Clothing pattern generation method and device, electronic equipment and storage medium

Abstract

The application is suitable for the technical field of data processing, and provides a garment paper pattern generation method and device, electronic equipment and a storage medium, including: obtaining a garment picture to be drafted; determining target garment design information, target human body feature information and target fabric feature information according to the garment picture to be drafted; determining a target paper pattern model according to the target garment design information; determining paper pattern adjustment parameters according to the target garment design information, the target human body feature information, the target fabric feature information and a preset paper pattern parameter knowledge graph; and generating a target paper pattern corresponding to the garment picture to be drafted according to the target paper pattern model and the paper pattern adjustment parameters. The embodiment of the application can efficiently and accurately generate a garment paper pattern.

Description

TECHNICAL FIELD

[0001] The present application belongs to the technical field of data processing, and particularly relates to a garment paper pattern generation method and device, an electronic device, and a storage medium. BACKGROUND

[0002] With the increasing diversification of users' demands for clothes, the trend of small batch, multi-variety, fashion and individualization of clothes is becoming more and more obvious, and the production mode of small single fast feedback and flexible production in the field of clothing production gradually becomes the main theme. This production mode has increasingly strong demand for rapid generation of garment paper patterns.

[0003] A garment paper pattern, also known as a garment template or a garment template, is the most specific form of garment structure, and the making of a garment paper pattern is the most important link in the production process. At present, a garment paper pattern is usually obtained by manual grading by a professional pattern maker. A skilled pattern maker generally needs several hours to grade a garment paper pattern. It can be seen that the current garment paper pattern generation process has the defects of low efficiency and high labor cost, and since the quality of the paper pattern depends heavily on the technical level of the pattern maker, the accuracy of the generated garment paper pattern is difficult to guarantee. SUMMARY

[0004] Therefore, the embodiments of the present application provide a garment paper pattern generation method and device, an electronic device and a storage medium to solve the problem of how to efficiently and accurately generate a garment paper pattern in the prior art.

[0005] The first aspect of the embodiments of the present application provides a garment paper pattern generation method, comprising:

[0006] obtaining a garment picture to be graded;

[0007] determining target garment design information, target human feature information and target fabric feature information according to the garment picture to be graded;

[0008] determining a target paper pattern model according to the target garment design information;

[0009] determining paper pattern adjustment parameters according to the target garment design information, the target human feature information, the target fabric feature information and a preset paper pattern parameter knowledge graph;

[0010] generating a target paper pattern corresponding to the garment picture to be graded according to the target paper pattern model and the paper pattern adjustment parameters.

[0011] Optionally, the determining target garment design information, target human feature information and target fabric feature information according to the garment picture to be graded comprises:

[0012] The image of the garment to be patterned is processed by a preset garment recognition algorithm to obtain the design information of the target garment;

[0013] The image of the garment to be patterned is processed by a preset target human body feature recognition algorithm to obtain the target human body feature information;

[0014] The image of the garment to be patterned is processed by a preset fabric feature recognition algorithm to obtain the target fabric feature information.

[0015] Optionally, the target garment design information includes garment style information, fit information, and garment component shape information; the garment recognition algorithm includes a garment style recognition algorithm, a garment component detection algorithm, and a component shape recognition algorithm; and the step of processing the garment image to be patterned using a preset garment recognition algorithm to obtain the target garment design information includes:

[0016] The clothing style recognition algorithm is used to process the clothing image to be patterned to obtain clothing style information and fit information;

[0017] The garment component detection algorithm is used to process the garment image to be patterned, thereby obtaining the position information of each garment component in the garment image to be patterned.

[0018] For each garment component, the component shape information corresponding to the garment component is determined based on the position information of the garment component and the component shape recognition algorithm corresponding to the garment component.

[0019] Optionally, the target pattern model includes a target garment body pattern model and target component pattern models. Determining the target pattern model based on the target garment design information includes:

[0020] Based on the clothing style information and the fit information, the target garment pattern model is determined from the preset garment pattern library;

[0021] Based on the component design information corresponding to the garment component, the target component pattern model is determined from the preset component pattern library.

[0022] Optionally, the fabric characteristic information includes any one or more of the following: fabric type, thickness, softness, elasticity, stiffness, luster, and drape.

[0023] Optionally, the pattern adjustment parameters include human body size adjustment parameters, ease adjustment parameters, and key point adjustment parameters for adjusting the garment shape.

[0024] Optionally, generating the target pattern corresponding to the garment image to be patterned based on the target pattern model and the pattern adjustment parameters includes:

[0025] If the personalized configuration information input by the user is obtained, a target personalized parameter is generated according to the personalized configuration information and the paper pattern adjustment parameter; wherein the personalized configuration information includes human body size modification information, relaxation amount modification information and fabric modification information.

[0026] According to the target paper pattern model and the target personalized parameter, a target paper pattern corresponding to the garment picture to be drafted is generated.

[0027] The second aspect of the embodiment of the present application provides a garment paper pattern generation device, comprising:

[0028] a picture acquisition unit configured to acquire a garment picture to be drafted;

[0029] a recognition unit configured to determine target garment design information, target human body feature information and target fabric feature information according to the garment picture to be drafted;

[0030] a paper pattern model determination unit configured to determine a target paper pattern model according to the target garment design information;

[0031] a paper pattern adjustment parameter determination unit configured to determine a paper pattern adjustment parameter according to the target garment design information, the target human body feature information, the target fabric feature information and a preset paper pattern parameter knowledge graph;

[0032] a paper pattern generation unit configured to generate a target paper pattern corresponding to the garment picture to be drafted according to the target paper pattern model and the paper pattern adjustment parameter.

[0033] The third aspect of the 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, when the processor executes the computer program, the electronic device realizes the steps of the garment paper pattern generation method.

