Garment sheet generation method and device, electronic equipment and storage medium

Abstract

The invention relates to the field of costume design, and discloses a costume sheet generation method and device, electronic equipment and a storage medium. The garment sheet generation method comprises the following steps: determining a garment design language input by a user; and calling a sheet generation model to generate a target clothing sheet based on the clothing design language: converting the clothing design language into a symbolized sheet making language, and outputting the target clothing sheet according to the symbolized sheet making language. Compared with the prior art, the scheme can reduce the requirement for professional ability of a user in the garment sheet manufacturing process, and improves the sheet manufacturing efficiency.

Description

Technical Field

[0001] The present application relates to the field of clothing design, and more specifically, to a method, device, electronic device, and storage medium for generating clothing patterns. Background Art

[0002] In the apparel industry, pattern making has always been a significant challenge. With the advancement of technology, traditional manual production methods are gradually revealing their limitations in the context of rapid digitalization and personalized demands. Market demand for more sophisticated and diverse designs continues to grow, but manual pattern making, due to its time-consuming nature, high error rates, and low efficiency, has become unable to meet the requirements of modern clothing production.

[0003] To address this challenge, pattern-making software such as Richpeace utilizes a procedural modeling approach based on symbolic programming. These software transforms the garment pattern-making process into symbolic drawing based on points, lines, and surfaces. This offers numerous advantages, including easy-to-understand design parameters, support for random variations, high-quality output, and compact representation.

[0004] Despite this, the template creation process remains complex, requiring not only in-depth pattern-making knowledge but also the complex coupling between steps and parameters, requiring users to possess advanced mathematical reasoning, geometric intuition, and programming skills. This limits its flexibility and universal applicability. This results in high professional requirements for garment pattern preparation and low pattern-making efficiency. Summary of the Invention

[0005] The purpose of this application is to provide a clothing pattern generation method, device, electronic device and storage medium, which can reduce the requirements for user professional ability in the clothing pattern preparation process and improve pattern making efficiency.

[0006] In a first aspect, an embodiment of the present application provides a method for generating a clothing pattern, comprising: determining a clothing design language input by a user; calling a pattern generation model to generate a target clothing pattern based on the clothing design language, comprising: converting the clothing design language into a symbolic pattern making language, and outputting the target clothing pattern according to the symbolic pattern making language.

[0007] Compared with the related art, in the clothing pattern generation method provided by the embodiment of the present application, after receiving the clothing design language input by the user, the pattern generation model can be directly called to generate the target clothing pattern based on the language. Therefore, when making a pattern, the user only needs to input clothing design languages such as natural language, design manuscripts, photos, etc., without having to perform complex professional operations such as typesetting or programming. This not only significantly reduces the requirements for the user's professional ability in the clothing pattern preparation process; at the same time, due to the high degree of automation of the entire pattern making process, the user can perform a large amount of clothing pattern making work at the same time, thereby improving the overall pattern making efficiency. In addition, this solution splits the generation process of the target clothing pattern into a language generation process and a pattern output process through language conversion, reducing the degree of coupling between the steps, making the complete generation process of the target clothing pattern clearer and more orderly, thereby allowing the user to accurately control the parameters such as the size, size, shape, etc. of the pattern through symbolic pattern making language, which helps to improve the accuracy of the generated pattern.

[0008] In the second aspect, an embodiment of the present application provides a method for training a target language parsing model, comprising: obtaining a training data set and a standard multimodal large model, wherein each training data in the training data set includes a sample clothing design language and a sample symbolic pattern making language; and using the training data set to fine-tune the standard multimodal large model to obtain the target language parsing model.

[0009] Compared with related technologies, the embodiment of the present application fine-tunes the standard multimodal large model to train the target language parsing model. It is understandable that, on the one hand, the standard multimodal large model itself already has powerful analysis and reasoning capabilities for multimodal data, so the target language parsing model obtained by fine-tuning the model naturally also has this capability; on the other hand, using the sample symbolic pattern-making language as the label of the sample clothing design language to construct a sample and fine-tune the standard multimodal large model can ensure that the generated target language parsing model can fully analyze the clothing design language and generate a relatively accurate symbolic pattern-making language, which helps to improve the accuracy and quality of the target clothing pattern finally generated.

[0010] On the third aspect, an embodiment of the present application provides a method for training a target symbol translation model, including: obtaining an initial model and a training sample; using the training sample to train the initial model to obtain the target symbol translation model, which is used to convert the symbolic pattern-making language into a drawing command, and / or convert the modified symbolic pattern-making language into a modified drawing command. Similar to the training process of the aforementioned pattern generation model, the initial model itself already has powerful data analysis and reasoning capabilities, so the target symbol translation model obtained by training with the help of this model naturally also has this capability; on the other hand, training the initial model with training samples can enable the generated target language parsing model to comprehensively analyze the clothing design language and generate a relatively accurate symbolic pattern-making language, which also helps to improve the accuracy and pattern quality of the target clothing pattern finally generated.

[0011] In a fourth aspect, an embodiment of the present application provides a target language parsing model software, which is used to: convert clothing design language into symbolic pattern making language; and / or convert modified clothing design language into modified symbolic pattern making language.

[0012] In a fifth aspect, an embodiment of the present application provides a target symbol translation model software, which is used to: convert a symbolic board language into a drawing command, and / or convert a modified symbolic board language into a modified drawing command.

[0013] In a sixth aspect, an embodiment of the present application provides a clothing pattern generation device, comprising: a design language determination module for determining a clothing design language input by a user; and a clothing pattern generation module for calling a pattern generation model to generate a target clothing pattern based on the clothing design language.

