Artificial intelligence-based 3D simulation of clothing customization and automatic generation of process sheets

Abstract

The present application is applicable to the field of clothing customization technology, and in particular relates to a method for automatic generation of 3D simulation and process sheets for clothing customization based on artificial intelligence. The method comprises: providing basic data support by acquiring clothing customization parameters; performing three-dimensional parameter modeling based on the clothing customization parameters, determining a three-dimensional fitting model corresponding to the clothing customization parameters, and intuitively displaying the effect of clothing on the body; performing discrete mesh division on the three-dimensional fitting model to obtain a first triangular mesh topology corresponding to the three-dimensional fitting model; performing physical constraint iteration on the first triangular mesh topology based on the clothing customization parameters to obtain a second triangular mesh topology corresponding to the three-dimensional fitting model; obtaining a customized process sheet corresponding to the clothing customization parameters based on the second triangular mesh topology, realizing the integration of simulation and process sheet, effectively improving clothing production efficiency and quality, and reducing cutting and sewing errors.

Description

Technical Field

[0001] The present application belongs to the technical field of clothing customization, and in particular relates to a method for automatic generation of clothing customization 3D simulation and process sheets based on artificial intelligence. Background Art

[0002] Custom clothing is a personalized service based on the specific needs of customers, from demand communication, body measurement, design confirmation, fabric selection to production, quality inspection, delivery and after-sales service. Its advantages lie in its ability to meet personalized needs, ensure fit and provide high-quality craftsmanship. It is widely used in various scenarios such as business formal wear, wedding and special occasion clothing, professional uniforms and fashion items.

[0003] Traditional clothing customization technology often uses static model simulation or sparse grids and mass springs for dynamic simulation in 3D simulation. It only focuses on the conversion between 3D simulation and two-dimensional paper patterns, and ignores the integrated process of 3D simulation and process sheets. The lack of accurate and effective conversion methods between simulation models and process sheets leads to lengthy customization cycles and high error rates in cutting and sewing pieces, which limits the online development space for complex styles. Summary of the Invention

[0004] The embodiment of the present application provides a method for automatic generation of 3D simulation and process sheets for clothing customization based on artificial intelligence, which can solve the problem of lack of effective conversion means between simulation models and process sheets during clothing customization, resulting in a long customization cycle, a high error rate in cutting and sewing pieces, and limiting the online development space for complex styles.

[0005] In a first aspect, an embodiment of the present application provides a method for 3D simulation and automatic generation of a process sheet for clothing customization based on artificial intelligence, comprising:

[0006] Get clothing customization parameters;

[0007] Performing three-dimensional parameter modeling based on the clothing customization parameters to determine a three-dimensional fitting model corresponding to the clothing customization parameters;

[0008] Performing discrete mesh division on the three-dimensional fitting model to obtain a first triangular mesh topology corresponding to the three-dimensional fitting model;

[0009] Performing physical constraint iteration on the first triangular mesh topology based on the clothing customization parameters to obtain a second triangular mesh topology corresponding to the three-dimensional fitting model;

[0010] Based on the second triangular mesh topology, a customization process sheet corresponding to the clothing customization parameters is obtained.

[0011] The embodiment of the present application provides an artificial intelligence-based clothing customization 3D simulation and process sheet automatic generation method, which provides basic data support by obtaining clothing customization parameters. Based on the clothing customization parameters, three-dimensional parameter modeling is performed to determine the three-dimensional fitting model corresponding to the clothing customization parameters, intuitively preview the effect of the clothing on the body, adjust the design plan in advance, and avoid the cost of post-production modifications. The three-dimensional fitting model is discretely meshed to obtain the first triangular mesh topology corresponding to the three-dimensional fitting model, laying a mesh foundation for accurate physical simulation and cutting analysis. Based on the clothing customization parameters, the first triangular mesh topology is physically constrained and iterated to obtain the second triangular mesh topology corresponding to the three-dimensional fitting model, ensuring that the clothing version meets the physical properties of the fabric and the dynamic requirements of human wearing. Based on the second triangular mesh topology, a customized process sheet corresponding to the clothing customization parameters is obtained to guide the clothing production process, realize the integration of simulation and process sheet, effectively improve clothing production efficiency and quality, and reduce cutting and sewing errors.

[0012] In a second aspect, the embodiments of the present application provide an artificial intelligence-based clothing customization 3D simulation and process sheet automatic generation system, including:

[0013] An acquisition unit, used to acquire clothing customization parameters;

[0014] A modeling unit, configured to perform three-dimensional parameter modeling based on the clothing customization parameters and determine a three-dimensional fitting model corresponding to the clothing customization parameters;

[0015] a dividing unit, configured to perform discrete mesh division on the three-dimensional fitting model to obtain a first triangular mesh topology corresponding to the three-dimensional fitting model;

[0016] an iterative unit, configured to perform physical constraint iteration on the first triangular mesh topology based on the clothing customization parameters to obtain a second triangular mesh topology corresponding to the three-dimensional fitting model;

[0017] A generating unit is configured to obtain a customization process sheet corresponding to the clothing customization parameters based on the second triangular mesh topology.

[0018] In a third aspect, an embodiment of the present application provides an electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the computer program, the method described in any one of the first aspects above is implemented.

[0019] In a fourth aspect, an embodiment of the present application provides a computer program product, which, when executed on a terminal device, enables the terminal device to execute any one of the methods described in the first aspect above.

[0020] It can be understood that the beneficial effects of the second to fourth aspects mentioned above can be found in the relevant description of the first aspect mentioned above, and will not be repeated here. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the following briefly introduces the drawings required for use in the embodiments or descriptions of the prior art. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without any creative work.

[0022] Figure 1 This is a flow chart of a method for 3D simulation and automatic generation of process sheets for clothing customization based on artificial intelligence provided by an embodiment of the present application;

[0023] Figure 2 This is a schematic diagram of a three-dimensional fitting model in a method for 3D simulation and automatic generation of a process sheet for clothing customization based on artificial intelligence provided in one embodiment of the present application;

[0024] Figure 3 This is a schematic diagram of a process sheet in a method for automatically generating a process sheet and 3D simulation of clothing customization based on artificial intelligence provided by an embodiment of the present application;

[0025] Figure 4 This is a structural diagram of an artificial intelligence-based clothing customization 3D simulation and process sheet automatic generation system provided in one embodiment of the present application;

[0026] Figure 5 It is a structural diagram of an electronic device provided in one embodiment of the present application. DETAILED DESCRIPTION

[0027] In the following description, specific details such as specific system structures and techniques are provided for purposes of illustration rather than limitation to facilitate a thorough understanding of the embodiments of the present application. However, it will be apparent to those skilled in the art that the present application may be implemented in other embodiments without these specific details. In other cases, detailed descriptions of well-known systems, devices, circuits, and methods are omitted to avoid obscuring the description of the present application with unnecessary detail.