[0034] The fourth aspect of the embodiment of the present application provides a computer readable storage medium, the computer readable storage medium stores a computer program, when the computer program is executed by the processor, the electronic device realizes the steps of the garment paper pattern generation method.

[0035] The fifth aspect of the embodiment of the present application provides a computer program product, when the computer program product is executed on the electronic device, the electronic device executes the garment paper pattern generation method of any one of the first aspect.

[0036] Compared with the prior art, the embodiment of the present application has the beneficial effects that: in the embodiment of the present application, a to-be-patterned garment picture is acquired, and current target garment design information, target human body feature information and target fabric feature information are determined according to the to-be-patterned garment picture. Then, a target pattern model of the current to-be-patterned garment picture is determined according to the target garment design information, and a pattern adjustment parameter is determined according to the target garment design information, the target human body feature information, the target fabric feature information and a preset pattern parameter knowledge graph. Finally, a target pattern corresponding to the to-be-patterned garment picture is generated according to the target pattern model and the pattern adjustment parameter. Since the target garment design information of the to-be-patterned garment picture can reflect the basic design structure of the to-be-patterned garment, the basic target pattern model can be automatically and accurately determined according to the target garment design information. On this basis, the influences of garment design, human body feature and fabric on the details of the pattern are further considered, and the quantified pattern adjustment parameter corresponding to the qualitative characteristic information of the target garment design information, the target human body feature information and the target fabric feature information of the to-be-patterned garment picture is determined by using the preset pattern parameter knowledge graph. The pattern adjustment parameter can accurately adjust the details of the target pattern model. That is, by using the method of the embodiment of the present application, the target pattern model can be accurately determined without the participation of a professional pattern maker, and the details can be accurately adjusted by combining the quantified pattern adjustment parameter, so that the automatic generation of the garment pattern can be efficiently and accurately realized on the premise of saving labor cost. BRIEF DESCRIPTION OF DRAWINGS

[0037] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings needed to be used in the embodiments or the prior art description will be briefly introduced.

[0038] Figure 1 is an implementation process schematic diagram of a garment pattern generation method provided by the embodiment of the present application;

[0039] Figure 2 is a schematic diagram of a garment pattern generation device provided by the embodiment of the present application;

[0040] Figure 3 is a schematic diagram of an electronic device provided by the embodiment of the present application. DETAILED DESCRIPTION

[0041] In the following description, specific details such as specific system structures, techniques, etc. are presented in order to thoroughly understand the embodiments of the present application, but it should be clear to those skilled in the art that the present application can also 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 unnecessary details that hinder the description of the present application.

[0042] In order to illustrate the technical solutions described in the present application, the following will be described by specific embodiments.

[0043] It should be understood that the term "comprising" as used in the specification and the appended claims indicates the presence of the recited 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 groups thereof.

[0044] It should also be understood that the terms used in the present application specification are only for the purpose of describing specific embodiments and are not intended to limit the present application. As used in the present application specification and the appended claims, the singular forms "a," "an," and "the" are intended to include the plural forms unless the context clearly indicates otherwise.

[0045] It should be further understood that the term "and / or" as used in the present application specification and the appended claims means any combination of one or more of the associated listed items and all possible combinations, and includes these combinations.

[0046] As used in the present application 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 [a described condition or event] is detected" can be interpreted to mean "upon determining" or "in response to determining" or "upon detecting [the described condition or event]" or "in response to detecting [the described condition or event]" depending on the context.

[0047] In addition, in the description of the present application, the terms "first", "second", "third", etc. are only used for differentiation in description and cannot be understood as indicating or implying relative importance.

[0048] Traditional garment pattern generation usually relies on professional pattern makers to manually grade, which is low in efficiency, high in labor cost and difficult to guarantee accuracy.

[0049] In order to improve the efficiency of garment pattern generation, the following two methods of automatic grading are considered:

[0050] 1) Parameterized grading: Formulate and parameterize the pattern formula, input the corresponding size information when using, and the pattern can be automatically generated.

[0051] 2) Automatic grading: Establish a basic pattern library and grading rule model, input the corresponding size and adjustment parameters when grading, and the system adjusts the basic pattern according to the rules output by the rule model.

[0052] Both methods have a certain degree of automation, but a garment is usually composed of multiple garment components, and both methods require manual selection of the correct pattern component or formula from a massive database. For example, in actual operation, the operator usually needs to refer to a standard garment design picture to find the corresponding pattern from the database, and even on a graphical interface software, the operator is still required to have a certain pattern basis to select the correct pattern component according to the garment design drawing. That is, although the above two automatic pattern making methods can improve the pattern making speed to a certain extent, they still require professional pattern makers and still require a certain amount of labor cost.

[0053] Therefore, in order to solve the problem of how to efficiently and accurately generate a garment pattern, on the basis of the above automatic pattern making method, further considering the demand for intelligent pattern making throughout the process without relying on pattern makers, the garment pattern generation method of the embodiments of the present application is proposed. The garment pattern generation method comprises: obtaining a garment picture to be patterned; determining target garment design information, target human feature information and target fabric feature information according to the garment picture to be patterned; determining a target pattern model according to the target garment design information; determining a pattern adjustment parameter according to the target garment design information, the target human feature information, the target fabric feature information and a preset pattern parameter knowledge graph; and generating a target pattern corresponding to the garment picture to be patterned according to the target pattern model and the pattern adjustment parameter.

[0054] Since the target garment design information of the garment picture to be patterned can reflect the basic design structure of the garment to be patterned, the basic target pattern model can be automatically and accurately determined according to the target garment design information. On this basis, further considering the detailed influence of garment design, human features and fabric on the pattern, the preset pattern parameter knowledge graph is used to determine the quantified pattern adjustment parameter corresponding to the qualitative characteristic information of the target garment design information, the target human feature information and the target fabric feature information of the garment picture to be patterned, which can accurately adjust the details of the target pattern model. That is, through the method of the embodiments of the present application, efficient and accurate intelligent pattern making throughout the process from picture to pattern can be realized, and the whole process does not require the participation of professional pattern makers, so that the garment pattern generation efficiency can be improved on the premise of saving labor cost.

