Personalized clothing recommendation and matching system and method based on artificial intelligence

Abstract

The invention relates to the technical field of clothing matching, and discloses a personalized clothing recommendation and matching system and method based on artificial intelligence, and the system comprises a clothing recommendation information processing module, a clothing recommendation information analysis module, and an optimal clothing matching module. Based on a user shape space model and a user favorite clothing product standard wearing space model, in combination with three-dimensional model assembly software, accurate matching of a user shape and a favorite clothing space model and intelligent detection of a matching interference state of the user shape and the favorite clothing space model are carried out. Clothing information is accurately recommended based on user preferences and clothing sizes; and according to the user favorite garment product information, the user shape and favorite garment matching space interference state detection information and the user favorite garment wearing satisfaction judgment information, intelligent screening and evaluation of optimal matching garments of the user are carried out, and the matching garments are visually and truly recommended based on user preferences, garment sizes and garment wearing visual effects.

Description

Technical Field

[0001] This invention relates to the technical field of clothing matching, specifically to a personalized clothing recommendation and matching system and method based on artificial intelligence. Background Technology

[0002] Clothing coordination mainly refers to the harmony in style and color to achieve an overall appropriate and elegant effect. A crucial issue is how to quickly and accurately provide style type results that align with individual personal clothing styles and dressing methods for various styles, making these results understandable and manageable for the individual, thus making everyday dressing more scientific and systematic. Existing clothing coordination systems cannot intelligently recommend clothing based on personalized user needs, nor can they achieve visualized intelligent clothing matching based on user body spatial characteristics, thus reducing the efficiency and quality of clothing coordination.

[0003] Chinese invention patent application CN119515509A, published on February 25, 2025, discloses a method and system for clothing matching recommendation based on big data. It obtains users' clothing browsing history data and input body shape data from e-commerce platforms, groups clothing types to obtain time-series grouped clothing type data. Based on this, it extracts color and pattern information for different clothing types, generates color-pattern data between clothing types, and analyzes the matching effect of user body shape with different clothing types based on this data, obtaining matching ratio segmentation effect data. However, this technical solution relies entirely on data analysis for intelligent clothing matching recommendations and cannot achieve customized intelligent recommendations for clothing styles and sizes based on the user's actual body shape and spatial characteristics. Summary of the Invention

[0004] (a) Technical problems to be solved To address the issues of existing clothing matching systems failing to intelligently recommend clothing based on personalized user needs or to provide visualized intelligent clothing matching based on user body spatial characteristics, thus reducing the efficiency and quality of clothing matching, this system aims to achieve the following: scientifically establish a user body spatial model; intelligently recommend clothing products that users prefer; accurately construct a spatial model of the coordination state between the user's body shape and preferred clothing; intelligently detect the spatial interference state between the user's body shape and preferred clothing; dynamically analyze user satisfaction information regarding preferred clothing; and visualize the optimal clothing matching based on the user's visual perception.

[0005] (II) Technical Solution This invention is achieved through the following technical solution: a personalized clothing recommendation and matching method based on artificial intelligence, the method comprising the following steps: S1. Collect user's preferred clothing features and user appearance space model information; S2. Based on the user appearance space model information, process the user's body space entity model information to generate user body space model information. S3. Based on the user's preferred clothing feature information and the latest clothing product information, perform product information search processing for the user's preferred clothing to generate user's preferred clothing product information. S4. Based on the user's preferred clothing product information and the latest standard wearing space model information of clothing products, perform spatial three-dimensional model information search and processing under the standard wearing state of the user's preferred clothing to generate standard wearing space model information of the user's preferred clothing products. S5. Based on the user body space model information and the user's preferred clothing product standard wearing space model information, perform spatial model matching processing between the user body and preferred clothing to generate spatial model information of matching state between the user body and preferred clothing, and perform matching interference state detection processing between the user body and preferred clothing space model to generate spatial interference state detection information of matching space between the user body and preferred clothing. S6. Based on the user's body shape and preferred clothing matching state space model information, process the display of the user's preferred clothing product wearing status, and collect user's opinion information on the wearing effect of preferred clothing. S7. Based on the user's opinion information on the effect of wearing the clothing and the user's satisfaction keywords for wearing the clothing, perform the satisfaction judgment process for the effect of wearing the clothing to generate user satisfaction judgment information for wearing the clothing. S8. Based on the user's preferred clothing product information, the user's body shape and preferred clothing spatial interference state detection information, and the user's preferred clothing wearing satisfaction judgment information, perform the user's optimal clothing matching evaluation process to generate the user's optimal clothing matching product information.

[0006] Preferably, the steps for collecting user-preferred clothing feature information and user appearance spatial model information are as follows: S11. Collect text information on the style, type, and size of the target user's preferred clothing online through a voice input device, and generate user preferred clothing feature information. The system uses a 3D full-body scanner to collect the three-dimensional spatial entity model information of the target user's human body in a clothed state online, and generates user appearance spatial model information; the user appearance spatial model information includes the user's clothing spatial model information and human body spatial model information.

[0007] Preferably, the steps for constructing the user's physical space model information based on the user's appearance space model information are as follows: S21. Using human body space model processing software, the user's appearance space model information is used to separate the user's outer clothing from the human body features, and user body space model information is constructed. The user body space model information represents the human body space model information when the user is wearing underwear. The human body space model processing software includes any one of 3DS Max, Maya and Rhino.