[0028] It should be understood that when used in the present specification and the appended claims, the term "comprising" indicates the presence of described features, integers, steps, operations, elements and / or components, but does not preclude the presence or addition of one or more other features, integers, steps, operations, elements, components and / or collections thereof.

[0029] It will also be understood that the term "and / or" used in this specification and the appended claims refers to and includes any and all possible combinations of one or more of the associated listed items.

[0030] As used in this specification and the appended claims, the term "if" can be interpreted as "when" or "upon" or "in response to determining" or "in response to detecting," depending on the context. Similarly, the phrases "if it is determined" or "if the described condition or event is detected" can be interpreted as meaning "upon determination" or "in response to determining" or "upon detection of the described condition or event" or "in response to detecting the described condition or event," depending on the context.

[0031] In addition, in the description of the present application specification and the appended claims, the terms "first", "second", "third", etc. are only used to distinguish the descriptions and cannot be understood as indicating or implying relative importance.

[0032] References to "one embodiment" or "some embodiments" in this specification mean that a particular feature, structure, or characteristic described in conjunction with that embodiment is included in one or more embodiments of the present application. Thus, phrases such as "in one embodiment," "in some embodiments," "in other embodiments," and "in other embodiments" appearing in various places in this specification do not necessarily refer to the same embodiment, but rather mean "one or more but not all embodiments," unless otherwise specifically emphasized. The terms "including," "comprising," "having," and variations thereof all mean "including but not limited to," unless otherwise specifically emphasized.

[0033] In traditional 3D simulation of custom apparel, static model simulation dominates, with an excessive emphasis on converting 3D simulations into 2D patterns, while completely neglecting the integrated process of 3D simulation and process sheets. The mass-spring technology commonly used in dynamic simulation also has flaws. While mass-spring technology can simulate garment deformation to a certain extent, because it simplifies garments into a combination of mass points and springs, it overly crudely simulates the complex structure of garments and the curvature of the human body. When dealing with complex styles, it is difficult to accurately simulate the deformation of each garment component, resulting in significant deviations between 3D simulation results and actual wear. The lack of accurate and effective conversion methods between simulation models and process sheets results in lengthy customization cycles and high errors in cutting and sewing, limiting the online development of complex styles.

[0034] To solve the above problems, an embodiment of the present application provides a method for automatic generation of 3D simulation and process sheets for clothing customization based on artificial intelligence. In this method, basic data support is provided by obtaining clothing customization parameters. Three-dimensional parameter modeling is performed based on the clothing customization parameters, and the three-dimensional fitting model corresponding to the clothing customization parameters is determined. The effect of the clothing on the body is visually previewed, and the design plan is adjusted in advance to avoid the cost of post-production modifications. The three-dimensional fitting model is discretely meshed to obtain the first triangular mesh topology corresponding to the three-dimensional fitting model, laying a mesh foundation for accurate physical simulation and cutting analysis. Based on the clothing customization parameters, the first triangular mesh topology is physically constrained and iterated to obtain the second triangular mesh topology corresponding to the three-dimensional fitting model, ensuring that the clothing pattern meets the physical properties of the fabric and the dynamic requirements of human wearing. Based on the second triangular mesh topology, a customized process sheet corresponding to the clothing customization parameters is obtained to guide the clothing production process, realize the integration of simulation and process sheets, effectively improve clothing production efficiency and quality, and reduce cutting and sewing errors.

[0035] The artificial intelligence-based clothing customization 3D simulation and process sheet automatic generation method provided in the embodiment of the present application can be applied to a terminal device. In this case, the terminal device is the executor of the artificial intelligence-based clothing customization 3D simulation and process sheet automatic generation method provided in the embodiment of the present application. The embodiment of the present application does not impose any restrictions on the specific type of the terminal device.

[0036] For example, the terminal device can be a mobile phone, tablet computer, wearable device, vehicle-mounted device, augmented reality (AR) / virtual reality (VR) device, laptop computer, ultra-mobile personal computer (UMPC), netbook, personal digital assistant (PDA), desktop computer, smart large screen, smart TV and other terminal devices, handheld devices with wireless communication function, computing devices or other processing devices connected to a wireless modem, etc.

[0037] In order to better understand the artificial intelligence-based clothing customization 3D simulation and process sheet automatic generation method provided in the embodiment of the present application, the specific implementation process of the artificial intelligence-based clothing customization 3D simulation and process sheet automatic generation method provided in the embodiment of the present application is exemplarily introduced below.

[0038] Figure 1 A schematic flow chart of an artificial intelligence-based 3D simulation and automatic process sheet generation method for clothing customization provided in an embodiment of the present application is shown. The artificial intelligence-based 3D simulation and automatic process sheet generation method for clothing customization includes:

[0039] S100, obtaining clothing customization parameters.

[0040] As you can understand, clothing customization parameters are the foundational data for personalized clothing customization, encompassing a wide range of key information. A key component is the user's body measurements, such as height, weight, chest circumference, waist circumference, hip circumference, shoulder width, arm length, and leg length. These precise measurements ensure that the garment fits the user's body curves perfectly. Furthermore, clothing style parameters are essential, including garment type (e.g., shirt, dress, suit), collar type (e.g., lapel, crew neck, stand-up collar), sleeve type (e.g., long sleeve, short sleeve, sleeveless, lantern sleeve), and skirt style (e.g., straight skirt, A-line skirt, pleated skirt). These parameters determine the overall style and appearance of the garment. Fabric selection parameters also play a role, including material (cotton, wool, silk, synthetic fiber, etc.), color, and pattern (solid color, striped, floral, etc.), which influence the garment's texture, comfort, and visual appeal. To obtain clothing customization parameters, users can use online questionnaires and offline measurements, using HTML forms and back-end data processing programs. Users fill in or select information such as body measurements and clothing style preferences on the front-end interface. The back-end uses Python's Flask framework to parse the data and store it in the database. The clothing customization parameters provide a comprehensive and accurate basis for the subsequent clothing customization process.

[0041] S200 , performing three-dimensional parameter modeling based on clothing customization parameters to determine a three-dimensional fitting model corresponding to the clothing customization parameters.

[0042] It can be understood that 3D parameter modeling is to transform clothing customization parameters into a visual 3D model. Figure 2 , allowing users to intuitively experience the effect of customized clothing on the upper body in advance. The acquired clothing customization parameters can be processed as input data through computer graphics algorithms and professional modeling software, such as open source libraries such as Open3D. Key information, such as human body size information and clothing style parameters, can be extracted from clothing customization parameters. Based on human body size information and clothing style parameters, a three-dimensional fitting model that can truly reflect the wearing state of clothing on the human body is gradually constructed. The three-dimensional fitting model can not only show the appearance of the clothing, but also simulate the changes in the shape of the clothing when the human body moves, providing strong support for subsequent design adjustments and user confirmation, effectively reducing the modification costs caused by unreasonable design, and improving the accuracy and efficiency of clothing customization.

