generating the segmented regions and boundary bands and completing the reconstruction of the segmented boundaries, a portion of the segmented regions is separated from the object, and the separated segmented regions are mechanically connected to generate a manufacturable mesh and connection structure, including the following steps: Based on the determined segmentation regions and boundary bands, as well as the reconstructed segmentation boundary data, the connection surfaces and interface regions of the objects are determined. The acquisition of connection surfaces includes: determining whether two adjacent surfaces need to be connected by a connection structure, and the surfaces connected by the connection structure are taken as connection surfaces. The acquisition of interface regions includes: identifying and evaluating the regions corresponding to the connection surfaces generated by the segmentation process as interface regions. The mesh of the connection structure is generated based on the gap parameters and the type of the connection structure. Connecting structure chains are constructed based on meshes with interconnected structures. The generated mesh and chain of connection structures are verified. Obtain the final mesh of the object's segmented regions, boundary bands, segmented boundaries, and connection structures; The connection structure is used to connect detachable components and fixed components separated by a dividing boundary; The geometry of the connection structure used to connect the two components is generated based on the mesh of the detachable component and the fixed component. The generation methods include: deriving the connection structure geometry from the vertex and triangle data at the dividing boundary between the detachable component and the fixed component; or generating the connection structure geometry by constructing solid geometry or Boolean operations.

[0014] In a preferred embodiment of the present invention, the three-dimensional model of the object includes a combination of geometric elements and appearance attributes, wherein the appearance attributes are used to define the object's visual appearance information and material functional information; And / or, geometric elements include: faces, polygons, vertices, and edges; a vertex is a single point in three-dimensional space defined by X, Y, and Z coordinates, and some vertices constitute the corner points of an object; an edge is a straight line connecting two vertices, and multiple edges constitute the wireframe skeleton of an object; a face is a polygon formed by connecting multiple edges; And / or, some of the segmented regions are detachable components, and some adjacent detachable components are physically connected through a connecting structure; complementary positive and negative feature meshes are created for some connecting structures and adjacent segmented regions to form a mating connection; And / or, before performing the segmentation operation, the 3D model is preprocessed; the preprocessing includes verifying and repairing geometric elements to enforce the manifold topology, close open surfaces, and remove invalid or degenerate elements; And / or, the boundary band is generated according to generation criteria, which include a predetermined geodesic distance from the surface of the 3D model to the boundary, a predetermined number of mesh edge rings or adjacent layers around the boundary, and an adaptive standard based on local curvature, mesh density, and feature size compatible with the nozzle resolution of the 3D printer; the boundary band includes flexible material and buffers to facilitate part separation or coloring.

[0015] In a preferred embodiment of the present invention, an object mesh generation system based on a three-dimensional model includes a computer system, the computer system including one or more interconnected processors, a display and a memory, the memory being used to store computer programs / instructions; A monitor is used to display visual content output by a computer system; A processor is used to execute computer programs / instructions to implement a method for generating object meshes based on a 3D model.

[0016] This invention addresses the deficiencies in the technical background, and the beneficial technical effects of this invention are: A method and system for generating object meshes based on 3D models is disclosed. The method generates object meshes based on the 3D models of objects, realizing the structural decomposition and reconstruction optimization of objects in additive manufacturing. This improves the production compatibility of additive manufacturing, the stability and smoothness of structural connection and assembly, and the rationality of the reconstruction form setting. Attached Figure Description

[0017] The present invention will be further described below with reference to the accompanying drawings and embodiments.

[0018] Figure 1 Figures 1A to 1H illustrate objects in three-dimensional space and corresponding printing methods in a preferred embodiment of the present invention.

[0019] Figure 2 This is a flowchart of a method for generating a mesh for an object based on a three-dimensional model, as described in a preferred embodiment of the present invention.

[0020] Figure 3 This is a flowchart illustrating the determination of segmentation boundaries and segmentation regions in a preferred embodiment of the present invention.

[0021] Figure 4 This is a flowchart of constructing the boundary band and reconstructing the segmentation boundary in a preferred embodiment of the present invention.

[0022] Figure 5 This is a flowchart illustrating the process of determining the connection structure type, generating the connection structure mesh, and completing the final reconstruction of the object mesh in a preferred embodiment of the present invention.

[0023] Figure 6 This is a schematic diagram of an object with a connection structure in a preferred embodiment of the present invention. Figure 1 .

[0024] Figure 7 This is a schematic diagram of an object with a connection structure in a preferred embodiment of the present invention. Figure 2 .

[0025] Figure 8 This is a schematic diagram of an object with a connection structure in a preferred embodiment of the present invention. Figure 3 .

[0026] Figure 9 This is a schematic diagram of an object mesh generation system based on a three-dimensional model, according to a preferred embodiment of the present invention.

[0027] Among them, 10. Object; 101. 3D model; 12. Cuboid I; 14. Cuboid II; 16. Cuboid III; 18. Dividing boundary I; 19. Dividing boundary II; 20. Side region I; 22. Side region II; 24. Hole; 26. Cylindrical structure; 30. Boundary zone I; 32. Boundary zone II; 50. Duck; 51. Head; 52. Body; 53. Duck dividing boundary; 54A. Duck boundary band one; 54B. Duck boundary band two; 55. Protruding tongue; 56. Slot; 57. Interlocking ring one; 58. Interlocking ring two; 200. Computer system; 202. Processor; 204. Display; 206. Memory. Detailed Implementation

[0028] The present invention will now be described in further detail with reference to the accompanying drawings and embodiments. These drawings are simplified schematic diagrams, which are only used to illustrate the basic structure of the present invention and therefore only show the components relevant to the present invention.

[0029] It should be noted that if directional indicators (such as up, down, bottom, top, etc.) are involved in the embodiments of the present invention, these directional indicators are only used to explain the relative positional relationship and movement of the components in a specific posture. If the specific posture changes, the directional indicators will also change accordingly. The terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of indicated technical features. Therefore, features defined with "first" and "second" may explicitly or implicitly include one or more of the aforementioned features. Unless otherwise explicitly specified and limited, the terms "set," "connected," and "linked" should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral connection; they can refer to a direct connection or an indirect connection through an intermediate medium; they can also refer to the internal connection of two components. For those skilled in the art, the specific meaning of the above terms in the present invention can be understood according to the specific circumstances.

[0030] This invention provides a method and system for performing segmentation on a 3D model of an object and generating segmentation boundaries, segmentation regions, boundary bands, and connecting structure meshes. The generated mesh facilitates additive manufacturing of the object, such as 3D printing.

[0031] When using 3D printing technology to manufacture an object (object 10), the conventional 3D model 101 of the object 10 typically requires certain preprocessing before it can be used. In this invention, the terms "3D digital model" and "3D model," as well as "3D digital model" and "3D model 101," have the same meaning and can be used interchangeably. The conventional 3D model 101 is a geometric representation primarily used for visualization, animation, or simulation, where aesthetics and functional accuracy in digital space are crucial. In contrast, the printable 3D model 101 is a geometry that has been converted to fit the physical and mechanical constraints of 3D printing and can be used for manufacturing. Furthermore, the conversion process includes model segmentation, repairing mesh defects (such as holes or non-manifold edges), optimizing structure and orientation to improve strength and material efficiency, ensuring wall thickness meets printer resolution requirements, and preparing support structures for overhanging portions.

[0032] Example 1 A method for generating object meshes based on a 3D model includes the following steps: Obtain the object's 3D model 101; wherein the object's 3D model 101 includes: geometric elements and appearance attributes. Geometric elements define the object's shape and structure; geometric elements include: faces, polygons, vertices, and edges; vertices are single points in 3D space defined by X, Y, and Z coordinates, and some vertices constitute the object's corner points; edges are straight lines connecting two vertices, and multiple edges constitute the object's wireframe skeleton; faces are polygons formed by connecting multiple edges. Appearance attributes are used to define the object's visual appearance information and material functional information.