[0008] Preferably, the steps for generating user-preferred clothing product information by performing a product information search based on the user's preferred clothing features and the latest clothing product information are as follows: S31. Establish a collection of the latest apparel product information. , ;in Indicates the first The latest clothing product information corresponding to each type of clothing feature. This represents the maximum number of clothing feature types; the clothing feature types represent an index data type mainly composed of clothing style, pattern, type and size information, used to search for the latest clothing product parameters; the latest clothing product information represents the product feature information of the most popular clothing in the current season of this year; the latest clothing product information includes the manufacturer information, product name information, model information, material information, usage and maintenance information and hangtag information of the latest clothing. S32. Using the Aho-Corasick search algorithm, the user's preferred clothing feature information is compared with the current latest clothing product information set. The latest clothing product information described in the text Perform keyword matching based on clothing features to retrieve all the latest clothing product information corresponding to the user's preferred clothing features. And through data identification, a set of user-preferred clothing product information is generated. ,in and They represent the first and Information on user preferences for clothing products belong A subset of.

[0009] Preferably, the steps for generating standard wearing space model information for user-preferred clothing products by searching and processing the spatial 3D model information under the standard wearing state of the user-preferred clothing based on the user's preferred clothing product information and the latest standard wearing space model information of the clothing products are as follows: S41. Establish a current and latest clothing product standard wearing space model information set. ,in This indicates the latest clothing product information. The corresponding latest standard wearing space model information for clothing products, which represents the most standard three-dimensional space model information of the latest clothing products under the theoretical state of being worn by a user and fully unfolded. S42. The Sorensen-Morris-Pratt search algorithm is used to process the user's preferred clothing product information set. The user's preferred clothing product information mentioned in the text to Information set of the latest standard clothing product wearing space model The information described in the current latest clothing product standard wearing space model By performing keyword matching based on clothing features, information on clothing products that the user prefers can be retrieved. to The corresponding latest standard wearing space model information for clothing products And through data identification, a user preference clothing product standard wearing space model information set is generated. ,in and These respectively represent the user's preferred clothing product information. and Corresponding user preference clothing product standard wearing space model information, belong A subset of.

[0010] Preferably, based on the user's body shape spatial model information and the user's preferred clothing product standard wearing spatial model information, the spatial model matching processing of the user's body shape and preferred clothing is performed to generate the user's body shape and preferred clothing matching state spatial model information, and the matching interference state detection processing of the user's body shape and preferred clothing spatial model is performed. The operation steps for generating the user's body shape and preferred clothing matching space interference state detection information are as follows: S51. Using the assembly module in 3D model assembly software, the user-preferred clothing product standard wearing space model information set is assembled. The user preference clothing product standard wearing space model information described in the article to Based on coaxial and fitted alignment, the user's body shape is orderly aligned with the external user body space model corresponding to the user body space model information. Spatial model alignment processing of the user's body shape and preferred clothing in the wearing state is performed, and a spatial model information set of the user body shape and preferred clothing alignment state is generated. ,in This indicates the user's preferred clothing product standard wearing space model information. The spatial model information of user body and preferred clothing matching state is obtained by matching the spatial model of user body and preferred clothing in the wearing state with the user body spatial model information; This indicates the user's preferred clothing product standard wearing space model information. The spatial model information of the user's body shape and preferred clothing is obtained by matching the spatial model of the user's body shape and preferred clothing in the wearing state with the spatial model information of the user's body shape and preferred clothing; the spatial model information of the user's body shape and preferred clothing represents the three-dimensional model information of the user and clothing in the wearing state constructed by matching and combining the spatial model of the user's body shape and preferred clothing; the three-dimensional model assembly software includes any one of SOLIDWORKS and Zhongwang 3D. S52. Use the interference checking module in the 3D model assembly software to analyze the user's body shape and preferred clothing matching state space model information set. The user's body shape and preferred clothing matching state space model information described in the document to The system performs interference state detection processing on the spatial model of user body shape and preferred clothing, and generates a set of spatial interference state detection information on the user body shape and preferred clothing. ,in and These represent the state space model information of the user's body shape and preferred clothing. and Corresponding information on spatial interference between user body shape and preferred clothing; when or When there is interference between the spatial model of the user's body shape and their preferred clothing, it indicates that the currently selected clothing product information of the user's preferred clothing is... or If the clothing size is not suitable for the user's body shape, then output the spatial interference state detection information between the user's body shape and the preferred clothing. or For interference; when or When there is no interference between the spatial model of the user's body shape and their preferred clothing, it indicates that the currently selected clothing product information is the user's preferred clothing. or If the clothing size fits the user's body shape, then output the spatial interference state detection information of the user's body shape and preferred clothing. or No interference.

[0011] Preferably, the steps for processing the display of the user's preferred clothing product wearing status based on the user's body shape and preferred clothing matching state space model information, and collecting user feedback on the wearing effect of preferred clothing are as follows: S61. Combine the user's body shape with their preferred clothing in a state space model information set. The user's body shape and preferred clothing matching state space model information described in the document to The system imports the model into 3D model assembly software and displays the user's preferred clothing product wearing status on a screen. Simultaneously, a voice input device is used to collect user feedback on the clothing product. to The corresponding textual information on the user's preferred clothing wearing effect is collected, and a user preference clothing wearing effect information set is generated. ,in and These respectively represent the user's preferred clothing product information. and The corresponding user preference clothing wearing effect opinion information, which represents the user's opinion information on the wearing effect of clothing in the preferred wearing state.

