Clothing matching result generation method and device, storage medium and electronic device

Abstract

The invention discloses a clothes matching result generation method and device, a storage medium and an electronic device, and relates to the technical field of artificial intelligence, and the clothes matching result generation method comprises the steps: obtaining clothes data from a database, the clothes data comprising a clothes image and clothes attributes corresponding to clothes in the clothes image; a text matching scheme is generated according to a demand instruction of a target object and the clothes data, and the demand instruction of the target object represents a clothes matching demand, described by a natural language, of the target object; and generating a visual clothes matching result based on the text matching scheme and the clothes data. By adopting the technical scheme, the problem of how to realize personalized clothes matching recommendation is solved.

Description

Technical Field

[0001] This application relates to the field of artificial intelligence technology, and more specifically, to a method, apparatus, storage medium, and electronic device for generating clothing matching results. Background Technology

[0002] Existing digital wardrobe management applications rely heavily on pre-set, fixed rule bases when recommending clothing combinations, such as simple color matching tables or style matching rules. This approach is inherently static and cannot understand or reason about the complex and ever-changing real-time needs input by users through natural language. Lacking intelligent features, the generated combinations often remain superficial, lacking professionalism and practicality. Therefore, a key challenge in related technologies is how to achieve personalized clothing combination recommendations.

[0003] No effective solution has yet been proposed for the problem of how to achieve personalized clothing matching recommendations in related technologies. Summary of the Invention

[0004] This application provides a method, apparatus, storage medium, and electronic device for generating clothing matching results, so as to at least solve the problem of how to achieve personalized clothing matching recommendations in related technologies.

[0005] According to one embodiment of this application, a method for generating clothing matching results is provided, comprising: obtaining clothing data from a database, wherein the clothing data includes clothing images and clothing attributes corresponding to the clothing in the clothing images; generating a text matching scheme based on the demand instructions of a target object and the clothing data, wherein the demand instructions of the target object represent the clothing matching needs of the target object described in natural language; and generating a visualized clothing matching result based on the text matching scheme and the clothing data.

[0006] In an exemplary embodiment, before obtaining clothing data from the database, the method further includes: acquiring the clothing image, wherein the clothing image is acquired by the image acquisition device or uploaded by the target object; identifying the clothing image using an image recognition tool to obtain clothing attributes corresponding to the clothing in the clothing image; associating the clothing image and the clothing attributes and saving them to the database to obtain the clothing data.

[0007] In an exemplary embodiment, generating a text-based outfit matching scheme based on the target object's demand instruction and the clothing data includes: inputting preset prompt words into a large language model, wherein the preset prompt words are used to instruct the large language model to recommend clothing for the target object based on knowledge in the clothing knowledge base and the clothing data after receiving the demand instruction; inputting the target object's demand instruction into the large language model to obtain a target clothing combination selected by the large language model from the clothing data; and generating the text-based outfit matching scheme based on the target clothing combination.

[0008] In one exemplary embodiment, generating a visualized clothing combination result based on the text matching scheme and the clothing data image includes: upon receiving a preview instruction to visualize the text matching scheme, inputting the text matching scheme into a clothing visualization model; and obtaining the clothing combination result output by the clothing visualization model, wherein the clothing combination result is used to demonstrate the effect of the target object wearing the target clothing combination.

[0009] In an exemplary embodiment, before inputting the text matching scheme into the clothing visualization model, the method further includes: acquiring a first image of the target object wearing a first clothing combination, a second image of the target object wearing a second clothing combination, and a third image of the target object wearing a third clothing combination, wherein the clothing in the third clothing combination comes from clothing in the first clothing combination or clothing in the second clothing combination; training an image generation model using the first image and the second image as input data and the third image as output data to obtain the clothing visualization model.

[0010] In an exemplary embodiment, before inputting the text matching scheme into the clothing visualization model, the method further includes: acquiring a fourth image of the target object wearing a fourth clothing combination; using all clothing images corresponding to the clothing in the fourth clothing combination as input data and the fourth image as output data to train an image generation model to obtain the clothing visualization model.

[0011] In one exemplary embodiment, the method further includes: upon receiving a storage request for storing clothing data in the database, initiating a stabilization countdown; if a new storage request is received before the stabilization countdown ends, resetting the stabilization countdown; and if the stabilization countdown ends, batch executing the received storage requests.

[0012] According to another aspect of the embodiments of this application, an apparatus for generating clothing matching results is also provided, comprising: a data acquisition module for acquiring clothing data from a database, wherein the clothing data includes clothing images and clothing attributes corresponding to the clothing in the clothing images; a first generation module for generating a text matching scheme based on the demand instructions of a target object and the clothing data, wherein the demand instructions of the target object represent the clothing matching needs of the target object described in natural language; and a second generation module for generating a visualized clothing matching result based on the text matching scheme and the clothing data.

