Fashion design 3D modeling visualization display system
Through innovative technology of the three-dimensional modeling visual display system for clothing design, the problem of inaccurate matching of clothing models and model models is solved, efficient modeling error analysis and display is achieved, and the accuracy and efficiency of clothing design is improved.
Patent Information
- Application Number
- CN202411348044.3
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2024-09-26
- Publication Date
- 2024-11-29
- Estimated Expiration
- 2044-09-26
AI Technical Summary
The existing three-dimensional modeling technology of clothing design has problems such as low modeling accuracy and low correction efficiency, especially the inaccurate matching between clothing models and model models, which leads to frequent model wearing phenomena, affecting the display effect and design understanding.
By providing a three-dimensional modeling visual display system for clothing design, the three-dimensional model acquisition module, independent grid accuracy analysis module, integrated grid accuracy analysis module, modeling accuracy analysis module and visual display module are used to carry out clothing and model models. Raster division, index marking, random search and fusion search calculate modeling accuracy, and realize accurate modeling error analysis and visual display.
The matching accuracy between the three-dimensional model of clothing and the model model is improved, the systematicity and accuracy of modeling are enhanced, the frequency of modeling problems is reduced, and the efficiency and accuracy of clothing design modeling are improved.
Smart Images

Figure CN118864785B_ABST
Abstract
Description
Technical Field
[0001] The invention relates to the technical field of data modeling, and in particular to a three-dimensional modeling visualization display system for clothing design. Background Art
[0002] In the existing 3D modeling process of clothing design, designers need to fit the 3D model of clothing with the 3D model of the model to ensure the realism of the clothing wearing effect. There are some problems in 3D modeling of clothing design. For example, one of the common modeling problems is "penetration", that is, the mutual penetration between the clothing model and the model model. This problem not only affects the final clothing display effect, but also interferes with the designer's understanding of clothing structure and form, reducing the accuracy and efficiency of modeling.
[0003] Although the 3D modeling software currently available on the market has certain model checking and error recognition functions, most of them are limited to simply identifying the inaccurate parameters of the model and cannot systematically detect the inaccurate modeling of the entire clothing model. At the same time, the existing solutions lack sufficient visual recognition and display methods for the modeling accuracy of multiple clothing areas, such as the modeling effects of sleeves and hems. This limitation makes it difficult for designers to quickly and intuitively locate problem areas, and when adjusting clothing models, they often need to rely on repeated manual inspection and adjustment, resulting in low work efficiency and low clothing design modeling accuracy. Summary of the invention
[0004] The present invention aims to solve the technical problems that the three-dimensional modeling of clothing design in the prior art cannot visualize the modeling accuracy, has low modeling accuracy and low modeling correction efficiency.
[0005] In a first aspect, the present invention provides a clothing design three-dimensional modeling visualization display system, which includes: a three-dimensional model acquisition module, which is used to acquire a clothing three-dimensional model initially constructed by a user, and acquire a model three-dimensional model, and perform model grid division and index marking on the clothing three-dimensional model and the model three-dimensional model to obtain a plurality of clothing model grids and a plurality of model model grids that index each other;
[0006] An independent grid accuracy analysis module is used to set multiple search probabilities according to multiple historical modeling accuracies of multiple clothing areas, randomly search a preset number of multiple search model grid combinations in the multiple clothing model grids and multiple model model grids, and analyze and obtain multiple independent model grid accuracy rates according to multiple search grid model parameter combinations of the multiple search model grid combinations, wherein the independent model grid accuracy rate is analyzed according to the matching degree of clothing grid model parameters and model grid model parameters in each search grid model parameter combination;
[0007] A fusion grid accuracy analysis module is used to set multiple fusion rounds and multiple fusion quantities according to the multiple historical modeling accuracies, search and fuse the model grid accuracy of the adjacent model grids for the multiple search model grid combinations, and obtain multiple fusion model grid accuracy rates;
[0008] A modeling accuracy analysis module, used for calculating and obtaining regional modeling accuracy parameters of the plurality of clothing regions and model modeling accuracy parameters of the clothing three-dimensional model according to the plurality of independent model grid accuracy rates and the plurality of fusion model grid accuracy rates;
[0009] The visualization display module is used to update the multiple historical modeling accuracies according to the multiple regional modeling accuracy parameters, and to visually identify and display the clothing three-dimensional model according to the multiple regional modeling accuracy parameters and the model modeling accuracy parameters.
[0010] In a second aspect, the present invention provides an electronic device comprising: a memory for storing a computer software program; and a processor for reading and executing the computer software program, thereby realizing the clothing design three-dimensional modeling visualization display system of the first aspect.
[0011] In a third aspect, the present invention provides a computer-readable storage medium having a computer software program stored therein, and when the computer software program is executed by a processor, the clothing design three-dimensional modeling visualization display system of the first aspect is implemented.
[0012] The present invention can effectively improve the matching accuracy between the clothing three-dimensional model and the model three-dimensional model through one or more of the above-mentioned technical solutions, and make the modeling process more systematic and accurate through innovative grid division and index marking technology. Specifically, by grid division and index marking of the initially constructed clothing three-dimensional model and the model three-dimensional model, a detailed association between multiple clothing areas and the model is established, so that the system can analyze the local modeling error more accurately. Then, combined with the historical modeling accuracy, the clothing and model grid combination is randomly searched, and the accuracy of the independent model grid is calculated by the matching degree of the three-dimensional model parameters. In this way, local modeling inaccuracy problems can be detected in multiple areas of clothing, rather than being limited to a single area. In addition, the model accuracy of the adjacent grids is subjected to multiple rounds and multiple number of fusion searches by adopting the fusion search method, which further improves the comprehensiveness and stability of the modeling accuracy. By calculating the independent model grid accuracy and the fused model grid accuracy, the scheme can accurately classify the modeling accuracy parameters of different clothing areas and generate the overall model modeling accuracy parameters. This process not only provides dual accuracy analysis of the whole and the part, but also enables the system to continuously optimize itself and improve the accuracy and efficiency of subsequent modeling by updating the historical modeling accuracy data in real time. Finally, through the visual identification and display based on the regional modeling accuracy parameters and the overall modeling accuracy parameters, designers can intuitively identify the problem areas in the modeling and make targeted adjustments. This precise modeling evaluation and feedback mechanism greatly improves the accuracy of three-dimensional modeling of clothing design, reduces the frequency of problems such as model penetration, and achieves the technical effect of improving the efficiency and accuracy of clothing design modeling. BRIEF DESCRIPTION OF THE DRAWINGS
[0013] Figure 1 A schematic diagram of the structure of a clothing design three-dimensional modeling and visualization display system provided by an embodiment of the present invention;
[0014] Figure 2 A functional block diagram of an electronic device provided by an embodiment of the present invention;
[0015] Figure 3 A principle block diagram of the computer-readable storage medium provided by the present invention.