[0012] Preferably, the steps for processing user satisfaction with clothing wearing effect based on the user's opinion information and user satisfaction keywords are as follows: S71. Establish a keyword set for user preferences, clothing wearing satisfaction. , ;in Indicates the first User preferences and clothing satisfaction keywords This represents the maximum number of keywords indicating user satisfaction with clothing; these keywords include "satisfied," "good-looking," "likes," "fits well," and "pretty." S72, Collect the user's preferred clothing wearing effect opinion information set User preferences and feedback on clothing wearing effects are mentioned in the text. to The clothing products are ordered according to user preferences and the set of keywords related to user satisfaction with wearing the clothing. The keywords mentioned above describe user preferences for clothing and wearing satisfaction. Perform keyword matching for clothing satisfaction, and generate a user preference clothing satisfaction assessment information set based on the keyword matching results. ,in and These represent the user's preferred clothing product information. and Information on user preferences and satisfaction with clothing wearing; when or and Successful matching of the keyword "clothing satisfaction" indicates that the user is satisfied with the described clothing product information. or If the user perceives the desired effect of wearing the clothing, then the system will output the user's preferred clothing satisfaction rating. or Satisfactory; when or and No matching was found for the keywords related to clothing satisfaction, indicating that the user is not satisfied with the described clothing product information. or If the user perceives the desired effect of wearing the clothing, then the system will output the user's preferred clothing satisfaction rating. or Dissatisfied.

[0013] Preferably, the steps for evaluating the user's optimal clothing combination and generating the user's optimal clothing combination product information based on the user's preferred clothing product information, the spatial interference state detection information of the user's body shape and preferred clothing, and the user's satisfaction with wearing preferred clothing are as follows: S81. Using the BERT language model algorithm, based on the spatial interference state detection information set of the user's body shape and preferred clothing. The information set includes the spatial interference state detection information of the user's body shape and preferred clothing, and the information set of the user's satisfaction with wearing preferred clothing. The user preference clothing satisfaction judgment information mentioned in the user preference clothing product information set The system filters out all user-preferred clothing products whose spatial interference state detection information matches the user's body shape and preferred clothing is non-interference, and whose user-preferred clothing wearing satisfaction judgment information is satisfactory. This data is then used to construct a set of optimal clothing product information for the user's outfit. ,in and They represent the first and This refers to the optimal clothing product information for a user, which represents the best clothing product recommendations based on the user's personal preferences, body size, and wearing effect. belong A subset of.

[0014] An AI-based personalized clothing recommendation and matching system is used to implement the AI-based personalized clothing recommendation and matching method. The system includes a clothing recommendation information processing module, a clothing recommendation information analysis module, and an optimal clothing matching module. The clothing recommendation information processing module includes a user preference clothing feature information collection unit, a user appearance space model information collection unit, and a user body space model information construction unit. The user's preferred clothing feature information acquisition unit acquires user's preferred clothing feature information through a voice input device; the user's appearance space model information acquisition unit acquires user's appearance space model information through a 3D full-body scanner; the user's body space model information construction unit constructs and processes the user's body space entity model information based on the user's appearance space model information and in conjunction with human body space model processing software, thereby generating user body space model information. The clothing recommendation information analysis module includes a current and latest clothing product information storage unit, a user-preferred clothing product information search unit, a current and latest clothing product standard wearing space model information storage unit, a user-preferred clothing product standard wearing space model information matching unit, a user body shape and preferred clothing matching state space model construction unit, and a user body shape and preferred clothing matching space interference state detection unit. The current latest clothing product information storage unit is used to store the current latest clothing product information; the user preference clothing product information search unit performs product information search processing based on the user preference clothing feature information and the current latest clothing product information to generate user preference clothing product information; the current latest clothing product standard wearing space model information storage unit is used to store the current latest clothing product standard wearing space model information; the user preference clothing product standard wearing space model information matching unit performs spatial three-dimensional model information search processing based on the user preference clothing product information and the current latest clothing product standard wearing space model information under the standard wearing state of the user preference clothing to generate... The system generates user-preferred clothing product standard wearing space model information; the user body shape and preferred clothing matching state space model construction unit, based on the user body shape space model information, the user-preferred clothing product standard wearing space model information, and combined with 3D model assembly software, performs spatial model matching processing of user body shape and preferred clothing to generate user body shape and preferred clothing matching state space model information; the user body shape and preferred clothing matching space interference state detection unit, based on the user body shape and preferred clothing matching state space model information, and combined with 3D model assembly software, performs matching interference state detection processing of user body shape and preferred clothing space model to generate user body shape and preferred clothing matching space interference state detection information. The optimal clothing matching module includes a user's preferred clothing product wearing status display unit, a user's preferred clothing wearing effect opinion collection unit, a user's preferred clothing wearing satisfaction keyword storage unit, a user's preferred clothing wearing satisfaction judgment unit, and a user's optimal clothing matching evaluation unit. The user-preferred clothing product wearing status display unit displays the user's preferred clothing product wearing status based on the user's body shape and preferred clothing matching status space model information, in conjunction with 3D model assembly software and a display screen; the user-preferred clothing wearing effect opinion collection unit collects user-preferred clothing wearing effect opinion information through a voice input device; the user-preferred clothing wearing satisfaction keyword storage unit stores user-preferred clothing wearing satisfaction keywords; the user-preferred clothing wearing satisfaction judgment unit performs user-preferred clothing wearing effect satisfaction judgment processing based on the user-preferred clothing wearing effect opinion information and user-preferred clothing wearing satisfaction keywords, generating user-preferred clothing wearing satisfaction judgment information; the user-optimal matching clothing evaluation unit performs user-optimal matching clothing evaluation processing based on the user-preferred clothing product information, the user's body shape and preferred clothing matching space interference state detection information, and the user-preferred clothing wearing satisfaction judgment information, generating user-optimal matching clothing product information.