[0033] The 3D model 101 is segmented to determine at least one segmentation boundary. Before performing the segmentation operation, the 3D model 101 is preprocessed; the preprocessing includes verifying and repairing geometric elements to enforce the manifold topology, close open surfaces, and remove invalid or degenerate elements.

[0034] The segmentation boundary is determined based on the segmentation operation, and the 3D model 101 is cut or divided into multiple segmentation regions based on the segmentation boundary.

[0035] The step of determining the segmentation boundary based on the segmentation operation includes: generating at least one candidate segmentation boundary based on the three-dimensional model 101; selecting and determining the segmentation boundary from the candidate segmentation boundaries according to preset segmentation boundary determination conditions, the segmentation boundary determination conditions including: manufacturing constraints of the segmentation boundary, assembly compatibility of the segmentation boundary, and connection compatibility of the segmentation boundary.

[0036] The creation of segmented regions includes: dividing the surface of the 3D model 101 into multiple segments based on defined segmentation boundaries to create segmented regions; wherein the segmented regions include: a first type of segmented region, used to form parts that can be detached from the object body and manufactured independently; and a second type of segmented region, used to form parts that are manufactured as a whole in a single 3D printing process.

[0037] Boundary bands are constructed between adjacent segmented regions based on the segmentation boundaries, and the boundary bands are configured between adjacent segmented regions.

[0038] A remeshing operation is performed on the grid region of the boundary zone to reconstruct the segmented boundary, thereby reconstructing the grid region of the boundary zone and obtaining the reconstructed segmented boundary.

[0039] A new mesh is generated for the object based on the reconstructed segmentation boundaries; and a connection structure is generated between adjacent segmented regions based on the reconstructed segmentation boundaries. Further, some segmented regions are detachable components, and some adjacent detachable components are physically connected through connection structures; meshes with complementary positive and negative features are created for some connection structures and adjacent segmented regions to form mating connections.

[0040] The specific process includes: Figure 2 A flowchart of process S100 for generating a mesh for an object (not shown) according to an embodiment of the present invention is shown. Process S100 schematically illustrates the steps for generating a mesh from a conventional 3D model 101 for printing the object.

[0041] Step S11: Receive and acquire the 3D model 101 of the object. The 3D model 101 may contain geometric elements such as faces, polygons, vertices, and edges. A vertex is a single point in 3D space defined by X, Y, and Z coordinates; some vertices constitute the corner points of the object. An edge is a straight line connecting two vertices; multiple edges can form the wireframe skeleton of the object. A face is a polygon formed by connecting multiple edges. Triangles and quadrilaterals are common polygon types that constitute the surface of the 3D model 101.

[0042] Step S12: Segment the 3D model 101. Segmenting the 3D model 101 is typically a necessary step in additive manufacturing processes such as 3D printing. As mentioned above, many digital models cannot be reliably manufactured as a single, monolithic structure due to geometric complexity, manufacturing time constraints, material limitations, or equipment capabilities. Furthermore, segmentation enhances appearance control, such as multi-color manufacturing, multi-material combinations, and improved color or texture fidelity, effects that are difficult or impossible to achieve with a conventional 3D model 101.

[0043] Before segmentation, the 3D model 101 can be preprocessed to ensure its geometric validity meets manufacturing requirements. Preprocessing may include geometry verification and repair to enforce manifold topology, closed open surfaces, and removal of invalid or degenerate elements. This ensures that subsequent segmentation and downstream manufacturing operations are performed on a topologically valid and physically realizable model.

[0044] Step S13: Determine candidate segmentation boundaries. As described above, segmentation boundaries can be used to cut or divide the surface area of ​​the 3D model 101 into multiple regions. Therefore, the surface area of ​​the 3D model 101 can be separated into multiple segments or segmented regions through segmentation boundaries. In this step, candidate segmentation boundaries are identified and recorded as potential segmentation boundaries of the 3D model 101.

[0045] Step S14: Select and determine the segmentation boundaries from the candidate segmentation boundaries. After defining the segmentation boundaries, the segmented regions created by these segmentation boundaries are finally determined. Optionally, some segmented regions of the object correspond to parts or components that can be detached from the object and manufactured individually, while other segmented regions of the object correspond to parts or components that are integrated together and can be manufactured as a whole in a single 3D printing process.

[0046] Step S15: Generate boundary bands. As described above, a boundary band is a narrow surface region surrounding a segmentation boundary. As a 3D buffer, the boundary band is configured between adjacent segmented regions. Optionally, two boundary bands can be set at adjacent segmented regions separated by the segmentation boundary, and these two boundary bands each surround the segmentation boundary.

[0047] Step S16: Perform boundary reconstruction. After generating the boundary band, perform a remeshing operation to reconstruct the grid region of the boundary band using a finer grid, and further reconstruct the segmented boundary.

[0048] Step S17: Generate a new mesh for the object. As mentioned earlier, some segmented regions are detachable components, and some adjacent detachable components are physically connected through connection structures. After reconstructing the segmentation boundaries, a mesh for the connection structures is generated using a specific algorithm or other method. Optionally, meshes with complementary positive and negative features (e.g., dovetail grooves, pins / sockets, interlocking teeth, or alignment flanges) can be created for some connection structures and their adjacent regions to form mating connections.

[0049] Example 2 Based on Example 1, a method for generating object meshes based on a 3D model further includes the following steps: Obtaining candidate segmentation boundaries includes: The system receives a 3D model 101 containing geometric elements that define the shape and structure of an object; optionally, the 3D model 101 may also include appearance attributes such as color, texture coordinates, and material identifiers to indicate the materials involved. These appearance attributes provide additional visual information.

[0050] Candidate segmentation boundaries are obtained based on preset manufacturing requirements and morphological factors of 3D model 101. Manufacturing requirements include manufacturing constraints, which include gravity direction, preferred build orientation for printing, preferred orientation for separating or inserting and assembling separated parts, necessary support for 3D printing or other manufacturing processes, structural stability, and manufacturability, among one or more of these. Morphological factors include geometric discontinuities, surface features, curvature variations, appearance transitions, and seed region-based transitions. The candidate segmentation boundaries are represented as independent mesh structures.

[0051] Specifically, in the step of determining the segmentation boundaries: Manufacturing constraints on the segmentation boundary include: the orientational relationship of the generated connection surface relative to the direction of gravity or the orientation of the structure; the predicted stability of the segmented region after segmentation; and the expected requirements for manufacturing support structures. Assembly and connection compatibility of segmentation boundaries include: analyzing the compatibility of candidate segmentation boundaries with the assembly and reconnection process of several segmented areas in the object after individual manufacturing; the examination of some candidate segmentation boundaries includes: the feasibility of having sufficient material thickness along the boundary, the feasibility of generating connection structures, connectors or interlocking structures along the boundary, and the compatibility of candidate segmentation boundaries with chain, layered or multi-component assembly structures. The performance of candidate segmentation boundaries is characterized by the results of candidate segmentation boundaries after mesh reconstruction or remeshing, including: expected boundary smoothness based on curvature or other relevant geometric factors, robustness of candidate boundaries after remeshing operations, and ability to maintain appearance properties.

[0052] The specific process includes: Figure 3 A flowchart of process S200 for determining segmentation boundaries and generating segmented regions according to an embodiment of the present invention is shown. Process S200 describes the process after obtaining the three-dimensional model 101, except... Figure 2 In addition to the steps shown in process S100, there are additional operational steps for printing the object. In step S21, a three-dimensional model 101 containing geometric elements (e.g., triangles, vertices, edges, and surface normals) is received. The geometric elements define the shape and structure of the object. Optionally, the three-dimensional model 101 may also include appearance attributes, such as color, texture coordinates, and material identifiers to indicate the materials involved. Appearance attributes provide additional visual and functional information.