[0043] In one possible implementation, S200, performing three-dimensional parameter modeling based on clothing customization parameters to determine a three-dimensional fitting model corresponding to the clothing customization parameters, includes:

[0044] S210, extracting human body size information and clothing style parameters based on clothing customization parameters.

[0045] As you can understand, the user's body measurements can be parsed from clothing customization parameters. For example, height is measured to the nearest centimeter, and chest circumference must be measured horizontally and snugly. These measurements represent human dimensions, which accurately reflect the user's body type. The user's selected clothing style parameters can be identified, such as whether it's a casual hooded sweatshirt or a formal business suit, along with details like collar and sleeve shape. These parameters determine the design and fit of the garment. String matching and data classification algorithms can be used to classify and filter the input clothing customization parameters, converting complex parameter information into structured data for easy processing. This allows the extraction of human dimensions and clothing style parameters, laying the foundation for building a clothing model that meets the user's needs.

[0046] S220 , extracting anatomical features based on the human body size information, and determining a coordinate set of anatomical feature points corresponding to the human body size information.

[0047] It's understandable that human anatomical features are closely linked to the fit of clothing. Extracting a coordinate set of anatomical feature points can help clothing better fit the human body. Based on the acquired body size information, characteristic points of key body parts, such as the acromion point on the shoulder, the nipple point on the chest, and the iliac crest point on the waist, can be determined according to human anatomy principles. Anatomical feature points represent key locations of the human skeleton and muscles, and their coordinates accurately reflect the body's morphological structure. Using three-dimensional measurement technology and mathematical transformation algorithms, the positions of these feature points in three-dimensional space can be quantified to generate a coordinate set of anatomical feature points. This coordinate set of anatomical feature points provides a precise reference framework for the subsequent construction of human body mechanics models and clothing pattern design, helping to improve clothing comfort and wearing quality, and avoiding garments that don't fit properly in key areas.

[0048] S230: constructing a human body mechanics model based on the anatomical feature point coordinate set, and generating a dynamic body surface model corresponding to the human body size information.

[0049] Understandably, the morphology of various parts of the human body changes during movement, and clothing needs to adapt to these changes. The construction of a human body mechanics model is based on a set of anatomical feature point coordinates. This takes into account the movement patterns and mechanical properties of human muscles and bones, and applies mechanical principles and mathematical models to simulate the morphological changes of the human body under different motion states. By using the finite element method to mesh the human body model based on the anatomical feature point coordinates, and incorporating the mechanical principles of human muscle and skeletal movement, a dynamic simulation algorithm is employed to calculate the deformation of the model under different motion states, generating a dynamic body surface model. For example, when a person raises their arm, the dynamic body surface model can display the surface deformation around the shoulder and arm. This provides a dynamic human morphology reference during the clothing design phase, ensuring that customized clothing not only fits well in static conditions but also maintains comfort and aesthetics during dynamic wear, thereby enhancing the practicality and wearing experience of the garment.

[0050] S240 , mapping clothing style parameters to a dynamic body surface model and performing differential coordinate deformation to determine a three-dimensional fitting model corresponding to the clothing customization parameters.

[0051] It can be understood that clothing style parameters, such as collar type, sleeve type, and skirt style, can be mapped to a dynamic body surface model through a specific mapping algorithm, allowing the clothing style to be presented on the human body model. Differential coordinate deformation is then performed, a precise deformation method based on tiny coordinate changes. By fine-tuning the coordinates of the surface of the dynamic body surface model, details such as the shape and folds of the clothing are made more consistent with the actual wearing effect. For example, for slim-fitting clothing, differential coordinate deformation can be used to adjust the fit between the clothing and the human body, allowing the clothing to better reflect the body's curves. By integrating clothing style parameters into the dynamic body surface model, a three-dimensional fitting model is ultimately generated that can intuitively display the wearing effect of the clothing, providing users with a more realistic fitting experience.

[0052] Optionally, S240, mapping clothing style parameters to a dynamic body surface model and performing differential coordinate deformation to determine a three-dimensional fitting model corresponding to the clothing customization parameters, includes:

[0053] S241, performing geometric feature decomposition on the clothing style parameters to obtain geometric constraint features corresponding to the clothing style parameters.

[0054] As you can understand, clothing style parameters contain a variety of complex design elements. These can be decomposed into geometric features, transforming these design elements into easily manageable geometric constraint features. This can be achieved through shape analysis and geometric parameter extraction algorithms. Taking a clothing collar as an example, an edge detection algorithm is used to extract the collar edge contour. Then, a curve fitting algorithm is used to obtain geometric parameters such as the curvature and length of the curve. These parameters are used as geometric constraint features to describe the geometric shape and position constraints of the clothing style. For example, for a garment with a unique collar design, its collar shape can be decomposed into the geometric features of the curve, such as curvature and length. Similarly, a garment's pocket design can be decomposed into geometric shape features, such as rectangles or circles, along with constraints such as their position and size on the garment. Through geometric feature decomposition, the abstract design of a clothing style is transformed into specific geometric parameters and constraint conditions, including geometric set constraint features. These geometric constraint features serve as the foundation for the subsequent accurate mapping of the clothing style onto a dynamic body surface model, ensuring that the clothing style's representation on the model meets the design requirements and maintains the garment's original style and characteristics.

[0055] S242 , mapping the geometric constraint features to the dynamic body surface model through geodesic projection to obtain a feature line path corresponding to the geometric constraint features.

[0056] As can be understood, geodesic projection is an effective method for finding the shortest path in surface space. Geometric constraints can be mapped onto a dynamic body surface model through geodesic projection, accurately determining the position and orientation of clothing styles on the body model. The dynamic body surface model is a three-dimensional surface. Geodesic projection is used to first identify regions of high curvature, followed by a shortest path search. A curvature calculation algorithm can be used to identify meshes in high-curvature regions of the dynamic body surface model. A shortest path search algorithm, such as Dijkstra or A*, combined with geometric constraints, searches for the shortest path from the starting point to the end point on the dynamic body surface model, obtaining feature line paths. The shortest paths from one point to another on the dynamic body surface model are then found. These paths are the feature line paths corresponding to the geometric constraints. For example, geodesic projection can be used to distribute a decorative line on a garment along the natural contours of the human body surface on the dynamic body surface model, ensuring the garment's aesthetics and integrity while also providing an accurate reference path for subsequent model deformation.

[0057] Exemplarily, S242, mapping the geometric constraint feature to the dynamic body surface model through geodesic projection to obtain a feature line path corresponding to the geometric constraint feature, includes:

[0058] S2421, identifying curvature-sensitive regions of the dynamic body surface model, and determining high-curvature region meshes of the dynamic body surface model.