[0014] In the seventh aspect, an embodiment of the present application provides a training device for a target language parsing model, including: a standard model acquisition unit, used to acquire a training data set and a standard multimodal large model, each training data in the training data set includes a sample clothing design language and a sample symbolic pattern making language; a parsing model training unit, used to use the training data set to fine-tune the standard multimodal large model to obtain the target language parsing model.

[0015] In an eighth aspect, an embodiment of the present application provides a training device for a target symbol translation model, comprising: an initial model acquisition unit for acquiring an initial model and training samples; a translation model training unit for training the initial model using the training samples to obtain the target symbol translation model, which is used to convert a symbolic board language into a drawing command, and / or convert a modified symbolic board language into a modified drawing command.

[0016] The advantages of each of the above devices over related technologies can be found in the detailed analysis of the corresponding method embodiments above, and will not be described in detail here.

[0017] In the ninth aspect, an embodiment of the present application provides an electronic device, comprising: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor can implement the method described in the first, second or third aspect above.

[0018] In a tenth aspect, an embodiment of the present application provides a computer-readable storage medium storing a computer program, wherein the computer program is implemented by a processor as described in the first aspect, the second aspect or the third aspect above.

[0019] In the eleventh aspect, an embodiment of the present application provides a computer program product, which includes a computer program code. When the computer program code is executed by a computer device, the computer device implements the method described in the first aspect, the second aspect or the third aspect above. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the following is a brief introduction to the drawings required for use in the embodiments. It should be understood that the following drawings only illustrate certain embodiments of the present application and therefore should not be regarded as limiting the scope. For ordinary technicians in this field, other relevant drawings can also be obtained based on these drawings.

[0021] Figure 1 A flowchart of a method for generating garment patterns according to an embodiment of the present application is provided; Figure 2 A schematic diagram of design drawings and garment photos in a garment pattern generation method provided in one embodiment of the present application; Figure 3 This is a schematic diagram of the display effect of a target garment pattern provided by an embodiment of the present application; Figure 4 A flowchart of generating a target garment pattern based on a pattern generation model provided in one embodiment of the present application; Figure 5 A flowchart of a method for training a target language parsing model provided in one embodiment of the present application; Figure 6 A flowchart of a method for training a target symbol translation model provided in one embodiment of the present application; Figure 7 A schematic structural diagram of a garment pattern generating device provided in one embodiment of the present application; Figure 8 A schematic diagram of the structure of a training device for a target language parsing model provided in one embodiment of the present application; Figure 9 A schematic diagram of the structure of a training device for a target symbol translation model provided in one embodiment of the present application; Figure 10 A schematic diagram of the structure of an electronic device provided in one embodiment of the present application. DETAILED DESCRIPTION

[0022] To make the objectives, technical solutions, and advantages of the embodiments of the present application more clear, the technical solutions in the embodiments of the present application will be clearly and completely described below in conjunction with the accompanying drawings of the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Generally, the components of the embodiments of the present application described and shown in the drawings herein can be arranged and designed in various different configurations.

[0023] Therefore, the following detailed description of the embodiments of the present application provided in the accompanying drawings is not intended to limit the scope of the claimed application but is merely representative of selected embodiments of the present application.

[0024] It should be noted that similar reference numerals and letters denote similar items in the following drawings, and therefore, once an item is defined in one drawing, it does not need to be further defined or explained in subsequent drawings.

[0025] In addition, the terms "first", "second", etc., if used, are merely used to distinguish and describe, and should not be understood as indicating or implying relative importance.

[0026] It should be noted that, in the absence of conflict, the features in the embodiments of this application can be combined with each other.

[0027] The first embodiment of the present application provides a method for generating clothing patterns, which is applied to a clothing pattern generating device, such as Figure 1 As shown in the figure, the clothing pattern generation method specifically includes: Step S101: Determine the clothing design language input by the user.

[0028] In step S101, the clothing design language input by the user can be a natural language, a design drawing uploaded by the user, a photo or other image language. For example, the clothing design language can be a natural language description sentence input by the user when making a pattern, such as "A black, A-line dress with a fitted bodice and flaring skirt", "a black A-line dress with a fitted bodice and flaring skirt", etc.; or the clothing design language can be a natural language description sentence uploaded by the user in the clothing pattern generation device when making a pattern, such as "A black, A-line dress with a fitted bodice and flaring skirt", etc. Figure 2 The design drawings or garment photos and other images shown can be obtained according to actual design needs.

[0029] Among them, a text input interface can be displayed to the user, and the text manually input by the user at the preset position of the interface, the text copied and pasted, etc., can be determined, and the above text can be used as the clothing design language input by the user. Alternatively, a file upload control can also be provided to the user. At this time, the user can trigger the control to select the design drawings, photos and other files stored locally in the clothing pattern generation device and upload them, or download files such as tables or images from a preset online file platform and specify them as clothing design languages. Among them, the above clothing design language can be pre-made by the user or can be temporarily generated using a file generation plug-in provided by the clothing pattern generation device (such as the AI clothing picture generation function provided by the target clothing software, etc.), and the embodiments of the present application are not limited to this.

[0030] Step S102: calling a pattern generation model to generate a target garment pattern based on the garment design language, including: converting the garment design language into a symbolic pattern making language, and outputting the target garment pattern according to the symbolic pattern making language.

[0031] Compared with the prior art, in the clothing pattern generation method provided by an embodiment of the present application, after receiving the clothing design language input by the user, the pattern generation model can be directly called to generate the target clothing pattern based on the language. Therefore, when making a pattern, the user only needs to input clothing design languages such as natural language, design manuscripts, photos, etc., without having to perform complex professional operations such as typesetting or programming. This not only significantly reduces the requirements for the user's professional ability in the clothing pattern preparation process; at the same time, due to the high degree of automation of the entire pattern making process, the user can perform a large amount of clothing pattern making work at the same time, thereby improving the overall pattern making efficiency. In addition, this solution splits the generation process of the target clothing pattern into a language generation process and a pattern output process through language conversion, reducing the degree of coupling between the steps, making the complete generation process of the target clothing pattern clearer and more orderly, thereby allowing the user to accurately control the parameters such as the size, size, shape, etc. of the pattern through symbolic pattern making language, which helps to improve the accuracy of the generated pattern.