[0055] Example One

[0056] Figure 1 A flowchart of a garment pattern generation method provided by the embodiments of the present application is shown, and the execution subject of the garment pattern generation method is an electronic device, which is described in detail as follows:

[0057] In S101, a garment picture to be patterned is obtained.

[0058] In the embodiments of the present application, the garment picture to be graded can be a standard garment design drawing (design draft drawing), or a garment picture obtained by photographing a person wearing the garment to be graded (hereinafter referred to as a garment photograph). That is, the garment picture to be graded in the embodiments of the present application can be any form of picture containing complete information of the garment to be graded, and does not need to be strictly limited to a garment design drawing as in traditional garment grading.

[0059] In one embodiment, the electronic device in the embodiments of the present application can obtain a garment design drawing or a garment photograph uploaded by a user to obtain the garment picture to be graded. In another embodiment, the electronic device in the embodiments of the present application can obtain picture website information input by a user, and obtain the garment picture to be graded from a specified website according to the picture website information. In yet another embodiment, the electronic device can obtain a garment photograph by photographing a person wearing the garment to be graded through a camera carried by the electronic device or a photographing device connected to the electronic device, and obtain the garment picture to be graded.

[0060] In S102, target garment design information, target body feature information and target fabric feature information are determined according to the garment picture to be graded.

[0061] In the embodiments of the present application, after obtaining the garment picture to be graded, the garment picture to be graded is recognized to obtain target garment design information, target body feature information and target fabric feature information of the garment to be graded.

[0062] The target garment design information described above is information representing the garment design structure of the garment to be graded, and can generally include any one or more of garment style information, fit information, garment part modeling information of each garment part. The target body feature information is information related to the body feature carried in the garment picture to be graded; generally, the target body feature information can include any one or more of body shape, age range, gender and the like. When the garment picture to be graded is a garment photograph, the target body feature information represents the body feature information of the person wearing the garment picture to be graded, and therefore, when the garment picture to be graded in the embodiments of the present application is the garment photograph described above, the target body feature information can be more accurately obtained, and the accuracy of garment pattern generation is further improved.

[0063] The target fabric feature information of the embodiment of the present application is fabric information of the garment to be drafted identified from the garment picture to be drafted. In an embodiment, the target fabric feature information includes any one or more of fabric type, thickness, softness, elasticity, stiffness, luster, and drape. The fabric type can include cotton, hemp, silk, wool, leather, and the like; the thickness can be divided into thin, relatively thick, thick, and very thick; the softness can be divided into soft, relatively hard, and hard; the elasticity can be divided into no elasticity, slight elasticity, medium elasticity, and high elasticity; and the stiffness can be divided into poor, medium, and good. By identifying one or more fabric feature information of the garment picture to be drafted, the accuracy of garment pattern generation can be further improved.

[0064] In S103, a target pattern model is determined according to the target garment design information.

[0065] In the embodiment of the present application, after the target garment design information is determined, a pattern model matched with the target garment design information can be automatically obtained from a preset pattern model library as the target pattern model according to the target garment design information. The pattern model library stores various pattern models carrying labels of various garment design information in advance, and each pattern model can store related formulas, rules, and / or basic graphics of a pattern of a garment of the type.

[0066] In S104, a pattern adjustment parameter is determined according to the target garment design information, the target body feature information, the target fabric feature information, and a preset pattern parameter knowledge graph.

[0067] In the embodiment of the present application, the pattern adjustment parameter is a parameter for adjusting the basic target pattern model quantitatively (i.e., with specific quantitative values). Generally, the pattern adjustment parameter can include a body size adjustment parameter for adjusting the body size of the pattern, and / or a relaxation amount adjustment parameter for adjusting the relaxation amount of the pattern. The preset pattern parameter knowledge graph is a knowledge base of a network structure containing garment design information, body feature information, fabric feature information, and pattern adjustment parameters, which can represent the coupling relationship between the garment design information, the body feature information, the fabric feature information, and the pattern adjustment parameter.

[0068] This step can be performed simultaneously with step S103, or the execution order of the two can be arbitrarily changed. After the target garment design information, the target body feature information, and the target fabric feature information of the garment to be drafted are determined, the corresponding quantitative pattern adjustment parameter can be indexed from the preset pattern parameter knowledge graph according to the three qualitative feature information.

[0069] In S105, a target pattern corresponding to the to-be-graded garment picture is generated according to the target pattern model and the pattern adjustment parameter.

[0070] After the pattern adjustment parameter is determined, the pattern adjustment parameter can be input into the target pattern model determined in step S103. The target pattern model can call a preset drawing software interface to draw a target pattern according to the pattern adjustment parameter. In an embodiment, a pattern drawing in DWG or DXF format can be output as the target pattern. DWG is a proprietary file format used by computer-aided design software AutoCAD and AutoCAD-based software to save design data. DXF is a CAD data file format used for CAD data exchange between AutoCAD and other software.

[0071] In the embodiments of the present application, the target garment design information of the to-be-graded garment picture can reflect the basic design structure of the to-be-graded garment, so the basic target pattern model can be accurately determined according to the target garment design information. On this basis, the influences of garment design, human body characteristics and fabric on the details of the pattern are further considered, and the preset pattern parameter knowledge graph is used to determine the quantized pattern adjustment parameter corresponding to the target garment design information, the target human body characteristic information and the target fabric characteristic information of the to-be-graded garment picture. The pattern adjustment parameter can accurately adjust the details of the target pattern model. That is, by the method of the embodiments of the present application, the target pattern model can be accurately determined without the participation of professional pattern designers, and the details can be accurately adjusted by combining the quantized pattern adjustment parameter, so that the automatic generation of the garment pattern can be efficiently and accurately realized under the premise of saving labor costs.

