Clothing template parameter reconstruction method and equipment based on three-dimensional design and storage medium

Abstract

The invention discloses a garment template parameter reconstruction method and device based on three-dimensional design and a storage medium, and relates to the technical field of garment template generation, and the method comprises the following steps: scanning a standardized garment template mannequin to obtain a standardized garment mannequin model, carrying out the feature point extraction of the standardized garment mannequin model, and carrying out the feature point extraction of the standardized garment mannequin model; obtaining standard feature point coordinate data; a model deformation correction algorithm is established, and personalized deformation of the standardized clothing mannequin model is achieved; establishing a mapping generation algorithm from the three-dimensional model to the two-dimensional clothing template, and marking the mapping generation algorithm as a template two-dimensional algorithm; a personalized human body model is obtained, and personalized garment template parameters are obtained based on a model deformation correction algorithm and a template two-dimensional algorithm; the method is used for solving the problem that in the process of generating a personalized garment template through a human body three-dimensional scanning technology in an existing garment template generation technology, only part of human body characteristic curves are selected to establish a relation with a standard garment template, and the generated garment template cannot ensure that any part of the body is fit.

Description

Technical Field

[0001] The present invention relates to the technical field of clothing pattern generation, and specifically to a method, device, and storage medium for reconstructing clothing pattern parameters based on 3D design. Background Art

[0002] Clothing pattern generation technology refers to the method and technical means of making clothing patterns according to clothing style diagrams or human body size data; clothing pattern generation technology is a crucial part in clothing design and production, which involves transforming design concepts into actual producible patterns.

[0003] With the development of technology, human body 3D scanning technology provides important data support for clothing pattern generation; through scanning to obtain 3D human body scanning data, human body characteristic parts and their measurement methods can be defined, and then models of characteristic parts can be established to express the relationship between the forms of human body characteristic parts and pattern curves; however, the existing technology for generating clothing patterns based on human body 3D scanning technology often scans the human body to obtain the body shape data of the human body, and then selects certain characteristic parameters, such as waist circumference, shoulder width, and chest circumference, etc., and uses these characteristic parameters to establish the connection with clothing patterns, such as the human waist circumference and the clothing waist circumference, and changes certain parameters of the clothing pattern based on the characteristic parameters of the human body, or generates patterns using existing clothing parameters. For example, in the patent application with the publication number CN115758503A, a clothing intelligent pattern making method based on similar pattern matching is disclosed. This solution generates a new pattern by automatically matching the pattern in the pattern library that is most similar to the input clothing style diagram and parameterizing and deforming the pattern, making full use of the existing clothing pattern information, and making the generation of patterns more simple and efficient; however, certain characteristic parameters of the human body or existing clothing, such as waist circumference, shoulder width, and chest circumference, although they can relatively accurately represent the human body form, cannot completely represent the human body form; and the clothing generated by using these parameters to change the clothing pattern will result in the appropriate position of the selected characteristic parameters, while the other positions are inappropriate. For example, if the waist circumference and chest circumference are selected, the generated clothing may fit at the waist circumference and chest circumference positions, but not fit at the position between the waist and the chest. Moreover, the human body form varies greatly, and the clothing patterns generated by only focusing on certain characteristic parts cannot ensure fitness at any part of the body; therefore, in the process of generating personalized clothing patterns through human body 3D scanning technology in the existing clothing pattern generation technology, only some human body characteristic curves are selected to establish the connection with the standard clothing pattern, and the generated clothing pattern cannot ensure fitness at any part of the body. Summary of the Invention

[0004] The present invention aims to solve at least one of the technical problems in the prior art to some extent. By scanning a standardized clothing sample mannequin, coordinate data of standard feature points are obtained; a model deformation correction algorithm is established to realize the personalized deformation of the standardized clothing mannequin model; a mapping generation algorithm from a three-dimensional model to a two-dimensional clothing sample is established, and personalized clothing sample parameters are obtained based on the model deformation correction algorithm and the sample two-dimensionalization algorithm. This is to solve the problem that in the process of generating personalized clothing samples through human body three-dimensional scanning technology in the existing clothing sample generation technology, only some human body feature curves are selected to establish a connection with the standard clothing sample, and the generated clothing samples cannot ensure a perfect fit at any part of the body.

[0005] To achieve the above object, in the first aspect, the present application provides a method for reconstructing clothing sample parameters based on three-dimensional design, including the following steps:

[0006] Scan a standardized clothing sample mannequin to obtain a standardized clothing mannequin model, and extract feature points from the standardized clothing mannequin model to obtain standard feature point coordinate data;

[0007] Establish a model deformation correction algorithm to realize the personalized deformation of the standardized clothing mannequin model;

[0008] Establish a mapping generation algorithm from a three-dimensional model to a two-dimensional clothing sample, denoted as the sample two-dimensionalization algorithm;

[0009] Obtain a personalized human body model, and obtain personalized clothing sample parameters based on the model deformation correction algorithm and the sample two-dimensionalization algorithm.