[0053] Step S22: Manufacturing requirements can be obtained from, for example, the 3D model 101 or printing instructions, and key manufacturing constraints required for manufacturing are analyzed and determined. Optionally, manufacturing constraints may include the gravity direction or the preferred build orientation for printing; the gravity direction refers to the direction of gravity of object 10 relative to itself, and the build orientation refers to the printing direction of object 10. It should be noted that all segmented regions share the same gravity direction. In some cases, the gravity direction is also the build orientation of object 10; in other cases, the gravity direction is different from the build orientation of object 10. Optionally, manufacturing constraints may also include preferred orientations for separating, inserting, and assembling separated parts. In addition, manufacturing constraints may also include considerations for necessary support, structural stability, and manufacturability required for 3D printing or other manufacturing processes.

[0054] Step S23: Determine candidate segmentation boundaries. Candidate segmentation boundaries are the segmentation boundaries of potential segmented regions in the 3D model 101. To ensure sufficient coverage, candidate boundaries can be determined from various aspects and factors. These aspects and factors include: geometric discontinuities, surface features or curvature variations; appearance transitions, such as color variations, texture seams, or material boundaries; or transitions in 3D space from the original region to a seed-based region configured by the user, robotic system, or artificial intelligence (AI) agent. Furthermore, candidate boundaries can also be defined by projecting the 3D model 101 onto a 2D view and applying image-based segmentation techniques. In this invention, "AI agent" refers to a software program capable of autonomously interacting with its environment, taking actions, and performing tasks using acquired data. In some cases, 3D spatial computation algorithms can be used for segmentation. At this stage, each candidate segmentation boundary is considered as a preliminary segmentation hypothesis to be evaluated, rather than a final segmentation decision, thus providing flexibility and optimization for subsequent processing.

[0055] Step S24: Represent the candidate segmentation boundaries using a mesh. A boundary-centric approach represents each candidate segmentation boundary as an independent mesh structure, independent of the segmentation region. Optionally, candidate segmentation boundaries can be associated with and determined by specific factors, such as different adjacent segmentation regions on either side of the boundary, different local surface orientations or different normal directions, and different characteristics in terms of local material thickness, structural strength, or geometric support properties. Furthermore, regions can be segmented based on the definition of the segmentation boundaries; the determination of segmentation regions can be accomplished automatically by an algorithm, manually, or through a combination of the first two methods.

[0056] Step S25: Evaluate the manufacturing constraints of the candidate segmentation boundaries. This evaluation is performed based on a series of manufacturing-oriented criteria. Exemplary factors considered include: the orientational relationship of the generated connection surface relative to the direction of gravity or the build orientation; the predicted stability of certain components (or segmented regions) after segmentation; the anticipated requirements for manufacturing support structures; and the availability and convenience of certain components for subsequent assembly, post-processing, or other product manufacturing. Optionally, the connection surface of the connection structure may have a planar area compatible with the 3D printer build platform, allowing for manufacturing without a support structure; and the connection surface may also serve as a hidden interface between the segmented components after assembly to maintain the aesthetics of the visible outer surface. Based on the above analysis, the manufacturing score for each candidate segmentation boundary is calculated and determined.

[0057] Step S26: Evaluate the assembly and connection compatibility of the candidate segmentation boundaries. This evaluation aims to analyze the compatibility of the candidate segmentation boundaries with the assembly and reconnection processes of several components in the object after individual manufacturing. Components need to be assembled to form the object, with some components assembled using adhesive materials and others connected through insertion holes or slots. The examination of some candidate segmentation boundaries may include: the feasibility of having sufficient material thickness along the boundary (e.g., whether the local material thickness along the boundary exceeds a predetermined minimum threshold), the feasibility of generating connection structures, connectors, or interlocking structures along the boundary, and the compatibility of the boundary with chain-like, layered, or multi-component assembly structures. In some cases, if certain candidate segmentation boundaries are too narrow to be manufactured by the 3D printer nozzle, cannot meet certain manufacturing constraints, or cannot support sub-components or connections requiring support, these candidate segmentation boundaries may be assigned a lower score or eliminated, thereby improving the manufacturing process and enhancing the structural integrity and assembly convenience of the final product.

[0058] Step S27 evaluates the performance of the candidate segmentation boundaries, such as their performance after local mesh reconstruction (or remeshing). The analysis includes: expected boundary smoothness based on curvature or other relevant geometric factors, robustness of the candidate boundaries after remeshing, and the ability to maintain appearance attributes (e.g., color and texture continuity). Candidate segmentation boundaries that maintain or improve certain attributes or quality (e.g., becoming smoother, more stable, or having stronger consistency in appearance attributes after remeshing) will be assigned higher scores.

[0059] Step S28: Determine the segmentation boundary. Based on the comprehensive results of the above evaluation and analysis (e.g., the overall score), select and determine the segmentation boundary from the candidate segmentation boundaries. The selection can comprehensively consider manufacturing constraint adaptability, assembly and connection structure compatibility (e.g., mortise and tenon structures), and appearance consistency (e.g., color and texture similarity). Optimization strategies can be adopted, such as manufacturing detachable parts with the fewest connection surfaces.

[0060] Step S29: Finally, the segmented regions of the object are determined. Optionally, after determining the segmentation boundaries, segmented regions can be defined based on the segmentation boundaries, thereby finally determining and obtaining the segmented regions of the object. The configuration of the segmented regions is also applicable to subsequent cutting, reconstruction, assembly, and other manufacturing operations. In addition, region labels can be assigned to the segmented regions, and some segmentation boundaries can be optimized according to manufacturing constraints. The setting of segmented regions is not only used for visualization but also for subsequent geometric processing.

[0061] Example 3 Based on Example 1 or Example 2, a remeshing operation of an object mesh generation method based on a 3D model involves remeshing the mesh region of the segmented boundary surrounded by the boundary band using a finer mesh than the original mesh of the 3D model 101, including the following steps: Remeshing is performed on the surface and inside the boundary zone respectively. When the size of the existing triangle or longer side inside the boundary zone is greater than the threshold, the existing triangle or longer side inside the boundary zone is subdivided. The subdivision method includes: inserting additional vertices along the edge adjacent to the dividing boundary, and reducing or not subdividing the mesh area outside the boundary zone as needed to achieve local remeshing. Within the boundary band after local remeshing, additional vertices are located using a finer mesh than the original mesh of 3D model 101, including: aligning vertex positions to the spatial mesh of the fine mesh; and aligning the fine mesh by averaging small geometric perturbations to complete the remeshing operation.

[0062] After completing the remeshing operation, the segmentation boundary is re-derived based on the finer mesh representation with higher geometric resolution to obtain the reconstructed segmentation boundary. The reconstructed segmentation boundary is then evaluated to determine whether it meets the preset quality threshold. If it does not meet the threshold, it is discarded or further refined. The quality evaluation indicators include: boundary smoothness, curvature continuity, and compatibility with subsequent downstream processing.