[0015] (III) Beneficial Effects This invention provides a personalized clothing recommendation and matching system and method based on artificial intelligence. It has the following beneficial effects: 1. Accurately acquire users' preferred clothing features through voice input devices, providing real data support for personalized clothing products needed by users; scientifically construct user appearance space model information through 3D full-body scanner and dynamically establish user body space entity model by combining human body space model processing software, realize the customized and accurate establishment of user body 3D model, and improve the accuracy of personalized clothing matching.

[0016] Second, by combining user-preferred clothing feature information with intelligent search algorithms, we can accurately search for user-preferred clothing product information and user-preferred clothing product wearing space models, thus achieving scientific recommendation of clothing product information based on user preferences; based on user body space models, user-preferred clothing product standard wearing space models, and combined with 3D model assembly software, we can accurately match user body shape with preferred clothing space models, and intelligently detect the interference state of the matching between user body shape and preferred clothing space models, thus achieving accurate recommendation of clothing information based on user preferences and clothing size, improving the efficiency and quality of personalized clothing matching.

[0017] Third, by using a spatial model of the user's body shape and preferred clothing, combined with 3D model assembly software and a display screen, the system visualizes and outputs the user's preferred clothing product wearing status. Simultaneously, it uses voice input devices to dynamically collect user feedback on the wearing effect of preferred clothing, combining this with intelligent search algorithms and big data-based keywords indicating user satisfaction with the wearing effect. This achieves a personalized presentation of the user's clothing wearing effect based on a digital twin. Furthermore, based on user preferred clothing product information, spatial interference state detection information of the user's body shape and preferred clothing, and user satisfaction assessment information, the system intelligently filters and evaluates the optimal clothing combinations for the user. This enables visualized and realistic recommendations of clothing combinations based on user preferences, clothing size, and visual effects, improving the reliability and satisfaction of personalized clothing matching services. Attached Figure Description

[0018] Figure 1 A schematic diagram of the modules of the AI-based personalized clothing recommendation and matching system provided by the present invention; Figure 2 A flowchart illustrating the AI-based personalized clothing recommendation and matching method provided by this invention. Detailed Implementation

[0019] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0020] An example of the AI-based personalized clothing recommendation and matching system and method is as follows: Example 1: Please see Figure 1 - Figure 2 A personalized clothing recommendation and matching method based on artificial intelligence, which includes the following steps: S1. Collect user's preferred clothing features and user appearance space model information; S2. Based on the user's appearance space model information, process the user's body space entity model information to generate the user's body space model information. S3. Based on the user's preferred clothing features and the latest clothing product information, perform product information search and processing for the user's preferred clothing to generate the user's preferred clothing product information. S4. Based on user preference clothing product information and the latest standard wearing space model information of clothing products, perform spatial 3D model information search and processing of user preference clothing under standard wearing conditions to generate standard wearing space model information of user preference clothing products. S5. Based on the user's body shape spatial model information and the user's preferred clothing product standard wearing spatial model information, perform spatial model matching processing between the user's body shape and preferred clothing, generate spatial model information of the matching state between the user's body shape and preferred clothing, and perform matching interference state detection processing between the user's body shape and preferred clothing spatial model to generate spatial interference state detection information of the matching space between the user's body shape and preferred clothing. S6. Based on the user's body shape and preferred clothing matching state space model information, process the display of the user's preferred clothing product wearing status, and collect user opinions on the wearing effect of preferred clothing. S7. Based on user opinions on clothing wearing effect and user satisfaction keywords for clothing wearing effect, perform user satisfaction judgment on clothing wearing effect and generate user satisfaction judgment information for clothing wearing effect. S8. Based on user's preferred clothing product information, user's body shape and preferred clothing spatial interference state detection information, and user's preferred clothing wearing satisfaction judgment information, evaluate the user's optimal clothing combination and generate the user's optimal clothing combination product information.

[0021] For further details, please refer to Figure 1 - Figure 2 The steps for collecting user preference clothing feature information and user appearance space model information are as follows: S11. Collect text information on the style, type, and size of the target user's preferred clothing online through a voice input device, and generate user preferred clothing feature information. The system uses a 3D full-body scanner to collect the three-dimensional spatial entity model information of the target user's human body while it is dressed, and generates the user's appearance spatial model information, which includes the user's clothing spatial model information and human body spatial model information.