[0010] Further, scanning a standardized clothing sample mannequin to obtain a standardized clothing mannequin model includes the following sub-steps:

[0011] Use a three-dimensional scanner to scan a standardized clothing sample mannequin to obtain point cloud data of the standardized clothing sample mannequin, establish a three-dimensional model of the standardized clothing sample mannequin based on the point cloud data of the standardized clothing sample mannequin, and divide the three-dimensional model of the standardized clothing sample mannequin using triangular meshes. After completion, a standardized clothing mannequin model is obtained;

[0012] Place the standardized clothing mannequin model in a space rectangular coordinate system, define the horizontal vector direction from the left hand to the right hand of the model of the standardized clothing mannequin as the X-axis, the vertically upward direction in which the standardized clothing mannequin model stands as the Z-axis, and the directly forward direction faced by the standardized clothing mannequin model as the Y-axis.

[0013] Further, extracting feature points from the standardized clothing mannequin model to obtain standard feature point coordinate data includes the following sub-steps:

[0014] Using the layer cutting method, with the step size of the first layer cutting interval, horizontally scan and cut the standardized clothing mannequin model. The step size of the first layer cutting interval is set as dz, and the closed cross-section curves formed by the horizontal scan and cutting of the standardized clothing mannequin model are obtained, which are marked as the cross-section curve sequence;

[0015] Obtain the spatial coordinates of the symmetry centers of all the closed cross-section curves in the cross-section curve sequence, which are marked as the symmetry center coordinates; arrange all the closed cross-section curves in the cross-section curve sequence according to the corresponding symmetry center coordinates to obtain the sagittal plane cross-section curve of the standardized clothing mannequin model; the surface enclosed by the sagittal plane cross-section curve is the sagittal plane;

[0016] Extract the first feature points of the standardized clothing mannequin model based on the geometric features of the sagittal plane. The first feature points include: BP point, shoulder point, front neck point, back neck point, side neck point, armpit point, scapular prominence point, front waist point, and back waist point; extract the morphological feature curves based on the cross-section curve sequence and the first feature points. The morphological feature curves include: neckline, shoulder line, front center line, back center line, armhole line, chest line, waist line, hip line, bottom edge line, princess line, and side seam line;

[0017] Based on the morphological feature curves and the first feature points, equally divide all the morphological feature curves into n parts, and respectively obtain the equally divided points on each morphological feature curve, which are marked as the second feature points;

[0018] Mark the first feature points and the second feature points as the standard feature points; obtain the coordinate data of all the standard feature points, which are marked as the standard feature point coordinate data.

[0019] Furthermore, establish a model deformation correction algorithm to realize the personalized deformation of the standardized clothing mannequin model, including the following sub-steps:

[0020] The standardized clothing mannequin model contains T triangular meshes and the vertices of t triangular meshes;

[0021] Define any vertex of the triangular meshes contained in the standardized clothing mannequin model as vi, and the coordinates of vi are vi(xi, yi, zi), where i represents the i-th vertex of the triangular mesh, and i ≤ t;

[0022] Define the vertex of the triangular mesh directly connected to vi as vj, and the coordinates of vj are vj(xj, yj, zj), where j is the j-th vertex of the triangular mesh directly connected to vi, and j < t;

[0023] Define the calculation formula for the detailed information of vi as, Where li represents the detailed information of vi, xi represents the x-axis coordinate of vi, xj represents the x-axis coordinate of vj, N represents the number of vertices of the triangular network directly connected to vi; ωij represents the weight of vertex vj of the triangular network directly connected to vi.

[0024] Furthermore, establishing a model deformation correction algorithm to realize the personalized deformation of the standardized clothing mannequin model also includes the following sub-steps:

[0025] Set the constraint condition of the model deformation correction algorithm as: minimizing the sum of the differences in the detailed information of the vertices of the corresponding triangular networks before and after the personalized deformation of the standardized clothing mannequin model;

[0026] Establish an objective function for minimizing the sum of the differences in detailed information. The objective function for minimizing the sum of the differences in detailed information is as follows: Where f(x1, x2,..., xt) is the sum of the differences in detailed information, represents the detailed information of the vertices of the triangular network for the personalized deformation of the standardized clothing mannequin model; represents the detailed information of the vertices of the triangular network after the personalized deformation of the standardized clothing mannequin model, where xi’ represents the x-axis coordinate of vi after the personalized deformation of the standardized clothing mannequin model, and xj’ represents the x-axis coordinate of vj after the personalized deformation of the standardized clothing mannequin model;

[0027] Convert the objective function for minimizing the sum of the differences in detailed information into matrix form, defined as M. The formula for M is as follows: Let the standard feature point be vk before the personalized deformation of the standardized clothing mannequin model. The coordinates of vk are vk(xk, yk, zk), and vk after the personalized deformation of the standardized clothing mannequin model is a specified known point, denoted as pk. The coordinates of pk are (x’k, y’k, z’k). k represents the kth standard feature point. Establish a column vector B. The form of B is as follows: Based on the point vi(xi, yi, zi) of the triangular network before the personalized deformation of the standardized clothing mannequin model, solve the equation M*vi = B to obtain the coordinate data of the vertices of the triangular network after the personalized deformation of the standardized clothing mannequin model, marked as personalized model vertex data.

[0028] Furthermore, establish a mapping generation algorithm from the 3D model to the 2D clothing pattern, marked as the pattern 2D conversion algorithm, including the following sub-steps:

[0029] Divide the standardized clothing mannequin model along the structural lines of the clothing pattern to obtain Q 3D clothing pattern pieces of the standardized clothing mannequin model. The structural lines of the clothing pattern include the neckline, shoulder line, front center line, back center line, armhole line, bottom line, and side seam line.