[0063] The specific process includes: Figure 4 A flowchart of process S300 for creating a boundary band and reconstructing a segmented boundary according to an embodiment of the present invention is shown. Except Figure 2 and Figure 3In addition to the steps shown in processes S100 and S200, process S300 provides additional operational steps for printing the object. As previously described, the boundary band is a narrow 3D geometric band surrounding the segmentation boundary. In step S31, a boundary band corresponding to the segmentation boundary of the object is generated. The boundary band may be generated according to one or more specific criteria, such as a predetermined geodesic distance from the surface of the 3D model 101 to the boundary, a predetermined number of mesh edge loops or adjacent layers around the boundary, or an adaptive criterion based on local curvature, mesh density, or feature size compatible with the nozzle resolution of the 3D printer. In some embodiments, the boundary band may include a soft or flexible material and a buffer to facilitate operations such as part separation or coloring. Optionally, the geometry (or mesh area) outside the boundary band may remain unchanged to maintain the original properties of the object 10.

[0064] In step S32, remeshing is performed on the boundary band. After the boundary band is generated, remeshing operations can be performed on the surfaces on the boundary band and inside the boundary band, respectively. Optionally, the remeshing operation may include: subdividing existing triangles or longer sides within the boundary band (e.g., when the size of an existing triangle or side is greater than a threshold); inserting additional vertices along certain sides adjacent to the subdivision boundary (e.g., when the distance between adjacent vertices is greater than a threshold); and performing local retriangulation to improve the quality, uniformity, and suitability for manufacturing of triangles (e.g., when local triangles are not optimized). In some embodiments, the mesh region outside the boundary band is not remeshed or only minimally remeshed, which helps to reduce the computational overhead of processing the entire model while preserving the original overall shape and design intent. In other embodiments, a certain degree of flexible remeshing can be performed on the mesh region outside the boundary band, for example, to improve the overall shape and optimize the overall appearance.

[0065] In step S33, additional vertices are added within the boundary band with a finer mesh density. Optionally, the additional vertices added within the boundary band in step S32 can be located using a mesh finer than the original mesh of the 3D model 101. The corresponding operations may include aligning the vertex positions to a finer spatial mesh than the original mesh and applying manufacturing-appropriate constraints (e.g., minimum edge length, minimum angle, or minimum aspect ratio). This fine mesh alignment process reduces boundary noise or irregularities by averaging small geometric perturbations and improves the numerical stability of the boundary geometry, thereby improving boundary smoothness and reducing the workload of global smoothing.

[0066] In step S34, the segmentation boundary is reconstructed and key boundary points are identified. After the remeshing operation is completed, the segmentation boundary is re-derived based on a finer mesh representation with higher geometric resolution. In some embodiments, reconstruction can be accomplished by tracing the refined edge chains or ordered vertex sequences corresponding to the original segmentation boundary. The reconstructed segmentation boundary is constrained to form a closed profile and maintains topological consistency with the underlying mesh, thereby ensuring the effectiveness of subsequent segmentation, extraction, or assembly operations.

[0067] Furthermore, multiple key points are identified on the segmentation boundary. In some embodiments, key points can be determined using one or more boundary analysis techniques, including uniform sampling along the boundary length, detecting geometric features such as curvature extrema or turning angles, and detecting appearance-based features such as color transitions or texture changes. Optionally, one or more key points can also be manually specified by the user or automatically selected through algorithmic processes or artificial intelligence-based tools.

[0068] Optionally, a stable boundary curve can be generated using numerical fitting methods, such as spline-based interpolation methods (e.g., B-splines or related curve representations), or constraint smoothing following the boundary topology. In some cases, the operations shown in steps S33 and S34 can be repeated until the boundary quality meets manufacturing and assembly requirements. Repeated operations may include: setting vertices in a finer mesh, aligning vertex positions to the finer mesh, performing boundary reconstruction, and boundary smoothing. The reconstructed boundary has a smoother geometry, more uniform vertex spacing, and is more suitable for subsequent operations such as cutting, stretching, or generating connection structures.

[0069] Step S35: Apply some appearance attributes to the boundary band. In some embodiments, appearance attributes include propagating appearance attributes such as vertex color, translucency, texture coordinates, and material identifiers. In some cases, increasing vertex density within the boundary band can improve cross-boundary color interpolation accuracy and texture continuity after cutting, splitting, or assembling operations. Optionally, the appearance attributes of the geometry outside the boundary band can remain unchanged to maintain the visual effect of the original design.

[0070] Step S36: Evaluate the reconstructed segmentation boundary. In some embodiments, the reconstructed segmentation boundary may be evaluated based on one or more quality metrics. These quality metrics may include boundary smoothness, curvature continuity, and specific compatibility with subsequent downstream processing (e.g., connection structure generation or gap generation). If the segmentation boundary does not meet a preset quality threshold, further local refinement processing may be performed within that boundary region. In some cases, if the segmentation boundary does not meet the relevant quality threshold, it may be removed in subsequent processing stages.

[0071] Example 4 Based on any of the embodiments 1 to 3, a method for generating object meshes based on a 3D model further includes: After generating the segmented regions and boundary bands and reconstructing the segmented boundaries, some of the segmented regions are separated from the object, and mechanical connections are made to the separated regions to generate a manufacturable mesh and connection structure, including the following steps: Based on the determined segmentation regions and boundary bands, as well as the reconstructed segmentation boundary data, the connection surfaces and interface regions of the objects are determined. The acquisition of connection surfaces includes: determining whether two adjacent surfaces need to be connected by a connection structure, and the surfaces connected by the connection structure are taken as connection surfaces. The acquisition of interface regions includes: identifying and evaluating the regions corresponding to the connection surfaces generated by the segmentation process as interface regions. The mesh of the connection structure is generated based on the gap parameters and the type of the connection structure. Connecting structure chains are constructed based on meshes with interconnected structures. The generated mesh and chain of connection structures are verified. Obtain the final mesh of the object's segmented regions, boundary bands, segmented boundaries, and connection structures; The connection structure is used to connect detachable components and fixed components separated by a dividing boundary; The geometry of the connection structure used to connect the two components is generated based on the mesh of the detachable component and the fixed component. The generation methods include: deriving the connection structure geometry from the vertex and triangle data at the dividing boundary between the detachable component and the fixed component; or generating the connection structure geometry by constructing solid geometry or Boolean operations.

[0072] The specific workflow includes: Figure 5 A flowchart of process S400 for determining the connection structure type, generating the connection structure mesh, and generating the segmented region mesh according to an embodiment of the present invention is shown. Except Figure 2 , Figure 3 and Figure 4 In addition to processes S100, S200, and S300 shown, process S400 provides further operational steps for printing the object. Optionally, processes S100 to S400 can combine the object to be printed in various ways. For example, the steps and operations in processes S100 to S400 can be combined in different ways to print the object.

[0073] After generating the segmented regions and boundary bands and reconstructing the segmented boundaries, some components (i.e., some segmented regions) can be separated from the object. After processing the separated components accordingly, they can be repositioned onto the object and mechanically connected. Manufacturable meshes and connection structures can be generated while meeting certain assembly clearance, assembly direction, and structural stability requirements, without needing to achieve precise geometric consistency with the 3D model 101.

[0074] In step S41, data on the segmented region and reconstructed segment boundaries are acquired. In addition to reconstructing the segment boundaries, this data also includes remeshable regions within and near the boundary bands, as well as appearance attributes such as color or texture. Optionally, some assembly and manufacturing constraint information may also be acquired, including target manufacturing processes, relevant material properties or dimensional tolerances, expected assembly orientation, and the required connection durability for the connection structure.

[0075] In step S42, the connection surfaces and interface regions of the object are determined. First, it is determined whether two adjacent segmented regions need to be connected by a connection structure. The connection structure forms a connection surface (or mating surface), and in some cases, it also forms an interface. Then, the regions corresponding to the connection surfaces and interfaces generated by the segmentation process are identified and evaluated. For each connection surface or interface, evaluation metrics may include analyzing its surface orientation relative to the direction of gravity, build orientation, or assembly direction, for example, to reduce the required support structure or improve assembly stability. Other evaluation metrics may include local material thickness, structural strength, load-bearing capacity, and geometric features such as boundary smoothness, curvature variation, and shape complexity. In some embodiments, the connection surface is considered not only as a separation surface but also as a functional interface capable of supporting mechanical connection structures, connectors, or alignment features for subsequent assembly.