[0013] According to another aspect of the embodiments of this application, a computer-readable storage medium is also provided, wherein a computer program is stored in the computer program, and the computer program is configured to execute the above-described method for generating clothing matching results when running.

[0014] According to another aspect of the embodiments of this application, an electronic device is also provided, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the above-described method for generating clothing matching results through the computer program.

[0015] According to another aspect of the embodiments of this application, a computer program product is also provided, including a computer program that, when executed by a processor, implements the steps in any of the above method embodiments.

[0016] In this embodiment, by structurally storing users' clothing images and attributes in data, and then generating clothing matching schemes for users based on their clothing matching needs described in natural language and the clothing data, the system further displays the visualization effects of these schemes to the users. This solves the problem of how to achieve personalized clothing matching recommendations, thereby achieving the desired effect. Attached Figure Description

[0017] The accompanying drawings, which are incorporated in and form part of this specification, illustrate embodiments consistent with this application and, together with the description, serve to explain the principles of this application.

[0018] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, for those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0019] Figure 1 This is a schematic diagram of the hardware environment for a method of generating clothing matching results according to an embodiment of this application;

[0020] Figure 2 This is a flowchart of a method for generating clothing matching results according to an embodiment of this application;

[0021] Figure 3 This is a system architecture diagram of an intelligent electronic wardrobe management and matching recommendation system according to an embodiment of this application;

[0022] Figure 4 This is a flowchart illustrating an intelligent clothing recognition method according to an embodiment of this application;

[0023] Figure 5 This is a flowchart of an intelligent clothing matching recommendation according to an embodiment of this application;

[0024] Figure 6 This is a flowchart of a shake-resistant storage mechanism according to an embodiment of this application;

[0025] Figure 7 This is a schematic diagram illustrating cross-platform data synchronization according to an embodiment of this application;

[0026] Figure 8 This is a structural block diagram of an apparatus for generating clothing matching results according to an embodiment of this application. Detailed Implementation

[0027] To enable those skilled in the art to better understand the present application, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present application, and not all embodiments. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative effort should fall within the scope of protection of the present application.

[0028] It should be noted that the terms "first," "second," etc., in the specification, claims, and accompanying drawings of this application are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of this application described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion; for example, a process, method, apparatus, product, or device that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or devices.

[0029] According to one aspect of the embodiments of this application, a method for generating clothing matching results is provided. This method for generating clothing matching results is widely used in application scenarios such as software development. Optionally, in this embodiment, the above-mentioned method for generating clothing matching results can be applied to, for example... Figure 1 The hardware environment shown consists of terminal device 102 and server 104. For example... Figure 1 As shown, server 104 is connected to terminal device 102 via a network and can be used to provide services (such as application services) to the terminal or clients installed on the terminal. A database can be set up on the server or independently of the server to provide data storage services for server 104. Cloud computing and / or edge computing services can be configured on the server or independently of the server to provide data processing services for server 104.

[0030] The aforementioned networks may include, but are not limited to, at least one of the following: wired network, wireless network. The aforementioned wired network may include, but is not limited to, at least one of the following: wide area network (WAN), metropolitan area network (MAN), local area network (LAN). The aforementioned wireless network may include, but is not limited to, at least one of the following: Wi-Fi (Wireless Fidelity), Bluetooth. Terminal device 102 may not be limited to PCs, mobile phones, tablets, smart air conditioners, smart range hoods, smart refrigerators, smart ovens, smart stoves, smart washing machines, smart water heaters, smart washing equipment, smart dishwashers, smart projectors, smart TVs, smart clothes racks, smart curtains, smart audio-visual equipment, smart sockets, smart speakers, smart speakers, smart fresh air systems, smart kitchen and bathroom equipment, smart bathroom equipment, smart robot vacuum cleaners, smart window cleaning robots, smart mopping robots, smart air purifiers, smart steam ovens, smart microwave ovens, smart water heaters, smart air purifiers, smart water dispensers, smart door locks, etc.

[0031] This embodiment provides a method for generating clothing matching results. Figure 2 This is a flowchart of a method for generating clothing matching results according to an embodiment of this application. The process includes the following steps:

[0032] Step S202: Obtain clothing data from the database, wherein the digital clothing data includes clothing images and clothing attributes corresponding to the clothing in the clothing images;

[0033] Optional, for example, one of the clothing data records is: {Clothing ID:123, Name:"Gray Suit Pants", Type:"Pants", Color:"Gray", Material:"Wool", Style:"Business", Occasion:"Commuting", Image Path:" / images / 123.jpg"}.