[0016] In the accompanying drawings, the components represented by the reference numerals are listed as follows:
[0017] Three-dimensional model acquisition module 11, independent grid accuracy analysis module 12, fused grid accuracy analysis module 13, modeling accuracy analysis module 14, visualization display module 15, electronic device 100, memory 110, processor 120, computer program 111, computer readable storage medium 200, computer program 211. DETAILED DESCRIPTION
[0018] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in 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. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative work are within the scope of protection of this application.
[0019] In the description of this application, the terms "first" and "second" are used for descriptive purposes only and should not be understood as indicating or implying relative importance or implicitly indicating the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of the features. In the description of this application, "plurality" means two or more, unless otherwise clearly and specifically defined.
[0020] In the description of the present application, the term "for example" is used to mean "used as an example, illustration or description". Any embodiment described as "for example" in the present application is not necessarily to be construed as being more preferred or advantageous than other embodiments. The following description is given to enable any technician in the field to implement and use the present invention. In the following description, details are listed for the purpose of explanation. It should be understood that a person of ordinary skill in the art can recognize that the present invention can be implemented without using these specific details. In other examples, well-known structures and processes will not be elaborated in detail to avoid unnecessary details to obscure the description of the present invention. Therefore, the present invention is not intended to be limited to the embodiments shown, but is consistent with the widest scope consistent with the principles and features disclosed in the present application.
[0021] Example 1
[0022] like Figure 1 As shown, an embodiment of the present invention provides a clothing design three-dimensional modeling visualization display system, including: a three-dimensional model acquisition module 11, an independent grid accuracy analysis module 12, a fused grid accuracy analysis module 13, a modeling accuracy analysis module 14 and a visualization display module 15.
[0023] In the embodiment of the present application, the three-dimensional model acquisition module 11 is used to acquire the clothing three-dimensional model initially constructed by the user, and acquire the model three-dimensional model, perform model grid division and index marking on the clothing three-dimensional model and the model three-dimensional model, and obtain a plurality of clothing model grids and a plurality of model model grids that index each other;
[0024] In the embodiment of the present application, the three-dimensional model of clothing and the three-dimensional model of the model preliminarily designed and constructed by the user are obtained. First, the three-dimensional model acquisition module 11 extracts the three-dimensional model of clothing preliminarily designed and constructed by the user and the three-dimensional model of the model used to display the three-dimensional model of clothing through the design file provided by the user or the model data stored in the modeling software. Generally, the three-dimensional model of clothing needs to be constructed according to the design and material fitting model three-dimensional model to improve the effect of displaying the clothing design.
[0025] After obtaining the model, the three-dimensional clothing model and the three-dimensional model of the model are grid-divided. This step is to split the complex three-dimensional model into multiple grid units, each of which represents an independent small area in the model. The purpose of grid division is to simplify the calculation of the model so that matching analysis and accuracy detection can be performed on each small area separately. After grid division, index marking is performed for each clothing model grid and model model grid. Index marking refers to the construction of an index relationship for the corresponding clothing model grid and model model grid, so that the clothing model grid and the model model grid can index each other, thereby establishing an accurate association between the clothing model and the model model, facilitating the subsequent analysis of modeling accuracy and improving analysis accuracy.
[0026] In the embodiment of the present application, the clothing 3D model and the model 3D model are divided into model grids and indexed to obtain a plurality of clothing model grids and a plurality of model model grids that index each other, including:
[0027] Performing grid division on the clothing three-dimensional model to obtain a plurality of clothing model grids and clothing grid coordinate information of the plurality of model model grids;
[0028] Dividing the model three-dimensional model into grids to obtain a plurality of model model grids and model grid coordinate information of the plurality of model model grids;
[0029] According to the plurality of clothing grid coordinate information and the model grid coordinate information, an index relationship between the plurality of clothing model grids and the plurality of model model grids is constructed.
[0030] In an embodiment of the present application, when processing a three-dimensional clothing model, the three-dimensional clothing model is first grid-divided. Grid division refers to splitting a complex three-dimensional clothing model into multiple small grid units, such as dividing the model surface of the three-dimensional clothing model into multiple grid areas of the same size, and each grid unit represents an independent part of the model, which can improve the analysis accuracy and calculation efficiency of the three-dimensional clothing model. In the process of grid division, corresponding coordinate information will be generated for each grid of the clothing model. For example, coordinate information is assigned to each grid according to the coordinate system of the three-dimensional clothing model. These coordinate information are used to accurately locate the position of each grid unit in three-dimensional space. These coordinate information enable each grid to be bound to a specific spatial position, thereby supporting subsequent precision analysis and model matching. In this way, multiple clothing model grids and multiple clothing grid coordinate information of multiple model model grids are obtained.
[0031] Next, the model 3D model is similarly divided into grids to generate multiple model model grids. Each model model grid also generates corresponding model grid coordinate information, which is also used to calibrate the position of each grid unit in the model 3D model.
[0032] After obtaining the grids and coordinate information of clothing and models, the index relationship between multiple clothing model grids and multiple model model grids is constructed based on these coordinate information. The index relationship is constructed by comparing the spatial coordinates of clothing grids and model grids to determine their relative positions and matching degrees in three-dimensional space. By analyzing the relative position relationship between clothing grids and model grids, the matching accuracy between clothing and model models can be effectively tracked and identified, especially the possible interaction (such as "piercing") between adjacent clothing model grids and model model grids.