[0076] In step S43, gap parameters are determined. In some embodiments, the gap can be defined as a geometric parameter generated by connecting the structure and the interface. Optionally, the gap parameters may include a uniform offset distance, a direction-dependent gap (e.g., using different gap values ​​in different directions), and an adaptive gap set differently based on a specific local geometry, a specific curvature, or a specific material property. Optionally, the gap parameters may be applied during the mesh generation process of the connecting features, rather than just for adjustments in the post-processing stage.

[0077] In step S44, the connection structure type is determined. For example, after determining the connection structure used to connect two adjacent segmented regions, its type can be selected. Common connection structure types may include standard geometric connection structures (e.g., circular or square mortise and tenon matting). Tenons and mortises can form a strong interlocking connection, where the tenon (i.e., the protruding tongue) and mortise (i.e., the corresponding hole, slot, or opening to provide the interlocking connection) fit tightly together. Connection structure types may also include complex shape-adaptive connection structures, which can be derived from the local geometry of the interface region, including irregular or non-uniform boundary shapes. Such shape-adaptive connection structures can fit complex interface contours and can be generated by analyzing boundary features, curvature variations, or topological properties of the connection surfaces. The connection structure type can be determined based on a variety of factors, including the shape, size, and area of ​​the interface, the required mechanical strength, assembly or disassembly direction constraints, and compatibility with chain assemblies or multi-part assemblies. For example, the connection geometry can be derived or created using inscribed regions within the connection surface, boundary offsets, or skeletal representations, or by utilizing and matching other geometric features.

[0078] In step S45, a mesh of the connection structure is generated. As previously mentioned, some connection structures are mortise and tenon type, which, or other similar connection structures, include complementary convex (tenon) and concave (mortise) components, and set a clearance-controlled offset between mating surfaces to ensure proper fit. The generated connection structure is represented by a mesh of connection areas and is configured to conform to boundary geometry, avoid self-intersection, and be compatible with manufacturing processes. Furthermore, the clearance remains consistent across the depth, width, and contact surfaces of the connection. For example, one or more connection parameters (e.g., wall thickness or insertion depth) can be determined based on inscribed geometric primitives (e.g., inscribed circles or inscribed polygons) on the connection surfaces to ensure the geometric validity, structural integrity, and manufacturability of the generated connection.

[0079] In step S46, a chain of connecting structures is constructed. This chain provides support for multiple segmented regions (e.g., fixed or movable segments). The chain can be used to form complex assemblies, define assembly sequences, or enable modular replacement or customization of individual components. Complex assemblies may contain parts connected by one or more connecting structures. A single connecting structure can connect two, three, or even more parts. For example, a single connection interface can connect three or more components simultaneously. Furthermore, the interactions between the chained connecting structures can be evaluated to prevent geometric interference and avoid creating overly constrained assemblies, thus preventing difficulties in manufacturing operations.

[0080] In step S47, the generated mesh and connection structure are verified. In some embodiments, the verification of the generated geometry and connection structure may include: evaluating and analyzing assembly feasibility, gap adequacy, post-assembly structural robustness, and compatibility with downstream processes such as slicing or printing path generation. If the mesh of the segmented region fails verification, one or more adjustment operations may be performed. Optionally, the adjustment operations may include gap adjustment, connection structure type resetting, and segmentation boundary reconstruction or remeshing of the segmented region.

[0081] In step S48, the final mesh of the object's segmented regions, boundary bands, segmentation boundaries, and connection structures is determined. The final mesh data corresponds to the clearance-controlled 3D geometry and contains manufacturing-ready data for printing the object. The final data also includes definitions and parameters of the connection structures and assembly metadata (e.g., assembly sequence instructions and part labels). The final mesh data can be used by technicians, robotic systems, and AI agents to perform automated or semi-automated assembly, manufacturing, or post-processing operations.

[0082] Connector structures are typically used to link detachable and fixed components separated by a dividing boundary. In some embodiments, the geometry of the connector structure connecting the two components can be generated based on the meshes of those two components. For example, a first method of connector generation derives the connector geometry from the vertices and triangles at the dividing boundary between the two components; alternatively, a second method generates the connector geometry by constructing solid geometry (CSG) or Boolean operations. Boolean operations are techniques for constructing complex shapes by performing union, difference, or intersection operations on two or more 3D objects.

[0083] The first method for generating the connection structure is as follows: The dividing boundary is a closed line (or curve) defined by an ordered sequence of mesh vertex indices. The closed line divides the original mesh into internal and external regions. Thus, the original mesh is divided into multiple mesh components, each corresponding to one side of the closed line. The closed line is shared by mutually independent mesh components.

[0084] After partitioning, vertex reindexing is performed on each mesh component. Only triangles belonging to that mesh component or vertices referenced by the partition boundaries are retained, and a compact vertex array is constructed by remapping the original vertex indices. The reindexing operation aims to make each mesh component self-consistent and independent, and to prevent vertex duplication. In some embodiments, the reindexing operation may be implemented using a table-based index remapping, thereby achieving linear time execution and adapting to constrained execution environments.

[0085] For the segmentation boundary, an inward-pointing direction vector is calculated at the boundary vertex, for example, based on the tangent-weighted surface normal derived from adjacent mesh triangles. The inward-pointing direction is used to define the semantic interior of the connected region and is used uniformly during the construction of the connected structure, including gap adjustment, wall thickness generation, and extrapolation direction (or stretching direction).

[0086] To construct the geometry of the connection structure, the boundaries can be projected onto one or more planes to form a bottom loop. Planar normals can be calculated based on the normals at the vertices of the boundary lines (e.g., using Newell-style normals), thus providing stable extrapolation directions even if the boundary lines lie on curved or irregular surfaces. For each boundary line vertex, it is projected along the plane normal direction with a projection depth determined by the connection parameters, generating the corresponding bottom loop vertex. The resulting bottom loop forms a closed loop, defining the end faces of the connection structure geometry.

[0087] The sidewalls of the connecting structure are constructed by generating triangles between corresponding edges of the top boundary and the bottom ring. For each pair of adjacent boundary line vertices, two triangles are generated to form quadrilateral sidewalls. The vertex order of the triangles is set according to whether the connecting structure geometry is an outward-convex or inward-containing structure, ensuring that the outward normal direction is consistent with the overall mesh orientation.

[0088] The extrapolated structure is closed by generating a bottom end face. Optionally, the bottom ring used to define the boundary of the bottom end face can be reshaped into a regular geometry, such as a circle or polygon. The reshaped bottom ring can be triangulated, for example, by fan-shaped triangulation centered on the centroid, or by subdividing the polygonal region to provide structural support. Alternatively, the ear-clipping algorithm can be directly applied to the vertex index of the bottom ring to triangulate it. The winding order of the bottom end face triangles will be adjusted to maintain the correct outward orientation normal relative to the surrounding mesh.

[0089] A gap between complementary connecting parts can be introduced by inwardly contracting one or more boundary profiles before extrapolation. Furthermore, when the segmentation boundary is sufficiently flat, an inner profile can be generated by inwardly offsetting the original profile, and sidewalls can be constructed between the outer and inner profiles to form a strip structure with a certain wall thickness. In some embodiments, this strip structure can provide mechanical interlocking and controllable assembly clearance. For example, the strip structure can be flat or have a mushroom-shaped connection profile. The above operations modify only the local connection geometry and do not require Boolean operations such as volume difference.