[0034] Step S204: Generate a text matching scheme based on the target object's demand instructions and the digital clothing data, wherein the target object's demand instructions represent the target object's clothing matching needs described in natural language;

[0035] Optionally, a user could input their request in natural language on the interface: "Help me put together an outfit for work that looks professional yet approachable, while still showing off my gray suit pants." Upon receiving this instruction, it is fed into a recommendation system based on a large language model, along with all clothing data retrieved from the database (including tops, pants, shoes, etc.). After the system runs, it generates the following textual outfit suggestion: "Top: Light blue cotton shirt; Pants: Gray suit pants; Shoes: Brown loafers. Reason: A light blue shirt and gray suit pants are a classic business casual combination; the colors are soft and approachable, the cotton material is comfortable, and the brown loafers add a touch of style, resulting in an overall professional yet energetic look."

[0036] Step S206: Generate a visualized clothing matching result based on the text matching scheme and the clothing data.

[0037] Optionally, the system can parse the above text to identify the recommended clothing IDs (e.g., shirt ID 456, shoe ID 789). Then, the system retrieves clothing images with IDs 123, 456, and 789 from the database, and combines this with a clothing visualization model to output an image or 3D model of the user wearing the recommended clothing.

[0038] Through the above steps, by structurally storing users' clothing images and attributes in the data, and then generating clothing matching schemes for users based on their clothing matching needs described in natural language and the clothing data, the system further presents the visual effects of these schemes to the users. This solves the problem of how to achieve personalized clothing matching recommendations, thereby realizing the desired personalized clothing matching effect.

[0039] In an exemplary embodiment, before obtaining clothing data from the database, the method further includes: acquiring the clothing image, wherein the clothing image is acquired by the image acquisition device or uploaded by the target object; identifying the clothing image using an image recognition tool to obtain clothing attributes corresponding to the clothing in the clothing image; associating the clothing image and the clothing attributes and saving them to the database to obtain the clothing data.

[0040] Optionally, in the above embodiment, for example, a user opens the smart wardrobe app, clicks the "Add Clothes" button, takes a picture of a shirt laid flat on the bed using their phone's camera, and uploads it. After receiving the image, the system first preprocesses it (adjusting the image size to 500x500 pixels and converting the format to JPG). Then, it calls the API of the integrated image recognition tool and sends the processed image. The image recognition tool returns the following result: {Type: "Shirt", Main Color: "Blue", Pattern: "Stripes", Material: "Cotton"}. Through further in-depth analysis of the returned basic attributes using AI, advanced attributes are inferred: style is "business casual", and suitable occasions are "commuting, daily wear". Next, the system automatically generates tags for this garment, such as "blue", "stripes", "cotton", and "business casual". Finally, the system creates a structured data object containing all identified attributes, inferred attributes, generated tags, and the URL address of the image stored in the cloud, and then saves this complete record to the clothing table in the database. At this point, the user's shirt becomes "clothing data" that can be used for subsequent outfit recommendations.

[0041] This embodiment automatically extracts clothing attributes and generates structured data through image recognition, realizing the automation and batch processing of clothing information digitization. It completely solves the core pain points of traditional methods, which rely on manual user input, such as low efficiency, strong subjectivity, and easy errors. It reduces the input work that may have taken several minutes to complete for one item to seconds, and the attribute information is more objective, comprehensive, and standardized, laying a reliable data foundation for subsequent high-quality intelligent recommendations.

[0042] In an exemplary embodiment, generating a text-based outfit matching scheme based on the target object's demand instruction and the clothing data includes: inputting preset prompt words into a large language model, wherein the preset prompt words are used to instruct the large language model to recommend clothing for the target object based on knowledge in the clothing knowledge base and the clothing data after receiving the demand instruction; inputting the target object's demand instruction into the large language model to obtain a target clothing combination selected by the large language model from the clothing data; and generating the text-based outfit matching scheme based on the target clothing combination.