[0032] In step S102, the garment design language is first converted into a symbolic pattern-making language, and then the target garment pattern is output according to the symbolic pattern-making language. In this way, the target garment pattern generation process can be split into a language generation process and a pattern output process, making the complete target garment pattern generation process clearer and more organized. The symbolic pattern-making language may, for example, include at least one of structural commands, symbolic programs, and program parameters. For example, a symbolic program may be a function such as "make top (_make_bodice)" or "make skirt (_make_skirt)"; a structural command may be a function set such as "stitching" or "paste" that defines the combination relationship and logical order between different symbolic programs, such as "stitch top and skirt" or "paste collar and hat"; and a program parameter may be, for example, relevant parameters for function execution, such as the size parameters of the top to be produced, the size parameters of the skirt to be produced, the stitch spacing for stitching, the width of the overlapped portion of pasting, etc.

[0033] In one embodiment, the target garment pattern can be outputted by target pattern-making software according to the symbolic pattern-making language. Pattern-making software in related art typically has the function of generating patterns based on symbolic language. By converting the garment design language into a symbolic pattern-making language recognizable by the pattern-making software, this solution is compatible with pattern-making software in related art to quickly generate target garment patterns, thereby reducing the implementation difficulty of this solution and helping to improve pattern generation efficiency.

[0034] Furthermore, to help users more intuitively understand the progress and intermediate stages of target garment pattern generation and enhance their user experience, at least a portion of the symbolic pattern-making language can be displayed to the user. For example, the aforementioned structural commands can be displayed to help the user understand the garment production process and the connections between patterns; and / or the aforementioned program parameters can be displayed to help the user understand detailed information such as the garment dimensions and the relative sizes of the individual patterns.

[0035] It should be understood that the symbolic pattern-making language, including structural commands, symbolic programs, and program parameters, is merely a specific example in the embodiments of this application and serves only as a specific, general, logical abstraction of pattern-making rules. It defines the garment objects and operations involved in the pattern-making process using the structure of structural commands, symbolic programs, and program parameters, and is not specific to the specific syntax of any industrial pattern-making language. Furthermore, the symbolic pattern-making language is independent of the specific syntax of the pattern-making software. Therefore, different pattern-making software can use the symbolic pattern-making language that follows the same syntax, which helps improve the compatibility of this solution with different pattern-making software.

[0036] Furthermore, when outputting the target garment pattern according to the symbolic pattern-making language, the symbolic pattern-making language can be first converted into a drawing command, and then the target garment pattern is output according to the drawing command. The target garment pattern can be generated according to the drawing command by the target pattern-making software, such as by inputting the drawing command into the target pattern-making software and using the output result of the target pattern-making software as the target garment pattern. If the pattern-making software has the function of generating patterns according to the drawing command, by converting the symbolic pattern-making language into a drawing command that can be recognized by the pattern-making software, the solution can be compatible with various types of pattern-making software to quickly generate target garment patterns, thereby reducing the difficulty of implementing the solution and helping to improve the efficiency of pattern generation.

[0037] The drawing command is a code segment for generating a plate, which can be applied to CAD plate-making software such as ET and Assyst. CAD plate-making software such as ET and Assyst can generate corresponding plates according to the drawing command.

[0038] Furthermore, before converting the symbolic pattern-making language into drawing commands, the target pattern-making software can be determined to facilitate conversion of the symbolic pattern-making language into drawing commands corresponding to the target pattern-making software. This ensures that the converted drawing commands can be accurately recognized by the target pattern-making software used subsequently, thereby improving the quality of pattern generation. It is understood that this solution can generate drawing commands in different formats (e.g., different syntax formats, different standards, etc.) by selecting different target pattern-making software, thereby achieving high compatibility with software formats.

[0039] In the aforementioned embodiment, the target garment pattern can be generated by the target pattern making software. The target pattern making software can be a pre-set pattern making software or a pattern making software included in the garment design language, such as "use A pattern making software to generate patterns for a black A-line dress, a corset and a flared skirt", etc., and can be set according to actual needs. Exemplarily, the pattern making software can be the aforementioned CAD pattern making software such as ET and Assyst. In addition, the target pattern making software can be run in the garment pattern generating device or in an external device connected to the garment pattern generating device, and can be used flexibly according to actual needs.

[0040] In one embodiment, after generating the target garment pattern based on the garment design language in step S102, part or all of the target garment pattern can be displayed to the user so that the user can view the display effect and then determine whether the current target garment pattern meets the user's pattern generation requirements. In order to facilitate the user to view and make decisions more intuitively and accurately, the individual patterns can be displayed in order according to the relative positional relationship between the patterns, such as displaying the individual planar patterns according to a planar relationship, or displaying the individual three-dimensional patterns in a three-dimensional space according to a spatial relationship (such as displaying the individual patterns and their connection relationships around a three-dimensional human body model), and even displaying the three-dimensional clothing wearing effect after the patterns are synthesized in a three-dimensional space according to the three-dimensional connection relationship of the individual patterns, etc., which will not be repeated here.

[0041] In one embodiment, after the user inputs the clothing design language, the clothing pattern generation device can display the program parameters and target clothing patterns generated by the above embodiment in the relevant interface of the pattern generation function. Figure 3 As shown, the relevant interface includes a parameter display area 301 and a pattern display area 302, wherein the parameter display area 301 displays multiple program parameters 303 generated by the target language parsing model; the pattern display area 302 displays the target clothing pattern 304 finally generated by the target pattern making software.

