Personalized interior design method based on user floor plan

Abstract

The invention relates to a personalized interior design method based on a user floor plan, which comprises the following steps of: analyzing behavior data of a target user to obtain preference parameters of the target user; performing vector conversion on the house plane graph to obtain vector space relation feature data; inputting the vector space relation feature data and the space function arrangement parameters into a preset graph neural network for space layout, and generating an initial furniture layout scheme; performing scheme screening on the initial furniture layout scheme to obtain a screened furniture layout scheme; matching design is carried out on matching parameters and the screened furniture layout scheme, a matching design scheme is obtained, initial visualization is carried out on the matching design scheme in a preset interactive interface, and a three-dimensional visualization effect picture is generated. The technical problems that traditional interior design services are often high in cost and long in consumed time, common consumers are difficult to afford, and large supply and demand imbalance exists in the market are solved.

Description

Technical Field

[0001] The invention relates to the technical field of interior design, and in particular to a personalized interior design method based on a user's house plan. Background Art

[0002] With the acceleration of urbanization, people's requirements for living environment are increasing, not only limited to basic living needs, but also pursuing personalization and comfort. However, traditional interior design services are often costly and time-consuming, which is unaffordable for ordinary consumers. This has led to a large imbalance between supply and demand in the market. On the one hand, consumers have a strong demand for high-quality and personalized interior design; on the other hand, the existing design service model cannot efficiently meet these needs, especially in terms of quickly responding to users' personalized preferences.

[0003] In addition, in the traditional interior design process, designers usually need to spend a lot of time and energy communicating with customers to understand their specific needs and preferences. This process is not only inefficient, but also prone to information transmission bias, which in turn affects the final design effect. With the development of information technology, especially the application of big data, artificial intelligence and other technologies, new possibilities have been provided for solving the above problems. By analyzing user behavior data to predict their preferences, the accuracy of the design and service efficiency can be significantly improved, making personalized interior design more convenient and popular.

[0004] Although there are many advantages in using advanced technology to achieve personalized interior design, there are still some challenges in practical applications. For example, how to ensure that the recommendation algorithm can accurately capture the real needs of users rather than just superficial preferences; how to balance the differences between design creativity and user expectations; and how to protect user privacy. If these problems are not effectively solved, they will directly affect the user experience and market acceptance of personalized interior design systems. Therefore, exploring a design method that can both meet the personalized needs of users and ensure their privacy security has become one of the important directions of current research. Summary of the invention

[0005] The main purpose of the present invention is to provide a personalized interior design method based on the user's floor plan, which solves the technical problem that traditional interior design services are often costly and time-consuming, which are unaffordable for ordinary consumers and lead to a large imbalance between supply and demand in the market.

[0006] To achieve the above object, the present invention provides a personalized interior design method based on a user's floor plan, comprising the following steps: By using a preset recommendation algorithm, the behavior data of the target user is analyzed to obtain the target user's preference parameters; wherein the preference parameters include space function arrangement parameters and matching parameters; Obtaining a house plan uploaded by a target user, and performing vector conversion on the house plan using a plan image vectorization technology to obtain vector space relationship feature data; Inputting the vector space relationship feature data and the space function arrangement parameters into a preset graph neural network to perform space layout and generate an initial furniture layout plan; Screening the initial furniture layout schemes according to preset design hard rules to obtain a screened furniture layout scheme; Performing matching design on the matching parameters and the screened furniture layout scheme to obtain a matching design scheme, and initially visualizing the matching design scheme on a preset interactive interface; It is detected in real time whether the target user performs interface interaction operations on the interactive interface. If so, the matching design scheme targeted by the interface interaction operation is taken as the target scheme, and the target scheme is edited and rendered according to the interface interaction operation through a preset AI editing system to obtain a three-dimensional visualization effect diagram corresponding to the house floor plan.

[0007] Furthermore, the behavior data includes furniture browsing records and input text data. The behavior data of the target user is analyzed by a preset recommendation algorithm to obtain the target user's preference parameters, including: By using a preset recommendation algorithm, the furniture product browsing record and input text data of the target user are analyzed to obtain user-product interaction information; wherein the user-product interaction information is the browsing time, click times and collection status of different furniture products by the target user; Performing cluster analysis on the user-product interaction information to obtain a target user's preferred furniture style feature vector; Analyze the co-occurrence relationship of furniture products in the user-product interaction information to obtain a furniture matching pattern vector; Analyzing the furniture product description text in the user-product interaction information, and extracting the spatial function keywords in the furniture product description text to obtain a spatial function keyword vector; Constructing a target user preference matrix based on the preferred furniture style feature vector, the furniture matching pattern vector and the space function keyword vector; Inputting the target user preference matrix into a preset Bayesian inference algorithm to obtain the target user's preference probability distribution; The furniture selection behavior of the target user is predicted based on the preference probability distribution to obtain the preference parameters of the target user; wherein the preference parameters of the target user include space function arrangement parameters and matching parameters.

[0008] Furthermore, the floor plan image vectorization technology is used to convert the floor plan into a vector to obtain vector space relationship feature data, including: Identifying and marking each functional area in the house plan to obtain functional area marking data; Preprocessing the house plan using plan image vectorization technology to obtain a preprocessed image; Extracting contour lines in the preprocessed image to obtain a contour line image, and classifying and attribute-marking the contour lines in the contour line image based on the functional area annotation data to obtain a contour line image with functional area information; Converting the outline line image with the functional area information into a vector graphic to obtain vector graphic data; The vector graphics data is subjected to spatial relationship analysis to obtain vector spatial relationship feature data.

