Automatic flattening method and system for three-dimensional long-sleeve mesh model
By preprocessing, boundary detection, and geodesic calculation of the 3D long-sleeved mesh model, the optimal cutting line is selected. Combined with the measurement of area change, the problems of cutting line deviation from process requirements and insufficient topological stability in the existing technology are solved. This achieves efficient and automated cutting and flattening of long-sleeved mesh models, which is suitable for customized clothing production.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- 卓尚服饰(杭州)有限公司
- Filing Date
- 2026-01-22
- Publication Date
- 2026-05-05
AI Technical Summary
Existing technologies fail to accurately incorporate the structural features of long sleeves during the cutting and flattening process of 3D long-sleeved mesh models. This results in cutting lines deviating from garment manufacturing requirements, low automation, and insufficient topological stability, impacting production efficiency and flexibility, and making it difficult to adapt to customized and small-batch production.
By preprocessing the 3D long-sleeved mesh model, detecting the boundary, calculating the geodesic distance, and selecting the optimal cutting line, combined with the area change measurement, automatic cutting and flattening are achieved. The system modules work together to ensure that the cutting line fits the process characteristics and avoids topological defects.
It achieves high-precision cutting and flattening without manual adjustment, adapts to long-sleeved models of different styles and conditions, improves production efficiency and flexibility, provides high-quality two-dimensional pattern data, and supports digital garment production.
Smart Images
Figure CN121982252A_ABST
Abstract
Description
Technical Field
[0001] This invention belongs to the field of reverse engineering technology for clothing, and in particular relates to an automatic flattening method and system for a three-dimensional long-sleeved mesh model. Background Technology
[0002] In the fields of 3D garment digitization, computer graphics, and intelligent garment manufacturing, the cutting process of 3D garment models (especially models with triangular meshes as their core representation) is a key technology connecting digital garment design with downstream processes such as physical production and virtual fitting. Long sleeves, as a core functional component of garments, have a significantly unique structure—they must conform to the natural shape of the human arm, including multiple characteristic areas such as the sleeve cap, sleeve body, cuff, and sleeve seam. Furthermore, different styles of long sleeves (such as straight sleeves, bell sleeves, and puff sleeves) exhibit significant differences in surface curvature and topological structure, which places stringent requirements on cutting precision, the rationality of cutting lines, and the degree of automation.
[0003] Currently, the cutting technology for 3D triangular mesh models has made some progress, but automatic cutting for the specific target of long sleeves still faces many bottlenecks. In existing technologies, most cutting schemes tend towards generalized mesh cutting, failing to fully consider the structural characteristics of long sleeves, leading to numerous problems in practical applications.
[0004] Traditional automatic cutting algorithms are mostly based on mesh geometry information (such as vertex distance and face normal vector) for cutting path planning. They fail to accurately combine the structural characteristics of long sleeves and have difficulty identifying key process features of long sleeves such as sleeve cap curves and sleeve seams. This leads to the cutting lines deviating from the garment process requirements, requiring a lot of manual adjustment to meet the needs of sheet unfolding, layout and production, which seriously affects the efficiency of digital garment production.
[0005] Some cutting schemes rely on manual preset cutting paths or marking of feature points, resulting in extremely low automation. They cannot meet the needs of rapid cutting and flattening of long-sleeved mesh models of different styles and sizes in the context of customized clothing and small-batch production, which greatly limits the flexibility and response speed of digital clothing production.
[0006] The cutting of long-sleeved triangular mesh models needs to take into account the mesh topology consistency to avoid problems such as cracks, redundant vertices, and missing faces after cutting. However, the existing general cutting algorithm is not topologically stable enough when dealing with such special parts with complex curved surface features. The above defects are easy to occur after cutting, which further affects the flattening effect and the normal progress of downstream processes (such as pattern sewing and virtual fitting), thus restricting the application of 3D clothing digitization technology.
[0007] Therefore, an automatic flattening method and system for three-dimensional long-sleeved mesh models is needed to solve the above problems. Summary of the Invention
[0008] The purpose of this invention is to provide an automatic flattening method and system for three-dimensional long-sleeved mesh models to solve the problems mentioned in the background art.