[0042] It is worth noting that the above-mentioned program parameters 303 correspond to the target clothing pattern 304 and can be displayed in a linked manner. For example, if the user is not satisfied with certain details of the currently displayed target clothing pattern 304, the user can adjust the program parameters 303 corresponding to the details in the parameter display area 301 on the left, and click the "Reset" button after the adjustment is completed to trigger the target pattern making software to regenerate the target clothing pattern according to the adjusted program parameters; conversely, the user can also directly adjust the size, relative position and other information of one or more patterns 304 by dragging, pulling, and dragging. At this time, the one or more program parameters 303 related to the adjusted pattern displayed in the left interface will change in real time as the operation proceeds, thereby realizing the linkage between the parameters and the pattern display effect, which helps the user to view the adjustment effect in real time, improve the adjustment efficiency and user experience.

[0043] Of course, if the user is not satisfied with the above program parameters 303 and / or target clothing pattern 304, he or she may click the "Re-enter" button and re-enter a new clothing design language in the interface that is subsequently displayed, and trigger the device to regenerate the corresponding target clothing pattern according to the language. No further details will be given.

[0044] After adjusting in the above way, if the user is satisfied with the target garment pattern effect, he can click the "Save" button to save the relevant data, such as directly inputting a pattern file in a standard format, etc., which will not be repeated here.

[0045] The user may not be satisfied with the target garment pattern generated by the aforementioned method. In this regard, this solution allows the user to further modify the target garment pattern. Specifically, the user can input a modified garment design language. At this time, the garment pattern generation device can determine the modified garment design language input by the user, and then convert the modified garment design language into a modified symbolic pattern-making language, and then modify the target garment pattern according to the modified symbolic pattern-making language. After generating the target garment pattern, this solution can continue to receive the modified garment design language input by the user, and implement convenient modification of the target garment pattern according to the modified garment design language, which helps to simplify the user's modification operation on the target garment pattern and improve modification efficiency.

[0046] Furthermore, when modifying the target garment pattern according to the modified symbolic pattern-making language, the modified symbolic pattern-making language is directly input into the aforementioned target pattern-making software, and the pattern output by the model is used as the modified target garment pattern. Alternatively, the modified symbolic pattern-making language may be first converted into a modified drawing command, and then the target garment pattern is modified according to the modified drawing command, such as inputting the modified drawing command into the aforementioned target pattern-making software, and then using the pattern output by the model as the modified target garment pattern.

[0047] Among them, the specific operation process of the steps of determining the modified clothing design language input by the user, converting the modified clothing design language into a modified symbolic pattern-making language, and modifying the target according to the modified symbolic pattern-making language is no different in essence from the previous steps of determining the clothing design language input by the user, converting the clothing design language into a symbolic pattern-making language, and generating a target clothing pattern according to the symbolic pattern-making language. Please refer to the previous records and will not be repeated here.

[0048] In addition, the language type of the modified clothing design language and the clothing design language described above may be the same or different. For example, both may be natural languages or image languages such as design drawings. For another example, the aforementioned clothing design language may be an image language such as design drawings, while the modified clothing design language in this embodiment may be a natural language.

[0049] As previously mentioned, in the process of generating a target garment pattern, it may be necessary to convert the garment design language into a symbolic pattern-making language; and / or, in the process of modifying a target garment pattern, it may be necessary to convert the modified garment design language into a modified symbolic pattern-making language. In one embodiment, the above conversion can be performed by the aforementioned pattern generation model. For example, when the pattern generation model includes a target language parsing model, the model can be used to convert the garment design language into the symbolic pattern-making language; and / or, to convert the modified garment design language into the modified symbolic pattern-making language.

[0050] Furthermore, during the process of generating a target garment pattern, it may be necessary to convert the symbolic pattern-making language into drawing commands; and / or, during the process of modifying a target garment pattern, it may be necessary to convert the modified symbolic pattern-making language into modified drawing commands. In one embodiment, the above conversion can also be performed by the aforementioned pattern generation model. For example, if the pattern generation model also includes a target symbol translation model, the model can be used to convert the symbolic pattern-making language into the drawing commands; and / or to convert the modified symbolic pattern-making language into the modified drawing commands.

[0051] As can be seen from the above embodiments, this solution splits the target clothing pattern generation process into three sub-processes: "the target language parsing model converts the clothing design language into the symbolic pattern-making language", "the target symbolic translation model converts the symbolic pattern-making language into the drawing command", and "the target pattern-making software generates the target clothing pattern according to the drawing command". It can be understood that because the conversion logic of converting the clothing design language into the symbolic pattern-making language is quite different from the conversion logic of converting the symbolic pattern-making language into the drawing command, this solution uses the symbolic pattern-making language as an intermediary and decouples the two conversion logics through the above splitting, so that the respective operating logics of the target language parsing model and the target symbolic translation model are clearer, the corresponding training process is simpler, the amount of data required for training is smaller, and the reasoning process is more efficient.

[0052] In one embodiment, before calling the target symbol translation model to convert the symbolic pattern-making language into drawing commands and drawing parameters, the target symbol translation model can also be determined based on the target pattern-making software. This approach can ensure that the target symbol translation model used for conversion matches the target pattern-making software used when generating the target garment pattern. That is, the pattern-making commands generated by the target symbol translation model are more compatible with the target pattern-making software used in subsequent steps, which helps improve the pattern-making quality of the target pattern-making software.