[0059] It's understandable that the curvature of a dynamic body surface model varies across different locations. High-curvature areas often correspond to joints, bony protrusions, and other areas of the human body. These areas significantly impact the deformation and fit of clothing. By calculating the curvature values ​​at each point in the dynamic body surface model and utilizing a curvature calculation algorithm and threshold determination method, regions with curvature values ​​greater than a threshold can be identified. These high-curvature areas can be divided into a grid, forming a high-curvature area grid. For example, joints like the elbow and knee, due to their significant bending during movement and high surface curvature, are high-curvature areas. This helps ensure that clothing in these key areas better adapts to human movement during subsequent shortest path searches and model deformation, enhancing comfort and flexibility.

[0060] S2422: Perform a shortest path search based on the geometric constraint feature and the high curvature area grid to obtain a feature line path corresponding to the geometric constraint feature.

[0061] Understandably, finding an optimal path that conforms to geometric constraints on the complex surfaces of dynamic body models is crucial. This is the purpose of shortest path search. Geometric constraints, such as the start and end points and shape requirements of garment decorative lines, as well as mesh information in high-curvature regions, can be used to determine the search start and end points. Using these geometric constraints as the search start and end points, a shortest path search algorithm (such as Dijkstra or A*) is employed to find the shortest path from start to end on the dynamic body model. The presence of meshes in high-curvature regions can affect path selection, as garment deformation is more complex in these areas. The shortest path obtained through the search is the characteristic line path corresponding to the geometric constraints, ensuring that the garment style is properly arranged on the body model. Furthermore, as the body moves, the garment's lines and structure naturally follow the body's movements, avoiding unnatural wrinkles or stretching.

[0062] S243: construct a local deformation coordinate system based on the feature line path and adjust the dynamic body surface model through differential coordinate deformation to determine a three-dimensional fitting model corresponding to the clothing customization parameters.

[0063] It can be understood that in order to make the dynamic body surface model better present the effect of clothing style, a local deformation coordinate system can be constructed and differential coordinate deformation can be performed. Based on the feature line path, the coordinate axis direction of the local coordinate system is defined, that is, precise deformation operations can be performed on local areas related to clothing style. By generating differential coordinate constraints, the average offset of the model vertex relative to the neighborhood vertex is calculated to obtain a reference quantity, which reflects the degree of deformation of the local area. According to this reference quantity, the vertex displacement constraint equation is constructed, and the iterative solution is performed in the local deformation coordinate system to gradually adjust the vertex position of the dynamic body surface model to make the model more in line with the design requirements of the clothing style. The final three-dimensional fitting model can accurately show the wearing effect of the clothing on the human body, including details such as the fit of the clothing and the distribution of wrinkles, providing an intuitive reference for clothing customization.

[0064] Exemplarily, S243, constructing a local deformation coordinate system based on the feature line path and adjusting the dynamic body surface model through differential coordinate deformation to determine a three-dimensional fitting model corresponding to the clothing customization parameters, includes:

[0065] S2431, define the u-axis of the local coordinate system based on the tangent direction of the feature line path, define the v-axis based on the normal direction of the dynamic body surface model, and construct a local deformation coordinate system.

[0066] It can be understood that the local deformation coordinate system is a reference framework for adjusting the dynamic body surface model. The local deformation coordinate system can be constructed using vector calculation and coordinate system construction algorithms. The tangent direction of the feature line path reflects the extension direction of the clothing style on the surface. It can be defined as the u-axis, and the model can be stretched, shrunk, and other deformation operations along the u-axis. The normal direction of the dynamic body surface model is perpendicular to the surface of the surface. It can be defined as the v-axis, which can control the displacement and deformation of the model in the direction perpendicular to the surface. By defining the u-axis and v-axis, a two-dimensional local deformation coordinate system can be constructed. In the local deformation coordinate system, it is easier to make local adjustments to the model. For example, precise deformation operations can be performed on a specific part of the clothing, such as the neckline or cuffs, to ensure that the details of the clothing style can be accurately presented on the model, improving the authenticity and accuracy of the model.

[0067] S2432: Generate differential coordinate constraints based on the local deformation coordinate system, calculate the average offset of the dynamic body surface model vertices relative to the neighborhood vertices, and obtain a reference value of the differential coordinate deformation.

[0068] It can be understood that in a local deformation coordinate system, each vertex has a certain positional relationship with its neighboring vertices. This can be achieved through vertex displacement calculation and averaging algorithms. In the local deformation coordinate system, the displacement vector of each vertex and its neighboring vertices is calculated. The displacement difference is obtained through vector operations. These displacement differences are then averaged to obtain the average offset, which serves as a benchmark for differential coordinate deformation and is used to measure the degree of deformation in the local area. By calculating the average offset of the vertices of the dynamic body surface model relative to their neighboring vertices, a benchmark can be obtained that reflects the deformation trend of the local area. For example, in the wrinkles of clothing, the relative offset between vertices is large. By calculating the average offset, the degree of deformation can be quantified. This benchmark provides a standard for subsequent differential coordinate deformation, making the model deformation more orderly and controllable, avoiding excessive or uneven deformation, and ensuring the quality and accuracy of the model.

[0069] S2433, constructing a vertex displacement constraint equation through the reference quantity, iteratively solving the vertex displacement constraint equation in the local deformation coordinate system, and determining the three-dimensional fitting model corresponding to the clothing customization parameters.

[0070] As can be understood, equation construction and iterative solution algorithms can be used. Based on the relationship between the reference quantity and vertex displacement, vertex displacement constraint equations are constructed. Iterative algorithms such as the Gauss-Seidel method or the conjugate gradient method can be used to continuously adjust vertex positions in the local deformed coordinate system. Iterations are terminated when the change in vertex displacement falls below a preset threshold, resulting in a 3D fitting model that meets the garment customization parameters. The vertex displacement constraint equation reflects the relationship between vertex displacement and the reference quantity in the local deformed coordinate system. This equation is solved iteratively in the local deformed coordinate system, with each iteration adjusting the vertex position based on the previous calculation results, gradually bringing the model to meet the garment customization parameters. After multiple iterations, when the change in vertex displacement is minimal and meets the preset convergence criteria, the final 3D fitting model is determined. This 3D fitting model accurately demonstrates the effect of clothing on the human body, providing a reliable visualization basis for garment customization.

[0071] S300 , performing discrete mesh division on the three-dimensional fitting model to obtain a first triangular mesh topology corresponding to the three-dimensional fitting model.