[0009] Furthermore, the vector space relationship feature data and the space function arrangement parameters are input into a preset graph neural network to optimize the space layout and generate an initial furniture layout plan, including: Performing spatial feature extraction on the vector spatial relationship feature data to obtain extracted spatial features; wherein the extracted spatial features include the area, shape, window positions and door opening direction information of the house plan; Classify and process the spatial function arrangement parameters to obtain functional area division information; wherein the functional area division information includes area requirements and relative position requirements of different functional areas; Based on the spatial feature vector and the functional area division information, construct a spatial layout relationship diagram; Based on a multi-scale graph neural network, feature propagation and fusion are performed on the spatial layout relationship graph to obtain a fused feature graph; Inputting the fused feature graph into a preset graph neural network to perform furniture layout exploration and space layout optimization to obtain a candidate furniture layout solution; The candidate furniture layout plans are evaluated and screened for rationality to obtain an initial furniture layout plan.

[0010] Furthermore, the initial furniture layout scheme is screened by using preset hard design rules to obtain a screened furniture layout scheme, including: Performing spatial relationship analysis on the initial furniture layout plan using a spatial topology analysis algorithm to obtain a spatial topology relationship diagram between the furniture; wherein the spatial topology relationship diagram includes relative positions, distances, and connectivity between the furniture; Based on the ergonomic principles and the spatial topological relationship diagram, an ergonomic evaluation is performed on the initial furniture layout plan to obtain an ergonomic scoring matrix; By using a multi-objective optimization algorithm, the initial furniture layout scheme is optimized and calculated based on the ergonomic scoring matrix to obtain an ergonomic layout scheme set; wherein the ergonomic layout scheme set is a candidate layout scheme set that meets ergonomic requirements; Dividing the ergonomic layout solution set into functional areas to obtain a functional area semantic map; By using preset hard design rules, the functional area semantic graph is subjected to rule constraint check to obtain a set of ergonomic layout solutions after constraint check; The ergonomic layout scheme set after the constraint check is comprehensively evaluated and screened by a multi-criteria decision analysis method to obtain a screened furniture layout scheme.

[0011] Furthermore, matching and designing the matching parameters and the screened furniture layout scheme to obtain a matching design scheme includes: Deconstructing the matching parameters and the selected furniture layout scheme respectively, and obtaining a deconstructed parameter set and a layout element set; Performing semantic encoding on the deconstructed parameter set to obtain an encoded semantic vector; Performing spatial relationship encoding on the layout element set to obtain a spatial relationship vector; Based on the attention mechanism network and matching algorithm, the correlation between the encoded semantic vector and the spatial relationship vector is calculated to obtain the matching weight matrix; Reorganize the layout of the screened furniture layout scheme based on the matching weight matrix to obtain a layout reorganization scheme; The layout reorganization scheme is checked and adjusted for style consistency to obtain a matching design scheme, and the matching design scheme is initially visualized in a preset interactive interface.

[0012] Furthermore, the target scheme is edited and rendered according to the interface interaction operation through a preset AI editing system to obtain a three-dimensional visualization effect diagram corresponding to the house plan, including: Performing real-time monitoring of the target user's interface interaction operations on the interaction interface to obtain the target user's operation instruction sequence for the matching design solution; wherein the operation instruction sequence includes recording the user's clicks, drags, and zooms; Using a multi-objective optimization algorithm, adjusting the parameters in the target solution based on the operation instruction sequence to obtain the adjusted design parameters and the adjusted design solution corresponding to the adjusted design parameters; wherein the parameters in the target solution include furniture layout parameters, color scheme parameters and material parameters; Performing spatial mapping on the adjustment design scheme through a preset spatial mapping algorithm to obtain a three-dimensional spatial layout model; Performing material and lighting rendering on the three-dimensional space layout model to obtain a three-dimensional space layout model with material texture and lighting effects; Through virtual reality technology, a three-dimensional visualization effect diagram is generated based on a three-dimensional space layout model with material texture and lighting effects.

[0013] The present invention also provides a personalized interior design system based on a user's floor plan, comprising: The recommendation module is used to analyze the behavior data of the target user through a preset recommendation algorithm to obtain the preference parameters of the target user; wherein the preference parameters include space function arrangement parameters and matching parameters; An acquisition module is used to acquire a house plan uploaded by a target user, and to perform vector conversion on the house plan using a plan image vectorization technology to obtain vector space relationship feature data; Inputting the vector space relationship feature data and the space function arrangement parameters into a preset graph neural network to perform space layout and generate an initial furniture layout plan; A screening module, used to screen the initial furniture layout scheme according to preset hard design rules to obtain a screened furniture layout scheme; A matching module, used for matching and designing the matching parameters and the screened furniture layout scheme to obtain a matching design scheme, and initially visualizing the matching design scheme on a preset interactive interface; The editing module is used to detect in real time whether the target user performs interface interaction operations on the interactive interface. If so, the matching design scheme targeted by the interface interaction operation is taken as the target scheme, and the target scheme is edited and rendered according to the interface interaction operation through a preset AI editing system to obtain a three-dimensional visualization effect diagram corresponding to the house floor plan.

[0014] The present invention also provides a computer device, comprising a memory and a processor, wherein the memory stores a computer program, and the processor implements the steps of any one of the above methods when executing the computer program.

[0015] The present invention also provides a computer-readable storage medium on which a computer program is stored. When the computer program is executed by a processor, the steps of any of the above-mentioned methods are implemented.