[0072] Optionally, the target garment design information, the target human body characteristic information and the target fabric characteristic information are determined according to the to-be-graded garment picture, and the method comprises:

[0073] The target garment design information is obtained by processing the to-be-graded garment picture through a preset garment recognition algorithm.

[0074] The target human body characteristic information is obtained by processing the to-be-graded garment picture through a preset target human body characteristic recognition algorithm.

[0075] The target fabric characteristic information is obtained by processing the to-be-graded garment picture through a preset fabric characteristic recognition algorithm.

[0076] In the embodiments of the present application, the preset garment recognition algorithm can be implemented based on a garment recognition network model. The garment recognition network model can be a neural network model trained in advance using garment pictures carrying garment design information labels as sample data. The current garment picture to be patterned is input into the trained garment recognition network model for processing, and the current target garment design information can be obtained.

[0077] In the embodiments of the present application, the preset target human body feature recognition algorithm can be a multi-label classification algorithm preset for identifying multi-dimensional information such as body shape, age range, and gender of a human body from a garment picture. That is, the multi-label classification algorithm is an algorithm that combines multiple dimensions of human body feature recognition tasks into one network for completion. Through the multi-label classification algorithm, multiple dimensions of recognition results can be directly output. For ease of description, the multi-label classification algorithm is referred to as a human body feature multi-label classification model. The human body feature multi-label classification model is trained using garment pictures carrying three label information of body shape, age range, and gender as sample data. In this step, the current obtained garment picture to be patterned is input into the trained human body feature multi-label classification model for processing, and the current human body features such as body shape, age range, and gender are obtained as target human body feature information. The body shape can be divided into A type, B type, C type, and Y sex; the age range can be divided into youth, youth, middle age, and old age; and the gender feature can be divided into male, female, and neutral. Further, since the output target human body feature information only includes body shape, age range, and gender, and does not include specific height information, the subsequent generation of a paper pattern is not limited to generating a paper pattern exactly the same as the size of the garment to be patterned in the current garment picture to be patterned. Instead, the size of all heights or a specified height can be flexibly selected from a human body standard body shape library to realize automatic paper pattern grading. For example, the height can be from 160 cm to 195 cm, with each 5 cm as a code. In the human body standard body shape library, the following size feature information is usually included: height, cervical vertebra point height, sitting cervical vertebra point height, full arm length, waist height, chest circumference, neck circumference, total shoulder width, waist circumference, hip circumference, thigh length, and thigh circumference.

[0078] Similarly, the fabric feature recognition algorithm in the embodiments of the present application can be a multi-label classification algorithm for identifying the fabric type, thickness, softness, elasticity, stiffness, luster, drape, and other multi-dimensional fabric features of the garment from the garment picture, which is referred to as a fabric feature multi-label classification model. In an embodiment, when creating a data set, seven labels of fabric type, thickness, softness, elasticity, stiffness, luster, and drape are identified for each garment picture; then the garment pictures in the data set are used as sample data for model training to obtain the trained fabric feature multi-label classification model. In this step, the currently obtained garment picture to be patterned is input into the trained fabric feature multi-label classification model for processing to obtain the current target fabric feature information.

[0079] Exemplarily, the backbone network of the multi-label classification algorithm described above adopts a structure combining depthwise separable convolution and attention mechanism, which ensures the recognition accuracy of the model under a smaller parameter amount, so as to ensure a faster inference speed of the algorithm while fully extracting image features; and a sigmoid activation function is used at the end of the network to determine each classification result.

[0080] In the embodiments of the present application, unlike directly recognizing the target garment design information, target human feature information, and target fabric feature information of the garment to be patterned by one recognition algorithm, the target garment design information, target human feature information, and target fabric feature information are recognized by the three algorithms of the preset garment recognition algorithm, target human feature recognition algorithm, and fabric feature recognition algorithm respectively, so as to reduce the complexity of recognition and improve the feature recognition efficiency.

[0081] Optionally, the target garment design information includes garment style information, fit information, and garment component modeling information, the garment recognition algorithm includes a garment style recognition algorithm, a garment component detection algorithm, and a component modeling recognition algorithm, and the processing of the garment to be patterned by the preset garment recognition algorithm to obtain the target garment design information includes:

[0082] processing the garment to be patterned by the garment style recognition algorithm to obtain garment style information and fit information;

[0083] processing the garment to be patterned by the garment component detection algorithm to obtain the position information of each garment component in the garment to be patterned;

[0084] For each garment component, the component modeling information corresponding to the garment component is determined according to the position information of the garment component and the component modeling recognition algorithm corresponding to the garment component.

[0085] In the embodiments of the present application, the target garment design information specifically includes garment style information, fit information and garment component modeling information, and the garment recognition algorithm is also different from the general image segmentation algorithm. The garment recognition algorithm in the embodiments of the present application includes a garment style recognition algorithm, a garment component detection algorithm and a component modeling recognition algorithm.

[0086] Specifically, the garment style recognition algorithm is used to perform garment style recognition processing on the garment picture to be drafted, and the garment style information and the fit information can be obtained. In the embodiments of the present application, the garment style information can be any one of the following 21 garment styles: a sweater, a T-shirt, a vest, a shirt, a Polo shirt (also known as a tennis shirt or a golf shirt), a knit shirt, a sweater, a vest, a jacket, a coat, a windbreaker, a down jacket, a suit, an overcoat, long pants, shorts, a skirt, a dress, a jumpsuit, a swimsuit and a pajamas. The fit information can be divided into: tight, slim, fit, loose and oversized. In an embodiment, the garment style recognition algorithm is implemented based on a garment style recognition model, which can be a neural network model trained in advance with garment pictures carrying garment style information labels and fit labels as sample data.