[0043] Optionally, in the above embodiments, for example, if a user inputs a request: "What should I wear for a weekend picnic in the park? It needs to be comfortable and windproof." The system first parses the instruction and extracts the key intents "picnic," "comfort," and "windproof." Then, the system retrieves knowledge fragments related to "outdoor activities," "casual wear," and "windproof jacket" from a pre-built clothing matching knowledge base (which contains over 10,000 pieces of professional knowledge stored in Elasticsearch, including color matching principles, style matching rules, and occasion suitability). For example, it retrieves a piece of knowledge: "For spring picnics, it is recommended to wear layered clothing; a breathable inner cotton T-shirt and an outer windbreaker or jacket for windproof protection." The prompt first defines the role of the large language model: "You are a professional outdoor clothing stylist." Then, it combines the user's request instruction, the windproof clothing knowledge retrieved from the knowledge base, and a structured description of all the clothing in the user's wardrobe (e.g., "The user owns the following clothing: 1. Clothing ID: 100, hooded sweatshirt, gray... 2. Clothing ID: 101, denim jacket, blue..."). Finally, it specifies the output format: "Please generate 3 outfit combinations, each including a top, bottoms, shoes, and a brief reason." The system prompts are then input into a large language model to generate the combinations: the system prompts are sent to the large language model. The model reasones within the context of understanding personal wardrobe constraints and expertise, ultimately returning a text-based outfit combination containing "grey hoodie + denim jacket + sweatpants + sneakers."

[0044] This embodiment introduces cue word engineering and RAG (Retrieval Enhancement Generation) technology to precisely guide the general capabilities of the large language model to the vertical field of clothing matching. This provides a dual guarantee for the professionalism and personalization of the recommendation results: on the one hand, by injecting professional knowledge, the recommended solutions are no longer simple item compilations, but reasonable suggestions supported by professional theory; on the other hand, by closely integrating with the user's personal wardrobe data, it ensures that all recommendations are feasible solutions that the user actually owns and can immediately adopt, thereby significantly improving the practicality, accuracy, and user satisfaction of the recommendations.

[0045] In one exemplary embodiment, generating a visualized clothing combination result based on the text matching scheme and the clothing data image includes: upon receiving a preview instruction to visualize the text matching scheme, inputting the text matching scheme into a clothing visualization model; and obtaining the clothing combination result output by the clothing visualization model, wherein the clothing combination result is used to demonstrate the effect of the target object wearing the target clothing combination.

[0046] Optionally, in the above embodiments, after receiving the text scheme returned by the model, a dedicated parser is used for parsing. The clothing names mentioned in the scheme are accurately extracted, matched with records in the database to find the corresponding clothing IDs, and then high-resolution images corresponding to these clothing IDs are retrieved from the database or cloud storage. Finally, a clothing visualization model is used to generate an image of the user wearing these clothes.

[0047] This embodiment transforms AI-generated text suggestions into concrete clothing images for display, thus converting abstract text into tangible visual representations. This significantly improves the understandability and intuitiveness of the recommendations. Users no longer need to mentally visualize the effects of the textual descriptions; they can quickly judge whether a suggestion is suitable through intuitive image combinations, greatly shortening their decision-making process.

[0048] In an exemplary embodiment, before inputting the text matching scheme into the clothing visualization model, the method further includes: acquiring a first image of the target object wearing a first clothing combination, a second image of the target object wearing a second clothing combination, and a third image of the target object wearing a third clothing combination, wherein the clothing in the third clothing combination comes from clothing in the first clothing combination or clothing in the second clothing combination; training an image generation model using the first image and the second image as input data and the third image as output data to obtain the clothing visualization model.

[0049] Optionally, in the above embodiments, a model training method is provided to implement advanced virtual try-on functionality (i.e., generating an image of the user wearing the clothes). The specific process is as follows:

[0050] Image acquisition for training: Images of the user's everyday clothing are acquired through physical smart wardrobes or other IoT devices such as televisions and refrigerators, including:

[0051] First image: The user is wearing the first clothing combination (red T-shirt + blue jeans).

[0052] Second image: The user is wearing the second clothing combination (white shirt + black casual pants).

[0053] Third image: The user is wearing the third clothing combination (red T-shirt + black casual pants), meaning the clothing comes from the first two combinations.

[0054] Training the image generation model: The first image (red T-shirt + blue jeans) and the second image (white shirt + black casual pants) are used as input data, and the third image (red T-shirt + black casual pants) is used as the target output data. These images are fed into a generative adversarial network (GAN) structure for training.

[0055] The resulting clothing visualization model is trained on a large number of such datasets. The model learns to maintain the user's body shape and posture while replacing clothing items on the user's body based on the input clothing image. The final trained model is a clothing visualization model capable of generating a virtual try-on image of the user based on a suggested outfit.

[0056] The model training method provided in this embodiment can generate a model that offers users a personalized virtual try-on service. Its technical effect solves the industry problem of users being unable to predict how clothing will look on them. By generating realistic wearing effect images, users can intuitively see the overall effect of the outfit on themselves in advance.