[0016] The personalized interior design method based on user floor plan provided by the present invention comprises the following steps: analyzing the behavior data of the target user through a preset recommendation algorithm to obtain the preference parameters of the target user; wherein the preference parameters include space function arrangement parameters and matching parameters; obtaining the house floor plan uploaded by the target user, and performing vector conversion on the house floor plan using the floor plan image vectorization technology to obtain vector space relationship feature data; inputting the vector space relationship feature data and the space function arrangement parameters into a preset graph neural network for space layout to generate an initial furniture layout plan; screening the initial furniture layout plan through a preset design hard rule to obtain a screened furniture layout plan; matching and designing the matching parameters and the screened furniture layout plan to obtain a matching design plan, and The matching design scheme is initially visualized in the preset interactive interface; it is detected in real time whether the target user performs interface interaction operations in the interactive interface. If so, the matching design scheme targeted by the interface interaction operation is used as the target scheme, and the target scheme is edited and rendered according to the interface interaction operation through the preset AI editing system to obtain the three-dimensional visualization effect map corresponding to the house floor plan, which solves the technical problem that traditional interior design services are often costly and time-consuming, which is difficult for ordinary consumers to afford, resulting in a large imbalance between supply and demand in the market. It realizes the optimization of spatial layout using advanced floor plan image vectorization technology and graph neural network, which can not only generate a reasonable furniture layout plan based on the user's actual floor plan, but also further screen out the optimal solution through preset design hard rules. More importantly, it allows users to adjust the design plan directly on the interactive interface and view the adjusted three-dimensional visualization effect map in real time, which greatly enhances the beneficial effects of user participation and satisfaction. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] Figure 1 is a schematic diagram of steps of a personalized interior design method based on a user's floor plan in one embodiment of the present invention; Figure 2 is a structural block diagram of a personalized interior design system based on a user's floor plan in one embodiment of the present invention; Figure 3 It is a schematic block diagram of the structure of a computer device according to an embodiment of the present invention.

[0018] The realization of the purpose, functional features and advantages of the present invention will be further explained in conjunction with embodiments and with reference to the accompanying drawings. DETAILED DESCRIPTION

[0019] In order to make the purpose, technical solution and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0020] like Figure 1 As shown, Figure 1 It is a schematic diagram of steps of a personalized interior design method based on a user's floor plan in one embodiment of the present invention; In one embodiment of the present invention, a personalized interior design method based on a user's floor plan is provided, comprising the following steps: Step S1, analyzing the behavior data of the target user through a preset recommendation algorithm to obtain the target user's preference parameters; wherein the preference parameters include space function arrangement parameters and matching parameters.

[0021] Specifically, in this personalized interior design method, the target user's behavioral data is first analyzed through a preset recommendation algorithm, which is the key starting point of the entire design process. The "behavioral data" here can include the user's online browsing design style preferences, past purchase records, interactive content on social media, etc., which can reflect the user's specific preferences for interior design. For example, if a user often browses modern minimalist style home decoration cases, the system will recognize that the user may prefer this style. Next, based on these behavioral data, the recommendation algorithm will calculate and extract the target user's preference parameters, which are specifically divided into two parts: one is the "spatial function arrangement parameter", which is the user's functional requirements for different rooms or areas, such as whether an open kitchen or a closed kitchen is needed, whether a work area is needed in the bedroom, etc.; the second is the "matching parameter", which refers to the user's personal preferences for colors, materials, decorations, etc., such as preference for cool colors or warm colors, preference for wooden furniture or metal furniture, etc. In order to achieve this process, the system may use a machine learning model to train a large number of labeled data sets so that the model can accurately capture the correlation between user preferences and specific design elements, and then provide each user with unique design suggestions. For example, in an actual application scenario, when a user uploads his or her home floor plan and authorizes the system to access his or her online behavior data, the system can quickly generate an interior design plan that is in line with the user's living habits and reflects personal aesthetic preferences.

[0022] Step S2, obtaining the house floor plan uploaded by the target user, and performing vector conversion on the house floor plan using floor plan image vectorization technology to obtain vector space relationship feature data.

[0023] Specifically, in the personalized interior design method, obtaining the house floor plan uploaded by the target user is a crucial step, because it directly determines the basis of subsequent design work. When the user uploads his house floor plan through an application or website, the system will immediately process the floor plan. The "floor plan image vectorization technology" mentioned here refers to the process of converting the floor plan uploaded by the user, which may be in image format (such as JPEG, PNG, etc.) into vector graphics. Vector graphics are composed of geometric elements such as points, lines, and surfaces, which can more accurately represent spatial structure and size information, are not limited by resolution, and are convenient for subsequent editing and calculation. During the conversion process, the system automatically identifies and extracts key elements such as room boundaries, door and window positions, and then converts this information into vector data to form the so-called "vector space relationship feature data". For example, suppose a user wants to design a modern living room layout for his new home. He first uploads a floor plan of his home to the design platform. After receiving this floor plan, the platform will immediately start the vectorization technology to process it. In this process, the system will not only identify the outlines of each room, but also accurately mark the location and size of the doors and windows, and even distinguish different wall materials, such as load-bearing walls and partition walls. In this way, the system can obtain detailed vector space relationship feature data, which not only contains the basic structural information of the room, but also provides accurate data support for subsequent space layout planning and furniture placement. Through such processing, designers or AI systems can design space layout more efficiently, ensuring that the design plan not only meets the actual needs of users, but also makes full use of every inch of space to achieve the best design effect.

[0024] Step S3, inputting the vector space relationship feature data and the space function arrangement parameters into a preset graph neural network to perform space layout and generate an initial furniture layout plan.