[0087] In the embodiments of the present application, the garment component detection algorithm is a target detection model trained in advance to detect garment components in a garment picture. The garment picture to be drafted is input into the target detection model for processing, and the type and position information of each garment component in the garment picture to be drafted can be obtained. The garment components can include a collar, a sleeve, a hat, a pocket, a waistband, a hem, a decorative component and the like.

[0088] Exemplarily, the target detection model of the embodiment of the present application is composed of two parts of a backbone and a detection header. Among them, the backbone part adopts a convolutional neural network with a residual structure, and combines attention mechanism and skip-connection enhancement algorithm to improve the extraction ability of clothing features. The header part adopts a feature pyramid structure to extract multi-layer features, while ensuring semantic information, more location information is extracted; the classification branch uses an improved loss function (generalized focal loss) to improve the accuracy; the regression branch uses an anchor-free method to replace the anchor-based method, improving the operation speed. The loss function of network training can use GIoU Loss (a kind of regression loss function for target detection). In the target detection model training process, Mosaic (a data enhancement method of four images in 2x2 splicing mode and indefinite splicing center splicing in one image) and Mixup (a data enhancement method of two images in a certain proportion channel superposition mixing) can be used to enhance the algorithm accuracy.

[0089] After detecting the types and positions of each clothing component in the to-be-patterned clothing picture by the above-mentioned clothing component detection algorithm, for each detected clothing component, a local picture corresponding to the clothing component is extracted from the to-be-patterned clothing picture according to the position of the clothing component in the to-be-patterned clothing picture; and a component modeling recognition algorithm corresponding to the clothing component is determined according to the type of the clothing component; then, the component modeling recognition algorithm corresponding to the clothing component is used to process the local picture, and component modeling information corresponding to the clothing component is obtained. The component modeling recognition algorithm can also be implemented by the above-mentioned multi-label classification algorithm, that is, the component modeling recognition algorithm can be a component modeling multi-label classification model; by using a clothing picture carrying a component modeling label as sample data for model training, a trained component modeling multi-label classification model can be obtained.

[0090] In one embodiment, in order to improve the data interaction speed between the above-mentioned clothing component detection algorithm and the component modeling recognition algorithm, a multi-thread parallel acceleration technology based on CUDA (Compute Unified Device Architecture) core can be used to parallelly fuse multiple algorithm models together.

[0091] In the embodiments of the present application, the garment style information and fit information are recognized through the garment style recognition algorithm, and the component modeling information corresponding to each garment component is recognized through the garment component detection algorithm and the component modeling recognition algorithm. Compared with the current garment recognition method realized by a single image segmentation algorithm (i.e., a multi-label classification algorithm is added to the image segmentation algorithm to determine the style and component modeling), the data labeling speed can be greatly accelerated, and the recognition accuracy can be greatly improved through the independent and cooperative recognition algorithms and detection algorithms.

[0092] Optionally, the target paper pattern model includes a target body paper pattern model and a target component paper pattern model, and the target paper pattern model is determined according to the target garment design information, including:

[0093] The target body paper pattern model is determined from a preset body paper pattern library according to the garment style information and the fit information;

[0094] The target component paper pattern model is determined from a preset component paper pattern library according to the component modeling information corresponding to the garment component.

[0095] In the embodiments of the present application, the preset paper pattern model library stores parameterized paper pattern models, including a body paper pattern library storing various body paper pattern models and a component paper pattern library storing various component paper pattern models. In some embodiments, the preset paper pattern model library also stores uniformly defined interface parameters, including front sleeve armhole lines, back sleeve armhole lines, front neckline lines, back neckline lines, waistline curves, and joint lines. The connection of each part of the garment can be realized through these interface parameters.

[0096] In the embodiments of the present application, each body paper pattern model in the preset body paper pattern library carries a corresponding garment style label and fit label. After the garment style information and fit information corresponding to the current garment picture to be graded are determined, the body paper pattern model with a garment style label matching the current garment style information and a fit label matching the current fit information is searched from the body paper pattern library as the target body paper pattern model of the current garment picture to be graded.

[0097] Each component paper pattern model in the preset component paper pattern library carries a corresponding component modeling label. After the component modeling information corresponding to each garment component in the current garment to be graded is determined, for each garment component, the component paper pattern model with a component modeling label matching the current component modeling information is searched from the component paper pattern library corresponding to the garment component as the target component paper pattern model of the current garment component.

[0098] In the embodiments of the present application, the target body paper pattern model can be determined according to the garment style information and the fit information, the target component paper pattern model can be determined according to the component modeling information corresponding to the garment component, and the paper pattern model of the corresponding part can be determined according to the characteristic information of different parts of the garment, so that the target paper pattern model can be more accurately obtained, and the accuracy of the generated garment paper pattern is further improved.

[0099] Optionally, the paper pattern adjustment parameter includes a human body size adjustment parameter, a relaxation amount adjustment parameter, and a key point adjustment parameter for adjusting the garment modeling.

[0100] In the embodiments of the present application, the human body size adjustment parameter is a parameter for adjusting the size of the paper pattern, which is determined according to the target human body characteristic information such as body shape, age, gender, etc.; and the relaxation amount adjustment parameter is a parameter for adjusting the relaxation amount of the paper pattern, which is determined according to the target human body characteristic information, the fit information in the target garment design information, and the target fabric characteristic information. For example, the relaxation amount of a fabric with large elasticity is smaller than that of a fabric without elasticity, the relaxation amount should be appropriately increased when the fabric is thicker, and the relaxation amount of an outerwear should be appropriately increased than that of an innerwear. Exemplarily, the relaxation amount of a female garment chest circumference is shown in Table 1.