[0072] It can be understood that discrete meshing is the process of converting a continuous three-dimensional fitting model into a discrete triangular mesh representation. The first triangular mesh topology describes the connection relationship and layout structure between these triangular meshes. Through discrete meshing, the surface of the three-dimensional fitting model can be divided into many small triangular facets, which are the basic units that make up the mesh. The first triangular mesh topology records the vertex information of each triangular facet and the adjacency relationship between them. Through this topological structure, various parts of the model can be quickly accessed and manipulated, facilitating further optimization of the model and generating clothing customization process sheets. It can also improve computing efficiency and ensure the smooth progress of the entire clothing customization process.

[0073] In one possible implementation, S300 , performing discrete mesh division on the three-dimensional fitting model to obtain a first triangular mesh topology corresponding to the three-dimensional fitting model, includes:

[0074] S310, performing triangulation based on the three-dimensional fitting model to generate an initial triangular mesh, and performing binary subdivision on triangular facets in the initial triangular mesh whose side lengths exceed a preset threshold, until the side lengths of all triangular facets are less than or equal to the threshold; wherein the triangular facets are the basic building blocks of the initial triangular mesh.

[0075] It can be understood that triangulation is to convert a three-dimensional fitting model into a triangular mesh, and through a specific triangulation algorithm, divide the model surface into multiple triangles to generate an initial triangular mesh. The Delaunay triangulation algorithm can be used to generate the initial mesh. The initial triangular mesh is constructed by calculating the Delaunay triangulation of the surface point set of the three-dimensional fitting model; the edge length comparison algorithm can be used to compare the edge length with the preset threshold. For triangular facets with side lengths exceeding the threshold, the bisection method is used to split them along the midpoint of the longest side, and the process is repeated until the side lengths of all triangular facets meet the requirements, ensuring that the density and accuracy of the triangular mesh meet the requirements, making subsequent processing of the model more accurate. For example, when performing physical simulation and cutting analysis, more accurate results can be obtained, thereby improving the quality of clothing customization.

[0076] S320 , optimizing the mesh vertex distribution of the subdivided triangular mesh to obtain a first triangular mesh topology corresponding to the three-dimensional fitting model.

[0077] It is understandable that the vertex distribution of the subdivided triangle mesh may not be uniform enough, which will affect the quality of the model and the subsequent processing effect, so it needs to be optimized. The positions of the triangle mesh vertices can be adjusted through mesh vertex distribution optimization algorithms, such as the Laplacian smoothing algorithm and the Taubin algorithm, to make the vertex distribution more uniform. During the optimization process, the new position of each vertex can be calculated based on the relationship between the vertex and the neighboring vertices, making the mesh surface smoother and reducing local unevenness. The optimized vertex distribution can improve the stability and computational efficiency of the model, and can more accurately reflect the mechanical properties of clothing when performing physical simulations; when generating a cutting process sheet, it can more accurately calculate the shape and size of the cutting piece, ensure the accuracy and consistency of clothing customization, and obtain a higher quality first triangle mesh topology.

[0078] S400 , performing physical constraint iteration on the first triangular mesh topology based on the clothing customization parameters to obtain a second triangular mesh topology corresponding to the three-dimensional fitting model.

[0079] It is understandable that clothing is subject to various physical forces during the wearing process. In order to more realistically simulate the wearing effect of clothing, the first triangular mesh topology can be iterated with physical constraints. Based on clothing customization parameters, such as fabric elasticity and weight, anisotropic mechanical constraints are imposed on the first triangular mesh topology, taking into account the differences in mechanical properties of clothing in different directions. The motion and deformation of mesh vertices under the action of physical forces can be described by constructing projected dynamic equations, and the vertex displacement field is solved through local-global alternating iterations. During the iterative process, the positions of the mesh vertices are continuously adjusted so that the mesh topology gradually conforms to the physical state of the clothing during actual wear, and finally the second triangular mesh topology is obtained. The second triangular mesh topology can more accurately reflect the actual shape of the clothing on the human body, providing a more reliable basis for the subsequent generation of customized process orders that meet actual wearing needs.

[0080] In one possible implementation, S400 performs physical constraint iteration on the first triangular mesh topology based on the clothing customization parameters to obtain a second triangular mesh topology corresponding to the three-dimensional fitting model, including:

[0081] S410, based on the clothing customization parameters, applies anisotropic mechanical constraints to the first triangular mesh topology and constructs the projected dynamic equations, and solves the vertex displacement field through local-global alternating iteration; wherein, the local iteration is used to calculate the ideal displacement of the vertex under the action of elastic force and gravity, and the global iteration is used to correct the vertex position through implicit surface constraints and update the mesh topology.

[0082] As can be understood, anisotropic mechanical constraints account for the mechanical differences in different directions of clothing fabrics, making simulations more realistic. Based on information such as the fabric's elastic modulus and Poisson's ratio from the garment customization parameters, elastic force constraints in different directions are determined. Mechanical models and iterative algorithms are used to establish an anisotropic elastic mechanical model based on the fabric parameters. Mechanical constraints are applied using the gravity formula, and projected dynamic equations are constructed to describe vertex motion. Local iterations employ the explicit Euler method to calculate ideal vertex displacements. Global iterations utilize implicit surface constraint equations to correct vertex positions and update the mesh topology, solving the vertex displacement field through alternating iterations. Projected dynamic equations are used to translate mechanical constraints into mathematical form, describing the motion and deformation of mesh vertices. Local iterations focus on the mechanical response of a single vertex and its neighborhood, calculating the ideal displacement of the vertex under elastic forces and gravity, and determining the local deformation. Global iterations, starting from the overall model, use implicit surface constraints to correct vertex positions, ensuring that the overall mesh shape conforms to the surface characteristics of the human body, and updating the mesh topology. Through the local-global alternating iteration method, the vertex displacement field is gradually solved, so that the mesh model can more realistically simulate the state of clothing in the physical environment, improving the solution efficiency and accuracy.

[0083] S420 , when the vertex displacement fields of consecutive preset number of iterations are all less than a preset convergence threshold, the iteration is stopped to obtain a second triangular mesh topology corresponding to the three-dimensional fitting model.

[0084] It is understandable that during the physical constraint iteration process, it is necessary to determine whether the iterations have converged to determine when to stop and obtain the final second triangular mesh topology. A convergence judgment algorithm can be used to preset a convergence threshold to measure the degree of change in the vertex displacement field. After each iteration, the change in the vertex displacement field is calculated. If the vertex displacement field is less than this convergence threshold for a preset number of consecutive times (such as 5 or 10 times), it indicates that the position changes of the mesh vertices have been minimal and the model has essentially reached a stable state. At this point, the iterations are stopped, and the resulting triangular mesh topology is the second triangular mesh topology. This second triangular mesh topology accurately reflects the final form of the garment under physical constraints, providing an accurate model foundation for the subsequent generation of custom process sheets, ensuring the accuracy and reliability of garment customization and avoiding garment production issues caused by inaccurate models.