[0033] Exemplarily, according to each clothing grid coordinate information, model grid coordinate information with overlapping coordinates and the largest repetition amplitude, or the closest model grid coordinate information, can be screened out from multiple model grid coordinate information as indexed model grid coordinate information; or, an index relationship between multiple clothing grid coordinate information and multiple model grid coordinate information can be manually constructed until the index relationship between multiple clothing grid coordinate information and multiple model grid coordinate information is completed, and an index relationship between multiple clothing model grids and multiple model model grids is constructed according to the grids corresponding to the coordinates.
[0034] Through the above method, it is ensured that each unit of the clothing grid and the model grid can be accurately located and matched, providing solid technical support for subsequent model accuracy analysis and optimization.
[0035] The independent grid accuracy analysis module 12 is used to set multiple search probabilities according to multiple historical modeling accuracies of multiple clothing areas, randomly search a preset number of multiple search model grid combinations in the multiple clothing model grids and multiple model model grids, and analyze and obtain multiple independent model grid accuracy rates according to multiple search grid model parameter combinations of the multiple search model grid combinations, wherein the independent model grid accuracy rate is analyzed according to the matching degree of clothing grid model parameters and model grid model parameters in each search grid model parameter combination;
[0036] In the embodiment of the present application, the multiple clothing areas are multiple areas in the clothing design process, such as sleeves, collars, hems, etc., and multiple search probabilities are set for different areas according to the historical modeling accuracy of the multiple clothing areas. The size of the historical modeling accuracy is negatively correlated with the size of the search probability. The multiple search probabilities are used to randomly search the clothing model grids and the model model grids in the multiple clothing areas, so that the clothing areas with lower historical modeling accuracy can search more clothing model grids, thereby improving the number of modeling accuracy analysis grids and analysis accuracy of clothing areas with lower modeling accuracy. In this way, the detection scale and detection accuracy of key clothing areas are improved.
[0037] In the embodiment of the present application, according to multiple search probabilities, a random search is performed in multiple clothing model grids and multiple model model grids to obtain clothing model grids and model model grids corresponding to multiple indexes as multiple search model grid combinations. The embodiment of the present application does not calculate each grid one by one, but by setting a preset number, randomly searching and selecting multiple search model grid combinations, which can improve the efficiency of clothing design modeling accuracy analysis and reduce the analysis and calculation costs.
[0038] Further, according to the multiple search grid model parameter combinations of the multiple search model grid combinations, the accuracy rates of multiple independent model grids are analyzed and obtained. The search grid model parameters include the three-dimensional model grid parameters in the search model grid, such as the geometry, position, size and other parameters of the three-dimensional model, each search model grid combination corresponds to a search grid model parameter combination, and each search grid model parameter combination includes the clothing grid model parameters and the model grid model parameters of the clothing model grid and the model model grid corresponding to the index.
[0039] Among them, by judging whether the clothing model grid and the model model grid are consistent in spatial position and shape according to the difference between the clothing grid model parameters and the model grid model parameters, and then analyzing the accuracy of this group of clothing model grids and model model grids in clothing three-dimensional modeling, which is used as the independent modeling accuracy parameter of the clothing model grid, that is, the independent model grid accuracy rate.
[0040] Through the above-mentioned random search and matching analysis of model grid combinations, the accuracy of multiple search model grid combinations can be generated. This process can not only identify the specific grids with errors in modeling, but also provide designers with detailed accuracy analysis results to help them quickly locate problem areas and then optimize and adjust them.
[0041] In the embodiment of the present application, according to multiple historical modeling accuracies of multiple clothing areas, multiple search probabilities are set, and a preset number of multiple search model grid combinations are randomly searched in the multiple clothing model grids and the multiple model model grids, including:
[0042] Acquire multiple clothing areas in the clothing three-dimensional model, and extract multiple historical modeling accuracies of the multiple clothing areas according to clothing design three-dimensional modeling records in historical time;
[0043] Calculating and setting a plurality of search probabilities according to the plurality of historical modeling accuracies;
[0044] According to the multiple search probabilities and the preset numbers, a random search of model grids is performed within the multiple clothing model grids and the multiple model model grids to obtain multiple search clothing model grids and multiple search model model grids, which are combined as multiple search model grid combinations.
[0045] Wherein, according to the multiple historical modeling accuracies, multiple search probabilities are calculated and set, including:
[0046] According to each historical modeling accuracy and the sum of the multiple historical modeling accuracies, a calculation is set to obtain multiple search probabilities, wherein the size of the search probability is negatively correlated with the size of the historical modeling accuracy.
[0047] In the embodiment of the present application, multiple clothing regions are obtained from the clothing three-dimensional model, and the modeling data of the multiple clothing regions can be obtained according to the pre-modeled region identifiers or the building blocks of different clothing regions. The multiple clothing regions may include specific clothing parts such as collars, sleeves, and trouser legs. The modeling complexity of each clothing region is different, so targeted analysis is performed according to the characteristics of each clothing region. For example, the sleeve part has a greater construction complexity due to the large number of details and angles of the model's three-dimensional model arm.
[0048] Refer to the three-dimensional modeling records of clothing design in historical time, and extract the historical modeling accuracy of each clothing area. The historical modeling accuracy refers to the modeling accuracy performance of different clothing areas in the previous modeling process. For example, multiple historical modeling accuracies can be obtained through the modeling accuracy evaluation data of each clothing area recorded by the technicians in the three-dimensional modeling records of clothing design in historical time, or multiple historical modeling accuracies of multiple clothing areas can be obtained based on the modeling data in the last clothing design modeling combined with the method provided in the embodiment of the present application. Exemplarily, multiple historical modeling accuracies can be obtained by calculating the mold penetration area of the three-dimensional model of each clothing area in the three-dimensional modeling records of clothing design in historical time, and the size of the mold penetration area is inversely proportional to the historical modeling accuracy. For example, the size range of the historical modeling accuracy is (0,1).