[0030] Further, a mapping generation algorithm from a three-dimensional model to a two-dimensional garment pattern is established. The algorithm for flattening the pattern into a two-dimensional form also includes the following sub-steps:

[0031] Based on the three-dimensional garment pattern piece, consider the vertices of the triangular mesh of the three-dimensional garment pattern piece as mass points, regard the area of the triangle in the mass point domain as the mass of the mass point, consider the mesh edges of the triangular mesh as springs, establish a mass point-spring model of the three-dimensional garment pattern piece, calculate the mass and acting force of each mass point, keep the mass of each mass point unchanged, use the XY plane as a constraint, set the z-axis coordinate value of the mass point to 0, and perform time integration iteration of the mass point-spring model. When the total internal energy of the mass point-spring model reaches the minimum within the XY plane, the two-dimensional graph at this time is the two-dimensional garment pattern piece after the three-dimensional garment pattern piece is flattened.

[0032] Further, to obtain a personalized human body model and get personalized garment pattern parameters based on the model deformation correction algorithm and the pattern two-dimensionalization algorithm, the following sub-steps are included:

[0033] A three-dimensional scanner acquires the human body surface model, denoted as the personalized human body model, extracts feature point coordinates from the personalized human body model to obtain the standard feature point coordinate data of the personalized human body model. Then, based on the standard feature point coordinate data of the personalized human body model, use the model deformation correction algorithm to realize the personalized deformation of the standardized garment form model, obtain the personalized garment model. Then, segment the personalized garment model according to the structure lines of the garment pattern to get the three-dimensional garment pattern pieces of the personalized garment model. Then, use the mapping generation algorithm from the three-dimensional model to the two-dimensional garment pattern to obtain the two-dimensional personalized garment pattern, and extract the garment pattern parameters of the personalized garment pattern.

[0034] In a second aspect, the present application provides an electronic device, including a processor and a memory. The memory stores computer-readable instructions. When the computer-readable instructions are executed by the processor, the steps in the above method are run.

[0035] In a third aspect, the present application provides a storage medium, on which a computer program is stored. When the computer program is executed by a processor, the steps in the above method are run.

[0036] Advantages of the present invention: By scanning the standardized garment form mannequin, the present invention obtains the standardized garment form model, extracts feature points from the standardized garment form model to obtain the standard feature point coordinate data; establishes a model deformation correction algorithm to realize the personalized deformation of the standardized garment form model; establishes a mapping generation algorithm from the three-dimensional model to the two-dimensional garment pattern, denoted as the pattern two-dimensionalization algorithm; obtains the personalized human body model, and gets the personalized garment pattern parameters based on the model deformation correction algorithm and the pattern two-dimensionalization algorithm; the generated garment pattern can ensure a perfect fit at any part of the body.

[0037] The present invention establishes a model deformation correction algorithm. By obtaining the changes in the body characteristic parameters of the customizer before and after, all parts of the standardized clothing mannequin model are deformed to ensure that the obtained personalized human body model can truly reflect the morphological characteristics of any part of the customizer's body. Furthermore, a clothing pattern is generated, which can ensure a perfect fit at any part of the customizer's body. By establishing a particle spring model to flatten the personalized human body model, the quality of the model unfolding can be improved. Moreover, this model can judge the error before and after energy release, reduce the accumulation of errors, ensure the topological integrity of the initial unfolding plane, make the obtained unfolding plane more accurate, and the algorithm steps of this model are simple, improving the efficiency of flattening and thus the overall efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0038] Figure 1 is a flowchart of the steps of the method of the present invention;

[0039] Figure 2 is a schematic diagram of a cross-sectional curve sequence and a sagittal plane structure of the present invention

[0040] Figure 3 is a schematic diagram of the structure of the particle spring model of the present invention;

[0041] Figure 4 is a schematic diagram of the structure of the electronic device of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0042] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention.

[0043] Embodiment 1. Please refer to Figure 1 As shown, the present application provides a method for reconstructing clothing pattern parameters based on three-dimensional design, including the following steps:

[0044] Step S1: Scan the standardized clothing pattern mannequin to obtain a standardized clothing mannequin model, and extract feature points from the standardized clothing mannequin model to obtain standard feature point coordinate data. Step S1 includes the following sub-steps:

[0045] Step S101: Use a 3D scanner to scan a standardized clothing sample mannequin to obtain the point cloud data of the standardized clothing sample mannequin. In this embodiment, a SmartScan handheld scanner is used for scanning. This device is based on the principle of multi-angle stereo vision and uses ECHO software for 3D point cloud reconstruction of multi-angle images, registration of 3D point cloud models, as well as triangle mesh generation and data processing operations. A mannequin refers to a human model made according to human proportions, and the standardized clothing sample mannequin correspondingly refers to a standard clothing sample.

[0046] Step S102: Based on the point cloud data of the standardized clothing sample mannequin, establish a 3D model of the standardized clothing sample mannequin, and use triangle meshes to divide the 3D model of the standardized clothing sample mannequin. After completion, obtain the standardized clothing mannequin model.

[0047] Step S103: Place the standardized clothing mannequin model in a spatial rectangular coordinate system. Define the horizontal vector direction from the left hand to the right hand of the standardized clothing mannequin model as the X-axis, the vertically upward direction in which the standardized clothing mannequin model stands as the Z-axis, and the directly forward direction that the standardized clothing mannequin model faces as the Y-axis.