[0085] S500: Obtaining a customization process sheet corresponding to the clothing customization parameters based on the second triangular mesh topology.

[0086] As you can see, the second triangle mesh topology is the key data foundation for generating custom process sheets, containing detailed information about the garment's structure and shape in three-dimensional space. Further processing can translate it into the technical parameters required for garment production, guiding the actual production of the garment. Custom process sheets are a crucial guide for garment production, encompassing specific requirements for every step of the process, from fabric cutting and sewing to ironing and packaging. This ensures garments are precisely manufactured according to custom parameters, achieving personalized clothing customization.

[0087] In one possible implementation, S500, based on the second triangular mesh topology, obtaining a custom process sheet corresponding to the clothing custom parameters includes:

[0088] S510 , generating a piece alignment mark on the second triangular mesh topology according to the anatomical feature point coordinate set, and flattening the second triangular mesh topology into a two-dimensional geometric outline through conformal mapping.

[0089] It can be understood that the second triangular mesh topology is the key data basis for generating customized process orders, and contains detailed structural and shape information of the garment in three-dimensional space. In the existing technology, the mass spring technology is usually used to simplify the garment into a combination of mass points and springs. When simulating garment deformation and determining cutting pieces, the conversion accuracy of complex human body surfaces and asymmetric cutting is insufficient. Therefore, a marker generation algorithm can be used first, and then a conformal mapping algorithm can be used to improve the conversion accuracy between the three-dimensional model and the two-dimensional cutting piece. Using the coordinate matching algorithm, the coordinates of the anatomical feature points are matched with the points on the second triangular mesh topology to generate cutting piece alignment marks. The second triangular mesh topology can be flattened into a two-dimensional geometric contour by using an algorithm based on discrete conformal mapping, such as the Laplace-Beltrami operator method, to maintain the shape and angle relationship of the graphic.

[0090] Optionally, S510, generating a piece alignment mark on the second triangular mesh topology according to the anatomical feature point coordinate set, and flattening the second triangular mesh topology into a two-dimensional geometric outline by conformal mapping, includes:

[0091] S511 , generating a piece alignment mark on the second triangular mesh topology according to the anatomical feature point coordinate set, and determining a piece coordinate position corresponding to the piece alignment mark.

[0092] It can be understood that the coordinate set of anatomical feature points corresponds closely to the key parts of the human body, and is crucial to the accuracy of piece splicing in clothing production. Through coordinate matching and position determination algorithms, coordinate transformation and nearest neighbor search algorithms, the coordinates of anatomical feature points can be mapped to the second triangular mesh topology, and the nearest grid point can be found as the piece alignment mark. According to the index and topological relationship of the grid point, the piece coordinate position corresponding to the piece alignment mark is determined to provide positioning for piece splicing. Each piece alignment mark corresponds to a specific piece coordinate position, and the piece coordinate position accurately defines the relative position relationship of the piece in the three-dimensional grid. By determining the piece alignment mark and its corresponding piece coordinate position, a clear positioning standard is provided for subsequent piece sewing, ensuring the accuracy of the various parts of the garment when splicing, and improving the production quality and fit of the garment.

[0093] S512 , flattening the second triangular mesh topology into a two-dimensional geometric outline according to the coordinate positions of the cutting pieces corresponding to the cutting piece alignment marks.

[0094] It can be understood that the coordinate positions of the pieces corresponding to the piece alignment marks provide a positioning basis for flattening the second triangular mesh topology. A conformal mapping algorithm, based on the principle of discrete conformal mapping, can be used to flatten the three-dimensional mesh into two dimensions by adjusting the positions of the mesh vertices while maintaining the internal angles. The Laplace-Beltrami operator is used to calculate vertex displacements, and the flattening process is constrained according to the piece coordinate positions to obtain a two-dimensional geometric outline. This two-dimensional geometric outline can intuitively display the shape of the garment piece, facilitating cutting, layout, and process planning. It can effectively reduce sewing problems caused by inaccurate piece shapes and improve the efficiency and quality of garment production.

[0095] S520: Input the two-dimensional geometric outline into a process order generation model to obtain a customized process order corresponding to the garment customization parameters. The process order generation model is a pre-trained machine learning model.

[0096] While a 2D geometric outline encompasses key information such as the shape and size of a garment piece, its conversion into a complete custom process sheet requires further processing with the aid of a process sheet generation model. This model is a machine learning model trained on a large amount of garment production data. It learns the relationships between different garment styles, fabrics, and production processes. The 2D geometric outline can be used as input, and based on this knowledge, the model automatically generates a richly detailed custom process sheet. This custom process sheet details specific process parameters such as fabric cutting method, sewing stitches and sequence, seam width, and ironing temperature and time. This provides comprehensive and accurate guidance for garment production, transforming garment customization parameters into executable processes and facilitating intelligent integration of the entire garment customization process, from style design to delivery.

[0097] For example, to obtain the actual production process sheet, please refer to Figure 4 , containing detailed information on the cutting, sewing, and ironing processes. The 2D geometric contour data corresponding to the process sheet is also collected, which can be obtained by digitizing the garment pattern design drawings. For example, data on various garments, such as shirts, dresses, and suits, is collected, each with multiple sizes and different fabric options to ensure data diversity. The collected 2D geometric contour data and process sheet are annotated, and the 2D geometric contours are associated with key information in the corresponding custom process sheet. The annotations include the shape, size, sewing sequence, stitching, seam width, ironing temperature, and time of the pieces. For example, the boundaries of various parts, such as the neckline and cuff pieces, are annotated on the 2D geometric contour, as well as information such as whether the stitching used for the neckline in the corresponding process sheet is a lockstitch or a hemming stitch, and the seam width is 0.5 cm. Based on the characteristics of the process sheet generation task, an appropriate machine learning model, such as a convolutional neural network (CNN) or a recurrent neural network (RNN), is selected. CNNs excel at processing image features and can effectively extract shape, structure, and other features from 2D geometric contour data. RNNs, on the other hand, are more capable of processing data with sequential characteristics and are suitable for processing the sequential information of process steps in process sheets. For example, CNNs are preferred for tasks that primarily rely on 2D geometric contour shapes to generate process sheets. Standardized data should be divided into a training set (60%-80%), a validation set (10%-20%), and a test set (10%-20%). The training set allows the model to learn the mapping between 2D geometric contours and customized process sheets. The validation set is used to evaluate model performance and adjust hyperparameters during training. The test set is used to assess the generalization ability of the trained model. Training parameters such as the learning rate, number of iterations, and batch size are also determined. The training set data is fed into the model, and the backpropagation algorithm is used to calculate the loss between the predicted results and the actual process sheet (e.g., the mean squared error loss function is used for regression tasks). The model weights and biases are adjusted based on the loss, and iterations are repeated. Evaluate the trained model using the test set, calculating metrics such as accuracy, recall, and mean squared error (MSE) for predicting process sheet parameters. For example, evaluate the accuracy of seam width prediction and the accuracy of stitch prediction. Optimize the model based on these evaluation results. Overfitting can be addressed through regularization, increasing the data size, or stopping training early. Underfitting can be addressed through increasing model complexity, such as by adding network layers or adjusting the convolution kernel size. After multiple optimizations, the optimal model is obtained and used for actual process sheet generation.