[0009] To achieve the above objectives, the present invention provides the following technical solution: an automatic flattening method for a three-dimensional long-sleeved mesh model, comprising the following steps:
[0010] S1. Preprocess the 3D long-sleeved mesh model;
[0011] S2. Detect the boundaries of the three-dimensional long-sleeved mesh model and determine the double boundaries;
[0012] S3. Perform a double-layer traversal on the double boundary and calculate the geodesic distance between each pair of vertices in the double boundary.
[0013] S4. Sort the geodesic distances in ascending order and select the n vertex combinations with the smallest geodesic distances.
[0014] S5. Convert the geodesics corresponding to the n sets of vertex combinations into automatic cutting lines to cut the three-dimensional long-sleeved mesh model;
[0015] S6. Measure the change in area of triangles in the flattened model after cutting, determine the optimal cutting scheme, and complete the flattening of the three-dimensional long-sleeved mesh model.
[0016] A further technical solution is that the pretreatment includes filling micro-pores and homogenizing the area of triangles;
[0017] The micro-hole filling method involves retaining the cuff and hem holes of the three-dimensional long-sleeve mesh model, while filling the remaining micro-holes in the model.
[0018] The triangle area homogenization is achieved by gradually adding vertices, so that the areas of all triangles in the model are within a preset small range and that the areas of each triangle are similar.
[0019] A further technical solution involves the boundary detection process of selecting the vertices of the cuff and hem portions of the three-dimensional long-sleeved mesh model, which are then used as the boundary points of the cuff and hem boundaries, respectively. The dual boundaries are the hem boundary and the cuff boundary.
[0020] A further technical solution is to implement the double-layer traversal through the following logic:
[0021] (Number of vertices in the sleeve root boundary) Iterate through all vertices in the sleeve root boundary. (Number of vertices in the sleeve cuff boundary) Iterate through all vertices in the sleeve cuff boundary. Calculate the geodesic distance between vertices.
[0022] ;
[0023] The geodesic distance is the minimum path along a vertex or edge of a three-dimensional long-sleeved mesh model to move from one boundary point to another, and this path consists of multiple lines connecting vertices.
[0024] A further technical solution is to measure the change in area by calculating the area of all triangles before and after the model is flattened. Based on the characteristic that the number of triangles and the topology remain unchanged before and after flattening, the absolute value of the change in area of each triangle is calculated and accumulated. The cutting scheme with the smallest absolute value of the total change in area is determined as the optimal cutting scheme.
[0025] The area change refers to the percentage change in the area of all triangles before and after the long-sleeved mesh model is flattened.
[0026] An automatic flattening system for a three-dimensional long-sleeved mesh model is applied to any of the automatic flattening methods for three-dimensional long-sleeved mesh models described above. The system includes a preprocessing module, a boundary detection module, a geodesic calculation module, a screening module, a cutting module, and a flattening optimization module. The modules work together to complete the entire process of flattening the three-dimensional long-sleeved mesh model from preprocessing to two-dimensional template flattening.
[0027] In a further technical solution, the preprocessing module includes a hole filling unit and an area homogenization unit;
[0028] The hole filling unit is used to retain the cuff holes and sleeve root holes, and to fill the remaining tiny holes in the model.
[0029] The area homogenization unit is used to make the areas of all triangles in the model similar within a preset small range by gradually adding vertices.
[0030] In a further technical solution, the boundary detection module is used to filter the vertices of the cuff and hem of the three-dimensional long-sleeved mesh model, and mark them as the cuff boundary and hem boundary, respectively.
[0031] The geodesic calculation module is used to calculate the geodesic distance between each pair of vertices of the two boundary lines through a double-layer traversal logic.
[0032] In a further technical solution, the filtering module is used to sort the geodesic distances and filter out the n vertex combinations with the smallest geodesic distances;
[0033] The cutting module is used to convert the selected geodesic lines into automatic cutting lines, delete the vertices involved in the cutting lines, and convert the semi-closed model into a fully open, flattenable model.