[0090] Through the above operations, the geometry of the connecting structure can be constructed by vertex generation, triangle stitching, and contour or boundary line projection, generating grooved or protruding connecting components without the need for Boolean union or difference. The resulting connecting structure mesh is manifold in construction and maintains a direct geometric correspondence with the dividing boundaries.

[0091] Example 5 Based on any of the embodiments in Examples 1 to 4 Figure 1 Figures 1A to 1H illustrate an object 10 in three-dimensional space and a three-dimensional printing method disclosed in this invention.

[0092] like Figure 1 1A in the diagram is a perspective view of object 10, with its gravitational force directed opposite to the Z-direction. Object 10 comprises three cuboids: cuboid 12, cuboid 24, and cuboid 36. Cuboid 12 is the main body of object 10. Cuboid 24 is positioned on cuboid 12 and protrudes from its top surface along the Z-direction. Cuboid 36 is positioned on the side surface of cuboid 12 and protrudes from its side along the X-direction. Figure 1 The image of 1A can be generated from the 3D model 101 of object 10. In summary, this digital model is typically used to visually represent object 10, but is not suitable for direct 3D printing.

[0093] Figure 1 Figure 1B schematically illustrates the representation of the outer surface of object 10 in three-dimensional space. 3D printing primarily focuses on constructing the outer surface areas, while the internal space is typically only a secondary part of the printing process, used to fill the printing material. For example, the internal space of object 10 can be completely or partially filled. Constructing object 10 requires a mesh of its outer surface, which can be derived from the three-dimensional model 101. Except for some structurally simple objects, many objects cannot be printed in a single step. Therefore, these objects can be divided into several parts, which can be fabricated separately and then assembled together to form the final object.

[0094] Figure 1The 3D printing process in this invention typically only allows for a limited degree of unsupported bridging and overhanging structures. A bridging structure spans the gap between two supported regions, while an overhanging structure is supported only at one end. Assume that object 10 is not suitable for monolithic 3D printing, specifically because cuboid 16 is an overhanging structure and lacks sufficient support during printing. Also assume that cuboid 14 is similarly unsuitable for monolithic 3D printing (e.g., due to specific material or structural requirements). Therefore, cuboids 12, 14, and 16 can be printed separately. Before dividing object 10 into several cuboids, the dividing boundaries need to be determined. In this invention, the dividing boundary refers to the closed surface boundary line that divides the surface area of ​​object 10 into two regions, or the mesh of the surface area of ​​object 10 into two meshes. Figure 1 As shown in 1C, two dividing boundaries, 18 and 19, are generated for cuboid 2 14 and cuboid 3 16, respectively. When cuboid 2 14 and cuboid 3 16 are considered as surface regions covering their interiors, dividing boundaries 18 and 19 separate cuboid 2 14 and cuboid 3 16 from cuboid 1 12.

[0095] Take a cuboid 3.16 as an example. Figure 1 As shown in Figure 1D, the separation of cuboid 3 16 from cuboid 12 indicates that cuboid 3 16 can be manufactured independently. After cuboid 3 16 is manufactured, it can be reattached to cuboid 12 to form object 10. There are several options for reattaching cuboid 3 16, and these options determine the structure of cuboid 3 16.

[0096] like Figure 1 Figure 1E schematically illustrates a method for constructing object 10. The structure of cuboid 3 16 can be formed by cutting along the side surface of cuboid 12 at the location of cuboid 3 16. The cut will form or expose a segmented side region 1 20 on the side of cuboid 12, and form side region 2 22 of cuboid 3 16. Optionally, side region 2 22 and side region 1 20 can form a connecting structure. For example, side region 2 22 (or the 3D surface area of ​​side region 2 22) is a connector (or plug), and side region 20 (or the 3D surface area of ​​side region 20) is the slot portion of the connector. After cuboid 3 16 is made, it can be reattached to cuboid 12 by aligning side region 1 20 with side region 2 22 and using adhesive material.

[0097] Alternatively, such as Figure 1 Figure 1F schematically illustrates another method for constructing object 10. In some cases, the following approach is used... Figure 1The object 10 assembled by the method shown in E may not meet certain requirements. For example, the adhesive strength between cuboid 12 and cuboid 3 16 may be insufficient, or cuboid 3 16 may need to be able to rotate along the X-axis. Figure 1 As shown in Figure 1F, a cylindrical structure 26 is added to the side 28 of cuboid 3 16 as a rotary joint structure, and a matching hole 24 is provided on the side of cuboid 12 facing cuboid 3 16 to accommodate the cylindrical structure 26. The cylindrical structure 26 and the hole 24 serve as a joint plug and a slot, respectively, forming a matching connection structure. Cuboid 3 16 and the cylindrical structure 26 can be used as a single component and printed as a whole in one printing process. After printing, the cylindrical structure 26 can be inserted into the hole 24, allowing cuboid 3 16 to connect to cuboid 12, thus enabling cuboid 3 16 to rotate while connected.

[0098] Dividing boundary 18 and dividing boundary 2 19 divide object 10 into three parts. Furthermore, dividing boundary 18 and dividing boundary 2 19 also divide object 10 into three segmented mesh regions. In this invention, the terms "segmented mesh region" and "segmented region" have the same meaning and can be used interchangeably. After exporting or generating the mesh of the segmented region, the mesh of that segmented region can be used for printing.

[0099] Optionally, the mating surface area between two adjacent segmented regions may be referred to as the connecting surface. The connection between the adjacent segmented regions can be fixed (e.g., fixed by adhesive material) or movable (e.g., allowing relative movement, such as rotation). Figure 1 In 1E, the connecting surface includes side region 1 20 and side region 2 22, and is defined by dividing boundary 2 19. In such a case... Figure 1 In 1F, the connecting surfaces include the side and bottom regions of the hole 24, the outer surface of the cylindrical structure 26, and the side region 28. These connecting surfaces are not defined by the dividing boundary 19.

[0100] like Figure 1 Figures 1G and 1H show side views of the cuboid 16 and cylindrical structure 26 in the Y and X directions, respectively. Figure 1 In 1H, the edge or outer contour of side 28 corresponds to the position of the dividing boundary 2 19 on cuboid 3 16. Optionally, boundary band 1 30 and boundary band 2 32 can be created (e.g., Figure 1 Middle 1F and Figure 1As shown in Figure 1G, it is a narrow 3D surface region adjacent to and surrounding the second dividing boundary 19. For example, boundary band 30 can have a rectangular ring shape surrounding the entrance of hole 24. The inner edge of the rectangular ring corresponds to the position of the second dividing boundary 19 on cuboid 12. The meshes of boundary band 30 and boundary band 32 can be specially derived for structural enhancement, color continuity improvement, texture continuity improvement, etc.

[0101] Example 6 The printable 3D model 101 structure is obtained by using the object mesh generation method based on the 3D model in any of the embodiments 1 to 3.

[0102] Figure 6 A duck 50 with a connecting structure is shown. (Example) Figure 6 As shown, the duck 50 includes a head 51 and a body 52. ​​The geometry of the duck 50 can be divided by a duck dividing boundary 53 into a head segmentation region of the head 51 and a body segmentation region of the body 52. ​​Boundary bands 54A and 54B are respectively set to surround the duck dividing boundary 53. The duck dividing boundary 53, as well as duck boundary band 54A and duck boundary band 54B, cover the neck region of the duck 50.

[0103] Figure 7 A duck 50 with connection structure two is shown. (Example) Figure 7 As shown, the duck 50 includes a head 51 and a body 52, and the head 51 can be manufactured separately and assembled together. For example, the head 51 and the body 52 can be connected by a connection structure including a protruding tongue 55 and a slot portion 56 (or hole). The duck 50 is assembled after the protruding tongue 55 is inserted into the slot portion 56.