[0101] Table 1:

[0102]

[0103] The key point adjustment parameter in the embodiments of the present application is a parameter for adjusting the key points affecting the garment modeling. These key points can be defined when the paper pattern model is established, and each paper pattern model first determines its possible change modeling, and then defines the key points according to the position of the required change. In an embodiment, the key point adjustment parameter is specifically used to specify the action and value of the key point adjustment, the action includes: translation, scaling, rotation; and the value is the specific adjustment amplitude. The action and value in the key point adjustment parameter are specifically determined according to the target garment design information, the target human body characteristic information, and the target fabric characteristic information in combination with the preset paper pattern parameter knowledge picture, and the key points are adjusted according to the action and value, so that the modeling style of the generated paper pattern can be changed.

[0104] Exemplarily, the key points can include: front middle point, back middle point, side neck point, shoulder point, underarm point, sleeve cap curve control point, elbow point, sleeve opening point, bust point, waist point, hip point, knee point, front dart bend control point, etc. According to any one or more of the identified target garment design information, target human body feature information, and target fabric feature information, the key point adjustment parameters are determined to flexibly and accurately adjust the garment style. For example, the sleeve cap curve control point and the underarm point in the key points can control the shape of the sleeve cap curve to adjust the angle of the sleeve forward inclination or backward inclination; if the currently identified garment style information is a suit, the sleeve usually needs to be slightly forward inclined, and at this time, the garment style can be adjusted by adjusting the two key points; if the currently identified body type information determines that the sleeve opening needs to be slightly backward inclined, the garment style can also be adjusted by adjusting the two key points, but the adjustment action and the value size are different. For another example, the elbow point and the sleeve opening point in the key points can control the shape of the sleeve curve; the back middle point and the side neck point can control the lifting or lowering amount of the back piece of the top to adapt to the muscular or slim body type, and prevent the garment back from being raised or too tight.

[0105] In the embodiments of the present application, since the pattern adjustment parameters specifically include the human body size adjustment parameters that can be used to adjust the size of the pattern, the ease adjustment parameters that can be used to adjust the ease of the pattern, and the key point adjustment parameters that can be used to adjust the garment style, after the basic target pattern model is determined, the size, ease, and detail style of the pattern can be further adjusted in combination with the detail features of the garment, so that the accuracy of the garment pattern generation is further improved.

[0106] Optionally, the generating the target pattern corresponding to the garment picture to be marked out according to the target pattern model and the pattern adjustment parameters comprises:

[0107] If the personalized configuration information input by the user is acquired, the target personalized parameters are generated according to the personalized configuration information and the pattern adjustment parameters; wherein the personalized configuration information includes human body size modification information, ease modification information, and fabric modification information.

[0108] The target pattern corresponding to the garment picture to be marked out is generated according to the target pattern model and the target personalized parameters.

[0109] In the embodiments of the present application, the electronic device is provided with a personalized configuration interface, and the user can input the human body size modification information, the ease modification information, and the fabric modification information as the personalized configuration information through the personalized configuration interface. Then, the generated pattern adjustment parameters are adjusted according to the personalized configuration information to generate the target personalized parameters.

[0110] After the target individualized parameter is generated, the target individualized parameter is input into a target pattern model for processing to generate a target pattern corresponding to the picture of the garment to be patterned.

[0111] In the embodiments of the present application, the target pattern is generated according to the individualized configuration information input by the user to meet the individualized customization requirement, so that the flexibility of the garment pattern generation is improved.

[0112] It should be understood that the size of the serial number of each step in the above embodiments does not mean the order of execution, and the execution order of each process should be determined according to its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present application.

[0113] Example Two

[0114] Figure 2 A structure schematic diagram of a garment pattern generation device provided by the embodiments of the present application is shown, and only the parts related to the embodiments of the present application are shown for ease of illustration:

[0115] The garment pattern generation device comprises a picture acquisition unit 21, an identification unit 22, a pattern model determination unit 23, a pattern adjustment parameter unit 24, and a pattern generation unit 25. Among them:

[0116] The picture acquisition unit 21 is configured to acquire a picture of a garment to be patterned.

[0117] The identification unit 22 is configured to determine target garment design information, target human feature information, and target fabric feature information according to the picture of the garment to be patterned.

[0118] The pattern model determination unit 23 is configured to determine a target pattern model according to the target garment design information.

[0119] The pattern adjustment parameter determination unit 24 is configured to determine a pattern adjustment parameter according to the target garment design information, the target human feature information, the target fabric feature information, and a preset pattern parameter knowledge graph.

[0120] The pattern generation unit 25 is configured to generate a target pattern corresponding to the picture of the garment to be patterned according to the target pattern model and the pattern adjustment parameter.

[0121] Optionally, the identification unit 22 comprises:

[0122] The first identification module is configured to process the picture of the garment to be patterned by a preset garment identification algorithm to obtain the target garment design information.

[0123] a second identification module configured to process the to-be-patterned garment picture by using a preset target human body feature identification algorithm to obtain the target human body feature information;

[0124] a third identification module configured to process the to-be-patterned garment picture by using a preset fabric feature identification algorithm to obtain the target fabric feature information.

[0125] Optionally, the target garment design information includes garment style information, fit information, and garment component modeling information, the garment identification algorithm includes a garment style identification algorithm, a garment component detection algorithm, and a component modeling identification algorithm, and the first identification module includes:

[0126] a style identification module configured to process the to-be-patterned garment picture by using the garment style identification algorithm to obtain garment style information and fit information;

[0127] a component detection module configured to process the to-be-patterned garment picture by using the garment component detection algorithm to obtain position information of each garment component in the to-be-patterned garment picture;

[0128] a component modeling identification module configured to, for each garment component, determine component modeling information corresponding to the garment component according to position information of the garment component and the component modeling identification algorithm corresponding to the garment component.