[0034] In a further technical solution, the flattening optimization module is used to calculate the area change of the triangles before and after flattening the model, select the best cutting scheme with the smallest absolute value of the total area change, and complete the final flattening of the three-dimensional long-sleeved mesh model based on the scheme.
[0035] Compared with the prior art, the beneficial effects of the present invention are:
[0036] This invention precisely locates the key boundaries of the sleeve root and cuff through boundary detection, calculates the geodesic distance of the "sleeve root-cuff" that conforms to the process logic by combining double boundary double layer traversal logic, and then selects the optimal solution by quantitatively filtering the area change. The final cutting line strictly fits the core process features such as the sleeve seam and sleeve cap arc. The flattened two-dimensional template can be used directly for sewing production without manual adjustment, which significantly improves the process adaptability and flattening accuracy of the cutting line.
[0037] This invention eliminates the need for manual specification of cutting lines and marking of feature points throughout the entire process; all modules of the system automatically and collaboratively complete all operations. Regardless of whether the long sleeve is straight or bent, loose or fitted, a 3D model can be quickly input and a 2D template output, perfectly adapting to the "rapid response and flexible adaptation" requirements of customized clothing and small-batch production, significantly improving production efficiency and flexibility. Simultaneously, the density of the triangular mesh can be flexibly set according to actual needs; a higher density is set for high precision requirements, and a lower density is set for low precision requirements to improve calculation speed, further adapting to the needs of different production scenarios.
[0038] This invention optimizes the model's topology in advance through hole filling and triangle area homogenization operations during the preprocessing stage, reducing the risk of cutting defects from the source. The cutting process is based on geodesics for precise cutting, avoiding problems such as cracks, redundant vertices, and missing faces after cutting. The model's topology remains consistent throughout the entire process, with no structural distortion after cutting and flattening. This provides high-quality basic data for downstream processes such as pattern layout, virtual fitting, and physical production, significantly improving the efficiency of downstream processes.
[0039] To more clearly illustrate the structural features and effects of the present invention, the present invention will be described in detail below with reference to the accompanying drawings and specific embodiments. Attached Figure Description
[0040] Figure 1 This is a schematic diagram of the process of the present invention;
[0041] Figure 2 This is a schematic diagram illustrating a comparison of the triangle homogenization results of the three-dimensional long-sleeved mesh model of the present invention;
[0042] Figure 3 This is a schematic diagram illustrating the automatic cutting result of the three-dimensional long-sleeved mesh model of the present invention in a straightened state;
[0043] Figure 4 This is a schematic diagram illustrating an example of the automatic cutting result of the three-dimensional long-sleeved mesh model of the present invention under a curved state;
[0044] Figure 5This is a schematic diagram illustrating the automatic flattening results of the three-dimensional long-sleeved mesh model of the present invention in both straight and bent states. Detailed Implementation
[0045] The present invention will be further described below with reference to embodiments.
[0046] The following embodiments are used to illustrate the present invention, but should not be used to limit the scope of protection of the present invention. The conditions in the embodiments can be further adjusted according to specific conditions, and simple improvements to the method of the present invention under the premise of the concept of the present invention are all within the scope of protection claimed by the present invention.
[0047] Please see Figure 1-5 This invention provides an automatic flattening method for a three-dimensional long-sleeved mesh model, comprising the following steps:
[0048] S1. Preprocess the 3D long-sleeved mesh model;
[0049] S2. Detect the boundaries of the three-dimensional long-sleeved mesh model and determine the double boundaries;
[0050] S3. Perform a double-layer traversal on the double boundary and calculate the geodesic distance between each pair of vertices in the double boundary.
[0051] S4. Sort the geodesic distances in ascending order and select the n vertex combinations with the smallest geodesic distances.
[0052] S5. Convert the geodesics corresponding to the n sets of vertex combinations into automatic cutting lines to cut the three-dimensional long-sleeved mesh model;
[0053] S6. Measure the change in area of triangles in the flattened model after cutting, determine the optimal cutting scheme, and complete the flattening of the three-dimensional long-sleeved mesh model.