[0048] Step S104: Adopt the layer cutting method and horizontally scan and cut the standardized clothing mannequin model with the first layer cutting interval step size. The first layer cutting interval step size is set as dz, and obtain the closed cross-section curves formed by the horizontal scan and cutting of the standardized clothing mannequin model, which are marked as the cross-section curve sequence.

[0049] Step S105: Refer to Figure 2 As shown, obtain the spatial coordinates of the symmetry centers of all the closed cross-section curves in the cross-section curve sequence, which are marked as the symmetry center coordinates. Arrange all the closed cross-section curves in the cross-section curve sequence according to the corresponding symmetry center coordinates to obtain the sagittal plane cross-section curve of the standardized clothing mannequin model. The surface enclosed by the sagittal plane cross-section curve is the sagittal plane. The sagittal plane refers to all the cross-sections that longitudinally cut the human body into left and right parts in the front-back direction. Among them, the cross-section that divides the human body into two equal left and right halves is called the median sagittal plane. In this embodiment, the sagittal plane is the median sagittal plane, and the median sagittal plane must pass through all the closed cross-section curves according to the corresponding symmetry centers.

[0050] Step S106: Extract the first feature points of the standardized clothing mannequin model based on the geometric features of the sagittal plane. The first feature points include: the BP point, shoulder point, front neck point, back neck point, side neck point, armpit point, scapular prominence point, front waist point, and back waist point. The BP point refers to the center point of the human nipple. The shoulder point refers to the outermost protruding point on the lateral edge of the acromion of the scapula. The front neck point refers to the midpoint of the connection line of the upper edges of the sternal ends of the left and right collarbones. The back neck point refers to the tip of the spinous process of the seventh cervical vertebra. The side neck point refers to the intersection point of the curve connecting the supraclavicular fossa point and the cervical vertebra point on the anterior edge of the trapezius muscle and the lateral cervical part on the lateral cervical triangle. The armpit point refers to the lowest endpoint of the attachment of the pectoralis major muscle on the anterior axillary fold. The scapular prominence point R is the outermost point on the outer edge of the scapula. The back waist point refers to the point on the back waist corresponding to the point with the largest degree of curvature of the trunk part between the lowest rib and the buttocks of the human body. The front waist point refers to the point on the front waist corresponding to the point with the largest degree of curvature of the trunk part between the lowest rib and the buttocks of the human body.

[0051] Step S107: Extract the morphological feature curves based on the cross-sectional curve sequence and the first feature points. The morphological feature curves include: the neckline, shoulder line, front center line, back center line, armhole line, chest line, waist line, hip line, bottom line, princess line, and side seam line. The neckline is the reference line for measuring the neck circumference of the human body, including the front neck point, back neck point, and side neck point. The shoulder line refers to the line at the junction of the human shoulder and the arm, including the side neck point and the shoulder point. The front center line is the center line of the front part of the human body, passing from the neck through the chest to the center of the abdomen, including the front neck point and the front waist point. The back center line is the center line of the back part of the human body, passing from the neck through the back to the center of the waist, including the back neck point and the back waist point. The armhole line is the part connecting the shoulder and the sleeve, including the shoulder point and the armpit point. The chest line is the reference line for measuring the chest circumference size of the human body, including the BP point. The waist line is the reference line for measuring the waist circumference size of the human body, including the front waist point and the back waist point. The hip line is the reference line for measuring the hip circumference size of the human body. The bottom line is the edge line of the hem or bottom of the clothing. The princess line is the dividing line extending from the shoulder through the highest point of the chest to the edge line, including the BP point and the scapular prominence point. The side seam line is the sewing line on the side of the clothing, including the armpit point.

[0052] Step S108: Based on the morphological feature curves and the first feature points, divide all the morphological feature curves into n equal parts, and respectively obtain the equal division points on each morphological feature curve, which are marked as the second feature points. In this embodiment, n is 8, that is, the morphological feature curves are divided into 8 equal parts. Then, for the equal division points on the morphological feature curves, for example, on the waist line with the front waist point and the back waist point, it needs to be divided into 8 equal parts, and 6 more points need to be taken. So, a total of 8 points are taken on the waist line, including two first feature points and two second feature points.

[0053] Step S109: Mark the first feature points and the second feature points as standard feature points; obtain the coordinate data of all the standard feature points, which are marked as the standard feature point coordinate data.

[0054] In the specific implementation process, the three-dimensional model of the standardized clothing sample mannequin is divided. Since the triangular mesh is composed of vertices, edges, and faces, where vertices are the most basic elements; each vertex contains spatial coordinate information, each edge is defined by two vertices, and each face is a triangle composed of three vertices; this simple data structure makes the triangular mesh easy to store, calculate, and process; any polygon mesh can be converted into a triangular mesh, and many operations are easier for triangular meshes compared to general polygon meshes.