[0049] Based on multiple historical modeling accuracies, multiple search probabilities are calculated and set. The size of the search probability is negatively correlated with the size of the historical modeling accuracy. A higher search probability means that the area has more modeling errors in the historical records, so it needs more attention and detection; while a lower search probability means that the modeling of the area is relatively accurate, and the search probability can be appropriately reduced.
[0050] Exemplarily, a plurality of search probabilities are calculated based on each historical modeling accuracy and the sum of multiple historical modeling accuracies. For example, the inverse of the ratio of each historical modeling accuracy to the sum of multiple historical modeling accuracies is calculated to obtain multiple inverses, and then the ratio of each inverse to the sum of multiple inverses is calculated as the search probability to obtain multiple search probabilities.
[0051] Thus, if the historical modeling accuracy of a certain clothing area is low, a higher search probability will be assigned so that this area can be focused on in subsequent modeling and more model grids can be checked and analyzed. On the contrary, if the historical modeling accuracy of a certain area is high, the search probability of this area is low, indicating that this area has been stable in the past modeling process and does not need to be checked with more model grids.
[0052] Further, according to the plurality of search probabilities and the preset number, a random search is performed in the clothing model grid and the mannequin model grid. The preset number is a preset search number, and the more the preset number is, the more clothing model grids and mannequin model grids are searched, and the preset number can be set according to the construction complexity of the clothing three-dimensional model, for example, 100.
[0053] According to multiple search probabilities and preset quantities, a random search of model grids is performed in clothing model grids and model model grids within multiple clothing areas, wherein the clothing area to which each model grid belongs can be determined according to the grid coordinates of the multiple clothing model grids and the multiple model model grids, and then, a random search is performed according to the multiple search probabilities to obtain multiple search clothing model grids and multiple search model model grids, which are combined as multiple search model grid combinations.
[0054] Preferably, after searching, the ratio of the number of search model grid combinations belonging to multiple clothing areas is the same as the multiple search probabilities, so as to achieve the purpose of low historical modeling accuracy, high search probability, and a large number of searched model grids. Optionally, the preset number can be allocated according to the multiple search probabilities to obtain multiple clothing area search numbers, and then the model grids are randomly searched in the clothing model grids and the model model grids in the multiple clothing areas to obtain the search model grid combinations of multiple clothing area search numbers, and finally obtain multiple search model grid combinations.
[0055] In the embodiment of the present application, according to the multiple search grid model parameter combinations of the multiple search model grid combinations, multiple independent model grid accuracy rates are analyzed and obtained, including:
[0056] Based on the clothing three-dimensional model and the model three-dimensional model, clothing grid model parameters and model grid model parameters of multiple search clothing model grids and multiple search model model grids within the multiple independent model grid accuracy rates are extracted to obtain multiple search grid model parameter combinations;
[0057] Collecting a set of sample search grid model parameter combinations, annotating the matching degree of clothing grid model parameters and model grid model parameters for each sample search grid model parameter combination, and obtaining a set of sample independent model grid accuracy rates;
[0058] Using the sample search grid model parameter combination set and the sample independent model grid accuracy set, training an independent model accuracy identifier;
[0059] The independent model accurate identifier is used to perform independent accurate analysis and identification of model grids on the multiple search grid model parameter combinations to obtain multiple independent model grid accuracy rates.
[0060] In the embodiment of the present application, based on multiple search grid model parameter combinations of multiple search model grid combinations, model grid modeling accuracy analysis of multiple search clothing model grids is performed to obtain multiple independent model grid accuracy rates.
[0061] Based on the clothing 3D model and the model 3D model, clothing grid model parameters and model grid model parameters of each model grid are extracted from multiple search clothing model grids and multiple search model model grids. The grid model parameters generally include the geometric shape, position coordinates, size information, etc. of the grid model, which together constitute a search grid model parameter combination, and multiple search grid model parameter combinations are obtained.
[0062] Furthermore, an independent model accuracy identifier is constructed through machine learning, and the model grid modeling accuracy analysis of the clothing model grid is performed on multiple search grid model parameter combinations. Among them, the independent model accuracy identifier is trained based on the modeling matching degree of the clothing grid model parameters and the corresponding model grid model parameters to analyze the modeling accuracy.
[0063] Collect a set of sample search grid model parameter combinations, annotate the matching degree of clothing grid model parameters and model grid model parameters for each sample search grid model parameter combination, and obtain a sample independent model grid accuracy set. Wherein, by way of sample collection, based on the historical clothing three-dimensional model and historical model three-dimensional model previously used for clothing design three-dimensional modeling, grid division and grid model parameter collection are performed by the above method to obtain a set of sample search grid model parameter combinations, each of which includes sample clothing grid model parameters and sample model grid model parameters.
[0064] Further, according to the matching degree of the sample clothing grid model parameters and the sample model grid model parameters in each sample search grid model parameter combination, annotation is performed to obtain a sample independent model grid accuracy set. Matching degree annotation refers to assigning a matching degree score to each grid model parameter combination to evaluate the degree of fit between the clothing model and the model model in the model grid unit, thereby reflecting the degree to which the clothing three-dimensional model is constructed to fit the model three-dimensional model. Exemplarily, the sample distance between the sample clothing model grid and the sample model model grid is calculated according to the sample clothing grid model parameters and the sample model grid model parameters, and the matching degree is annotated according to the deviation between the sample distance and the preset distance. The larger the deviation, the lower the accuracy of the clothing three-dimensional model construction and the smaller the matching degree. When the deviation is 0, the matching degree is 1 (the value range of the matching degree is, for example, 0 to 1), and the matching degree corresponding to the maximum deviation is 0. The preset distance is the ideal distance between the clothing model grid and the model model grid when modeling, for example, 1 mm. Through the above annotation, a sample independent model grid accuracy set is obtained.