[0054] In this embodiment, a fully automated flattening system is constructed through six progressive steps. The core logic of this design lies in achieving a complete transformation from a 3D model to a 2D template through a closed-loop process of "preprocessing-probe-calculation-screening-cutting-optimization," without requiring manual intervention in specifying cutting lines. Each step is interconnected: preprocessing provides high-quality model data for subsequent steps, boundary detection clarifies the calculation object, geodesic calculation and screening lock in candidate cutting lines, and cutting and area optimization ensure the flattening effect, thus guaranteeing the automation, consistency, and accuracy of the method.
[0055] Specifically, the pretreatment includes filling micropores and homogenizing the area of triangles;
[0056] The micro-hole filling method involves retaining the cuff and hem holes of the three-dimensional long-sleeve mesh model, while filling the remaining micro-holes in the model.
[0057] The triangle area homogenization is achieved by gradually adding vertices, so that the areas of all triangles in the model are within a preset small range and that the areas of each triangle are similar.
[0058] In this embodiment, the hole filling operation preserves core functional holes (cuff and hem holes) and removes interfering micro holes, ensuring the integrity of the model surface and avoiding boundary detection deviations and geodesic calculation errors caused by micro holes. The triangle area uniformization is achieved by adding and adjusting vertices to make the areas of all triangles more consistent, providing a data basis for the accuracy of subsequent geodesic distance calculations and the objectivity of area change measurement, and ensuring the reliable execution of subsequent steps.
[0059] Specifically, the boundary detection involves selecting the vertices of the cuff and hem of the three-dimensional long-sleeved mesh model, which are used as the boundary points of the cuff boundary and hem boundary, respectively. The dual boundaries are the hem boundary and the cuff boundary.
[0060] In this embodiment, the core significance of this design lies in accurately locking the key structural boundaries of the long-sleeve mesh model, so that subsequent geodesic calculations focus on the core path of "sleeve root-cuff" which conforms to the logic of garment manufacturing, ensuring that the cutting line is always generated around the key features of the long sleeve, avoiding deviation from core process areas such as sleeve seam and sleeve cap, and laying the foundation for the rationality of the cutting line.
[0061] Specifically, the double-layer traversal is implemented through the following logic:
[0062] (Number of vertices in the sleeve root boundary) Iterate through all vertices in the sleeve root boundary. (Number of vertices in the sleeve cuff boundary) Iterate through all vertices in the sleeve cuff boundary. Calculate the geodesic distance between vertices.
[0063] ;
[0064] The geodesic distance is the minimum path along a vertex or edge of a three-dimensional long-sleeved mesh model to move from one boundary point to another, and this path consists of multiple lines connecting vertices.
[0065] In this embodiment, the traversal logic ensures that the geodesic distance calculation is complete and avoids the loss of the optimal cutting line due to the failure to calculate some combinations. The geodesic distance is defined as the shortest path on the model surface, which not only conforms to the process habits of clothing cutting (cutting along the shortest path on the surface can reduce material waste and facilitate sewing), but also provides a clear geometric basis for the generation of the cutting line, ensuring the feasibility and rationality of the cutting line.
[0066] Specifically, the area change is measured by the area of all triangles before and after the calculation model is flattened. Based on the characteristic that the number of triangles and the topology remain unchanged before and after flattening, the absolute value of the area change of each triangle is calculated and accumulated. The cutting scheme with the smallest absolute value of the total area change is determined as the optimal cutting scheme.
[0067] The area change refers to the percentage change in the area of all triangles before and after the long-sleeved mesh model is flattened.
[0068] In this embodiment, by utilizing the characteristic that "the number of triangles and the topological structure remain unchanged before and after flattening", an objective evaluation system for cutting schemes is established by quantitatively calculating the absolute value of the change in triangle area (the change in area is the percentage change in the area of all triangles before and after flattening). The scheme with the smallest total area change is selected as the best scheme. Essentially, this ensures that the flattened two-dimensional pattern restores the structural features of the three-dimensional model to the greatest extent, so that the pattern conforms to the sewing rules of clothing and avoids discomfort or process defects caused by area distortion.