[0104] Alternatively, in other embodiments, the protruding tongue 55 may be adhered to the slot portion 56 to fix the head 51. In some cases, the protruding tongue 55 may be movable within the slot portion 56, thereby allowing the head 51 to rotate relative to the body 52.

[0105] Figure 8 A duck 50 with a connecting structure three is shown. (Example) Figure 8 As shown, the duck 50 includes a head 51 and a body 52. ​​The head 51 and the body 52 are connected by a connecting structure three, which is more complex than connecting structure one or connecting structure two. The connecting structure three of the duck 50 includes interlocking ring one 57 and interlocking ring two 58, thereby allowing the head 51 to move.

[0106] In summary, the first method for generating connection structures involves deriving the connection structure geometry from the vertex and triangle data at the dividing boundary between two components. Alternatively, the second method involves generating the connection structure geometry through constructing solid geometry (CSG) or Boolean operations. Boolean operations are techniques for constructing complex shapes by performing union, difference, or intersection operations on two or more 3D objects. The first method focuses on boundary-based connection structure construction without requiring Boolean operations, while the second method employs Boolean-based connection structure construction. The second method is suitable when the connection structure is complex, independent of the dividing boundary, or difficult to represent well through extrapolation generation aligned along the boundary.

[0107] Specifically, some relatively simple connection structure types (e.g.) Figure 1 1E and Figure 7 The connection structure shown can be constructed using the first connection structure generation method. Connection structure types with complex designs can also be constructed using the first connection structure generation method, including ball-and-socket connections and rotational connections (e.g.,...). Figure 1 The connection structure shown in 1F), multi-axis hinges, connection structures with undercut or concave internal features, and Figure 8 The interlocking ring structure is shown. Complex designs involve volumetric geometries that may not extend along segmented boundaries, whose connecting surfaces are not defined by segmented boundaries, or have multiple extrapolated extension directions. In these cases, Boolean operations can assist in the construction of the connection structure.

[0108] In some embodiments, certain connection structure types can be pre-defined to be constructed using a first connection structure generation method, and other connection structure types can be constructed using a second connection structure generation method. For example, a lookup table can be set up to list the connection structure types and their corresponding first or second connection structure generation methods. The second connection structure generation method uses Boolean operations, which may require higher computing power and a larger data volume; while the first connection structure generation method has relatively lower cost and resource requirements. Therefore, the first connection structure generation method is usually preferred when feasible. The second connection structure generation method is described later.

[0109] As described above, the second connection structure generation method is based on Boolean operations. When using this method, the connection structure of an object can be generated independently without relying on the object's mesh. This connection structure can have a solid structure and can be defined according to preset schemes, models, and rules. For example, a revolved connection structure can have a spherical or cylindrical structure. The revolved connection structure can be determined based on the mesh of adjacent segmented regions and information derived from the corresponding segmentation boundaries, profile centroids, average surface normals, and predefined connection axes.

[0110] After determining the location and orientation of the connection structure, the mesh of the connection structure is combined with the mesh of the object using Boolean union, difference, or intersection operations. Boolean operations are used to create protruding connection members in the connection structure and inwardly recessing receiving grooves on the object, respectively. A clearance tolerance can be set between the protruding connection members and the receiving grooves; for example, the connection members can be offset before performing the Boolean operation to achieve the clearance tolerance.

[0111] Optionally, additional conditions can be set to determine whether to use the first connection structure generation method or the second connection structure generation method to construct the connection structure. For example, if any of the following conditions are detected or determined to be met, the first connection structure generation method can be used to construct the connection structure without Boolean operations: the mesh of the connection structure can be derived based on the split boundary; all connection faces of the connection structure are defined by the split boundary; all outer connection faces of the connection structure are completely defined by the split boundary; the connection structure has only a single connection face, and the corresponding split boundary is the boundary of that single connection face; the connection structure has multiple connection faces, and the corresponding split boundary is the boundary of one of the multiple connection faces; the connection structure does not have any connection faces not defined by the split boundary. Furthermore, if any of the following conditions are detected or determined to exist, the second connection structure generation method can be used to construct the connection structure and Boolean operations can be used: the mesh of the connection structure cannot be derived based on the split boundary; none of the connection faces of the connection structure are defined by the split boundary; the connection structure does not have any outer connection faces completely defined by the split boundary; the connection structure has only a single connection face, and the corresponding split boundary is not the boundary of that single connection face; the connection structure has multiple connection faces, and the corresponding split boundary is not the boundary of any of the multiple connection faces. Furthermore, if the connection structure has an internal solid structure, components that can move in multiple directions, or components that are not physically connected to the dividing boundary, Boolean operations may be required, and the second connection structure generation method can be used to construct the connection structure. Additionally, both the first and second connection structure generation methods can be used simultaneously to construct different connection structures for an object.

[0112] As described above, when the segmentation boundary can serve as a reliable geometric reference for defining connection surfaces, dimensions, and gaps, the first connection structure generation method is preferred. In such cases, complementary connection (or joint) features can be formed between the segmented components by parameterizing, offsetting, extrapolating, or otherwise transforming the segmentation boundary, thereby constructing the connection structure geometry. When the segmentation boundary is insufficient to provide adequate geometric support for defining the required connection structure geometry, or when the connection structure geometry is a result of volumetric interactions between entities within the region surrounded by the segmentation boundary, a Boolean-based connection structure generation method is selected.

[0113] Example 7 This embodiment is an object mesh generation system based on a 3D model 101, including a computer system 200. The computer system 200 includes one or more interconnected processors 202, a display 204, and a memory 206. The memory 206 is used to store computer programs / instructions; the display 204 is used to display the visual content output by the computer system 200; and the processor 202 is used to execute the computer program / instructions to implement the steps of the object mesh generation method based on the 3D model in any of the embodiments 1-6.

[0114] Figure 9 A block diagram of an object mesh generation system based on a 3D model 101 is shown, illustrating a computer system 200 that generates meshes for objects based on a 3D model 101 according to an embodiment of the present invention. The computer system 200 can be an electronic device with certain computing capabilities, including a personal computer (e.g., a laptop, desktop, or tablet computer), a 3D printer, a robot, or a system containing a robotic device. The computer system 200 may include one or more processors 202 or microprocessors (not shown), a display 204, and a memory 206. In this invention, "memory" refers to a computer-readable medium suitable for storing computer program instructions (or computer programs) and data. The memory 206 can be a non-volatile memory, including, for example, EPROM, EEPROM, flash memory devices, magnetic disks, magneto-optical disks, CD-ROMs, DVD-ROMs, etc. When the computer program instructions (or computer programs) stored in the memory 206 of the computer system 200 are executed by one or more processors 202, the computer system 200 can perform the processes S100 to S400 described above (e.g., ...). Figures 2 to 5 This refers to one or more operations shown in the diagram, as well as operations for constructing various connection structures. For example, computer system 200 can obtain the mesh of an object according to steps S11 to S17 in process S100. Computer system 200 can also execute operations set in processes S200, S300, and S400, which include more detailed steps for generating the object's segmented regions, boundary bands, segmented boundaries, and connection structure meshes. Computer system 200 can also determine the connection structure type, retrieve a connection structure type lookup table, and select a first or second connection structure generation method to derive the connection structure mesh. Furthermore, computer system 200 can use the first connection structure generation method and derive the connection structure mesh based on the segmented boundaries, or use the second connection structure generation method and derive the connection structure mesh through Boolean operations.

[0115] The above specific embodiments are specific support for the concept proposed in this invention, and should not be used to limit the scope of protection of this invention. Any equivalent changes or modifications made on the basis of this technical solution in accordance with the technical concept proposed in this invention shall still fall within the scope of protection of this invention.