[0065] Furthermore, the sample search grid model parameter combination set and the sample independent model grid accuracy set are used to train the independent model accurate identifier through machine learning methods. The sample search grid model parameter combination set and the sample independent model grid accuracy set are used as training data sets. For example, the independent model accurate identifier is constructed using a BP neural network, the search grid model parameter combination is used as input, and the independent model grid accuracy is used as output. The sample search grid model parameter combination is input into the independent model accurate identifier, and the parameters of the independent model accurate identifier are adjusted according to the deviation between the output and the corresponding sample independent model grid accuracy, so that the error can be minimized and the matching degree of the grid model parameter combination can be accurately predicted. Repeated iterative training and parameter adjustment are performed, and each iteration allows the independent model accurate identifier to adjust the weights and parameters according to the feedback of the sample data, gradually reduce the prediction error, and improve the accuracy of the model until the requirements are met, such as the error is less than 4%, and the training is completed.
[0066] The trained independent model accurate identifier is used to perform independent and accurate analysis and identification of multiple search grid model parameter combinations. The multiple search grid model parameter combinations are input into the independent model accurate identifier, and multiple independent model grid accuracy rates are obtained by identification, which reflects the matching degree between the clothing grid model parameters and the model grid model parameters in the multiple search grid model parameter combinations, and reflects the model construction accuracy of multiple search clothing model grids in the multiple search model grid combinations.
[0067] Through the above content, the model construction accuracy of multiple search clothing model grids within multiple search model grid combinations was identified, which can refine the search and identify the accuracy of model construction in multiple clothing areas, and then reflect the construction accuracy of different clothing areas and the overall model grid, which serves as the data basis for the subsequent display of the construction accuracy of different areas of the model, and improves the accuracy and efficiency of clothing three-dimensional model design and construction.
[0068] The fusion grid accuracy analysis module 13 is used to set multiple fusion rounds and multiple fusion quantities according to the multiple historical modeling accuracies, search and fuse the model grid accuracy of the adjacent model grids for the multiple search model grid combinations, and obtain multiple fusion model grid accuracy rates;
[0069] In an embodiment of the present application, after analyzing and obtaining the independent model grid accuracy of the searched multiple search grid model parameter combinations, the independent model grid accuracy can only reflect the modeling accuracy of a single clothing model grid, and it is also necessary to analyze the overall modeling accuracy of each clothing model grid.
[0070] According to the historical modeling accuracy extracted previously, set appropriate fusion rounds and fusion quantities for the search clothing model grids in the combination of multiple search model grids in different clothing areas. The fusion rounds include the number of times the neighboring model grids are searched and analyzed for modeling accuracy and the number of search fusions performed each time when the neighboring model grids are analyzed for modeling accuracy. Clothing areas with lower historical modeling accuracy require more rounds and a larger number of fusions to ensure that the overall modeling accuracy of the clothing model grids neighboring model grids in the clothing area can be more accurately fused.
[0071] Preferably, based on the graph neural network algorithm, multiple fusion rounds and multiple fusion numbers are used to search for neighboring model grids for multiple fusion rounds for multiple search model grid combinations. In each search, multiple fusion numbers of neighboring model grids are searched separately, and the model grid accuracies of the searched neighboring model grids are analyzed and then fused to finally obtain multiple fusion model grid accuracies.
[0072] According to the multiple historical modeling accuracies, multiple fusion rounds and multiple fusion quantities are set, and the model grid accuracy of the adjacent model grids is searched and fused for the multiple search model grid combinations to obtain multiple fusion model grid accuracy rates, including:
[0073] According to the multiple historical modeling accuracies, a preset number of fusion rounds and a preset number of fusions are corrected and calculated to obtain multiple numbers of fusion rounds and multiple numbers of fusions;
[0074] Taking the multiple search model grid combinations as the search origin, within a preset proximity range, according to the multiple fusion quantities, randomly search to obtain multiple first fusion model grid combination sets, and complete a first round of fusion search;
[0075] Continue to use the multiple first fusion model grid combination sets as search origins, and search for the second fusion model grid combination in a preset proximity range according to the multiple fusion quantities, until multiple fusion rounds of searching are completed;
[0076] The average model grid accuracy of multiple fused model grid combinations obtained by the fusion search of each model grid combination is used as the fused model grid accuracy of the model grid combination of the search origin, and based on the fusion search results of the multiple fusion rounds, multiple fused model grid accuracy rates are obtained by calculation and processing.
[0077] In an embodiment of the present application, in the fusion grid accuracy analysis, the preset number of fusion rounds and the preset number of fusions are first corrected and calculated based on multiple historical modeling accuracies. If the historical modeling accuracy of a certain clothing area is low, the number of fusion rounds and the number of fusions in that area will be increased to ensure that more modeling information of neighboring grids is included in the analysis, thereby improving the overall modeling accuracy.
[0078] Exemplarily, the mean of multiple historical modeling accuracies is calculated, and the ratio of each historical modeling accuracy to the mean is multiplied by the preset number of fusion rounds and the preset number of fusions and rounded to complete the correction calculation to obtain multiple fusion rounds and multiple fusion numbers. The preset number of fusion rounds and the preset number of fusions are the preset number of rounds and the number of fusions for searching the neighboring model grids using the graph neural network. Those skilled in the art can set them by themselves, for example, the preset number of fusion rounds is 3 and the preset number of fusions is 6.
[0079] Furthermore, with multiple search model grid combinations as search origins, neighboring model grid combinations are searched within a preset proximity range, and the preset proximity range is the range for each neighboring model grid combination search, for example, the 10 neighboring model grid combinations with the smallest distance from the search model grid combination of the search origin.
[0080] In the preset proximity range of multiple search model grid combinations, multiple fusion numbers of first fusion model grid combinations are randomly selected according to multiple fusion numbers, and random search is completed to obtain multiple first fusion model grid combination sets. Multiple first fusion model grid combination sets include multiple groups of first fusion model grid combinations obtained by searching within the preset proximity range with multiple search model grid combinations as search origins. The first round of fusion search is completed.
[0081] Continue to use each of the multiple first fusion model grid combination sets obtained in the first round of search as the search origin, and search for the second fusion model grid combination in the preset proximity range of each first fusion model grid combination set according to multiple fusion quantities, until multiple fusion rounds of search are completed, each round searches for multiple fusion numbers of fusion model grid combinations, and continue the above steps to complete multiple fusion rounds and multiple fusion numbers of fusion searches, respectively.