[0069] An automatic flattening system for a three-dimensional long-sleeved mesh model is applied to the automatic flattening method for a three-dimensional long-sleeved mesh model described in the above embodiments. The system includes a preprocessing module, a boundary detection module, a geodesic calculation module, a screening module, a cutting module, and a flattening optimization module. The modules work together to complete the entire process of flattening the three-dimensional long-sleeved mesh model from preprocessing to two-dimensional template flattening.
[0070] In this embodiment, the core of the system design lies in the "one-to-one correspondence between modules and steps". Through modular decomposition, the abstract steps of the method are transformed into specific executable functional units, and the modules work together to form a complete technical system. The modular design of the system not only facilitates actual deployment and maintenance, but also ensures the strict execution of the method, avoids omission of steps or execution deviation, and provides flexible expansion space for subsequent technology iterations.
[0071] Specifically, the preprocessing module includes a hole filling unit and an area homogenization unit;
[0072] The hole filling unit is used to retain the cuff holes and sleeve root holes, and to fill the remaining tiny holes in the model.
[0073] The area homogenization unit is used to make the areas of all triangles in the model similar within a preset small range by gradually adding vertices.
[0074] In this embodiment, the core function of the hole filling unit is to filter and retain the core holes, fill the interfering holes, and output a complete model; the core function of the area uniformization unit is to adjust the triangle area distribution through the vertex addition algorithm and output a model with uniform area. The two units work together to complete the preprocessing task and provide high-quality input data for subsequent modules.
[0075] Specifically, the boundary detection module is used to filter the vertices of the cuff and hem of the three-dimensional long-sleeved mesh model, and mark them as the cuff boundary and hem boundary, respectively.
[0076] The geodesic calculation module is used to calculate the geodesic distance between each pair of vertices of the two boundary lines through a double-layer traversal logic.
[0077] In this embodiment, the core function of the boundary detection module is to identify and mark the vertices of the cuff and cuff edges, providing a clear calculation object for the geodesic calculation module. The geodesic calculation module automatically completes the geodesic distance calculation between each pair of the two boundary vertices through a built-in double-layer traversal algorithm. The two modules work together to complete the core tasks of boundary identification and critical distance calculation, achieving a seamless connection from "model boundary identification" to "critical distance calculation".
[0078] Specifically, the filtering module is used to sort the geodesic distances and filter out the n vertex combinations with the smallest geodesic distances;
[0079] The cutting module is used to convert the selected geodesic lines into automatic cutting lines, delete the vertices involved in the cutting lines, and convert the semi-closed model into a fully open, flattenable model.
[0080] In this embodiment, the filtering module sorts the geodesic distances using a sorting algorithm to select the optimal candidate combinations (e.g., n=10 groups), providing the cutting module with precise cutting line basis. The cutting module converts the candidate combinations into solid cutting lines and deletes the relevant vertices, realizing the conversion from a semi-closed model to an open model. This conversion is a key prerequisite for the model to be directly flattened. The two modules work together to complete the distance sorting and filtering and model cutting operations, transforming abstract distance data into specific model cutting results.
[0081] Specifically, the flattening optimization module is used to calculate the change in area of triangles before and after flattening the model, select the best cutting scheme with the smallest absolute value of total area change, and complete the final flattening of the three-dimensional long-sleeved mesh model based on the scheme.
[0082] In this embodiment, the module optimizes the cutting scheme through quantitative calculations to ensure that the output two-dimensional template meets the process requirements. At the same time, it automatically completes the flattening operation without manual intervention, ultimately achieving the goal of "inputting a three-dimensional model and outputting a two-dimensional template" with full automation.
[0083] By designing the logical steps and deploying the corresponding system modules, a fully automated system of "data processing - feature recognition - path calculation - scheme optimization - output" was constructed, which effectively solved the three major defects of the existing technology, adapted to long-sleeve models with different flexion states and different patterns, and provided efficient technical support for digital garment production.