[0025] Specifically, after completing the vectorization processing of the house floor plan, the next step is to input the vector space relationship feature data and the space function arrangement parameters into the preset graph neural network for deeper spatial layout analysis and optimization. This process aims to generate the most suitable initial furniture layout plan based on the floor plan information and specific functional requirements provided by the user through advanced machine learning algorithms. Graph neural network is a deep learning model that is particularly suitable for processing graph structure data. It can effectively capture the complex relationship between nodes. For the house design application in this example, it can be understood as being able to well understand and optimize the spatial relationship and functional allocation between the various parts of the room. For example, continuing to take the living room design of the above user as an example, the house floor plan after vectorization has been converted into vector space relationship feature data containing information such as all room boundaries, door and window positions, and wall types. At the same time, the user also clearly pointed out his needs for the living room space, such as the need for sufficient leisure areas, TV viewing areas, and possible office corners, which constitute the spatial function arrangement parameters. When this data is input into the preset graph neural network, the network will comprehensively consider factors such as the shape, size, and orientation of each room, and combine the specific functional requirements proposed by the user, and finally output one or more optimal initial furniture layout solutions by simulating multiple layout possibilities. These solutions can not only meet the functional needs of users, but also maximize space utilization and aesthetics, providing users with an intuitive and practical design reference. In this way, even non-professional users can easily obtain professional-level interior design solutions, greatly improving the efficiency and satisfaction of the entire design process.

[0026] Step S4, screening the initial furniture layout schemes by using preset hard design rules to obtain screened furniture layout schemes.

[0027] Specifically, after the initial furniture layout scheme is generated, in order to ensure the rationality and practicality of the design scheme, it is necessary to further screen these schemes through preset design hard rules. The "design hard rules" mentioned here refer to a series of standards based on the basic principles of interior design and ergonomics, which are used to evaluate the feasibility and comfort of furniture layout schemes. These rules may include but are not limited to the minimum spacing requirements between furniture, barrier-free access areas in front of doors and windows, and safe distances between electrical sockets and switches, etc., in order to ensure that the design scheme is not only beautiful, but also safe and practical. Taking the living room design mentioned earlier as an example, assume that the graph neural network has generated several different initial furniture layout schemes. At this point, the system automatically calls the preset design hard rules to evaluate each scheme. For example, one rule may be to ensure that the sofa is at least 3 meters away from the TV to provide a comfortable viewing experience; another rule requires that the width of all main passages must not be less than 90 cm to ensure that family members can move smoothly. By applying these rules, the system can exclude layout schemes that do not meet the standards, such as a scheme in which the distance between the sofa and the coffee table is too small, or the placement of the TV cabinet blocks the path to the balcony. After this round of screening, the system will ultimately retain those furniture layout plans that meet both the user's personalized needs and the basic design specifications. These screened furniture layout plans can not only better adapt to the user's actual lifestyle, but also ensure the safety and convenience of the home environment. Through such a screening process, not only can the quality of the design plan be improved, but also the user's trust and satisfaction with the final design results can be enhanced.

[0028] Step S5, matching and designing the matching parameters and the screened furniture layout scheme to obtain a matching design scheme, and initially visualizing the matching design scheme on a preset interactive interface.

[0029] Specifically, after the screening of the furniture layout scheme is completed, the next step is to match the matching parameters and the screened furniture layout scheme to generate the final matching design scheme. The "matching parameters" here refer to the user's specific preferences in terms of color, material, decoration style, etc., while the "screened furniture layout scheme" is a layout scheme that meets functionality and safety after being tested by the hard rules of design. By combining the two, the system can provide users with a design scheme that meets their personalized needs and is practical. For example, in the application scenario of living room design, it is assumed that the user prefers a modern and simple style, tends to use white and gray as the main colors, and likes the natural feeling brought by wooden furniture. After obtaining these matching parameters, the system will combine them with the previously screened furniture layout scheme, and automatically select furniture styles, colors and materials that meet the user's style preferences, such as choosing a white sofa with a gray carpet, as well as a wooden coffee table and TV cabinet. In addition, the system will also consider the selection of decorations, such as hanging paintings, lamps, etc., to ensure the consistency and coordination of the overall style. After completing the matching design, the system will initially visualize the obtained matching design scheme on the preset interactive interface. This means that users can see the design effect of the living room in a virtual three-dimensional environment, including details such as the placement of furniture, color matching, and material texture, so as to get an intuitive feeling. If the user has any comments or suggestions on the preliminary design plan, they can also make instant adjustments through the interactive interface, such as replacing a different style of sofa or changing the color of the wall. The system will update the design plan in real time based on the user's feedback until the user is satisfied. This interactive design process not only improves the flexibility of the design and user participation, but also ensures that the final design results fully meet the user's expectations.

[0030] Step S6, real-time detection of whether the target user performs interface interaction operations on the interactive interface. If so, the matching design scheme targeted by the interface interaction operation is taken as the target scheme, and the target scheme is edited and rendered according to the interface interaction operation through a preset AI editing system to obtain a three-dimensional visualization effect diagram corresponding to the house floor plan.

[0031] Specifically, in the process of personalized interior design, the system will detect in real time whether the target user has performed any interface interaction operation on the preset interactive interface. This link is a key part of the user's participation in the design process. In this way, users can directly participate in the adjustment and improvement of the design plan. Once it is detected that the user has performed an interface interaction operation, the system will immediately take the matching design plan targeted by the operation as the target plan, and edit and render the target plan accordingly according to the user's operation through the preset AI editing system, and finally generate an updated three-dimensional visualization effect map. For example, in the aforementioned living room design case, the user first saw the living room design plan generated by the system for it through the interactive interface, including the placement of furniture, color matching, and the selection of decorations. Suppose the user feels that the color of the sofa is a bit monotonous and hopes to liven up the space atmosphere by adding some colors. So the user selects the sofa on the interactive interface and selects a dark blue from the color options. At this time, the system will detect the user's operation in real time, and immediately take the current design plan as the target plan, calling the preset AI editing system for editing. The AI ​​editing system will recalculate the color matching between the sofa and other furniture, walls, floors and other elements according to the new color selected by the user to ensure the harmony and unity of the overall visual effect. At the same time, the system will render the new design and generate the latest 3D visualization effect map, so that users can immediately see the actual effect of the adjustment. If the user is still not satisfied with the new plan, he can continue to make other adjustments through the interactive interface, such as changing to a different style of coffee table or adding wall decorations, and the system will respond and update in real time in the same way. This real-time interactive method not only greatly enhances the user's sense of participation, but also makes the design process more flexible and efficient, ensuring that the final design plan can meet the user's personalized needs to the greatest extent.