[0082] Among them, after the fusion search is completed, based on the clothing grid model parameters and model grid model parameters of the clothing model grids and model model grids in the multiple fusion model grid combinations obtained in the last round of search, and based on the independent model accurate identifier in the aforementioned content, independent and accurate analysis and identification of the input and model grids can be performed to obtain the model grid accuracy, and complete the modeling accuracy analysis of the model grid combination obtained in the last round of fusion search.
[0083] Furthermore, based on the model grid accuracy obtained by independent accurate analysis and identification of the model grids in the last round of fusion search, the average model grid accuracy calculated for multiple fused model grid combinations obtained by fusion search with each model grid combination as the search origin is used as the fused model grid accuracy of the model grid combination of the search origin. In this way, the average of the multiple model grid accuracy of the multiple model grid combinations obtained in the last round of search can be used as the fused model grid accuracy of the model grid combination of the search origin, and then the average of the fused model grid accuracy of the multiple fused model grid combinations obtained in the penultimate round of search is used as the fused model grid accuracy of the model grid combination of the search origin. In this way, the mean fusion of multiple rounds of model grid accuracy is completed, the fusion direction is opposite to the search direction, and the number of fusion rounds is the same as the number of fusion search rounds. Finally, the initial multiple search origins in the fusion search, that is, the multiple fused model grid accuracy of the multiple search model grid combinations, are obtained, and the search fusion of the model grid accuracy of the adjacent model grids is completed.
[0084] Through the above content, fusion search can not only perform modeling analysis on a single model grid combination, but also optimize the accuracy of the overall modeling by using the information of its neighboring grids, thereby improving the overall effect of the three-dimensional model of clothing design.
[0085] In the embodiment of the present application, the plurality of search model grid combinations are used as search origins, and within a preset proximity range, a plurality of first fusion model grid combination sets are randomly searched according to the plurality of fusion quantities, including:
[0086] Taking the plurality of search model grid combinations as search origins, within a preset proximity range, dividing and obtaining a plurality of proximity model grid combination sets;
[0087] According to the multiple fusion quantities, multiple first fusion model grid combination sets are randomly selected from the multiple neighboring model grid combination sets to obtain.
[0088] In the embodiment of the present application, multiple search model grid combinations are used as search origins, and within the preset proximity range of the multiple search model grid combinations, for example, the 10 model grid combinations with the smallest distances among the multiple search model grid combinations, multiple neighboring model grid combination sets are divided to obtain a range for random search. The distance of the model grid combination can be calculated based on the coordinate system of the clothing three-dimensional model and the model three-dimensional model, for example, the average of the distances between the clothing model grid and the model model grid in the two model grid combinations is calculated.
[0089] Furthermore, within a plurality of neighboring model grid combination sets, a plurality of fusion numbers of neighboring model grid combinations are randomly selected from the plurality of neighboring model grid combination sets according to a plurality of fusion numbers as a plurality of first fusion model grid combination sets to complete a first round of fusion search.
[0090] The modeling accuracy analysis module 14 is used to calculate and obtain the regional modeling accuracy parameters of the multiple clothing regions and the model modeling accuracy parameters of the clothing three-dimensional model according to the multiple independent model grid accuracy rates and the multiple fusion model grid accuracy rates;
[0091] In an embodiment of the present application, based on the accuracy rates of multiple independent model grids and multiple fusion model grids, according to the clothing area to which each search model grid combination belongs, regional modeling accuracy parameters of multiple clothing areas are calculated, and the model modeling accuracy parameters of the clothing three-dimensional model are calculated.
[0092] The accuracy of the independent model grid is obtained by analyzing the matching degree of a single clothing model grid and a model model grid, which reflects the modeling accuracy of each clothing model grid. The accuracy of the fusion model grid is obtained by analyzing the combination of neighboring model grids of each search model grid combination and the neighboring fusion search, which reflects the modeling accuracy of each clothing model grid in the neighboring area.
[0093] By fusing the accuracy rates of multiple independent model grids and multiple fused model grids, the regional modeling accuracy parameters of multiple clothing regions and the model modeling accuracy parameters of the clothing three-dimensional model are calculated, so that the overall modeling accuracy of each clothing model grid can be accurately evaluated, thereby improving the accuracy and systematicness of the evaluation of the construction accuracy of the clothing three-dimensional model, thereby improving the efficiency and accuracy of the construction of the clothing design three-dimensional model.
[0094] In the embodiment of the present application, according to the multiple independent model grid accuracy rates and the multiple fusion model grid accuracy rates, the regional modeling accuracy parameters of the multiple clothing regions and the model modeling accuracy parameters of the clothing three-dimensional model are calculated and obtained, including:
[0095] According to the multiple clothing regions, the multiple independent model grid accuracy rates and the multiple fusion model grid accuracy rates are divided, and multiple regional independent model accuracy rates and multiple regional fusion model accuracy rates are calculated;
[0096] Calculate the average of the multiple independent model grid accuracy rates and the multiple fusion model grid accuracy rates to obtain the independent model accuracy rate and the fusion model accuracy rate;
[0097] According to the independent model accuracy rates of the multiple regions and the fusion model accuracy rates of the multiple regions, the regional modeling accuracy parameters of the multiple clothing regions are calculated, and according to the independent model accuracy rates and the fusion model accuracy rates, the model modeling accuracy parameters of the three-dimensional clothing model are calculated.
[0098] In an embodiment of the present application, multiple independent model grid accuracies and multiple fusion model grid accuracies are divided according to the search model grid combinations in multiple clothing areas, and the averages are calculated based on several model grid accuracies and several fusion model grid accuracies corresponding to each clothing area to obtain multiple area independent model accuracies and multiple area fusion model accuracies.
[0099] Furthermore, the average of multiple independent model grid accuracies and multiple fusion model grid accuracies is calculated to obtain the independent model accuracy and fusion model accuracy of the entire clothing three-dimensional model.
[0100] Furthermore, according to the accuracy rates of multiple regional independent models and the accuracy rates of multiple regional fusion models, the average of the regional independent model accuracy and the regional fusion model accuracy of each clothing region is calculated to obtain the regional modeling accuracy parameters of the multiple clothing regions.