[0084] Working principle and usage process of this invention:
[0085] The automatic flattening of a 3D long-sleeved mesh model is achieved through six sequentially connected steps. The specific working logic is as follows:
[0086] Preprocessing stage: First, the input 3D long-sleeved mesh model undergoes dual optimization. On one hand, a micro-hole filling operation is performed. Through a model surface hole recognition algorithm, cuff holes and sleeve root holes directly related to the garment's function are accurately distinguished and retained. At the same time, all other micro-holes on the model surface that do not affect the overall structure are filled, eliminating the interference of micro-holes on the accuracy of subsequent boundary recognition and geodesic distance calculation, ensuring the integrity of the model surface. On the other hand, a triangle area homogenization operation is performed. Through a vertex addition algorithm, vertex interpolation is performed to supplement triangles with excessively large or small areas in the model, gradually adjusting the area of all triangles in the model so that the area of all triangle elements is within a preset small range, that is, all triangles have similar areas, providing a homogenized and high-quality data foundation for subsequent geodesic distance calculation and area change measurement.
[0087] Boundary detection stage: Based on the complete and homogenized 3D long-sleeve mesh model obtained after preprocessing, the set of vertices located at the "cuff edge" and "sleeve root edge" in the model is selected through the boundary vertex identification algorithm; the selected cuff edge vertices are defined as cuff boundaries, and the sleeve root edge vertices are defined as sleeve root boundaries, forming the double boundaries required for subsequent geodesic distance calculation, clarifying the starting set (sleeve root boundary vertices) and ending set (cuff edge boundary vertices) of geodesic distance calculation, laying the foundation for subsequent comprehensive calculation of geodesic distance between vertices;
[0088] Geodesic distance calculation stage: For the determined sleeve root boundary and sleeve cuff boundary, a double-layer traversal calculation logic is executed. The specific traversal logic is as follows:
[0089] (Number of vertices in the sleeve root boundary) Iterate through all vertices in the sleeve root boundary. (Number of vertices in the sleeve cuff boundary) Iterate through all vertices in the sleeve cuff boundary. Calculate the geodesic distance between vertices.
[0090] ;
[0091] Through the above double-loop logic, all vertex combinations of the sleeve root boundary and the cuff boundary are fully covered, ensuring that the geodesic distance of any vertex pair is not missed. In each loop iteration, the 3D mesh geodesic algorithm is called to calculate the shortest path along the surface of the 3D long sleeve mesh model between the sleeve root vertex and the cuff vertex corresponding to the current iteration. This shortest path is the geodesic distance between the two points, and finally n×m sets of complete geodesic distance data are obtained.
[0092] Sorting and filtering stage: All calculated geodesic distance data are systematically statistically analyzed and organized, and sorted in ascending order of distance value; based on the balance between accuracy requirements and computational efficiency in actual production applications, the number of filters is set to n (e.g., n=10), and n sets of vertex combinations with the smallest distance are extracted from the sorted geodesic distance data. Each set of vertex combinations contains one sleeve root boundary vertex and one sleeve cuff boundary vertex. These n sets of vertex combinations are potential optimal cutting line candidate schemes;
[0093] Cutting stage: The geodesics corresponding to the selected n sets of vertex combinations are converted into n automatic cutting lines. Each automatic cutting line is composed of the continuous vertices of the corresponding geodesic line connected in sequence. Through the model cutting algorithm, the 3D long-sleeved mesh model is cut, that is, all vertices involved in each automatic cutting line are deleted, so that the long-sleeved mesh model, which was originally in a semi-closed state, is cut into a completely open mesh structure. This open mesh structure has the conditions to directly perform flattening operation.
[0094] Flattening and Optimization Stage: Flattening operations are performed on the open mesh models obtained after n cuts. During the flattening process, the area of all triangles before flattening is calculated for each model, and then the area of the corresponding triangles after flattening is calculated. Utilizing the characteristic that "the number of triangles and the topology remain unchanged before and after flattening", the area change (i.e., the percentage change in area) of each triangle before and after flattening is calculated, and the absolute value of the area change of each triangle is obtained. The absolute values of the area change of all triangles in each model are summed to obtain the absolute value of the total triangle area change for each cutting scheme. The cutting scheme with the smallest absolute value of the total triangle area change is determined as the optimal cutting scheme. Based on the optimal scheme, the corresponding open mesh model is finally flattened to output a long-sleeved two-dimensional pattern that meets the requirements of garment technology and has the smallest area distortion.