[0098] It should be understood that the size of the serial numbers of the steps in the above embodiments does not mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of this application.

[0099] Corresponding to the artificial intelligence-based clothing customization 3D simulation and process sheet automatic generation method described in the above embodiment, the embodiment of the present application also provides an artificial intelligence-based clothing customization 3D simulation and process sheet automatic generation system, and the various units of the system can implement the various steps of the artificial intelligence-based clothing customization 3D simulation and process sheet automatic generation method. Figure 4 The structural block diagram of the artificial intelligence-based clothing customization 3D simulation and process sheet automatic generation system provided in an embodiment of the present application is shown. For the sake of convenience, only the parts related to the embodiment of the present application are shown.

[0100] Reference Figure 4 The artificial intelligence-based clothing customization 3D simulation and process sheet automatic generation system includes:

[0101] An acquisition unit, used to acquire clothing customization parameters;

[0102] A modeling unit, configured to perform three-dimensional parameter modeling based on the clothing customization parameters and determine a three-dimensional fitting model corresponding to the clothing customization parameters;

[0103] a dividing unit, configured to perform discrete mesh division on the three-dimensional fitting model to obtain a first triangular mesh topology corresponding to the three-dimensional fitting model;

[0104] an iterative unit, configured to perform physical constraint iteration on the first triangular mesh topology based on the clothing customization parameters to obtain a second triangular mesh topology corresponding to the three-dimensional fitting model;

[0105] A generating unit is configured to obtain a customization process sheet corresponding to the clothing customization parameters based on the second triangular mesh topology.

[0106] It should be noted that the information interaction, execution process, etc. between the above-mentioned systems / units are based on the same concept as the method embodiment of this application. Their specific functions and technical effects can be found in the method embodiment section and will not be repeated here.

[0107] Those skilled in the art can clearly understand that, for the convenience and brevity of description, only the division of the above-mentioned functional units and modules is used as an example for illustration. In actual applications, the above-mentioned functions can be distributed and completed by different functional units and modules as needed, that is, the internal structure of the system can be divided into different functional units or modules to complete all or part of the functions described above. The functional units and modules in the embodiment can be integrated into one processing unit, or each unit module can exist physically alone, or two or more unit modules can be integrated into one unit. The above-mentioned integrated unit can be implemented in the form of hardware or in the form of software functional units. In addition, the specific names of the functional units and modules are only for the convenience of distinguishing each other and are not used to limit the scope of protection of this application. The specific working process of the units and modules in the above-mentioned system can refer to the corresponding process in the aforementioned method embodiment, which will not be repeated here.

[0108] The embodiment of the present application also provides an electronic device, Figure 5 This is a schematic diagram of the structure of an electronic device provided in one embodiment of the present application. Figure 5 As shown, the electronic device 6 of this embodiment includes: at least one processor 60 ( Figure 5 Only one is shown), at least one memory 61 ( Figure 5 Only one is shown in the figure) and a computer program 62 stored in the at least one memory 61 and executable on the at least one processor 60. When the processor 60 executes the computer program 62, the electronic device 6 implements the steps of any of the above-mentioned embodiments of the method for automatic generation of 3D simulation and process sheets for customized clothing based on artificial intelligence, or implements the functions of each unit in the above-mentioned system embodiments.

[0109] For example, the computer program 62 may be divided into one or more units, which are stored in the memory 61 and executed by the processor 60 to implement the present application. The one or more units may be a series of computer program instruction segments capable of implementing specific functions, and the instruction segments are used to describe the execution process of the computer program 62 in the electronic device 6.

[0110] The electronic device 6 can be a computing device such as a desktop computer, a notebook, a PDA, or a cloud server. The electronic device may include, but is not limited to, a processor 60 and a memory 61. Those skilled in the art will understand that Figure 5 It is only an example of the electronic device 6 and does not constitute a limitation on the electronic device 6. It may include more or fewer components than shown in the figure, or a combination of certain components, or different components. For example, it may also include input and output devices, network access devices, buses, etc.

[0111] The processor 60 may be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor may be a microprocessor or any conventional processor.

[0112] In some embodiments, the memory 61 may be an internal storage unit of the electronic device 6, such as a hard drive or memory of the electronic device 6. In other embodiments, the memory 61 may also be an external storage device of the electronic device 6, such as a plug-in hard drive, a Smart Media Card (SMC), a Secure Digital (SD) card, a flash memory card, etc. equipped on the electronic device 6. Furthermore, the memory 61 may include both an internal storage unit of the electronic device 6 and an external storage device. The memory 61 is used to store an operating system, application programs, a boot loader, data, and other programs, such as the program code of the computer program. The memory 61 may also be used to temporarily store data that has been output or is about to be output.

[0113] An embodiment of the present application further provides a computer-readable storage medium, wherein the computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the steps in any of the above method embodiments are implemented.

[0114] An embodiment of the present application provides a computer program product. When the computer program product is run on an electronic device, the electronic device implements the steps of any of the above method embodiments.

[0115] If the integrated unit is implemented as a software functional unit and sold or used as a standalone product, it can be stored in a computer-readable storage medium. Based on this understanding, the present application implements all or part of the process steps in the above-mentioned method embodiments by instructing the relevant hardware through a computer program. The computer program can be stored in a computer-readable storage medium. When executed by a processor, the computer program can implement the steps of each of the above-mentioned method embodiments. The computer program includes computer program code, which can be in source code form, object code form, executable file, or some intermediate form. The computer-readable medium can include at least: any entity or device capable of carrying computer program code to an electronic device, recording medium, computer memory, read-only memory (ROM), random access memory (RAM), electric carrier signals, telecommunication signals, and software distribution media. Examples include USB flash drives, removable hard drives, magnetic disks, or optical disks. In some jurisdictions, based on legislation and patent practice, computer-readable media cannot be electric carrier signals or telecommunication signals.

[0116] In the above embodiments, the description of each embodiment has its own focus. For parts that are not described or recorded in detail in a certain embodiment, reference can be made to the relevant description of other embodiments.

[0117] Those skilled in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0118] In the embodiments provided in the present application, it should be understood that the disclosed artificial intelligence-based clothing customization 3D simulation and process sheet automatic generation system / electronic device and method can be implemented in other ways. For example, the artificial intelligence-based clothing customization 3D simulation and process sheet automatic generation system / electronic device embodiment described above is merely schematic. For example, the division of the units is merely a logical function division. In actual implementation, there may be other division methods, such as multiple units or components that can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of the system or unit, which can be electrical, mechanical or other forms.