[0053] When determining the target symbol translation model based on the target board software, a target symbol translation model can be temporarily generated, or a model that matches the target board software can be selected from multiple pre-trained symbol translation models as the target symbol translation model. Taking the latter as an example, a preset symbol translation model list can be first obtained. The list is used to record the correspondence between trained sample symbol translation models and sample board software. For example, the list includes multiple sample symbol translation models and sample board software corresponding to each of the sample symbol translation models. Based on this, the sample symbol translation model corresponding to the target board software recorded in the preset symbol translation model list can be determined as the target symbol translation model. The preset symbol translation model list includes records of multiple trained sample symbol translation models and their corresponding sample board software. The target symbol translation model is determined directly by looking up the table without the need for temporary model training, which helps to improve the efficiency of determining the target symbol translation model.

[0054] Any of the target language parsing models and target symbol translation models mentioned in the aforementioned embodiments may be an LLM (Large Language Model), or an LMM (Large Multimodal Model) such as a VLM (Vision Language Model). The target language parsing model and / or target symbol translation model trained based on the aforementioned large model naturally possesses the inherent high-efficiency analysis and reasoning capabilities for massive amounts of data inherent in large models. Therefore, it is possible to ensure that the symbolic pattern-making language generated using the target language parsing model and the drawing commands generated using the target symbol translation model are more accurate, thereby helping to improve the quality of the target garment pattern ultimately generated.

[0055] In one embodiment, when using the target pattern making software to determine the target garment pattern according to the drawing command and drawing parameters, the drawing command can be first input into the target pattern making software, and the garment pattern generated by the target pattern making software can be received; then the quality inspection model can be used to determine whether the garment pattern meets the preset pattern making rules: if the garment pattern does not meet the preset pattern making rules, the drawing command can be adjusted according to the preset pattern making rules, the adjusted drawing command can be input into the target pattern making software, and the target garment pattern generated by the target pattern making software can be received; if the garment pattern meets the preset pattern making rules, the garment pattern can be used as the target garment pattern. In this way, the garment pattern generated by the target pattern making software can be inspected with the help of the pre-trained quality inspection model and the pre-set pattern making rules, and adjusted and regenerated when necessary to ensure that the target garment pattern finally obtained meets the preset pattern making rules, thereby effectively improving the pattern quality of the target garment pattern.

[0056] Among them, the quality inspection model is a mathematical model trained according to the preset board making rules. The preset board making rules can be, for example, standardized board making rules in the board making industry, or board making rules set according to one's own needs. They can be flexibly set according to actual application needs.

[0057] In some embodiments of the present application, the quality inspection model can also be a large multimodal model. Furthermore, in some embodiments, the target language parsing model and the quality inspection model can be the same large multimodal model. In this case, the model training process and training requirements can be reduced, improving board production efficiency while reducing board production costs.

[0058] Corresponding to the above embodiment, Figure 4 As shown, the pattern generation model includes a target language parsing model and a target symbol translation model. When generating a target garment pattern based on the pattern generation model, a variety of methods can be used. For example, the clothing design language input by the user can be first determined, and then the language can be input into the target language parsing model to be converted into a symbolic pattern-making language; the symbolic pattern-making language can then be input into the target symbol translation model to be further converted into a drawing command; finally, the drawing command (and corresponding drawing parameters) can be input into the target pattern-making software, and the target garment pattern outputted by the software can be received. For another example, the clothing design language input by the user can also be first determined, and then the language can be input into the target language parsing model to be converted into a symbolic pattern-making language; the symbolic pattern-making language can then be input into the target pattern-making software, and the target garment pattern outputted by the software can be received.

[0059] The following is combined with Figure 5 and Figure 6 The training process of the target language parsing model and the target symbol translation model is explained in detail.

[0060] like Figure 5 As shown, the training method of the target language parsing model is applied to the model training device, and the method includes steps 501-502. Step 501: Acquire a training data set and a standard multimodal large model, wherein each training data in the training data set includes a sample clothing design language and a sample symbolic pattern making language.

[0061] Step 502: Fine-tune the standard multimodal large model using the training dataset to obtain the target language parsing model.

[0062] As previously mentioned, the target language parsing model can be a large multimodal model. The training process for the target language parsing model in the form of a large multimodal model can be to use the constructed standard large multimodal model as the initial language parsing model as the basis, and fine-tune the initial language parsing model of the standard large multimodal model using a pre-collected training dataset.

[0063] For example, a training dataset and a standard multimodal large model can be obtained first. The training dataset can include multiple training data, where each training data can be, for example, a sample clothing design language as data and a sample symbolic pattern language as a label. For example, for data in the format of "style description-symbol program-pattern", the "style description" and "symbol program" in this data set constitute a piece of training data.

[0064] This solution uses the sample symbolic pattern-making language as the label of the sample clothing design language to construct samples and fine-tune the standard multimodal large model, so that the trained target language parsing model can have powerful analysis and reasoning capabilities for multimodal data. It can output accurate symbolic pattern-making language through comprehensive and detailed analysis of multimodal (i.e., multiple formats) clothing design language, which helps to improve the accuracy and quality of the target clothing pattern finally generated.

[0065] Accordingly, the present application also proposes a target language parsing model, which is used to: Converting the clothing design language into the symbolic pattern-making language; and / or converting the modified clothing design language into the modified symbolic pattern-making language.

[0066] The specific training process of the model and the process of language conversion can be found in the description of the previous embodiment and will not be repeated here.

[0067] like Figure 6 As shown, the training method of the target symbol translation model is applied to the model training device, and the method includes steps 601-602. Step 601: Obtain an initial model and training samples.

[0068] Step 602: Use the training samples to train the initial model to obtain the target symbol translation model, which is used to convert the symbolic board language into drawing commands and / or convert the modified symbolic board language into modified drawing commands.