[0095] The system achieves automated execution of the above methods through the coordinated operation of six functional modules. The specific working logic is as follows:
[0096] After receiving the externally input 3D long-sleeved mesh model, the preprocessing module automatically starts the hole recognition program in its hole filling unit to accurately identify cuff holes, sleeve root holes and other tiny holes in the model. It retains the cuff and sleeve root holes and fills the remaining tiny holes. At the same time, the area homogenization unit starts the vertex addition algorithm to supplement the vertices of triangles with uneven area distribution in the model, adjusts the area of all triangles so that the area of all triangles is within a preset small range and remains similar, and finally outputs a complete and homogenized 3D model, and sends the model data to the boundary detection module.
[0097] After receiving the model data output by the preprocessing module, the boundary detection module automatically runs the boundary vertex recognition program to filter out the vertices of the cuff edge and the sleeve root edge in the model, and marks them as the cuff boundary and the sleeve root boundary respectively, forming double boundary data; then the double boundary data and the original model data are packaged together and sent to the geodesic calculation module.
[0098] After receiving the model data and the double boundary data, the geodesic calculation module starts running its built-in double-layer traversal program according to the following logic:
[0099] (Number of vertices in the sleeve root boundary) Iterate through all vertices in the sleeve root boundary. (Number of vertices in the sleeve cuff boundary) Iterate through all vertices in the sleeve cuff boundary. Calculate the geodesic distance between vertices.
[0100] ;
[0101] During the traversal, the geodesic calculation subroutine is called to calculate the geodesic distance for each group of vertices one by one, generating complete n×m groups of geodesic distance data, and then sending the distance data to the filtering module.
[0102] After receiving the geodesic distance data, the filtering module automatically starts the sorting program and sorts all the data in ascending order of distance. Based on the preset number of filters n (e.g., n=10), it extracts the n vertex combinations with the smallest distance to form a candidate cutting scheme set and sends the candidate set information to the cutting module.
[0103] After receiving the set of candidate cutting schemes, the cutting module automatically converts the geodesic corresponding to each set of vertex combinations into solid cutting lines, starts the model cutting subroutine, deletes the vertices involved in each cutting line, and converts the semi-closed 3D long-sleeved mesh model into a fully open flattenable model; it then packages the n sets of open mesh models and sends them to the flattening optimization module.
[0104] After receiving n sets of open mesh models, the flattening optimization module automatically starts the area calculation program to calculate the area of all triangles before and after flattening each model. Then, it calculates the area change (percentage of area change) and absolute value of each triangle, and accumulates them to obtain the absolute value of the total triangle area change for each scheme. By comparison, the best cutting scheme with the smallest absolute value of total change is selected, and the flattening subroutine is started to flatten the open mesh model corresponding to the scheme. Finally, a two-dimensional pattern that meets the requirements of garment technology is output.
[0105] Throughout the entire system operation, the modules achieve seamless collaboration and linkage through data transmission, requiring no manual intervention and automatically completing the entire process from 3D model input to 2D template output. This ensures the strict implementation and efficient application of the method, while also guaranteeing the accuracy of the flattening results and their adaptability to the process.
[0106] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.
Claims
1. An automatic flattening method for a three-dimensional long-sleeved mesh model, characterized in that, Includes the following steps: S1. Preprocess the 3D long-sleeved mesh model; S2. Detect the boundaries of the three-dimensional long-sleeved mesh model and determine the double boundaries; S3. Perform a double-layer traversal on the double boundary and calculate the geodesic distance between each pair of vertices in the double boundary. S4. Sort the geodesic distances in ascending order and select the n vertex combinations with the smallest geodesic distances. S5. Convert the geodesics corresponding to the n sets of vertex combinations into automatic cutting lines to cut the three-dimensional long-sleeved mesh model; S6. Measure the change in area of triangles in the flattened model after cutting, determine the optimal cutting scheme, and complete the flattening of the three-dimensional long-sleeved mesh model.