[0129] Optionally, the pattern model determination unit 23 includes:

[0130] a body pattern model determination unit configured to determine a target body pattern model from a preset body pattern library according to the garment style information and the fit information;

[0131] a component pattern model determination unit configured to determine a target component pattern model from a preset component pattern library according to component modeling information corresponding to the garment component.

[0132] Optionally, the fabric feature information includes any one or more of fabric type, thickness, softness, elasticity, stiffness, glossiness, and drape.

[0133] Optionally, the pattern adjustment parameter includes a human body size adjustment parameter, a relaxation amount adjustment parameter, and a key point adjustment parameter for adjusting garment modeling.

[0134] Optionally, the pattern generation unit 25 includes:

[0135] The configuration module is configured to, if personalized configuration information input by a user is acquired, generate target personalized parameters according to the personalized configuration information and the paper pattern adjustment parameters; wherein the personalized configuration information comprises human body size modification information, relaxation amount modification information and fabric modification information.

[0136] The generation module is configured to generate a target paper pattern corresponding to the to-be-cut garment picture according to the target paper pattern model and the target personalized parameters.

[0137] It is to be noted that the information interaction and execution process between the above apparatuses / modules are based on the same concept as the method embodiments, and the specific functions and technical effects thereof can be referred to the method embodiments, which will not be described here.

[0138] Example Three

[0139] Figure 3 is a schematic diagram of an electronic device provided in an embodiment of the present application. As shown in Figure 3 The electronic device 3 of this embodiment comprises a processor 30, a memory 31, and a computer program 32 stored in the memory 31 and executable on the processor 30, such as a garment paper pattern generation program. The processor 30 implements the steps in each of the above garment paper pattern generation method embodiments when executing the computer program 32, such as steps S101 to S105 shown in Figure 1 Alternatively, the processor 30 implements the functions of each module / unit in each of the above apparatus embodiments when executing the computer program 32, such as the functions of the picture acquisition unit 21 to the paper pattern generation unit 25 shown in Figure 2

[0140] For example, the computer program 32 can be divided into one or more modules / units, which are stored in the memory 31 and executed by the processor 30 to complete the present application. The one or more modules / units can be a series of computer program instruction segments capable of completing a specific function, which are used to describe the execution process of the computer program 32 in the electronic device 3.

[0141] The electronic device 3 can be a desktop computer, a notebook computer, a palm computer, a cloud server, and other computing devices. The electronic device can include, but is not limited to, the processor 30 and the memory 31. Those skilled in the art can understand that Figure 3 The electronic device 3 is only an example and does not constitute a limitation on the electronic device 3, which can include more or fewer components than those shown, or combine certain components, or different components, for example, the electronic device can also include an input / output device, a network access device, a bus, etc. ​

[0142] The processor 30 can be a central processing unit (CPU), and can also be other general-purpose processors, a digital signal processor (DSP), an application specific integrated circuit (ASIC), a field-programmable gate array (FPGA) or other programmable logic device, discrete gate or transistor logic device, discrete hardware components, etc. The general-purpose processor can be a microprocessor or the processor can also be any conventional processor.

[0143] The memory 31 can be an internal storage unit of the electronic device 3, such as a hard disk or a memory of the electronic device 3. The memory 31 can also be an external storage device of the electronic device 3, such as a plug-in hard disk, a smart media card (SMC), a secure digital (SD) card, a flash card, etc. provided on the electronic device 3. Further, the memory 31 can include both the internal storage unit and the external storage device of the electronic device 3. The memory 31 is used to store the computer program and other programs and data required by the electronic device. The memory 31 can also be used to temporarily store data that has been output or will be output.

[0144] Those skilled in the art can clearly understand that, for the convenience and brevity of description, only the division of the above functional units and modules is exemplified, and in actual application, the above functions can be completed by different functional units and modules according to needs, that is, the internal structure of the apparatus is divided into different functional units or modules to complete all or part of the above described functions. Each functional unit and module in the embodiment can be integrated in one processing unit, or each unit can exist physically, or two or more units can be integrated in one unit, and the integrated unit can be realized in the form of hardware or in the form of software. In addition, the specific names of each functional unit and module are only for easy distinction, and do not limit the protection scope of the present application. The specific working process of the units and modules in the above system can refer to the corresponding process in the foregoing method embodiments, which will not be described here.

[0145] In the above embodiments, the description of each embodiment has its own emphasis, and the parts not described or recorded in detail in a certain embodiment can be referred to the relevant description of other embodiments.

[0146] Those skilled in the art can understand that the units and algorithm steps of each example described in combination with the embodiments disclosed herein can be realized by electronic hardware or a combination of computer software and electronic hardware. Whether the functions are realized in hardware or software depends on the specific application and design constraints of the technical solution. A person skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of the present application.

[0147] In the embodiments provided in the present application, it should be understood that the disclosed apparatus / equipment and method can be implemented in other ways. For example, the apparatus / equipment embodiments described above are merely schematic, for example, the division of the modules or units is merely a logical function division, and actual implementation can have another division manner, for example, a plurality of units or components can be combined or integrated into another system, or some features can be ignored or not executed. In addition, the coupling or direct coupling or communication connection between the units shown or discussed can be indirect coupling or communication connection through some interface, device or unit, and can be electrical, mechanical or other forms.

[0148] The units described as separate components can or can not be physically separated, and the components shown as units can or can not be physical units, that is, they can be located in one place, or can be distributed on a plurality of network units. Part or all of the units can be selected according to actual needs to achieve the purpose of the embodiments.

[0149] In addition, each functional unit in each embodiment of the present application can be integrated in one processing unit, or each unit can exist physically, or two or more units can be integrated in one unit. The integrated unit can be realized in the form of hardware or in the form of a software functional unit.