[0069] In one embodiment, when obtaining the initial model and training samples, the initial symbol translation model and the target board making grammar rules of the target board making software can be obtained; accordingly, when using the training samples to train the initial model, the target board making grammar rules can be used to train the initial symbol translation model. This method obtains the target board making grammar rules of the target board making software in real time to train the target symbol translation model, so that when the target board making grammar rules change, the target symbol translation model can be changed accordingly, further making the board making commands generated by the target symbol translation model more compatible with the target board making software, thereby improving the board making quality of the target board making software.

[0070] In one embodiment, when the target board grammar rules are used to train the initial symbol translation model, a sample board language can be generated according to the target board grammar rules, and the initial symbol translation model can be trained using the sample board language. Model training of the initial symbol translation model using the target board grammar rules can, for example, generate a large number of sample board languages according to the target board grammar rules, and then use the generated sample board languages to perform model training on the constructed initialized initial symbol translation model. After the model training is completed, the target symbol translation model is obtained. This method can improve the accuracy and model quality of the trained target symbol translation model.

[0071] In one embodiment, when obtaining the initial model and training samples, an initial neural network model based on an encoder-decoder (i.e., Encoder-Decoder) conversion architecture can be obtained, and a data pair consisting of a corresponding sample symbolic pattern making language and a sample drawing command can be obtained; accordingly, when using the training samples to train the initial model, the data pair can be used as a training sample to train the initial neural network model.

[0072] Among them, the encoder can be, for example, a neural network encoder, which is used to convert natural language, design drawings, photos and other clothing design languages into feature space vector expressions, and the decoder can be, for example, a neural network decoder, which is used to parse the feature space vector into a symbolic pattern making language. The initial neural network model based on the encoder-decoder conversion architecture can be a neural network model based on the "text-text" conversion architecture, and the input of the target symbol translation model is the aforementioned symbolic pattern making language, and the output is the corresponding drawing command. In the training process of the target symbol translation model, the initial neural network model can be first constructed based on the encoder-decoder conversion architecture, and then the initial neural network model can be trained using a large amount of training data, wherein the training data can be a large amount of data pairs consisting of mutually corresponding symbolic pattern making languages and sample drawing commands. The target symbol translation model can be obtained by training the initial neural network model according to the training data.

[0073] Among them, when the target pattern-making software is running in the clothing pattern generation device, the target pattern-making software used in the above embodiments may be different. In view of the fact that different target pattern-making software usually corresponds to pattern-making commands of different grammatical formats, the sample drawing commands used for model training may also be different during the training process of the target symbol translation model. Specifically, the sample drawing commands used to train the target symbol translation model may be pattern-making commands of the grammatical format based on the target pattern-making software. In this way, different symbol translation models can be trained for different pattern-making software, so that in the use stage, the corresponding symbol translation model can be selected as the target symbol translation model according to the target pattern-making software to ensure that the symbolic pattern-making language of the unified format is converted into drawing commands that can be recognized by the target software.

[0074] Accordingly, this specification also proposes a target symbol translation model, which is used to: Converting the symbolic pattern making language into drawing commands; and / or converting the modified symbolic pattern making language into modified drawing commands.

[0075] The specific training process of the model and the process of performing the above-mentioned conversion can be found in the description of the previous embodiment and will not be repeated here.

[0076] Corresponding to the aforementioned embodiment, an embodiment of the present application provides a target language parsing model software, which is used to: convert clothing design language into symbolic pattern making language; and / or convert modified clothing design language into modified symbolic pattern making language.

[0077] Corresponding to the aforementioned embodiment, an embodiment of the present application provides a target symbol translation model software, which is used to: convert the symbolic board language into drawing commands, and / or convert the modified symbolic board language into modified drawing commands.

[0078] It should be noted that the target language parsing model software is the target language parsing model mentioned above, and the target symbol translation model software is the target symbol translation model mentioned above. Both are new software products proposed in this application, which are used to respectively realize the corresponding software functions and will not be described in detail. In addition, any of the target language parsing model software and the target symbol translation model software can be a callable independent functional software (such as clothing design software that can call the software through the calling interface opened by the independent functional software), or it can be a functional plug-in integrated into other software (such as a functional plug-in integrated in clothing design software, which can rely on clothing design software to run), and will not be described in detail.

[0079] The embodiment of the present application provides a clothing pattern generation device, such as Figure 7 Shown, including: The design language determination module 701 is used to determine the clothing design language input by the user; The garment pattern generation module 702 is used to call the pattern generation model to generate target garment patterns based on the garment design language, including: converting the garment design language into a symbolic pattern making language, and outputting the target garment pattern according to the symbolic pattern making language.

[0080] Optionally, the clothing design language includes at least one of the following: natural language, design drawings, and photos.

[0081] Optionally, the garment pattern generation module 702 is specifically configured to: convert the symbolic pattern-making language into drawing commands; and output a target garment pattern according to the drawing commands.

[0082] Optionally, the garment pattern generation module 702 is specifically configured to convert the symbolic pattern-making language into drawing commands corresponding to target pattern-making software.

[0083] Optionally, the apparatus further includes a software determining unit 703, configured to determine target plate-making software before converting the symbolic plate-making language into a drawing command corresponding to the target plate-making software.

[0084] Optionally, the symbolic board-making language includes at least one of the following: structural commands, symbolic programs, and program parameters.

[0085] Optionally, the system further includes a plate-making language display module 704 for displaying at least part of the symbolic plate-making language to the user.

[0086] Optionally, the garment pattern generation module 702 is specifically configured to output a target garment pattern according to the symbolic pattern making language using target pattern making software.

[0087] Optionally, the garment pattern generation module 702 is specifically configured to output a target garment pattern according to the drawing command using the target pattern making software.