[0055] Step S2, establish a model deformation correction algorithm to achieve the personalized deformation of the standardized clothing mannequin model; Step S2 includes the following sub-steps:

[0056] Step S201, the standardized clothing mannequin model contains T triangular meshes and the vertices of t triangular meshes; the number of triangular meshes and vertices depends on the size of the model and the size of the triangular meshes used for division;

[0057] Define the vertex of any triangular mesh in the triangular meshes contained in the standardized clothing mannequin model as vi, and the coordinates of vi are vi(xi, yi, zi), where i represents the vertex of the i-th triangular mesh, and i ≤ t;

[0058] Step S202, define the vertex of the triangular mesh directly connected to vi as vj, and the coordinates of vj are vj(xj, yj, zj), where j is the vertex of the j-th triangular mesh directly connected to vi, and j < t;

[0059] Step S203, define the calculation formula for the detail information of vi as, where li represents the detail information of vi, xi represents the x-axis coordinate of vi, xj represents the x-axis coordinate of vj, N represents the number of vertices of the triangular meshes directly connected to vi; ωij represents the weight of the vertex vj of the triangular mesh directly connected to vi; ωij can be set by itself, generally using uniform weights, that is, the weights of the vertices of the triangular meshes directly connected to vi are equal, and the sum of the weights is equal to 1. For example, if there are 4 vertices of the triangular meshes directly connected to vi, then each weight is 0.25;

[0060] Step S204, set the constraint condition of the model deformation correction algorithm as: minimizing the sum of the differences in the detail information of the corresponding vertices of the triangular meshes before and after the personalized deformation of the standardized clothing mannequin model;

[0061] Step S205, establish an objective function for minimizing the sum of the differences in detail information. The objective function for minimizing the sum of the differences in detail information is as follows: where f(x1, x2,..., xt) is the sum of the differences in detail information, Vertex detail information of the triangular network representing the personalized deformation of the standardized clothing dummy model; Vertex detail information of the triangular network after the personalized deformation of the standardized clothing dummy model, where xi’ represents the x-axis coordinate of vi after the personalized deformation of the standardized clothing dummy model, and xj’ represents the x-axis coordinate of vj after the personalized deformation of the standardized clothing dummy model; for example, t = 4, l1 = 0.25, l2 = 0.3, l3 = 0.15, l4 = 0.5, l’1 = 0.35, l’2 = 0.23, l’3 = 0.42, l’4 = 0.23, then f(x1, x2, x3, x4) = (0.35 - 0.25) 2 +(0.23 - 0.3) 2 +(0.42 - 0.15) 2 +(0.23 - 0.5) 2 = 0.1662;

[0062] Step S206, convert the objective function that minimizes the sum of detail information differences into a matrix form, defined as M, and the formula of M is as follows: For example, i = 1, j = 2, there are 4 vertices directly connected to vi, and the weight ωij of each vertex is 0.25; if i is directly connected to j, then M12 = -0.25; let the standard feature point be vk before the personalized deformation of the standardized clothing dummy model, the coordinates of vk are vk(xk, yk, zk), and vk after the personalized deformation of the standardized clothing dummy model is the specified known point, denoted as pk, the coordinates of pk are (x’k, y’k, z’k), k represents the kth standard feature point, establish a column vector B, and the form of B is as follows: Solve the equation M * vi = B based on the vertex vi(xi, yi, zi) of the triangular network before the personalized deformation of the standardized clothing dummy model to obtain the coordinate data of the vertices of the triangular network after the personalized deformation of the standardized clothing dummy model, marked as personalized model vertex data;

[0063] In the specific implementation process, based on the constraint of the sum of node information differences, the new coordinates of all vertices of the entire model are calculated, realizing the deformation from the standardized clothing mannequin model to the personalized clothing model. The deformed model has the same mesh structure as the original model, similar geometric morphology to the personalized human body scan model, and maintains approximate human body shape feature points and mesh topology structure. The reason for not using a 3D scanner to scan the human body and establish a clothing pattern is that the clothes worn daily do not fit perfectly to the skin and mostly have some allowance. The clothes made by directly scanning the human body to establish a clothing pattern can only be tight-fitting clothes, which are not suitable for wearing. The standardized clothing mannequin model represents the standard clothing pattern. By calculating and deforming it, the clothing pattern obtained through subsequent processing is a fitting and wearable clothing pattern.

[0064] Step S3, establish a mapping generation algorithm from the 3D model to the 2D clothing pattern, denoted as the pattern 2D conversion algorithm; Step S3 includes the following sub-steps:

[0065] Step S301, divide the standardized clothing mannequin model according to the structural lines of the clothing pattern to obtain Q 3D clothing pattern pieces of the standardized clothing mannequin model. The structural lines of the clothing pattern include the neckline, shoulder line, front center line, back center line, armhole line, bottom edge line, and side seam line; The structural line refers to the general term for the external and internal sewing lines of clothing parts that can cause changes in clothing styling. Since the clothing styles are different, the required structural lines for cutting are also different, and can be selected according to the actual scenario. For example, for a shirt, the structural lines involved in the division include the neckline, shoulder line, front center line, back center line, armhole line, bottom edge line, and side seam line;

[0066] Step S302, please refer to Figure 3 As shown, based on the 3D clothing pattern piece, regard the vertices of the triangular mesh of the 3D clothing pattern piece as mass points, regard the area of the triangle in the mass point domain as the mass of the mass point, regard the mesh edges of the triangular mesh as springs, and establish a mass point spring model of the 3D clothing pattern piece;

[0067] Step S303, calculate the mass and acting force of each mass point, keep the mass of each mass point unchanged, take the XY plane as the constraint, set the z-axis coordinate value of the mass point to 0, and perform time integral iteration of the mass point spring model;