[0057] In an exemplary embodiment, before inputting the text matching scheme into the clothing visualization model, the method further includes: acquiring a fourth image of the target object wearing a fourth clothing combination; using all clothing images corresponding to the clothing in the fourth clothing combination as input data and the fourth image as output data to train an image generation model to obtain the clothing visualization model.

[0058] Optionally, in the above embodiments, another more convenient virtual try-on model training scheme is provided, the specific process of which is as follows:

[0059] Capture a single image: The system captures an image of the user wearing the fourth clothing combination (e.g., a plain white T-shirt and a pair of solid-color basic pants), as well as a separate flat lay image of each item in the fourth clothing combination (white T-shirt, solid-color basic pants).

[0060] Training the model: The image generation model is trained by taking individual flat-lay images of each item of clothing in the fourth clothing combination (white T-shirt, solid color basic pants) as input data and the complete image of the user wearing these items (the fourth image) as output data.

[0061] The resulting clothing visualization model: Through learning, the model establishes a mapping relationship from "clothing lay-up image" to "user wearing image". After training, when a user is input a new, never-before-worn clothing lay-up image, the model can synthesize a realistic image of the user wearing that new clothing.

[0062] This embodiment provides another, more convenient virtual try-on model training scheme. Its technical advantage lies in significantly reducing the implementation threshold and user costs of virtual try-on functionality. Users do not need to take numerous photos of complex clothing combinations; only a few photos of basic garments worn and their corresponding flat lay images are required to train an effective model.

[0063] In one exemplary embodiment, the method further includes: upon receiving a storage request for storing clothing data in the database, initiating a stabilization countdown; if a new storage request is received before the stabilization countdown ends, resetting the stabilization countdown; and if the stabilization countdown ends, batch executing the received storage requests.

[0064] Optionally, in the above embodiments, for example, a user is quickly uploading photos of a batch of newly purchased clothing items to the electronic wardrobe system. After each photo is successfully uploaded and its attributes are automatically identified, the system generates a record of clothing data to be saved on the interface and triggers a storage request, preparing to write the complete data of that clothing item (including image path, identified attributes, tags, etc.) to the local database. Upon receiving the first storage request, the system does not immediately perform a database write operation but instead starts a 300-millisecond anti-shake countdown. Before the 300-millisecond countdown ends, the user successfully uploads and identifies another image of clothing, and the system immediately generates a new storage request. Upon detecting this new request, the system immediately resets the 300-millisecond countdown and continues waiting. When the user stops uploading, and the system does not receive any new storage requests within 300 milliseconds, and the countdown successfully ends, the system merges all accumulated clothing addition operations during this period (e.g., all data of 5 successfully identified and to-be-saved clothing items) into a single database transaction for batch writing. This process reduces the number of frequent I / O operations that might have previously been five to one, greatly reducing system resource consumption, avoiding interface lag caused by frequent read and write operations, and ensuring a smooth user experience when adding clothes in batches.

[0065] Meanwhile, the system is also designed with an exception handling mechanism. If the batch storage fails, it will automatically retry twice and display a friendly message to the user on the interface: "Clothes are being saved, please do not exit." The storage will continue to be completed after the network or system is restored, ensuring that the data is not lost.

[0066] This embodiment specifically optimizes performance bottlenecks caused by frequent I / O operations through anti-jitter latency and batch processing mechanisms. It significantly improves response speed and interface smoothness when performing data-intensive operations (such as batch adding clothing and batch editing tags). By merging multiple scattered storage requests into a single batch operation, it effectively reduces system resource consumption and avoids interface lag and operation delays, thus providing users with a stable and smooth interactive experience.

[0067] To better understand the process of generating the above clothing matching results, the implementation flow of the above clothing matching method will be described below in conjunction with optional embodiments, but this is not intended to limit the technical solution of the embodiments of this application.

[0068] In an optional embodiment, Figure 3 This is a system architecture diagram of an intelligent electronic wardrobe management and matching recommendation system according to an embodiment of this application, such as... Figure 3 As shown, it specifically includes:

[0069] Front-end presentation layer: developed based on Vue3 and uni-app framework, supporting deployment on multiple platforms including mini-programs, H5, and APP.

[0070] Business logic layer: includes three core business modules: wardrobe management, intelligent matching, and data statistics.

[0071] AI Service Layer: Integrates third-party AI platforms to provide image recognition and natural language processing capabilities.

[0072] Data storage layer: adopts a hybrid architecture of local storage + cloud storage.

[0073] In an optional embodiment, Figure 4 This is a flowchart of an intelligent clothing recognition method according to an embodiment of this application, such as... Figure 4 As shown, the specific steps include:

[0074] 1. Image preprocessing: Adjusting the size, converting the format, and optimizing the quality of user-uploaded images.