[0088] Optionally, a modification module 705 is further included, which is used to: determine the modified clothing design language input by the user; convert the modified clothing design language into a modified symbolic pattern making language; and modify the target clothing pattern according to the modified symbolic pattern making language.

[0089] Optionally, the modification module 705 is specifically configured to: convert the modification symbolic pattern-making language into a modification drawing command; and modify the target garment pattern according to the modification drawing command.

[0090] Optionally, the pattern generation model includes a target language parsing model, and the target language parsing model is used to convert the clothing design language into the symbolic pattern making language.

[0091] Optionally, the pattern generation model includes a target language parsing model, and the target language parsing model is used to: convert the clothing design language into the symbolic pattern making language; and / or convert the modified clothing design language into the modified symbolic pattern making language.

[0092] Optionally, the plate generation model further includes a target symbol translation model, and the target symbol translation model is used to convert the symbolic plate making language into the drawing command.

[0093] Optionally, the plate generation model further includes a target symbol translation model, and the target symbol translation model is used to: convert the symbolic plate making language into the drawing command; and / or convert the modified symbolic plate making language into the modified drawing command.

[0094] Optionally, the target language parsing model is a large multimodal model.

[0095] Optionally, the system further includes a pattern display unit 706 for displaying the target clothing pattern to the user.

[0096] The present application embodiment provides a training device for a target language parsing model, such as Figure 8 Shown, including: The standard model acquisition unit 801 is used to acquire a training data set and a standard multimodal large model, wherein each training data in the training data set includes a sample clothing design language and a sample symbolic pattern making language; The parsing model training unit 802 is used to fine-tune the standard multimodal large model using the training data set to obtain the target language parsing model.

[0097] The present application embodiment provides a training device for a target symbol translation model, such as Figure 9 Shown, including: Initial model acquisition unit 901, used to acquire an initial model and training samples; The translation model training unit 902 is used to train the initial model using the training samples to obtain the target symbol translation model, which is used to convert the symbolic board language into drawing commands and / or convert the modified symbolic board language into modified drawing commands.

[0098] Optionally, the initial model acquisition unit 901 is specifically used to: obtain an initial symbol translation model and target board making grammar rules of the target board making software; the translation model training unit 902 is specifically used to: use the target board making grammar rules to train the initial symbol translation model.

[0099] Optionally, the translation model training unit 902 is specifically configured to generate a sample board language according to the target board grammar rules, and perform training using the initial symbol translation model of the sample board language.

[0100] Optionally, the initial model acquisition unit 901 is specifically used to: obtain an initial neural network model based on the encoder-decoder conversion architecture, and obtain data pairs consisting of corresponding sample symbolic pattern making languages and sample drawing commands; the translation model training unit 902 is specifically used to: use the data pairs as training samples to train the initial neural network model.

[0101] The present application embodiment provides an electronic device, such as Figure 10 As shown, it includes: at least one processor 1001; and a memory 1002 that is communicatively connected to the at least one processor 1001; wherein the memory 1002 stores instructions that can be executed by the at least one processor 1001, and the instructions are executed by the at least one processor 1001 so that the at least one processor 1001 can execute the methods in the above embodiments.

[0102] The memory and processor are connected using a bus, which can include any number of interconnected buses and bridges. The bus connects various circuits of one or more processors and memories. The bus can also connect various other circuits such as peripheral devices, voltage regulators, and power management circuits. These are all well known in the art and are therefore not described further herein. The bus interface provides an interface between the bus and the transceiver. The transceiver can be a single component or multiple components, such as multiple receivers and transmitters, providing a unit for communicating with various other devices over a transmission medium. Data processed by the processor is transmitted over a wireless medium via an antenna. Furthermore, the antenna receives data and transmits it to the processor.

[0103] The processor is responsible for managing the bus and general processing, and can also provide various functions, including timing, peripheral interfaces, voltage regulation, power management, and other control functions. Memory can be used to store data used by the processor when performing operations.

[0104] In the above embodiments, they can be implemented in whole or in part by software, hardware, firmware or any combination thereof. When implemented using software, they can be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on the device, the process or function described in the embodiment of the present application is generated in whole or in part. The computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another computer-readable storage medium. For example, the computer instructions can be transmitted from one website, computer, server or data center to another website, computer, server or data center via a wired (e.g., coaxial cable, optical fiber, digital subscriber line) or wireless (e.g., infrared, wireless, microwave, etc.) method. The computer-readable storage medium can be any available medium that can be accessed by the device or a data storage device such as a server or data center that includes one or more available media. The available medium can be a magnetic medium (e.g., a floppy disk, a hard disk, and a tape, etc.), an optical medium (e.g., a digital video disk (DVD), etc.), or a semiconductor medium (e.g., a solid-state drive, etc.).

[0105] Those skilled in the art will understand that all or part of the steps to implement the above embodiments may be accomplished by hardware, or by a program to instruct the relevant hardware, and the program may be stored in a computer-readable storage medium, which may be a read-only memory, a disk, or an optical disk, etc.

[0106] The above are only specific embodiments of the present application, but the scope of protection of this application is not limited thereto. Any changes or substitutions that can be easily conceived by a person skilled in the art within the technical scope disclosed in this application should be included in the scope of protection of this application. Therefore, the scope of protection of this application should be based on the scope of protection of the claims.

Claims

1. A method for generating clothing patterns, characterized in that: include: Determine the clothing design language input by the user; The pattern generation model is called to generate a target garment pattern based on the garment design language, including: converting the garment design language into a symbolic pattern making language, and outputting the target garment pattern according to the symbolic pattern making language.

2. The method according to claim 1, characterized in that The clothing design language includes at least one of the following: natural language, design drawings, and photos.