Claims

1. A method for generating object meshes based on a 3D model, characterized in that, Includes the following steps: Obtain the 3D model of the object; The three-dimensional model is segmented to determine at least one segmentation boundary; Based on the segmentation operation, the segmentation boundaries are determined, and the three-dimensional model is divided into multiple segmentation regions. Construct boundary bands between adjacent segmented regions; Perform a remeshing operation on the grid region of the boundary band to reconstruct the segmentation boundary; Generate a new mesh for the object based on the reconstructed segmentation boundaries; And based on the reconstructed segmentation boundaries, a connection structure is generated between adjacent segmented regions.

2. The object mesh generation method based on a three-dimensional model according to claim 1, characterized in that: Determining segmentation boundaries based on segmentation operations includes: At least one candidate segmentation boundary is generated based on the 3D model; Based on preset segmentation boundary determination conditions, segmentation boundaries are selected and determined from candidate segmentation boundaries; the segmentation boundary determination conditions include: manufacturing constraints of the segmentation boundary, assembly compatibility of the segmentation boundary, and connection compatibility.

3. The object mesh generation method based on a three-dimensional model according to claim 2, characterized in that: The creation of segmented regions includes: Based on the defined segmentation boundaries, the surface of the 3D model is divided into multiple segments, creating segmented regions; wherein, the segmented regions include: The first type of segmentation region is used to form components that can be detached from the main body of the object and manufactured independently; The second type of segmentation region is used to form parts that are manufactured as a whole in a single 3D printing process.

4. The object mesh generation method based on a three-dimensional model according to claim 2, characterized in that: Obtaining candidate segmentation boundaries includes: Receive a 3D model containing geometric elements that define the shape and topology of the object; Candidate segmentation boundaries are obtained based on preset manufacturing requirements and morphological factors of the 3D model. The manufacturing requirements include manufacturing constraints, which include one or more of the following: gravity direction, preferred build orientation for printing, preferred orientation for separating or inserting and assembling separated parts, necessary support for 3D printing or other manufacturing processes, structural stability, and manufacturability. The morphological factors include: geometric discontinuities, surface features, curvature variations, appearance transitions, and seed region-based transitions. The candidate segmentation boundaries are represented as independent mesh structures.

5. The object mesh generation method based on a three-dimensional model according to claim 4, characterized in that: Multidimensional evaluation is performed during the step of determining the segmentation boundary: Manufacturing constraints on the segmentation boundary include: the orientational relationship of the generated connection surface relative to the direction of gravity or the orientation of the structure; the predicted stability of the segmented region after segmentation; and the expected requirements for manufacturing support structures. Assembly and connection compatibility of segmentation boundaries include: analyzing the compatibility of candidate segmentation boundaries with the assembly and reconnection process of several segmented areas in the object after individual manufacturing; the examination of some candidate segmentation boundaries includes: the feasibility of having sufficient material thickness along the boundary, the feasibility of generating connection structures, connectors or interlocking structures along the boundary, and the compatibility of candidate segmentation boundaries with chain, layered or multi-component assembly structures. The performance of candidate segmentation boundaries is characterized by the results of candidate segmentation boundaries after mesh reconstruction or remeshing, including: expected boundary smoothness based on curvature or other relevant geometric factors, robustness of candidate boundaries after remeshing operations, and ability to maintain appearance properties.

6. The object mesh generation method based on a three-dimensional model according to claim 5, characterized in that: The remeshing operation involves remeshing the mesh region surrounding the segmented boundary using a finer mesh than the original mesh of the 3D model, including: Remeshing is performed on the surface and inside the boundary zone respectively. When the size of the existing triangle or longer side inside the boundary zone is greater than a threshold, the existing triangle or longer side inside the boundary zone is subdivided. The subdivision method includes: inserting additional vertices along the edge adjacent to the dividing boundary, and reducing or not subdividing the mesh area outside the boundary zone as needed to achieve local remeshing. Within the boundary band after local remeshing, additional vertices are located using a finer mesh than the original mesh of the 3D model, including: aligning vertex positions to the spatial mesh of the finer mesh; and aligning the fine mesh by averaging small geometric perturbations to complete the remeshing operation.

7. The object mesh generation method based on a three-dimensional model according to claim 6, characterized in that: After completing the remeshing operation, the segmentation boundary is re-derived based on the finer mesh representation with higher geometric resolution to obtain the reconstructed segmentation boundary. The quality of the reconstructed segmentation boundary is then evaluated to determine whether it meets the preset quality threshold. If it does not meet the threshold, it is either eliminated or further refined. The quality evaluation indicators include: boundary smoothness, curvature continuity, and compatibility with subsequent downstream processing.

8. The object mesh generation method based on a three-dimensional model according to claim 7, characterized in that: After generating the segmented regions and boundary bands and reconstructing the segmented boundaries, some of the segmented regions are separated from the object, and mechanical connections are made to the separated regions to generate a manufacturable mesh and connection structure, including the following steps: Based on the determined segmentation regions and boundary bands, as well as the reconstructed segmentation boundary data, the connection surfaces and interface regions of the objects are determined. The acquisition of connection surfaces includes: determining whether two adjacent surfaces need to be connected by a connection structure, and the surfaces connected by the connection structure are taken as connection surfaces. The acquisition of interface regions includes: identifying and evaluating the regions corresponding to the connection surfaces generated by the segmentation process as interface regions. The mesh of the connection structure is generated based on the gap parameters and the type of the connection structure. Connecting structure chains are constructed based on meshes with interconnected structures. The generated mesh and chain of connection structures are verified. Obtain the final mesh of the object's segmented regions, boundary bands, segmented boundaries, and connection structures; The connection structure is used to connect detachable components and fixed components separated by a dividing boundary; The geometry of the connection structure used to connect the two components is generated based on the mesh of the detachable component and the fixed component. The generation methods include: deriving the connection structure geometry from the vertex and triangle data at the dividing boundary between the detachable component and the fixed component; or generating the connection structure geometry by constructing solid geometry or Boolean operations.

9. The object mesh generation method based on a three-dimensional model according to any one of claims 1-8, characterized in that: The three-dimensional model of an object includes geometric elements and appearance attributes. Appearance attributes are used to define the object's visual appearance information and material functional information. And / or, geometric elements include: faces, polygons, vertices, and edges; a vertex is a single point in three-dimensional space defined by X, Y, and Z coordinates, and some vertices constitute the corner points of an object; an edge is a straight line connecting two vertices, and multiple edges constitute the wireframe skeleton of an object; a face is a polygon formed by connecting multiple edges; And / or, some of the segmented regions are detachable components, and some adjacent detachable components are physically connected through a connecting structure; complementary positive and negative feature meshes are created for some connecting structures and adjacent segmented regions to form a mating connection; And / or, before performing the segmentation operation, the 3D model is preprocessed; the preprocessing includes verifying and repairing geometric elements to enforce the manifold topology, close open surfaces, and remove invalid or degenerate elements; And / or, the boundary band is generated according to generation criteria, which include a predetermined geodesic distance from the surface of the 3D model to the boundary, a predetermined number of mesh edge rings or adjacent layers around the boundary, and an adaptive standard based on local curvature, mesh density, and feature size compatible with the nozzle resolution of the 3D printer; the boundary band includes flexible material and buffers to facilitate part separation or coloring.

10. An object mesh generation system based on a three-dimensional model, characterized in that: Includes a computer system, the computer system comprising one or more interconnected processors, a display and a memory, the memory being used to store computer programs / instructions; The display is used to show visual content output by the computer system; The processor is configured to execute the computer program / instructions to implement the steps of the object mesh generation method based on a three-dimensional model as described in any one of claims 1-7.

Images