[0068] The strain energy generated during the spring expansion is expressed as: The force situation at a certain point during the expansion is expressed as: Where K is the elastic deformation coefficient. For the convenience of calculation, K can be taken as 1; d1 is the distance between the two points connected by the spring in the expansion plane; d2 is the actual distance between the two points connected by the spring in the initial 3D space; A is a vector representing the direction of the spring elastic force; u represents the u-th spring;

[0069] Let Δt be the specified unit time increment. When Δt is infinitesimal, the acceleration of a certain particle V during optimization on the particle spring model can be regarded as a constant. Then we have

[0070]

[0071] vp(t + Δt) = vp(t) + Δt * ap(t)

[0072] where mp represents the mass of a certain particle V, ρ is the triangular mesh density of the 3D garment pattern piece to be unfolded, Si represents the areas of all the triangles where particle V is located in the 3D garment pattern piece, and i represents the i-th triangular mesh; ap(t) represents the acceleration of particle V at time t; Fp(t) represents the spring strain force acting on particle V at time t; vp(t) represents the velocity of particle V at time t; Lp(t) represents the position of particle V at time t;

[0073] The overall process of optimizing the particle spring model is as follows: Set the initial unit time increment Δt to be optimized. Then calculate the masses of all the particles in the initial unfolding result; calculate the forces acting on all the particles in the initial unfolding result; calculate the accelerations of all the particles in the initial unfolding result; calculate the velocities and positions of all the particles on the shoe surface after the current iteration ends; check whether the average value DC of the side length errors in the triangulated mesh is less than the average value DL of the side length errors in the triangulated mesh after the previous iteration optimization. If it holds, use to replace Δt; if it does not hold, do not replace it. When the preset number of optimization iterations is reached, this process terminates;

[0074] Step S304. When the particle spring model reaches the minimum total internal energy within the particle spring model in the XY plane, the two-dimensional graph at this time is the two-dimensional garment pattern piece after the 3D garment pattern piece is flattened;

[0075] In the specific implementation process, the mass of a particle is determined by the area of the triangle it belongs to. Whether the spring is in a stretched or compressed state is determined by comparing the actual length of the edge in the unfolding result of the 3D garment pattern piece with the original length of the spatial edge in the unfolding of the 3D garment pattern piece. If the former is longer than the latter, the spring is in a stretched state; if the former is shorter than the latter, the spring is in a compressed state; if the former is equal to the latter, the spring is in an unloaded state. The significance of establishing this model is to release the spring energy along the direction opposite to the force direction of the stretched or compressed spring. By continuously iteratively releasing the spring energy, a particle spring model approaching an equilibrium state is finally obtained. At this time, it is considered that the unfolding result of the 3D garment pattern piece has obtained an ideal optimization result.

[0076] Step S4: Obtain a personalized human body model, and obtain personalized clothing pattern parameters based on the model deformation correction algorithm and the pattern two-dimensionalization algorithm. Step S4 includes the following sub-steps:

[0077] Step S401: A 3D scanner obtains a human body surface model, marked as a personalized human body model, and extracts feature points from the personalized human body model to obtain the standard feature point coordinate data of the personalized human body model. The standard feature points include the first standard point and the second standard point

[0078] Step S402: Then, based on the standard feature point coordinate data of the personalized human body model, use the model deformation correction algorithm to realize the personalized deformation of the standardized clothing mannequin model and obtain a personalized clothing model. That is, based on the sum of the node information differences, calculate the new coordinates of all vertices of the personalized human body model except the standard feature points, realizing the deformation from the standardized clothing mannequin model to the personalized clothing model. The deformed model has the same mesh structure as the original model;

[0079] Step S403: Then, segment the personalized clothing model according to the structure lines of the clothing pattern to obtain the 3D clothing pattern pieces of the personalized clothing model. The structure lines include

[0080] Step S404: Then, use the mapping generation algorithm from the 3D model to the 2D clothing pattern to obtain the 2D personalized clothing pattern and extract the clothing pattern parameters of the personalized clothing pattern;

[0081] In the specific implementation process, when using a 3D scanner to obtain a human body surface model, only the feature points and the nearby models can be obtained to improve the efficiency of obtaining the clothing pattern.

[0082] Example 2, please refer to Figure 4 as shown Figure 4 illustrates a schematic structural diagram of an electronic device. The electronic device may include: a processor, a communication interface, a memory, and a communication bus. Among them, the processor, the communication interface, and the memory complete mutual communication through the communication bus. The memory stores computer-readable instructions, and the processor can call the instructions in the memory. When the computer-readable instructions are executed by the processor, the steps in the clothing pattern parameter reconstruction method based on 3D design are run to achieve the following functions: scan the standardized clothing pattern mannequin to obtain a standardized clothing mannequin model, extract feature points from the standardized clothing mannequin model to obtain standard feature point coordinate data; establish a model deformation correction algorithm to realize the personalized deformation of the standardized clothing mannequin model; establish a mapping generation algorithm from the 3D model to the 2D clothing pattern, marked as the pattern two-dimensionalization algorithm; obtain a personalized human body model, and obtain personalized clothing pattern parameters based on the model deformation correction algorithm and the pattern two-dimensionalization algorithm.