3. The method according to claim 1, characterized in that Outputting a target garment pattern according to the symbolic pattern-making language includes: Converting the symbolic board-making language into drawing commands; Output the target garment pattern according to the drawing command.

4. The method according to claim 3, characterized in that The converting the symbolic board-making language into a drawing command comprises: The symbolic plate-making language is converted into drawing commands corresponding to the target plate-making software.

5. The method according to claim 4, characterized in that Also includes: Before converting the symbolic plate-making language into a drawing command corresponding to the target plate-making software, the target plate-making software is determined.

6. The method according to any one of claims 1 to 5, characterized in that The symbolic board design language includes at least one of the following: structural commands, symbolic programs, and program parameters.

7. The method according to claim 6, characterized in that Also includes: At least a portion of the symbolic platemaking language is presented to the user.

8. The method according to claim 1, characterized in that Outputting a target garment pattern according to the symbolic pattern-making language includes: Outputting target garment patterns according to the symbolic pattern making language through target pattern making software.

9. The method according to claim 3 or 4, characterized in that Outputting a target garment pattern according to the drawing command includes: Output the target garment pattern according to the drawing command through the target pattern making software.

10. The method according to claim 1, characterized in that Also includes: Determine the modified clothing design language input by the user; Converting the modified clothing design language into a modified symbolic pattern making language; The target garment pattern is modified according to the modified symbolic pattern-making language.

11. The method according to claim 10, characterized in that The modifying the target garment pattern according to the modified symbolic pattern making language includes: Converting the modified symbolic board language into a modified drawing command; The target garment pattern is modified according to the modification drawing command.

12. The method according to any one of claims 1 to 5 and 7 to 8, characterized in that The plate generation model includes a target language parsing model, and the target language parsing model is used to: The clothing design language is converted into the symbolic pattern making language.

13. The method according to claim 10 or 11, characterized in that The plate generation model includes a target language parsing model, and the target language parsing model is used to: Converting the clothing design language into the symbolic pattern making language; and / or, The modified clothing design language is converted into the modified symbolic pattern making language.

14. The method according to any one of claims 1 to 5 and 7 to 8, characterized in that The board generation model further includes a target symbol translation model, which is used to: The symbolic board language is converted into the drawing command.

15. The method according to claim 10 or 11, characterized in that The board generation model further includes a target symbol translation model, which is used to: Converting the symbolic board language into the drawing command; and / or, The modified symbolic plate making language is converted into the modified drawing command.

16. The method according to claim 12, characterized in that The target language parsing model is a large multimodal model.

17. The method according to any one of claims 1-5, 7-8 and 10-11, characterized in that Also includes: The target garment pattern is presented to the user.

18. A method for training a target language parsing model, characterized in that: include: Acquire a training data set and a standard multimodal large model, wherein each training data in the training data set includes a sample clothing design language and a sample symbolic pattern making language; The standard multimodal large model is fine-tuned using the training dataset to obtain the target language parsing model.

19. A method for training a target symbol translation model, characterized in that: include: Obtain the initial model and training samples; The initial model is trained using the training samples to obtain the target symbol translation model, which is used to convert the symbolic board language into drawing commands; and / or convert the modified symbolic board language into modified drawing commands.

20. The method according to claim 19, wherein The obtaining of the initial model and training samples includes: obtaining the initial symbol translation model and target board making grammar rules of the target board making software; The using the training samples to train the initial model includes: using the target board grammar rules to train the initial symbol translation model.

21. The method according to claim 20, characterized in that The training of the initial symbol translation model using the target board grammar rules includes: A sample board language is generated according to the target board grammar rule, and the initial symbol translation model of the sample board language is used for training.

22. The method according to claim 21, characterized in that The obtaining of the initial model and training samples includes: obtaining an initial neural network model based on an encoder-decoder conversion architecture, and obtaining data pairs consisting of corresponding sample symbolic pattern making languages and sample drawing commands; The using the training samples to train the initial model includes: using the data pairs as training samples to train the initial neural network model.

23. A target language parsing model software, characterized in that: The target language parsing model software is used to: Converting clothing design language into symbolic pattern making language; and / or, Convert the modified clothing design language into the modified symbolic pattern making language.

24. A target symbol translation model software, characterized in that: The target symbol translation model software is used to: Converting the symbolic drawing language into drawing commands; and / or, Convert the modified symbolic board language into modified drawing commands.

25. A clothing pattern generating device, characterized in that: include: A design language determination module, used to determine the clothing design language input by the user; The clothing pattern generation module is used to call the pattern generation model to generate target clothing patterns based on the clothing design language, specifically to convert the clothing design language into a symbolic pattern making language, and output the target clothing pattern according to the symbolic pattern making language.

26. A training device for a target language parsing model, characterized in that: include: A standard model acquisition unit, configured to acquire a training data set and a standard multimodal large model, wherein each training data in the training data set includes a sample clothing design language and a sample symbolic pattern making language; A parsing model training unit is used to fine-tune the standard multimodal large model using the training data set to obtain the target language parsing model.

27. A training device for a target symbol translation model, characterized in that: include: An initial model acquisition unit, used to acquire an initial model and training samples; The translation model training unit is used to train the initial model using the training samples to obtain the target symbol translation model, which is used to convert the symbolic board language into drawing commands and / or convert the modified symbolic board language into modified drawing commands.

28. An electronic device, characterized in that: include: at least one processor; and, a memory communicatively coupled to the at least one processor; The memory stores instructions that can be executed by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to perform the method according to any one of claims 1 to 22.

29. A computer-readable storage medium storing a computer program, characterized in that: The computer program is executed by a processor to implement the method according to any one of claims 1 to 22.

30. A computer program product, characterized in that The computer program product comprises computer program code. When the computer program code is executed by a computer device, the computer device performs the method according to any one of claims 1 to 22.

Images