[0101] The visualization display module 15 is used to update the multiple historical modeling accuracy parameters according to the multiple regional modeling accuracy parameters, and to visually identify and display the clothing three-dimensional model according to the multiple regional modeling accuracy parameters and the model modeling accuracy parameters.
[0102] In an embodiment of the present application, multiple historical modeling accuracies are updated based on multiple regional modeling accuracy parameters. For example, multiple regional modeling accuracy parameters are used as multiple modeling accuracies of new multiple clothing regions. When the next clothing three-dimensional model is constructed, the modeling accuracy analysis and visualization of the next clothing three-dimensional model are performed.
[0103] Furthermore, according to the modeling accuracy parameters of multiple regions, multiple clothing regions in the clothing 3D model are visually marked and displayed, so that clothing design modeling technicians can understand the current modeling accuracy of multiple clothing regions, which is convenient for tuning the clothing 3D model and improving the modeling quality. Furthermore, using the model modeling accuracy parameters, the entire clothing 3D model is visually marked, so that clothing design modeling technicians can understand the modeling accuracy of the entire clothing 3D model, which is convenient for tuning the clothing 3D model and improving the modeling tuning efficiency and accuracy.
[0104] The embodiments of the present application achieve the following technical effects through the above specific implementation methods:
[0105] The embodiment of the present application can effectively improve the matching accuracy between the clothing three-dimensional model and the model three-dimensional model through the above technical scheme, and make the modeling process more systematic and accurate through innovative grid division and index marking technology. Specifically, by grid division and index marking of the initially constructed clothing three-dimensional model and the model three-dimensional model, a detailed association between multiple clothing areas and the model is established, so that the system can analyze the local modeling error more accurately. Then, combined with the historical modeling accuracy, the clothing and model grid combination is randomly searched, and the accuracy of the independent model grid is calculated by the matching degree of the three-dimensional model parameters. In this way, local modeling inaccuracy problems can be detected in multiple areas of clothing, rather than being limited to a single area. In addition, the model accuracy of the adjacent grids is subjected to multiple rounds and multiple number of fusion searches by adopting the fusion search method, which further improves the comprehensiveness and stability of the modeling accuracy. By calculating the independent model grid accuracy and the fused model grid accuracy, the scheme can accurately classify the modeling accuracy parameters of different clothing areas and generate the overall model modeling accuracy parameters. This process not only provides dual accuracy analysis of the whole and the part, but also enables the system to continuously optimize itself and improve the accuracy and efficiency of subsequent modeling by updating the historical modeling accuracy data in real time. Finally, through the visual identification and display based on the regional modeling accuracy parameters and the overall modeling accuracy parameters, designers can intuitively identify the problem areas in the modeling and make targeted adjustments. This precise modeling evaluation and feedback mechanism greatly improves the accuracy of three-dimensional modeling of clothing design, reduces the frequency of problems such as model penetration, and achieves the technical effect of improving the efficiency and accuracy of clothing design modeling.
[0106] See also Figure 2 , Figure 2 Schematic diagram of an electronic device provided by an embodiment of the present invention. Figure 2 As shown, an embodiment of the present invention provides an electronic device 100, including a memory 110, a processor 120, and a computer program 111 stored in the memory 110 and executable on the processor 120. When the processor 120 executes the computer program 111, the system in Example 1 is implemented.
[0107] See also Figure 3 , Figure 3 A schematic diagram of an embodiment of a computer-readable storage medium provided in an embodiment of the present invention. Figure 3 As shown, this embodiment provides a computer-readable storage medium 200 on which a computer program 211 is stored. When the computer program 211 is executed by a processor, the system in Embodiment 1 is implemented.
[0108] It should be noted that in the above embodiments, the description of each embodiment has its own emphasis, and for parts that are not described in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.
[0109] It will be appreciated by those skilled in the art that embodiments of the present invention may be provided as methods, systems, or computer program products. Therefore, the present invention may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Furthermore, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.
[0110] The present invention is described with reference to flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded computer, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.
[0111] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.
[0112] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.
[0113] Although the preferred embodiments of the present invention have been described, those skilled in the art may make other changes and modifications to these embodiments once they have learned the basic creative concept. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments and all changes and modifications that fall within the scope of the present invention.
[0114] Obviously, those skilled in the art can make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if these modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalents, the present invention is also intended to include these modifications and variations.
Claims
1. The clothing design 3D modeling visualization display system is characterized by: The system comprises: A three-dimensional model acquisition module is used to acquire a clothing three-dimensional model initially constructed by a user, and acquire a model three-dimensional model, perform model grid division and index marking on the clothing three-dimensional model and the model three-dimensional model, and obtain a plurality of clothing model grids and a plurality of model model grids that index each other; An independent grid accuracy analysis module is used to set multiple search probabilities according to multiple historical modeling accuracies of multiple clothing areas, randomly search a preset number of multiple search model grid combinations in the multiple clothing model grids and multiple model model grids, and analyze and obtain multiple independent model grid accuracy rates according to multiple search grid model parameter combinations of the multiple search model grid combinations, wherein the independent model grid accuracy rate is analyzed according to the matching degree of clothing grid model parameters and model grid model parameters in each search grid model parameter combination; A fusion grid accuracy analysis module is used to set multiple fusion rounds and multiple fusion quantities according to the multiple historical modeling accuracies, search and fuse the model grid accuracy of the adjacent model grids for the multiple search model grid combinations, and obtain multiple fusion model grid accuracy rates; A modeling accuracy analysis module, used for calculating and obtaining regional modeling accuracy parameters of the plurality of clothing regions and model modeling accuracy parameters of the clothing three-dimensional model according to the plurality of independent model grid accuracy rates and the plurality of fusion model grid accuracy rates; The visualization display module is used to update the multiple historical modeling accuracies according to the multiple regional modeling accuracy parameters, and to visually identify and display the clothing three-dimensional model according to the multiple regional modeling accuracy parameters and the model modeling accuracy parameters.