[0032] In a specific embodiment, the behavior data includes furniture browsing records and input text data. The behavior data of the target user is analyzed by a preset recommendation algorithm to obtain the target user's preference parameters, including: By using a preset recommendation algorithm, the furniture product browsing record and input text data of the target user are analyzed to obtain user-product interaction information; wherein the user-product interaction information is the browsing time, click times and collection status of different furniture products by the target user; Performing cluster analysis on the user-product interaction information to obtain a target user's preferred furniture style feature vector; Analyze the co-occurrence relationship of furniture products in the user-product interaction information to obtain a furniture matching pattern vector; Analyzing the furniture product description text in the user-product interaction information, and extracting the spatial function keywords in the furniture product description text to obtain a spatial function keyword vector; Constructing a target user preference matrix based on the preferred furniture style feature vector, the furniture matching pattern vector and the space function keyword vector; Inputting the target user preference matrix into a preset Bayesian inference algorithm to obtain the target user's preference probability distribution; The furniture selection behavior of the target user is predicted based on the preference probability distribution to obtain the preference parameters of the target user; wherein the preference parameters of the target user include space function arrangement parameters and matching parameters.

[0033] Specifically, in the personalized interior design plan, in order to more accurately meet the needs of the target users, the system will collect and analyze the target users' behavioral data, which mainly includes furniture browsing records and text data entered by users. Through these data, the system can deeply understand the user's preference for furniture, so as to provide users with furniture recommendations and services that are more in line with their personal preferences. Specifically, the system first uses a preset recommendation algorithm to deeply analyze the target user's furniture product browsing records and input text data, aiming to obtain the interaction information between users and products. The user-product interaction information mentioned here covers many aspects such as the target user's browsing time, number of clicks, and collection status of different furniture products. For example, when a user stays on a modern minimalist style sofa page for a long time, clicks multiple times to zoom in to view details, or even collects it, it indicates that the user may have a high interest in this sofa, or even the entire modern minimalist style furniture. Next, the system will perform cluster analysis on these user-product interaction information, with the aim of extracting the target user's preferred furniture style feature vector. The process of cluster analysis is to classify user groups with similar browsing behaviors and preferences together. Through comparative analysis, it can be identified which style of furniture the target user prefers, such as Nordic style, industrial style or classical style. In this process, the system may find that the target user spends a long time browsing modern minimalist style furniture and has a high click-through rate, thus determining that the characteristic vector of the preferred furniture style tends to be modern minimalist style. At the same time, the system will also conduct an in-depth analysis of the co-occurrence relationship of furniture products in the user-product interaction information to obtain the furniture matching pattern vector. This analysis is mainly to understand which furniture has a high co-occurrence probability, that is, when users browse or buy a certain product, they often pay attention to or buy related furniture. For example, the system may find that while browsing sofas, users often check matching coffee tables, carpets, etc., which suggests that users have a certain matching tendency when purchasing furniture. Through the analysis of such co-occurrence relationships, the system can better understand the user's matching preferences and recommend more coordinated furniture combinations. In addition, the system will further analyze the furniture product description text in the user-product interaction information, extract spatial function keywords from it, and construct a spatial function keyword vector. This step is to capture the user's interest in specific spatial functions, such as office areas, leisure areas, or children's play areas. For example, if users frequently search for products containing keywords such as "bookshelf" and "workbench", this may mean that the user is looking for furniture suitable for the office area at home. Through such analysis, the system can more accurately understand the user's spatial needs and provide users with more targeted suggestions. Based on the preferred furniture style feature vector, furniture matching pattern vector, and spatial function keyword vector obtained above, the system will construct a comprehensive target user preference matrix.This matrix not only reflects the user's preference for furniture style and matching mode, but also includes the user's needs for specific space functions. It is a multi-dimensional data set that comprehensively displays the user's furniture selection preferences. Subsequently, the system inputs this preference matrix into the preset Bayesian inference algorithm and calculates the target user's preference probability distribution. The role of the Bayesian inference algorithm here is to perform probabilistic modeling on the user's various preference parameters and predict which style, matching method or spatial layout the user prefers in future furniture selection behavior. Finally, based on the preference probability distribution, the system predicts the furniture selection behavior of the target user, thereby obtaining the final target user preference parameters, including but not limited to spatial function arrangement parameters and matching parameters. These parameters will directly guide subsequent furniture recommendations and service optimization to ensure that the solutions provided can maximize the user's personal preferences and actual needs. For example, if the system predicts that the user has a strong preference for modern minimalist style furniture matching and the functional layout of the home office space, then when recommending furniture solutions to the user, modern minimalist style furniture will be given priority, and the design of the office area will be particularly emphasized to improve the user's satisfaction and user experience. Through such a systematic and personalized analysis process, not only can the accuracy of furniture recommendations be effectively improved, but also user engagement and satisfaction can be significantly enhanced, bringing users a more intimate and efficient home design service experience.