[0083] In addition, when the logical instructions in the above-mentioned memory are implemented in the form of software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on such an understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or a part of this technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions for causing a computer device (which may be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods described in various embodiments of this application. The aforementioned storage medium includes: various media that can store program codes, such as USB flash drives, mobile hard disks, read-only memories (ROM, Read-Only Memory), random access memories (RAM, Random Access Memory), magnetic disks, or optical discs.

[0084] Embodiment 3. This application also provides a computer-readable storage medium. This application provides a storage medium on which a computer program is stored. When the computer program is executed by a processor, it runs the steps in the above-mentioned method for reconstructing clothing sample parameters based on 3D design to achieve the following functions: scanning a standardized clothing sample mannequin to obtain a standardized clothing mannequin model, extracting feature points from the standardized clothing mannequin model to obtain standard feature point coordinate data; establishing a model deformation correction algorithm to achieve personalized deformation of the standardized clothing mannequin model; establishing a mapping generation algorithm from a 3D model to a 2D clothing sample, marked as a sample 2D conversion algorithm; obtaining a personalized human body model, and obtaining personalized clothing sample parameters based on the model deformation correction algorithm and the sample 2D conversion algorithm.

[0085] Through the description of the above embodiments, the embodiments of the present invention can be provided as a method, a system, or a computer program product. Based on such an understanding, the above technical solution, in essence, or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, magnetic disks, optical discs, etc., and includes several instructions for causing a computer device (which may be a personal computer, a server, or a network device, etc.) to execute the methods described in various embodiments or certain parts of the embodiments.

[0086] In the embodiments provided in the present application, it should be understood that the disclosed system or method can be implemented in other ways. The embodiments described above are merely illustrative. For example, the division of modules or units is only a logical function division, and there may be other division methods in actual implementation. For another example, multiple modules or units can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the displayed or discussed coupling or direct coupling or communication connection between each other can be through some communication interfaces. The indirect coupling or communication connection of systems, modules, and units can be in electrical, mechanical, or other forms.

[0087] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present application and are not intended to limit them. Although the present application has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that they can still modify the technical solutions described in the foregoing embodiments or perform equivalent replacements for some of the technical features. However, these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present application.

Claims

1. A method for reconstructing clothing pattern parameters based on three-dimensional design, characterized in that: The steps include: Scanning a standardized clothing sample mannequin to obtain a standardized clothing mannequin model, extracting feature points of the standardized clothing mannequin model, and obtaining standard feature point coordinate data; Establish a model deformation correction algorithm to achieve personalized deformation of standardized clothing mannequin models; Establish a mapping generation algorithm from a 3D model to a 2D clothing pattern, labeled as a pattern 2D algorithm; Obtain a personalized human body model, and obtain personalized clothing sample parameters based on the model deformation correction algorithm and the sample two-dimensional algorithm.

2. The method for reconstructing clothing pattern parameters based on three-dimensional design according to claim 1, characterized in that: Scanning a standardized clothing sample mannequin to obtain a standardized clothing mannequin model includes the following sub-steps: Scanning a standardized clothing model mannequin with a 3D scanner to obtain point cloud data of the standardized clothing model mannequin, establishing a 3D model of the standardized clothing model mannequin based on the point cloud data of the standardized clothing model mannequin, and dividing the 3D model of the standardized clothing model mannequin with a triangular mesh, and obtaining a standardized clothing model after completion; Place the standardized clothing mannequin model in a rectangular coordinate system, define the horizontal vector direction from the left hand to the right hand of the standardized clothing mannequin model as the X-axis, the vertical upward direction of the standardized clothing mannequin model as the Z-axis, and the forward direction of the standardized clothing mannequin model as the Y-axis.

3. The method for reconstructing clothing pattern parameters based on three-dimensional design according to claim 2, characterized in that: Extracting feature points from a standardized clothing mannequin model to obtain standard feature point coordinate data includes the following sub-steps: The layer cutting method is adopted, and the standardized clothing mannequin model is horizontally scanned and cut with the first layer cutting interval step length, and the first layer cutting interval step length is set to dz, and the closed section curve formed by the horizontal scanning and cutting of the standardized clothing mannequin model is obtained, and marked as a section curve sequence; The spatial coordinates of the symmetry centers of all closed section curves in the section curve sequence are obtained, and are marked as the symmetry center coordinates; all closed section curves in the section curve sequence are arranged according to the corresponding symmetry center coordinates, and the sagittal section curves of the standardized clothing mannequin model are obtained; the surface enclosed by the sagittal section curves is the sagittal plane; Extract the first feature points of the standardized clothing mannequin model based on the geometric features of the sagittal plane, the first feature points including: BP point, shoulder point, front neck point, back neck point, side neck point, armpit point, shoulder blade convex point, front waist point and back waist point; extract the morphological feature curve based on the cross-sectional curve sequence and the first feature point; The morphological characteristic curves include: neckline, shoulder line, front center line, back center line, armhole line, chest line, waist line, hip line, bottom line, princess line and side seam line; Based on the morphological characteristic curves and the first characteristic points, all the morphological characteristic curves are divided into n equal parts, and the equal division points on each morphological characteristic curve are obtained respectively and marked as second characteristic points; The first feature point and the second feature point are marked as standard feature points; and the coordinate data of all the standard feature points are obtained and marked as standard feature point coordinate data.