[0119] The units described as separate components may or may not be physically separate, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed across multiple network units. Some or all of these units may be selected to achieve the purpose of this embodiment according to actual needs.

[0120] The above-described embodiments are only used to illustrate the technical solutions of the present application, rather than to limit them. Although the present application has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. These modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the various embodiments of the present application, and should all be included in the scope of protection of the present application.

Claims

1. A method for 3D simulation and automatic generation of process sheets for clothing customization based on artificial intelligence, characterized in that: include: Get clothing customization parameters; Performing three-dimensional parameter modeling based on the clothing customization parameters to determine a three-dimensional fitting model corresponding to the clothing customization parameters; Performing discrete mesh division on the three-dimensional fitting model to obtain a first triangular mesh topology corresponding to the three-dimensional fitting model; Performing physical constraint iteration on the first triangular mesh topology based on the clothing customization parameters to obtain a second triangular mesh topology corresponding to the three-dimensional fitting model; Based on the second triangular mesh topology, a customization process sheet corresponding to the clothing customization parameters is obtained; wherein the customization process sheet is used to achieve intelligent integration of the entire process of clothing customization from style design to delivery; The performing three-dimensional parameter modeling based on the clothing customization parameters to determine a three-dimensional fitting model corresponding to the clothing customization parameters includes: Extracting human body size information and clothing style parameters based on the clothing customization parameters; Extracting anatomical features based on the human body size information to determine a coordinate set of anatomical feature points corresponding to the human body size information; Constructing a human body mechanics model based on the anatomical feature point coordinate set to generate a dynamic body surface model corresponding to the human body size information; Mapping the clothing style parameters to the dynamic body shape surface model and performing differential coordinate deformation to determine a three-dimensional fitting model corresponding to the clothing customization parameters; Obtaining a customization process sheet corresponding to the clothing customization parameters based on the second triangular mesh topology includes: generating a piece alignment mark on the second triangular mesh topology according to the anatomical feature point coordinate set, and flattening the second triangular mesh topology into a two-dimensional geometric outline through conformal mapping; Inputting the two-dimensional geometric outline into a process order generation model to obtain a customized process order corresponding to the garment customization parameters; wherein the process order generation model is a pre-trained machine learning model; Generating a piece alignment mark on the second triangular mesh topology according to the anatomical feature point coordinate set, and flattening the second triangular mesh topology into a two-dimensional geometric contour by conformal mapping, comprises: generating a piece alignment mark on the second triangular mesh topology according to the anatomical feature point coordinate set, and determining a piece coordinate position corresponding to the piece alignment mark; The second triangular mesh topology is flattened into a two-dimensional geometric outline according to the coordinate position of the cutting piece corresponding to the cutting piece alignment mark.

2. The method for 3D simulation and automatic generation of process sheets for clothing customization based on artificial intelligence according to claim 1, characterized in that: Mapping the clothing style parameters to the dynamic body shape surface model and performing differential coordinate deformation to determine a three-dimensional fitting model corresponding to the clothing customization parameters includes: Decomposing the clothing style parameters by geometric features to obtain geometric constraint features corresponding to the clothing style parameters; Mapping the geometric constraint feature to the dynamic body surface model through geodesic projection to obtain a characteristic line path corresponding to the geometric constraint feature; A local deformation coordinate system is constructed based on the characteristic line path, and the dynamic body surface model is adjusted through differential coordinate deformation to determine a three-dimensional fitting model corresponding to the clothing customization parameters.

3. The method for 3D simulation and automatic generation of process sheets for clothing customization based on artificial intelligence according to claim 2, characterized in that: Mapping the geometric constraint feature to the dynamic body surface model through geodesic projection to obtain a feature line path corresponding to the geometric constraint feature includes: Identifying curvature-sensitive regions of the dynamic body shape surface model and determining high-curvature region grids of the dynamic body shape surface model; A shortest path search is performed based on the geometric constraint feature and the high curvature area grid to obtain a characteristic line path corresponding to the geometric constraint feature.

4. The method for 3D simulation and automatic generation of process sheets for clothing customization based on artificial intelligence according to claim 2, characterized in that: The method of constructing a local deformation coordinate system based on the characteristic line path and adjusting the dynamic body surface model through differential coordinate deformation to determine a three-dimensional fitting model corresponding to the clothing customization parameters includes: A u-axis of a local coordinate system is defined based on a tangent direction of the characteristic line path, and a v-axis is defined based on a normal direction of the dynamic body surface model, to construct a local deformation coordinate system; Generating differential coordinate constraints according to the local deformation coordinate system, calculating the average offset of the vertices of the dynamic body surface model relative to the neighboring vertices, and obtaining a reference value of the differential coordinate deformation; A vertex displacement constraint equation is constructed using the reference quantity, and the vertex displacement constraint equation is iteratively solved in the local deformation coordinate system to determine a three-dimensional fitting model corresponding to the clothing customization parameters.

5. The method for 3D simulation and automatic generation of process sheets for clothing customization based on artificial intelligence according to claim 1, characterized in that: The discrete mesh division of the three-dimensional fitting model to obtain a first triangular mesh topology corresponding to the three-dimensional fitting model includes: Performing triangulation based on the three-dimensional fitting model to generate an initial triangular mesh, and performing bisection subdivision on triangular facets in the initial triangular mesh whose side lengths exceed a preset threshold, until the side lengths of all triangular facets are less than or equal to the threshold; wherein the triangular facets are the basic building blocks of the initial triangular mesh; The mesh vertex distribution of the subdivided triangular mesh is optimized to obtain a first triangular mesh topology corresponding to the three-dimensional fitting model.

6. The method for 3D simulation and automatic generation of process sheets for clothing customization based on artificial intelligence according to claim 1, characterized in that: The performing physical constraint iteration on the first triangular mesh topology based on the clothing customization parameters to obtain a second triangular mesh topology corresponding to the three-dimensional fitting model includes: Applying anisotropic mechanical constraints to the first triangular mesh topology based on the garment customization parameters and constructing a projected dynamics equation, and solving the vertex displacement field through local-global alternating iterations; wherein the local iteration is used to calculate the ideal displacement of the vertex under the action of elastic force and gravity, and the global iteration is used to correct the vertex position and update the mesh topology through implicit surface constraints; When the vertex displacement fields of consecutive preset number of iterations are all less than a preset convergence threshold, the iteration is stopped to obtain a second triangular mesh topology corresponding to the three-dimensional fitting model.

7. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein: When the processor executes the computer program, the method according to any one of claims 1 to 6 is implemented.

Images