[0022] The steps for constructing and generating the user's physical form model information based on the user's appearance space model information are as follows: S21. Use human body space model processing software to separate the user's outer clothing from the human body features in the user's appearance space model information to construct the user's body space model information. The user's body space model information represents the human body space model information when the user is wearing underwear. The human body space model processing software includes any one of 3DS Max, Maya and Rhino.

[0023] The user's preferred clothing feature information collection unit uses a voice input device to accurately acquire the user's preferred clothing feature information, providing real data support for personalized clothing products needed by users; the user appearance space model information collection unit and the user body space model information construction unit work together to scientifically construct the user's appearance space model information through a 3D full-body scanner and dynamically establish the user's body space entity model by combining human body space model processing software, so as to realize the customized and accurate establishment of the user's body 3D model and improve the accuracy of personalized clothing matching.

[0024] For further details, please refer to Figure 1 - Figure 2 The steps for generating user-preferred clothing product information by searching and processing user-preferred clothing features and the latest clothing product information are as follows: S31. Establish a collection of the latest apparel product information. , ;in Indicates the first The latest clothing product information corresponding to each type of clothing feature. This indicates the maximum number of clothing feature types. Clothing feature types represent index data types mainly composed of clothing style, pattern, type, and size information, used to search for the latest clothing product parameters. The latest clothing product information represents the product feature information of the most popular clothing in the current season of this year. The latest clothing product information includes the manufacturer information, product name information, model information, material information, usage and maintenance information, and hangtag information of the latest clothing. S32. Use the Aho-Corasick search algorithm to combine user-preferred clothing feature information with the latest clothing product information set. Latest apparel product information in China Perform keyword matching based on clothing features to search for all the latest clothing products that match the user's preferred clothing features. And through data identification, a set of user-preferred clothing product information is generated. ,in and They represent the first and Information on user preferences for clothing products belong A subset of.

[0025] The steps for generating standard wearing space model information for user-preferred clothing products by searching and processing the 3D spatial model information under the standard wearing state of the user's preferred clothing, based on user-preferred clothing product information and the latest standard wearing space model information, are as follows: S41. Establish a current and latest clothing product standard wearing space model information set. ,in This indicates the latest clothing product information. The corresponding latest standard wearing space model information for clothing products represents the most standard three-dimensional space model information of the clothing under the theoretical state of being worn by the user and fully unfolded. S42. Use the Sorensen-Morris-Pratt search algorithm to collect user-preferred clothing product information. Information on clothing products preferred by Chinese users to Information set of the latest clothing product standards and wearing space models Current and latest clothing product standard wearing space model information By matching clothing feature keywords, information on clothing products that users prefer can be found. to Corresponding latest clothing product standard wearing space model information And through data identification, a user preference clothing product standard wearing space model information set is generated. ,in and These represent user preferences for clothing products. and Corresponding user preference clothing product standard wearing space model information, belong A subset of.

[0026] Based on the user's body shape spatial model information and the user's preferred clothing product standard wearing spatial model information, the spatial model matching of the user's body shape and preferred clothing is processed to generate the spatial model information of the matching state of the user's body shape and preferred clothing. The operation steps for generating the spatial interference state detection information of the matching space model of the user's body shape and preferred clothing are as follows: S51. Use the assembly module in the 3D model assembly software to set up the standard wearing space model information of the user's preferred clothing products. Chinese users' preferred clothing product standard wearing space model information to Based on coaxial and fitted orderly alignment, the system is applied to the exterior of the user's body space model corresponding to the user's body space model information. This process performs spatial model alignment processing of the user's body and preferred clothing in the wearing state, and generates a set of spatial model information on the alignment state of the user's body and preferred clothing. ,in This indicates user preferences for clothing products and standard wearing space model information. The spatial model information of user body shape and preferred clothing matching state is obtained by matching the spatial model of user body shape and preferred clothing in the wearing state with the spatial model information of user body shape; This indicates user preferences for clothing products and standard wearing space model information. The spatial model information of user body and preferred clothing is obtained by matching the spatial model of user body and preferred clothing in the wearing state with the spatial model information of user body and preferred clothing; the spatial model information of user body and preferred clothing represents the three-dimensional model information of user and clothing in the wearing state constructed by matching and combining the spatial model of user body and preferred clothing. The three-dimensional model assembly software includes any one of SOLIDWORKS and Zhongwang 3D. S52. Use the interference checking module in the 3D model assembly software to analyze the user's body shape and the state space model information set of the clothing preference matching. Information on the state space model of user body shape and preferred clothing to The system performs interference state detection processing on the spatial model of user body shape and preferred clothing, and generates a set of spatial interference state detection information on the user body shape and preferred clothing. ,in and These represent the user's body shape and preferred clothing matching state space model information, respectively. and Corresponding information on spatial interference between user body shape and preferred clothing; when or When there is interference between the spatial model of the user's body shape and their preferred clothing, it indicates that the user is currently selecting clothing product information that they prefer. or If the clothing size is not suitable for the user's body shape, output the spatial interference state detection information between the user's body shape and the preferred clothing. or For interference; when or When there is no interference between the spatial model of the user's body shape and their preferred clothing, it indicates that the currently selected user's preferred clothing product information is being viewed. or If the clothing size fits the user's body shape, then output the spatial interference state detection information between the user's body shape and the preferred clothing. or No interference.