[0034] In a specific embodiment, the floor plan image vectorization technology is used to convert the floor plan of the house into a vector to obtain vector space relationship feature data, including: Identifying and marking each functional area in the house plan to obtain functional area marking data; Preprocessing the house plan using plan image vectorization technology to obtain a preprocessed image; Extracting contour lines in the preprocessed image to obtain a contour line image, and classifying and attribute-marking the contour lines in the contour line image based on the functional area annotation data to obtain a contour line image with functional area information; Converting the outline line image with the functional area information into a vector graphic to obtain vector graphic data; The vector graphics data is subjected to spatial relationship analysis to obtain vector spatial relationship feature data.

[0035] Specifically, in the personalized interior design process, using the floor plan image vectorization technology to convert the house floor plan into vector data is one of the key steps to achieve efficient and accurate design. This process can not only convert the floor plan uploaded by the user into vector data that can be processed by the computer, but also extract detailed vector space relationship feature data from it, providing a solid foundation for subsequent design optimization. Specifically, this step includes several closely related operations. The first is to identify and mark each functional area in the house floor plan uploaded manually or automatically to obtain the functional area marking data. The importance of this step is that it can help the system understand the purpose of different areas in the floor plan, such as living room, bedroom, kitchen, etc., and provide a basis for subsequent space layout design. For example, after the user uploads a floor plan containing multiple rooms, the system will automatically or through manual input by the user to identify and mark the names and functions of each room to ensure that each area can be accurately identified. Next, the system will use the floor plan image vectorization technology to preprocess the house floor plan to obtain a preprocessed image. The main purpose of preprocessing is to remove interference factors such as noise and blur in the image to make the image clearer and easier to process. This stage may involve a variety of technical means such as image denoising, contrast adjustment, and edge enhancement to ensure that the subsequent contour line extraction can be more accurate. Taking the living room design as an example, the preprocessed floor plan will present a cleaner and clearer effect, laying a good foundation for the next steps. After the preprocessing is completed, the system will enter the next important stage, which is to extract the contour lines in the preprocessed image to generate a contour line image. This process involves the application of edge detection algorithms. By detecting areas where pixel values ​​change significantly in the image, the system can accurately outline the contour lines of key elements such as room boundaries, door and window locations, etc. On this basis, the system will also classify and attribute the contour lines in the contour line image based on the previously acquired functional area annotation data, so as to obtain a contour line image with functional area information. For example, the system will mark which lines in the contour line image belong to the boundary of the living room, which lines represent the doors and windows of the bedroom, etc. This step is crucial for the subsequent spatial relationship analysis. Next, the system will convert the contour line image with functional area information into vector graphics to generate vector graphics data. The process of vectorization is to convert the lines, shapes and other elements in the image into mathematical geometric objects such as points, lines and surfaces. These objects not only have precise coordinate information, but can also be infinitely scaled without distortion. Through this conversion, the system can adjust and optimize the spatial layout more flexibly, and it also facilitates subsequent data processing and storage. For example, the boundary of the living room in the vector graphics will no longer be a string of pixels, but a line composed of a series of coordinate points, which can be easily edited and modified. Finally, the system will perform spatial relationship analysis on the generated vector graphics data to extract the vector spatial relationship feature data.This analysis process involves the analysis of multiple aspects such as spatial topological relationships and geometric relationships, with the aim of fully understanding the relative positions and connection methods between various areas inside the house. For example, the system will identify whether there is a direct doorway between the living room and the dining room, and the distance between the bedroom and the bathroom. This information is of great significance for designing a reasonable furniture layout and streamline planning. Through such spatial relationship analysis, the system can provide users with more scientific and practical design suggestions to ensure that the design plan is not only beautiful but also meets the actual use needs. In summary, by using the floor plan image vectorization technology to vectorize the floor plan of the house and extracting the vector space relationship feature data, the entire design process can not only greatly improve the design efficiency and accuracy, but also better meet the personalized needs of users. For example, in the aforementioned living room design case, the system can not only accurately identify the specific location and shape of the living room through this series of technical means, but also understand the user's needs for the living room function in detail, such as the need for sufficient leisure area, TV viewing area, etc., so as to provide users with a design plan that is both in line with personal preferences and practical. This process fully reflects the advancement and practicality of personalized interior design methods.

[0036] In a specific embodiment, the step of inputting the vector space relationship feature data and the space function arrangement parameters into a preset graph neural network to optimize the space layout and generate an initial furniture layout plan includes: Performing spatial feature extraction on the vector spatial relationship feature data to obtain extracted spatial features; wherein the extracted spatial features include the area, shape, window positions and door opening direction information of the house plan; Classify and process the spatial function arrangement parameters to obtain functional area division information; wherein the functional area division information includes area requirements and relative position requirements of different functional areas; Based on the spatial feature vector and the functional area division information, construct a spatial layout relationship diagram; Based on a multi-scale graph neural network, feature propagation and fusion are performed on the spatial layout relationship graph to obtain a fused feature graph; Inputting the fused feature graph into a preset graph neural network to perform furniture layout exploration and space layout optimization to obtain a candidate furniture layout solution; The candidate furniture layout plans are evaluated and screened for rationality to obtain an initial furniture layout plan.