2. The automatic flattening method for a three-dimensional long-sleeved mesh model according to claim 1, characterized in that, The pretreatment includes filling micro-pores and homogenizing the area of triangles; The micro-hole filling method involves retaining the cuff and hem holes of the three-dimensional long-sleeve mesh model, while filling the remaining micro-holes in the model. The triangle area homogenization is achieved by gradually adding vertices, so that the areas of all triangles in the model are within a preset small range and that the areas of each triangle are similar.
3. The automatic flattening method for a three-dimensional long-sleeved mesh model according to claim 1, characterized in that, The boundary detection involves selecting the vertices of the cuff and hem portions of the 3D long-sleeved mesh model, which are then used as the boundary points of the cuff and hem boundaries, respectively. The dual boundaries are the hem boundary and the cuff boundary.
4. The automatic flattening method for a three-dimensional long-sleeved mesh model according to claim 1, characterized in that, The double-layer traversal is implemented through the following logic: (Number of vertices in the sleeve root boundary) Iterate through all vertices in the sleeve root boundary. (Number of vertices in the sleeve cuff boundary) Iterate through all vertices in the sleeve cuff boundary. Calculate the geodesic distance between vertices. ； The geodesic distance is the minimum path along a vertex or edge of a three-dimensional long-sleeved mesh model to move from one boundary point to another, and this path consists of multiple lines connecting vertices.
5. The automatic flattening method for a three-dimensional long-sleeved mesh model according to claim 1, characterized in that, The area change is measured by the area of all triangles before and after the model is flattened. Based on the characteristic that the number of triangles and the topology remain unchanged before and after flattening, the absolute value of the area change of each triangle is calculated and accumulated. The cutting scheme with the smallest absolute value of the total area change is determined as the best cutting scheme. The area change refers to the percentage change in the area of all triangles before and after the long-sleeved mesh model is flattened.
6. An automatic flattening system for a three-dimensional long-sleeved mesh model, applied to the automatic flattening method for a three-dimensional long-sleeved mesh model as described in any one of claims 1-5, characterized in that, It includes a preprocessing module, a boundary detection module, a geodesic calculation module, a screening module, a cutting module, and a flattening optimization module. These modules work together to complete the entire process of a 3D long-sleeved mesh model from preprocessing to 2D template flattening.
7. The automatic flattening system for a three-dimensional long-sleeved mesh model according to claim 6, characterized in that, The preprocessing module includes a hole filling unit and an area homogenization unit; The hole filling unit is used to retain the cuff holes and sleeve root holes, and to fill the remaining tiny holes in the model. The area homogenization unit is used to make the areas of all triangles in the model similar within a preset small range by gradually adding vertices.
8. The automatic flattening system for a three-dimensional long-sleeved mesh model according to claim 6, characterized in that, The boundary detection module is used to filter the vertices of the cuff and hem of the 3D long-sleeved mesh model, and mark them as the cuff boundary and hem boundary, respectively. The geodesic calculation module is used to calculate the geodesic distance between each pair of vertices of the two boundary lines through a double-layer traversal logic.
9. The automatic flattening system for a three-dimensional long-sleeved mesh model according to claim 6, characterized in that, The filtering module is used to sort the geodesic distances and filter out the n vertex combinations with the smallest geodesic distances; The cutting module is used to convert the selected geodesic lines into automatic cutting lines, delete the vertices involved in the cutting lines, and convert the semi-closed model into a fully open, flattenable model.
10. The automatic flattening system for a three-dimensional long-sleeved mesh model according to claim 6, characterized in that, The flattening optimization module is used to calculate the change in area of triangles before and after flattening the model, select the best cutting scheme with the smallest absolute value of total area change, and complete the final flattening of the three-dimensional long-sleeved mesh model based on the scheme.