[0027] By cooperating between the user-preferred clothing product information search unit and the user-preferred clothing product standard wearing space model information matching unit, and combining user-preferred clothing feature information with intelligent search algorithms, accurate searches for user-preferred clothing product information and user-preferred clothing product wearing space models are performed, enabling scientific recommendations of clothing product information based on user preferences. Furthermore, by cooperating between the user body shape and preferred clothing matching state space model construction unit and the user body shape and preferred clothing matching space interference state detection unit, and combining the user body shape space model, the user-preferred clothing product standard wearing space model, and 3D model assembly software, accurate matching of user body shape and preferred clothing space models is performed, along with intelligent detection of interference states in the matching of user body shape and preferred clothing space models. This enables accurate recommendations of clothing information based on user preferences and clothing size, improving the efficiency and quality of personalized clothing matching.

[0028] For further details, please refer to Figure 1 - Figure 2 The steps for processing the display of user-preferred clothing products' wearing status based on the user's body shape and preferred clothing state space model information, and collecting user feedback on the wearing effect of preferred clothing, are as follows: S61. Combine user body shape with preferred clothing in a state space model information set. Information on the state space model of user body shape and preferred clothing to The system imports the model into 3D model assembly software and displays the user's preferred clothing product wearing status on a screen. Simultaneously, a voice input device is used to collect user feedback on the clothing product's preferences. to The corresponding textual information on the user's preferred clothing wearing effect is collected, and a user preference clothing wearing effect information set is generated. ,in and These represent user preferences for clothing products. and The corresponding user feedback information on the effect of wearing the clothing indicates the user's opinion on the effect of wearing the clothing in a certain state.

[0029] The steps for processing user feedback on clothing wearing effects and related keywords to determine user satisfaction with clothing wearing effects, and generating user satisfaction assessment information, are as follows: S71. Establish a keyword set for user preferences, clothing wearing satisfaction. , ;in Indicates the first User preferences and clothing satisfaction keywords This represents the maximum number of keywords indicating user satisfaction with clothing; user satisfaction keywords include "satisfied," "good-looking," "likes," "fits well," and "pretty." S72. Collect user feedback on clothing preferences and wearing effects. User feedback on clothing wearing effect to Ordered by user preference for clothing product type numbering and by user preference for clothing wearing satisfaction keyword set Chinese users' preferences for clothing and their satisfaction with the clothing Perform keyword matching for clothing satisfaction, and generate a user preference clothing satisfaction assessment information set based on the keyword matching results. ,in and These represent user preferences for clothing products. and Information on user preferences and satisfaction with clothing wearing; when or and Successful matching of keywords related to clothing satisfaction indicates user satisfaction and preference for clothing product information. or The system outputs information on user satisfaction with the clothing based on how it looks on them. or Satisfactory; when or and No matching was found for keywords related to user satisfaction with clothing, indicating user dissatisfaction with user-preferred clothing product information. or The system outputs information on user satisfaction with the clothing based on how it looks on them. or Dissatisfied.

[0030] The steps for evaluating and generating optimal clothing combinations for users based on their preferred clothing product information, spatial interference detection information regarding the combination of user body shape and preferred clothing, and user satisfaction with their preferred clothing are as follows: S81. Employing the BERT language model algorithm based on the spatial interference state detection information set of user body shape and preferred clothing. Information on spatial interference state detection of user body shape and preferred clothing, and information on user satisfaction with clothing wearing. Information on user preferences for clothing and their satisfaction with the clothing products they wear is included in the user preference clothing product information set. The system filters out all user-preferred clothing products whose spatial interference status between user body shape and preferred clothing is non-interference and whose user satisfaction with the clothing is satisfactory. This data is then used to construct a set of optimal clothing product information for each user. ,in and They represent the first and This refers to the optimal clothing combination information for each user, which represents the best clothing combination recommendations based on the user's personal preferences, body size, and wearing effect. belong A subset of.

[0031] By coordinating the user's preferred clothing product wearing status display unit, the user's preferred clothing wearing effect feedback unit, and the user's preferred clothing wearing satisfaction judgment unit, the system visualizes and outputs the user's preferred clothing product wearing status based on the user's body shape and preferred clothing matching state spatial model information, combined with 3D model assembly software and a display screen. Simultaneously, it uses voice input devices to dynamically collect user's preferred clothing wearing effect feedback information, combined with intelligent search algorithms and user preferred clothing wearing satisfaction keywords based on big data storage, to intelligently judge the user's preferred clothing wearing effect satisfaction, achieving a personalized presentation of the user's clothing wearing effect based on a digital twin. The user's optimal clothing matching evaluation unit intelligently filters and evaluates the user's optimal clothing matching based on the user's preferred clothing product information, user body shape and preferred clothing matching spatial interference state detection information, and user preferred clothing wearing satisfaction judgment information, achieving a visualized and realistic recommendation of clothing matching based on user preferences, clothing size, and visual effect of clothing wearing, improving the reliability and satisfaction of personalized clothing matching services.