[0075] 2. AI Image Recognition: Uses AI image recognition models to identify basic attributes of clothing such as type, color, and material.

[0076] 3. Feature Extraction: Extract visual feature vectors of clothing items for subsequent similarity calculation and outfit recommendations.

[0077] 4. Attribute Analysis: Based on the recognition results, analyze the advanced attributes of clothing such as style, applicable occasions, and season.

[0078] 5. Automatic classification: Based on the recognition results, the clothing is automatically classified into the corresponding categories.

[0079] 6. Tag Generation: Automatically generates descriptive tags for easy user searching and management.

[0080] 7. Data storage: Store the recognition results and original images locally and in the cloud.

[0081] In an optional embodiment, Figure 5 This is a flowchart of an intelligent clothing matching recommendation according to an embodiment of this application, such as... Figure 5 As shown, the specific steps include:

[0082] 1. User input: Users input their needs or preferences, which may be a specific description of clothing matching.

[0083] 2. Prompt word construction: Based on the user's input requirements, the system constructs corresponding prompt words, which will be used to guide the language model to the next step of processing.

[0084] 3. RAG Knowledge Retrieval: The RAG (Retrieval Enhanced Generation) system performs knowledge retrieval based on prompts to find matching knowledge or information related to user needs.

[0085] 4. Large Language Model Reasoning: The retrieved knowledge is input into the large language model for reasoning, and the model generates preliminary collocation schemes based on this information.

[0086] 5. Result Analysis: The results generated by the model need to be analyzed in order to extract specific matching schemes.

[0087] 6. Outfit Scheme Generation: Based on the analysis results, the system generates specific clothing outfit schemes.

[0088] 7. Post-processing optimization: The generated pairing schemes may require further optimization to ensure they meet user preferences and needs.

[0089] 8. Recommended Display: Finally, the optimized matching schemes are visualized and displayed to users for selection or reference.

[0090] In an optional embodiment, Figure 6 This is a flowchart of a de-jitter storage mechanism according to an embodiment of this application, such as... Figure 6 As shown, the specific steps include:

[0091] 1. User action: User actions trigger the need for data storage.

[0092] 2. Triggering storage: After receiving a user operation, the system triggers a storage operation request.

[0093] 3. Debouncing Detection: The system performs a debouncing detection to determine whether an immediate storage operation is necessary. Debouncing is a technique used to reduce the number of executions when events are triggered consecutively.

[0094] 4. Delay Waiting: If the debouncing system determines not to immediately perform storage, it will wait for 300 milliseconds. During this time, if the user performs another operation, the wait timer will be reset.

[0095] 5. Batch storage: After the waiting time is over, the system will perform a batch storage operation to process all the data to be stored at once, so as to reduce the frequency of I / O operations.

[0096] 6. Exception handling: If an exception occurs during the stored procedure, the system will perform exception handling, which may include retry mechanisms or error messages.

[0097] 7. User Feedback: After the storage operation is completed, the system will provide feedback to the user, informing them of the operation result.

[0098] In an optional embodiment, the above method for generating clothing matching results can be used synchronously across multiple platforms, making it convenient for users to use the application smoothly in different scenarios and on different devices. Figure 7 This is a schematic diagram illustrating cross-platform data synchronization according to an embodiment of this application, such as... Figure 7 As shown, the data flow begins with user interaction and, through different platforms such as mini-programs, H5 pages, and apps, first enters the local cache and then is synchronized to cloud storage. The data in the cloud storage is managed and synchronized through a data center to ensure data consistency and real-time updates. Finally, the synchronization service distributes the updated data to other devices, achieving cross-platform data synchronization and a consistent user experience.

[0099] Through the above description of the embodiments, those skilled in the art can clearly understand that the methods according to the above embodiments can be implemented by means of software plus necessary general-purpose hardware platforms. Of course, they can also be implemented by hardware, but in many cases the former is a better implementation method. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disk) and includes several instructions to cause a terminal device (which may be a mobile phone, computer, server, or network device, etc.) to execute the methods of the various embodiments of this application.

[0100] Figure 8 This is a structural block diagram of an apparatus for generating clothing matching results according to an embodiment of this application; as shown... Figure 8 As shown, it includes:

[0101] The data acquisition module 82 is used to acquire clothing data from the database, wherein the clothing data includes clothing images and clothing attributes corresponding to the clothing in the clothing images;

[0102] The first generation module 84 is used to generate a text matching scheme based on the target object's demand instructions and the clothing data, wherein the target object's demand instructions represent the target object's clothing matching needs described in natural language;

[0103] The second generation module 86 is used to generate a visualized clothing matching result based on the text matching scheme and the clothing data.