[0037] Specifically, in the personalized interior design method, inputting the vector space relationship feature data and space function arrangement parameters into the preset graph neural network for space layout optimization is one of the core steps to generate the initial furniture layout plan. This process not only relies on advanced graph neural network technology, but also requires fine processing of the input data to ensure that the final generated layout plan not only meets the functional requirements of the user, but also has good space utilization efficiency and aesthetics. First, the system will extract the spatial features of the vector space relationship feature data to obtain the extracted spatial features. The goal of this stage is to extract useful information for space layout optimization from the vector graphics data, including but not limited to the area, shape, window position and door opening direction of the house floor plan. For example, when processing the living room floor plan uploaded by the user, the system will accurately measure the total area of ​​the living room, identify the specific location and number of windows, and the opening direction of the door, which is crucial for the subsequent furniture layout design. Through these extracted spatial features, the system can better understand the basic structure and characteristics of the space, laying the foundation for the reasonable placement of furniture. Then, the system will classify and process the spatial function arrangement parameters to obtain functional area division information. The spatial function arrangement parameters reflect the user's specific needs for different functional areas, such as the area requirements and relative position requirements of each room. For example, the user may want the living room to be not only spacious enough to accommodate a large gathering, but also to have a reading corner near the window. According to these user needs, the system will divide the living room into different functional areas such as entertainment area, rest area and reading area, and clarify the approximate area and ideal location of each area. This classification process ensures that the user's functional needs can be fully considered during the subsequent layout optimization, avoiding the inconvenience caused by unreasonable space division. Based on the above-mentioned extracted spatial features and functional area division information, the system will construct a spatial layout relationship graph. The spatial layout relationship graph is a graph structure data, in which each node represents a functional area, and the edge represents the connection relationship and relative position between different functional areas. For example, in the design of the living room, the system will create multiple nodes in the graph, corresponding to the entertainment area, rest area and reading area, and then determine the connection mode and distance between these nodes according to the user's needs and the actual conditions of the space. By constructing the spatial layout relationship graph, the system can more intuitively express the logical relationship between the various functional areas, providing a clear framework for the subsequent spatial layout optimization. Subsequently, the system propagates and fuses the spatial layout relationship graph based on the multi-scale graph neural network to obtain a fused feature graph. The multi-scale graph neural network is a powerful tool for processing graph-structured data. It can capture local and global relationships between nodes at different scales, thereby generating richer and more accurate feature representations.In the example of living room design, the system uses multi-scale graph neural networks to capture the specific features of each functional area at the micro level, such as the size and shape of furniture, and understand the interaction between different areas at the macro level, such as the transition space between the entertainment area and the rest area. Through feature propagation and fusion, the system is able to generate a fused feature map, which not only contains the basic information of each functional area, but also reflects the complex relationship between them. Next, the system will input the fused feature map into the preset graph neural network to conduct furniture layout exploration and space layout optimization to generate candidate furniture layout plans. The task of this stage is to explore multiple possible furniture layout plans through simulation and calculation based on the fused feature map. The graph neural network will intelligently select suitable furniture and determine its best placement in the space based on the characteristics of each functional area and the needs of users. For example, in the design of a living room, the system may try to place the sofa near the window to make full use of natural light, and place a TV cabinet opposite the sofa to form an ideal viewing area. Through continuous iteration and optimization, the system is able to generate multiple candidate furniture layout plans for subsequent evaluation and selection. Finally, the system will evaluate and screen the rationality of the candidate furniture layout plans to determine the initial furniture layout plan. The rationality evaluation is mainly based on two aspects: hard design rules and user experience. Hard design rules include the minimum spacing between furniture, barrier-free areas in front of doors and windows, etc., to ensure the feasibility of the design plan; user experience focuses on the aesthetics, comfort and practicality of the plan to ensure that the design plan can meet the actual needs of users. For example, when evaluating candidate plans for living room design, the system will check whether each plan meets the basic design specifications, such as whether the distance between the sofa and the TV is appropriate, and will also consider the personalized needs of users, such as whether the reading corner is quiet and comfortable. Through comprehensive evaluation, the system will eventually select one or several optimal plans as the initial furniture layout plan for users to further review and adjust. In summary, by inputting vector spatial relationship feature data and spatial function arrangement parameters into the preset graph neural network for spatial layout optimization, the entire design process can not only efficiently generate furniture layout plans that meet user needs, but also ensure the rationality and aesthetics of the plan. This process fully demonstrates the intelligent and technical advantages of personalized interior design methods, bringing users a more convenient and high-quality design experience.

[0038] In a specific embodiment, the initial furniture layout scheme is screened by using preset hard design rules to obtain a screened furniture layout scheme, including: Performing spatial relationship analysis on the initial furniture layout plan using a spatial topology analysis algorithm to obtain a spatial topology relationship diagram between the furniture; wherein the spatial topology relationship diagram includes relative positions, distances, and connectivity between the furniture; Based on the ergonomic principles and the spatial topological relationship diagram, an ergonomic evaluation is performed on the initial furniture layout plan to obtain an ergonomic scoring matrix; By using a multi-objective optimization algorithm, the initial furniture layout scheme is optimized and calculated based on the ergonomic scoring matrix to obtain an ergonomic layout scheme set; wherein the ergonomic layout scheme set is a candidate layout scheme set that meets ergonomic requirements; Dividing the ergonomic layout solution set into functional areas to obtain a functional area semantic map; By using preset hard design rules, the functional area semantic graph is subjected to rule constraint check to obtain a set of ergonomic layout solutions after constraint check; The ergonomic layout scheme set after the constraint check is comprehensively evaluated and screened by a multi-criteria decision analysis method to obtain a screened furniture layout scheme.