[0032] Example 2: Please see Figure 1 - Figure 2 A personalized clothing recommendation and matching system based on artificial intelligence is used to realize personalized clothing recommendation and matching methods based on artificial intelligence. The system includes a clothing recommendation information processing module, a clothing recommendation information analysis module, and an optimal clothing matching module. The clothing recommendation information processing module includes a user preference clothing feature information collection unit, a user appearance space model information collection unit, and a user body space model information construction unit; The user preference clothing feature information collection unit collects user preference clothing feature information through a voice input device; the user appearance space model information collection unit collects user appearance space model information through a 3D full-body scanner; the user body space model information construction unit constructs and processes the user's body space entity model information based on the user appearance space model information and combined with human body space model processing software, generating user body space model information. The clothing recommendation information analysis module includes a current and latest clothing product information storage unit, a user preference clothing product information search unit, a current and latest clothing product standard wearing space model information storage unit, a user preference clothing product standard wearing space model information matching unit, a user body shape and preferred clothing matching state space model construction unit, and a user body shape and preferred clothing matching space interference state detection unit. The system includes: a current and latest clothing product information storage unit for storing the latest clothing product information; a user-preferred clothing product information search unit for searching and processing user-preferred clothing product information based on user-preferred clothing feature information and the current and latest clothing product information; a current and latest clothing product standard wearing space model information storage unit for storing current and latest clothing product standard wearing space model information; and a user-preferred clothing product standard wearing space model information matching unit for searching and processing user-preferred clothing product information and the current and latest clothing product standard wearing space model information to generate a 3D spatial model of the user-preferred clothing under standard wearing conditions. The system includes: a user-preferred clothing product standard wearing space model information unit; a user body shape and preferred clothing matching state space model construction unit, which, based on the user body shape space model information, the user-preferred clothing product standard wearing space model information, and combined with 3D model assembly software, performs spatial model matching processing between the user body shape and preferred clothing to generate user body shape and preferred clothing matching state space model information; and a user body shape and preferred clothing matching space interference state detection unit, which, based on the user body shape and preferred clothing matching state space model information, performs matching interference state detection processing between the user body shape and preferred clothing space model and combined with 3D model assembly software to generate user body shape and preferred clothing matching space interference state detection information. The optimal clothing matching module includes a user's preferred clothing product wearing status display unit, a user's preferred clothing wearing effect opinion collection unit, a user's preferred clothing wearing satisfaction keyword storage unit, a user's preferred clothing wearing satisfaction judgment unit, and a user's optimal clothing matching evaluation unit. The user-preferred clothing product wearing status display unit displays the user's preferred clothing product wearing status based on the user's body shape and preferred clothing matching status spatial model information, combined with 3D model assembly software and a display screen. The user-preferred clothing wearing effect feedback collection unit collects user feedback on the wearing effect of preferred clothing through a voice input device. The user-preferred clothing wearing satisfaction keyword storage unit stores user-preferred clothing wearing satisfaction keywords. The user-preferred clothing wearing satisfaction judgment unit judges user satisfaction with the wearing effect of preferred clothing based on the user's preferred clothing wearing effect feedback information and user-preferred clothing wearing satisfaction keywords, generating user-preferred clothing wearing satisfaction judgment information. The user's optimal clothing matching evaluation unit evaluates the user's optimal clothing matching based on user-preferred clothing product information, user body shape and preferred clothing matching spatial interference state detection information, and user-preferred clothing wearing satisfaction judgment information, generating user-preferred clothing matching product information.

[0033] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. A personalized clothing recommendation and matching method based on artificial intelligence, characterized in that, The method includes the following steps: S1. Collect user's preferred clothing features and user appearance space model information; S2. Perform the construction and processing of the user's physical space entity model information to generate the user's physical space model information; S3. Perform product information search and processing for user-preferred clothing, and generate user-preferred clothing product information. S4. Perform spatial 3D model information search and processing for user-preferred clothing under standard wearing conditions, and generate user-preferred clothing product standard wearing spatial model information. S5. Perform spatial model matching processing of user body shape and preferred clothing, generate spatial model information of user body shape and preferred clothing matching state, and perform matching interference state detection processing of user body shape and preferred clothing spatial model to generate spatial interference state detection information of user body shape and preferred clothing matching space. S6. Display the wearing status of users' preferred clothing products and collect user feedback on the wearing effect of their preferred clothing. S7. Perform user satisfaction assessment on the effect of wearing preferred clothing and generate user satisfaction assessment information on wearing preferred clothing. S8. Based on the user's preferred clothing product information, the user's body shape and preferred clothing spatial interference state detection information, and the user's preferred clothing wearing satisfaction judgment information, perform the user's optimal clothing matching evaluation process to generate the user's optimal clothing matching product information.

2. The personalized clothing recommendation and matching method based on artificial intelligence according to claim 1, characterized in that: S1 includes the following steps: S11. Collect text information on the style, type, and size of the target user's preferred clothing online through a voice input device, and generate user preferred clothing feature information. The system uses a 3D full-body scanner to collect the three-dimensional spatial model information of the target user's human body while it is dressed, and generates the user's appearance spatial model information.

3. The personalized clothing recommendation and matching method based on artificial intelligence according to claim 2, characterized in that: S2 includes the following steps: S21. Using human body space model processing software, the user's appearance space model information is used to separate the user's outer clothing from the human body features, and user body space model information is constructed. The user body space model information represents the human body space model information when the user is wearing underwear.