4. The method for reconstructing clothing pattern parameters based on three-dimensional design according to claim 3, characterized in that: Establishing a model deformation correction algorithm to achieve personalized deformation of a standardized clothing mannequin model includes the following sub-steps: The standardized clothing mannequin model includes T triangle networks and t triangle network vertices; Define any vertex of a triangular network in the triangular network included in the standardized clothing mannequin model as vi, and the coordinates of vi are vi(xi, yi, zi), where i represents the vertex of the i-th triangular network, i≤t; Define the vertex of the triangular network directly connected to vi as vj, and the coordinates of vj are vj(xj, yj, zj), where j is the jth vertex of the triangular network directly connected to vi, <t; The calculation formula for the detailed information of vi is defined as: Where li represents the detailed information of vi, xi represents the x-axis coordinate of vi, xj represents the x-axis coordinate of vj, N represents the number of vertices of the triangular network directly connected to vi; ωij represents the weight of vertex vj of the triangular network directly connected to vi.

5. The method for reconstructing clothing pattern parameters based on three-dimensional design according to claim 4, characterized in that: Establishing a model deformation correction algorithm to achieve personalized deformation of a standardized clothing mannequin model also includes the following sub-steps: The constraint condition of the model deformation correction algorithm is set as follows: minimizing the sum of the differences in vertex detail information of the triangular network corresponding to the standardized clothing mannequin model before and after personalized deformation; The objective function of minimizing the sum of the differences in detail information is established. The objective function of minimizing the sum of the differences in detail information is as follows: Among them, f(x1,x2,...,xt) is the sum of the differences in detail information, Vertex detail information of a triangulated network representing a personalized deformation of a standardized garment mannequin model; Represents the vertex detail information of the triangular network after the personalized deformation of the standardized clothing mannequin model, wherein xi' represents the x-axis coordinate of vi after the personalized deformation of the standardized clothing mannequin model, and xj' represents the x-axis coordinate of vj after the personalized deformation of the standardized clothing mannequin model; The objective function of minimizing the sum of detail information differences is changed to a matrix form and defined as M. The formula of M is as follows: Suppose the standard feature point before the personalized deformation of the standardized clothing mannequin model is vk, the coordinates of vk are vk(xk, yk, zk), and vk after the personalized deformation of the standardized clothing mannequin model is a designated known point, denoted as pk, the coordinates of pk are (x'k, y'k, z'k), k represents the kth standard feature point, and establishes a column vector B, the form of B is as follows: Based on the vertices vi(xi, yi, zi) of the triangular network of the standardized clothing mannequin model before personalized deformation, the equation M*vi=B is solved to obtain the coordinate data of the vertices of the triangular network of the standardized clothing mannequin model after personalized deformation, which is marked as personalized model vertex data.

6. The method for reconstructing clothing pattern parameters based on three-dimensional design according to claim 5, characterized in that: The mapping generation algorithm from the 3D model to the 2D clothing pattern is established, which is labeled as the pattern 2D conversion algorithm and includes the following sub-steps: The standardized clothing mannequin model is segmented according to the structure lines of the clothing pattern to obtain Q three-dimensional clothing pattern pieces of the standardized clothing mannequin model. The structure lines of the clothing pattern include the neckline, shoulder line, front center line, back center line, armhole line, bottom edge line and side seam line.

7. The method for reconstructing clothing pattern parameters based on three-dimensional design according to claim 6, characterized in that: The mapping generation algorithm from the three-dimensional model to the two-dimensional clothing sample is established, which is labeled as the sample two-dimensionalization algorithm and also includes the following sub-steps: Based on the three-dimensional clothing sample piece, the vertices of the triangular mesh of the three-dimensional clothing sample piece are regarded as particles, the area of ​​the triangle in the particle domain is regarded as the mass of the particle, and the mesh edge of the triangular mesh is regarded as a spring. A particle-spring model of the three-dimensional clothing sample piece is established, and the mass and force of each particle are calculated. The mass of each particle is kept unchanged, and the z-axis coordinate value of the particle is set to 0 with the XY plane as a constraint. The time integral iteration of the particle-spring model is performed. When the particle-spring model reaches the minimum overall internal energy of the particle-spring model in the XY plane, the two-dimensional figure at this time is the two-dimensional clothing sample piece after the three-dimensional clothing sample piece is flattened.

8. The method for reconstructing clothing pattern parameters based on three-dimensional design according to claim 7, characterized in that: Obtaining a personalized human body model and obtaining personalized clothing sample parameters based on a model deformation correction algorithm and a sample two-dimensional algorithm includes the following sub-steps: A three-dimensional scanner acquires a human body surface model, marks it as a personalized human body model, extracts feature points of the personalized human body model, obtains standard feature point coordinate data of the personalized human body model, and then uses a model deformation correction algorithm based on the standard feature point coordinate data of the personalized human body model to realize personalized deformation of the standardized clothing mannequin model to obtain a personalized clothing model. The personalized clothing model is then segmented according to the structural lines of the clothing sample to obtain a three-dimensional clothing sample piece of the personalized clothing model. A mapping algorithm from the three-dimensional model to the two-dimensional clothing sample is then used to obtain a two-dimensional personalized clothing sample, and clothing sample parameters of the personalized clothing sample are extracted.

9. An electronic device, characterized in that: The method comprises a processor and a memory, wherein the memory stores computer-readable instructions. When the computer-readable instructions are executed by the processor, the steps in the method according to any one of claims 1 to 8 are executed.

10. A storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 8 are executed.

Images