[0104] The aforementioned device stores users' clothing images and attributes in a structured data structure. Then, based on the user's clothing matching needs described in natural language and the clothing data, it generates clothing matching schemes for the user and further displays the visualization effects of these schemes. This solves the problem of how to achieve personalized clothing matching recommendations, thus realizing the desired personalized clothing matching effect.

[0105] In an exemplary embodiment, the data acquisition module 82 is used to acquire the clothing image, wherein the clothing image is acquired by the image acquisition device or uploaded by the target object; the clothing image is identified by an image recognition tool to obtain the clothing attributes corresponding to the clothing in the clothing image; the clothing image and the clothing attributes are associated and saved to the database to obtain the clothing data.

[0106] In an exemplary embodiment, the first generation module 84 is configured to input a preset prompt word into a large language model, wherein the preset prompt word is used to instruct the large language model to recommend clothing for the target object based on the knowledge in the clothing knowledge base and the clothing data after receiving a demand instruction; input the demand instruction of the target object into the large language model to obtain the target clothing combination selected by the large language model from the clothing data; and generate the text matching scheme based on the target clothing combination.

[0107] In one exemplary embodiment, the second generation module 86 is configured to, upon receiving a preview instruction to visualize the text matching scheme, input the text matching scheme into a clothing visualization model; and obtain the clothing matching result output by the clothing visualization model, wherein the clothing matching result is used to demonstrate the effect of the target object wearing the target clothing combination.

[0108] In an exemplary embodiment, the second generation module 86 is used to acquire a first image of the target object wearing a first clothing combination, a second image of the target object wearing a second clothing combination, and a third image of the target object wearing a third clothing combination, wherein the clothing in the third clothing combination comes from clothing in the first clothing combination or clothing in the second clothing combination; the image generation model is trained using the first image and the second image as input data and the third image as output data to obtain the clothing visualization model.

[0109] In an exemplary embodiment, the second generation module 86 is used to acquire a fourth image of the target object wearing the fourth clothing combination; and to train the image generation model using all clothing images corresponding to the clothing in the fourth clothing combination as input data and the fourth image as output data to obtain the clothing visualization model.

[0110] In one exemplary embodiment, the device is further configured to: initiate a stabilization countdown after receiving a storage request for storing clothing data in the database; reset the stabilization countdown if a new storage request is received before the stabilization countdown ends; and batch execute the received storage requests if the stabilization countdown ends.

[0111] Embodiments of this application also provide a storage medium including a stored program, wherein the program executes any of the methods described above when it is run.

[0112] Optionally, in this embodiment, the storage medium may be configured to store program code for performing the following steps:

[0113] S1, Obtain clothing data from the database, wherein the clothing data includes clothing images and clothing attributes corresponding to the clothing in the clothing images;

[0114] S2, Generate a text matching scheme based on the target object's demand instructions and the clothing data, wherein the target object's demand instructions represent the target object's clothing matching needs described in natural language;

[0115] S3, Generate a visualized clothing matching result based on the text matching scheme and the clothing data.

[0116] Embodiments of this application also provide an electronic device including a memory and a processor, wherein the memory stores a computer program and the processor is configured to run the computer program to perform the steps in any of the above method embodiments.

[0117] Optionally, the electronic device may further include a transmission device and an input / output device, wherein the transmission device is connected to the processor and the input / output device is connected to the processor.

[0118] Optionally, in this embodiment, the processor can be configured to perform the following steps via a computer program:

[0119] S1, Obtain clothing data from the database, wherein the clothing data includes clothing images and clothing attributes corresponding to the clothing in the clothing images;

[0120] S2, Generate a text matching scheme based on the target object's demand instructions and the clothing data, wherein the target object's demand instructions represent the target object's clothing matching needs described in natural language;

[0121] S3, Generate a visualized clothing matching result based on the text matching scheme and the clothing data.

[0122] Optionally, in this embodiment, the storage medium may include, but is not limited to, various media capable of storing program code, such as USB flash drives, read-only memory (ROM), random access memory (RAM), portable hard drives, magnetic disks, or optical disks.

[0123] Embodiments of this application also provide a computer program product, which includes a computer program that, when executed by a processor, implements the steps in any of the above method embodiments.

[0124] Embodiments of this application also provide another computer program product, including a non-volatile computer-readable storage medium storing a computer program that, when executed by a processor, implements the steps in any of the above method embodiments.

[0125] The embodiments described herein also provide a computer program that includes computer instructions stored in a computer-readable storage medium; a processor of a computer device reads the computer instructions from the computer-readable storage medium and executes the computer instructions, causing the computer device to perform the steps in any of the above method embodiments.