4. The personalized clothing recommendation and matching method based on artificial intelligence according to claim 3, characterized in that: S3 includes the following steps: S31. Establish a collection of the latest apparel product information. The include ;in Indicates the first The latest clothing product information corresponding to each clothing feature type; the clothing feature type represents an index data type mainly composed of clothing style, pattern, type and size information, which is used to search for the parameters of the latest clothing products. S32. Using the Aho-Corasick search algorithm, the user's preferred clothing feature information is compared with the... The above Perform clothing feature keyword matching to search for all the clothing feature information corresponding to the user's preferred clothing features. And through data identification, a set of user-preferred clothing product information is generated. The include and ;in and They represent the first and Information on user preferences for clothing products belong A subset of.

5. The personalized clothing recommendation and matching method based on artificial intelligence according to claim 4, characterized in that: S4 includes the following steps: S41. Establish a current and latest clothing product standard wearing space model information set. The include ;in Indicates the Corresponding to the latest and most current standard clothing product wearing space model information; S42. The Sorensen-Morris-Pratt search algorithm is used to search the... The above to With the The above Perform keyword matching based on clothing features to search for the aforementioned to The corresponding And through data identification, a user preference clothing product standard wearing space model information set is generated. The include and ,in and Each represents the and Corresponding user preference clothing product standard wearing space model information, belong A subset of.

6. The personalized clothing recommendation and matching method based on artificial intelligence according to claim 5, characterized in that: S5 includes the following steps: S51. Using the assembly module in the 3D model assembly software to assemble the... The above to Based on coaxial and fitted alignment, the user's body shape is orderly aligned with the external user body space model corresponding to the user body space model information. Spatial model alignment processing of the user's body shape and preferred clothing in the wearing state is performed, and a spatial model information set of the user body shape and preferred clothing alignment state is generated. The include and ;in Indicates the The spatial model information of user body and preferred clothing matching state is obtained by matching the spatial model of user body and preferred clothing in the wearing state with the user body spatial model information; Indicates the The spatial model information of user body and preferred clothing matching state is obtained by matching the spatial model of user body and preferred clothing in the wearing state with the user body spatial model information; S52. The interference check module in the three-dimensional model assembly software is used to check the... The above to The system performs interference state detection processing on the spatial model of user body shape and preferred clothing, and generates a set of spatial interference state detection information on the user body shape and preferred clothing. The include and ;in and Each represents the and Corresponding information on spatial interference between user body shape and preferred clothing; when or When there is interference between the spatial model of the user's body shape and their preferred clothing, the output will be as described. or For interference; when or When there is no interference between the spatial model of the user's body shape and preferred clothing, then the output is as described. or No interference.

7. The personalized clothing recommendation and matching method based on artificial intelligence according to claim 6, characterized in that: S6 includes the following steps: S61, the above The above to The system imports the model into a 3D model assembly software and displays the user's preferred clothing product wearing status on a screen. Simultaneously, a voice input device is used to collect user feedback on the displayed clothing product wearing status. to The corresponding textual information on the user's preferred clothing wearing effect is collected, and a user preference clothing wearing effect information set is generated. The include and ;in and Each represents the and Corresponding user preferences and opinions on the effect of wearing the clothing.

8. The personalized clothing recommendation and matching method based on artificial intelligence according to claim 7, characterized in that: S7 includes the following steps: S71. Establish a keyword set for user preferences, clothing wearing satisfaction. The include ;in Indicates the first Key words indicating user preferences and satisfaction with clothing; S72, the above The above to The clothing product types are numbered sequentially according to user preferences. The above Perform keyword matching for clothing satisfaction, and generate a user preference clothing satisfaction assessment information set based on the keyword matching results. The include and ;in and These represent the user's preferred clothing product information. and Information on user preferences and satisfaction with clothing wearing; when or and Successful matching of the keyword "clothing satisfaction" indicates that the user is satisfied with the described clothing product information. or If the user perceives the desired effect of wearing the clothing, then the system will output the user's preferred clothing satisfaction rating. or Satisfactory; when or and No matching was found for the keywords related to clothing satisfaction, indicating that the user is not satisfied with the described clothing product information. or If the user perceives the desired effect of wearing the clothing, then the system will output the user's preferred clothing satisfaction rating. or Dissatisfied.

9. The personalized clothing recommendation and matching method based on artificial intelligence according to claim 8, characterized in that: S8 includes the following steps: S81, Using the BERT language model algorithm based on the above The user's body shape and preferred clothing spatial interference state detection information and the aforementioned The user preference clothing wearing satisfaction judgment information mentioned in the document is in The system filters out all user-preferred clothing products whose spatial interference state detection information matches the user's body shape and preferred clothing is non-interference, and whose user-preferred clothing wearing satisfaction judgment information is satisfactory. This data is then used to construct a set of optimal clothing product information for the user's outfit. The include and ,in and They represent the first and Information on the best clothing combinations for each user. belong A subset of.

10. An AI-based personalized clothing recommendation and matching system, used to implement the AI-based personalized clothing recommendation and matching method as described in any one of claims 1-9, characterized in that: The system includes a clothing recommendation information processing module, a clothing recommendation information analysis module, and an optimal clothing matching module.

Images