[0150] The integrated module / unit, if implemented in the form of a software functional unit and sold or used as an independent product, can be stored in a computer readable storage medium. Based on such understanding, all or part of the processes in the above-mentioned embodiment methods can also be completed by a computer program instructing related hardware, and the computer program can be stored in a computer readable storage medium. The computer program can implement the steps of each method embodiment when executed by a processor. The computer program includes computer program code, which can be in the form of source code, object code, executable files, or some intermediate forms. The computer readable medium can include any entity or device capable of carrying the computer program code, recording medium, U disk, mobile hard disk, magnetic disk, optical disk, computer memory, read-only memory (ROM), random access memory (RAM), electrical carrier signal, telecommunication signal, and software distribution medium, etc. It should be noted that the contents included in the computer readable medium can be appropriately increased or decreased according to the requirements of legislation and patent practice in the jurisdiction, for example, in some jurisdictions, according to legislation and patent practice, the computer readable medium does not include electrical carrier signals and telecommunication signals.

[0151] The above-described embodiments are only used to illustrate the technical solutions of the present application, rather than limit them; although the present application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that: it can still modify the technical solutions recorded in the foregoing embodiments, or make equivalent replacement for part of the technical features; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present application, and should be included in the protection scope of the present application.

Claims

1. A garment pattern generation method characterized by, The method comprises the following steps: obtaining a picture of a garment to be patterned; determining target garment design information, target human body feature information and target fabric feature information according to the picture of the garment to be patterned; determining a target pattern model according to the target garment design information; determining pattern adjustment parameters according to the target garment design information, the target human body feature information, the target fabric feature information and a preset pattern parameter knowledge graph; generating a target pattern corresponding to the picture of the garment to be patterned according to the target pattern model and the pattern adjustment parameters. The method comprises the following steps: processing the picture of the garment to be patterned through a preset garment recognition algorithm to obtain the target garment design information, wherein the target garment design information comprises garment style information, fit information and garment component modeling information, and the garment recognition algorithm comprises a garment style recognition algorithm, a garment component detection algorithm and a component modeling recognition algorithm; processing the picture of the garment to be patterned through a preset target human body feature recognition algorithm to obtain the target human body feature information; processing the picture of the garment to be patterned through a preset fabric feature recognition algorithm to obtain the target fabric feature information. The method comprises the following steps: processing the picture of the garment to be patterned through the garment style recognition algorithm to obtain garment style information and fit information; processing the picture of the garment to be patterned through the garment component detection algorithm to obtain position information of each garment component in the picture of the garment to be patterned; for each garment component, determining component modeling information corresponding to the garment component according to the position information of the garment component and the component modeling recognition algorithm corresponding to the garment component. The target pattern model comprises a target body pattern model and a target component pattern model, and the method comprises the following steps: determining a target body pattern model from a preset body pattern library according to the garment style information and the fit information; determining a target component pattern model from a preset component pattern library according to the component modeling information corresponding to the garment component.

2. The garment pattern generation method of claim 1, wherein, The fabric feature information comprises any one or more of fabric type, thickness, softness, elasticity, stiffness, glossiness and drape.

3. The garment pattern generation method of claim 1, wherein, The pattern adjustment parameters comprise human body size adjustment parameters, ease adjustment parameters and key point adjustment parameters for adjusting garment modeling.

4. The garment pattern generation method according to any one of claims 1 to 3, characterized in that, The method comprises the following steps: if personalized configuration information input by a user is obtained, generating target personalized parameters according to the personalized configuration information and the pattern adjustment parameters, wherein the personalized configuration information comprises human body size modification information, ease modification information and fabric modification information. Generate a target pattern corresponding to the to-be-printed garment picture according to the target pattern model and the target individualized parameter.

5. A garment pattern generation apparatus characterized by comprising: Comprise: A picture acquisition unit is configured to acquire a to-be-printed garment picture. An identification unit is configured to determine target garment design information, target human body feature information, and target fabric feature information according to the to-be-printed garment picture. A pattern model determination unit is configured to determine a target pattern model according to the target garment design information. A pattern adjustment parameter determination unit is configured to determine a pattern adjustment parameter according to the target garment design information, the target human body feature information, the target fabric feature information, and a preset pattern parameter knowledge graph. A pattern generation unit is configured to generate a target pattern corresponding to the to-be-printed garment picture according to the target pattern model and the pattern adjustment parameter. The identification unit comprises: A first identification module is configured to process the to-be-printed garment picture through a preset garment identification algorithm to obtain the target garment design information, wherein the target garment design information comprises garment style information, fit information, and garment component modeling information, and the garment identification algorithm comprises a garment style identification algorithm, a garment component detection algorithm, and a component modeling identification algorithm. A second identification module is configured to process the to-be-printed garment picture through a preset target human body feature identification algorithm to obtain the target human body feature information. A third identification module is configured to process the to-be-printed garment picture through a preset fabric feature identification algorithm to obtain the target fabric feature information. The first identification module comprises: A style identification module is configured to process the to-be-printed garment picture through the garment style identification algorithm to obtain garment style information and fit information. A component detection module is configured to process the to-be-printed garment picture through the garment component detection algorithm to obtain position information of each garment component in the to-be-printed garment picture. A component modeling identification module is configured to, for each garment component, determine component modeling information corresponding to the garment component according to the position information of the garment component and the component modeling identification algorithm corresponding to the garment component. The pattern model determination unit comprises: A body pattern model determination unit is configured to determine a target body pattern model from a preset body pattern library according to the garment style information and the fit information. A component pattern model determination unit is configured to determine a target component pattern model from a preset component pattern library according to the component modeling information corresponding to the garment component.

6. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the computer program, the electronic device implements the steps of the method of any one of claims 1 to 4.

7. A computer-readable storage medium storing a computer program, wherein the computer program comprises the following steps of: receiving a request for a resource from a client; determining whether the client is authorized to access the resource; and if the client is authorized to access the resource, providing the resource to the client. When the computer program is executed by the processor, the electronic device implements the steps of the method of any one of claims 1 to 4.

Images