[0126] Optionally, specific examples in this embodiment can refer to the examples described in the above embodiments and optional implementations, and will not be repeated here.

[0127] Obviously, those skilled in the art should understand that the modules or steps of this application described above can be implemented using general-purpose computing devices. They can be centralized on a single computing device or distributed across a network of multiple computing devices. Optionally, they can be implemented using computer-executable program code, thereby storing them in a storage device for execution by a computing device. In some cases, the steps shown or described can be performed in a different order than those presented here, or they can be fabricated as separate integrated circuit modules, or multiple modules or steps can be fabricated as a single integrated circuit module. Thus, this application is not limited to any particular combination of hardware and software.

[0128] The above description is only a preferred embodiment of this application. It should be noted that for those skilled in the art, several improvements and modifications can be made without departing from the principle of this application, and these improvements and modifications should also be considered within the scope of protection of this application.

Claims

1. A method for generating clothing matching results, characterized in that, include: Retrieve clothing data from a database, wherein the clothing data includes clothing images and clothing attributes corresponding to the clothing in the clothing images; A text-based matching scheme is generated based on the target object's requirement instructions and the clothing data, wherein the target object's requirement instructions represent the target object's clothing matching needs described in natural language; Based on the text-based matching schemes and the clothing data, a visualized clothing matching result is generated.

2. The method for generating clothing matching results according to claim 1, characterized in that, Before retrieving clothing data from the database, the method further includes: The clothing image is acquired, wherein the clothing image is acquired by the image acquisition device or uploaded by the target object; The clothing image is identified using an image recognition tool to obtain the clothing attributes corresponding to the clothing in the clothing image; The clothing image and its attributes are associated and saved to the database to obtain the clothing data.

3. The method for generating clothing matching results according to claim 1, characterized in that, Generate text-based outfit matching schemes based on the target user's requirements and the clothing data, including: The preset prompt words are input into the large language model, wherein the preset prompt words are used to instruct the large language model to recommend clothing for the target object based on the knowledge in the clothing knowledge base and the clothing data after receiving the demand instruction; The demand instruction of the target object is input into the large language model to obtain the target clothing combination selected by the large language model from the clothing data; The text matching scheme is generated based on the target clothing combination.

4. The method for generating clothing matching results according to claim 3, characterized in that, Based on the text-based matching scheme and the clothing data, a visualized clothing matching result is generated, including: Upon receiving a preview instruction to visualize the text-based outfit matching scheme, the text-based outfit matching scheme is input into the clothing visualization model; Obtain the clothing matching result output by the clothing visualization model, wherein the clothing matching result is used to display the effect of the target object wearing the target clothing combination.

5. The method for generating clothing matching results according to claim 4, characterized in that, Before inputting the text matching scheme into the clothing visualization model, the method further includes: A first image of the target object wearing a first clothing combination, a second image of the target object wearing a second clothing combination, and a third image of the target object wearing a third clothing combination are acquired, wherein the clothing in the third clothing combination is derived from the clothing in the first clothing combination or the clothing in the second clothing combination. The image generation model is trained using the first image and the second image as input data and the third image as output data to obtain the clothing visualization model.

6. The method for generating clothing matching results according to claim 4, characterized in that, Before inputting the text matching scheme into the clothing visualization model, the method further includes: Acquire a fourth image of the target object wearing the fourth clothing combination; Using all clothing images corresponding to the clothing in the fourth clothing combination as input data and the fourth image as output data, the image generation model is trained to obtain the clothing visualization model.

7. The method for generating clothing matching results according to claim 1, characterized in that, The method further includes: Upon receiving a storage request to store clothing data in the database, the anti-shake countdown is initiated; If a new storage request is received before the stabilization countdown ends, the stabilization countdown is reset; Once the anti-shake countdown ends, the received storage requests are executed in batches.

8. A device for generating clothing matching results, characterized in that, include: The data acquisition module is used to acquire clothing data from the database, wherein the clothing data includes clothing images and clothing attributes corresponding to the clothing in the clothing images; The first generation module is used to generate a text matching scheme based on the target object's demand instructions and the clothing data, wherein the target object's demand instructions represent the target object's clothing matching needs described in natural language; The second generation module is used to generate visualized clothing matching results based on the text matching scheme and the clothing data.

9. A computer-readable storage medium, characterized in that, The computer-readable storage medium includes a stored program, wherein the program, when executed, performs the method described in any one of claims 1 to 7.

10. An electronic device comprising a memory and a processor, characterized in that, The memory stores a computer program, and the processor is configured to execute the method described in any one of claims 1 to 7 through the computer program.

Images