2. The clothing design three-dimensional modeling visualization display system according to claim 1 is characterized in that: The clothing 3D model and the model 3D model are divided into model grids and indexed to obtain a plurality of clothing model grids and a plurality of model model grids that index each other, including: Performing grid division on the clothing three-dimensional model to obtain a plurality of clothing model grids and clothing grid coordinate information of the plurality of model model grids; Dividing the model three-dimensional model into grids to obtain a plurality of model model grids and model grid coordinate information of the plurality of model model grids; According to the plurality of clothing grid coordinate information and the model grid coordinate information, an index relationship between the plurality of clothing model grids and the plurality of model model grids is constructed.
3. The clothing design three-dimensional modeling visualization display system according to claim 1 is characterized in that: According to multiple historical modeling accuracies of multiple clothing regions, multiple search probabilities are set, and a preset number of multiple search model grid combinations are randomly searched in the multiple clothing model grids and the multiple model model grids, including: Acquire multiple clothing areas in the clothing three-dimensional model, and extract multiple historical modeling accuracies of the multiple clothing areas according to clothing design three-dimensional modeling records in historical time; Calculating and setting a plurality of search probabilities according to the plurality of historical modeling accuracies; According to the multiple search probabilities and the preset numbers, a random search of model grids is performed within the multiple clothing model grids and the multiple model model grids to obtain multiple search clothing model grids and multiple search model model grids, which are combined as multiple search model grid combinations.
4. The clothing design three-dimensional modeling visualization display system according to claim 3 is characterized in that: According to the multiple historical modeling accuracies, multiple search probabilities are calculated and set, including: According to each historical modeling accuracy and the sum of the multiple historical modeling accuracies, a calculation is set to obtain multiple search probabilities, wherein the size of the search probability is negatively correlated with the size of the historical modeling accuracy.
5. The clothing design three-dimensional modeling visualization display system according to claim 1 is characterized in that: Analyzing and obtaining multiple independent model grid accuracy rates according to multiple search grid model parameter combinations of the multiple search model grid combinations includes: Based on the clothing three-dimensional model and the model three-dimensional model, clothing grid model parameters and model grid model parameters of multiple search clothing model grids and multiple search model model grids within the multiple independent model grid accuracy rates are extracted to obtain multiple search grid model parameter combinations; Collecting a set of sample search grid model parameter combinations, annotating the matching degree of clothing grid model parameters and model grid model parameters for each sample search grid model parameter combination, and obtaining a set of sample independent model grid accuracy rates; Using the sample search grid model parameter combination set and the sample independent model grid accuracy set, training an independent model accuracy identifier; The independent model accurate identifier is used to perform independent accurate analysis and identification of model grids on the multiple search grid model parameter combinations to obtain multiple independent model grid accuracy rates.
6. The clothing design three-dimensional modeling visualization display system according to claim 1, characterized in that: According to the multiple historical modeling accuracies, multiple fusion rounds and multiple fusion quantities are set, and the multiple search model grid combinations are searched and fused for the model grid accuracy of the adjacent model grids, including: According to the multiple historical modeling accuracies, a preset number of fusion rounds and a preset number of fusions are corrected and calculated to obtain multiple numbers of fusion rounds and multiple numbers of fusions; Taking the multiple search model grid combinations as the search origin, within a preset proximity range, according to the multiple fusion quantities, randomly search to obtain multiple first fusion model grid combination sets, and complete a first round of fusion search; Continue to use the multiple first fusion model grid combination sets as search origins, and search for the second fusion model grid combination in a preset proximity range according to the multiple fusion quantities, until multiple fusion rounds of searching are completed; The average model grid accuracy of multiple fused model grid combinations obtained by the fusion search of each model grid combination is used as the fused model grid accuracy of the model grid combination of the search origin, and based on the fusion search results of the multiple fusion rounds, multiple fused model grid accuracy rates are obtained by calculation and processing.
7. The clothing design three-dimensional modeling visualization display system according to claim 6, characterized in that: Taking the multiple search model grid combinations as search origins, within a preset proximity range, and according to the multiple fusion quantities, randomly searching to obtain multiple first fusion model grid combination sets, includes: Taking the plurality of search model grid combinations as search origins, within a preset proximity range, dividing and obtaining a plurality of proximity model grid combination sets; According to the multiple fusion quantities, multiple first fusion model grid combination sets are randomly selected from the multiple neighboring model grid combination sets to obtain.
8. The clothing design three-dimensional modeling visualization display system according to claim 1, characterized in that: According to the multiple independent model grid accuracy rates and the multiple fusion model grid accuracy rates, the regional modeling accuracy parameters of the multiple clothing regions and the model modeling accuracy parameters of the clothing three-dimensional model are calculated, including: According to the multiple clothing regions, the multiple independent model grid accuracy rates and the multiple fusion model grid accuracy rates are divided, and multiple regional independent model accuracy rates and multiple regional fusion model accuracy rates are calculated to obtain; Calculate the average of the multiple independent model grid accuracy rates and the multiple fusion model grid accuracy rates to obtain the independent model accuracy rate and the fusion model accuracy rate; According to the independent model accuracy rates of the multiple regions and the fusion model accuracy rates of the multiple regions, the regional modeling accuracy parameters of the multiple clothing regions are calculated, and according to the independent model accuracy rates and the fusion model accuracy rates, the model modeling accuracy parameters of the clothing three-dimensional model are calculated.
9. An electronic device, characterized in that: include: Memory for storing computer software programs; A processor is used to read and execute the computer software program, thereby realizing the clothing design three-dimensional modeling visualization display system as described in any one of claims 1-8.
10. A computer-readable storage medium, characterized in that: The storage medium stores a computer software program, which, when executed by a processor, implements the clothing design three-dimensional modeling visualization display system as described in any one of claims 1 to 8.
Citation Information
Patent Citations
Three-dimensional human body modeling optimization method and system
CN110910491A
Sketch-based intelligent clothing three-dimensional modeling method
CN114373059A
Cited By
Garment design three-dimensional model visual display method based on AI model
CN120257401A
A visualization display method for clothing design 3D models based on AI model
CN120257401B