[0039] Specifically, in the personalized interior design process, screening the initial furniture layout plan through preset design hard rules is an important link to ensure that the design plan is both ergonomic and meets the actual use needs. This process involves multiple steps, from the analysis of spatial topological relationships to the comprehensive evaluation and screening of the final plan, and each step is aimed at improving the rationality of the design plan and user satisfaction. First, the system uses the spatial topological analysis algorithm to analyze the spatial relationship of the initial furniture layout plan and generate a spatial topological relationship diagram between the furniture. The spatial topological relationship diagram records in detail the relative position, distance, and connectivity between the furniture, providing basic data for subsequent evaluation and optimization. For example, in the living room design, the system analyzes the distance between the sofa and the TV, the relative position of the coffee table and the sofa, and the path from the door to each piece of furniture. Through these analyses, the system can fully understand the spatial relationship of the furniture layout and provide accurate data support for subsequent evaluation. Next, based on the ergonomic principles, the system uses the generated spatial topological relationship diagram to perform an ergonomic evaluation of the initial furniture layout plan to obtain an ergonomic scoring matrix. The ergonomic scoring matrix reflects the scores of each furniture layout scheme in terms of human comfort, accessibility, and safety. For example, the system will evaluate whether the visual distance between the sofa and the TV is appropriate, whether the height of the coffee table is convenient to use, and whether the passage from the door to each piece of furniture is barrier-free. Through these evaluations, the system can quantify the ergonomic performance of each scheme and provide a basis for subsequent optimization. Based on the generated ergonomic scoring matrix, the system will optimize the initial furniture layout scheme through a multi-objective optimization algorithm to generate an ergonomic layout scheme set. The multi-objective optimization algorithm can find the best balance between multiple optimization goals to ensure that the generated layout scheme not only meets ergonomic requirements, but also takes into account aesthetics and practicality. For example, the system may generate multiple layout schemes, each of which has different performances in terms of visual distance, passage path, furniture spacing, etc., but all meet the basic requirements of ergonomics. These schemes constitute the ergonomic layout scheme set, providing a wealth of choices for subsequent screening. Subsequently, the system will divide the ergonomic layout scheme set into functional areas and generate a functional area semantic map. The functional area semantic map not only marks the specific location of each functional area, but also clarifies the function and purpose of each area. For example, in the design of a living room, the system will divide the living room into an entertainment area, a rest area, and a reading area, and mark the main furniture and functions of each area in the functional area semantic map. Through the functional area division, the system can ensure that each layout plan meets the ergonomic requirements while also meeting the functional needs of the user. Next, the system will perform a rule constraint check on the functional area semantic map through the preset design hard rules, and generate an ergonomic layout plan set after the constraint check.The hard design rules include, but are not limited to, the minimum spacing between furniture, barrier-free areas in front of doors and windows, and safe distances between electrical sockets and switches. These rules are intended to ensure the feasibility and safety of the design scheme. For example, the system will check whether the distance between the sofa and the coffee table in each layout scheme is greater than 60 cm, whether the passage from the door to each piece of furniture is wider than 90 cm, etc. Only the schemes that pass these rule constraint checks can enter the next step of evaluation and screening. Finally, the system will use the multi-criteria decision analysis method to comprehensively evaluate and screen the ergonomic layout scheme set after the constraint check, and finally obtain the screened furniture layout scheme. The multi-criteria decision analysis method can comprehensively consider multiple evaluation indicators, such as ergonomic score, space utilization, aesthetics, etc., to score each scheme. For example, the system will comprehensively score each scheme based on the user's needs and preferences, and finally select the schemes with the highest scores as the screened furniture layout schemes. These schemes not only meet ergonomic requirements, but also maximize the functional needs and aesthetic preferences of users. In summary, the initial furniture layout plan is screened through preset hard design rules. The whole process can not only ensure the rationality and safety of the design plan, but also significantly improve the user's satisfaction and user experience. For example, in the above-mentioned living room design case, the system not only generates multiple layout plans that meet ergonomic requirements through this series of steps, but also provides users with the most optimized design options through comprehensive evaluation and screening. This process fully demonstrates the scientificity and intelligence of personalized interior design methods, and brings users more intimate and efficient design services.

[0040] In a specific embodiment, matching and designing the matching parameters and the screened furniture layout scheme to obtain a matching design scheme includes: Deconstructing the matching parameters and the selected furniture layout scheme respectively, and obtaining a deconstructed parameter set and a layout element set; Performing semantic encoding on the deconstructed parameter set to obtain an encoded semantic vector; Performing spatial relationship encoding on the layout element set to obtain a spatial relationship vector; Based on the attention mechanism network and matching algorithm, the correlation between the encoded semantic vector and the spatial relationship vector is calculated to obtain the matching weight matrix; Reorganize the layout of the screened furniture layout scheme based on the matching weight matrix to obtain a layout reorganization scheme; The layout reorganization scheme is checked and adjusted for style consistency to obtain a matching design scheme, and the matching design scheme is initially visualized in a preset interactive interface.

[0041] Specifically, in the personalized interior design process, matching the matching parameters and the selected furniture layout scheme is one of the key steps to generate the final matching design scheme. This process requires not only in-depth analysis of user preferences and existing layout schemes, but also advanced calculation methods to ensure that the final design scheme not only meets the user's personalized needs, but also has a high degree of style consistency and practicality. First, the system will perform feature deconstruction on the matching parameters and the selected furniture layout scheme respectively, and obtain the deconstructed parameter set and layout element set accordingly. The matching parameters include the user's preferences in color, material, decoration style, etc., while the selected furniture layout scheme contains optimized space layout information. For example, in the living room design, the user's matching parameters may include preference for modern minimalist style, preference for white and gray as the main colors, and tendency to use wooden furniture. The system will deconstruct these parameters into specific color codes, material categories and other information to form a deconstructed parameter set. At the same time, the system will deconstruct each piece of furniture and its location information in the selected furniture layout scheme into a layout element set, such as the position of the sofa, the size of the coffee table, etc. Next, the system will perform semantic encoding on the deconstructed parameter set to obtain the encoded semantic vector. The purpose of semantic encoding is to convert the user's preference parameters into a form that can be processed by the computer for subsequent calculation and matching. For example, the system will encode "modern minimalist style" as a specific vector, "white" and "gray" as color vectors, and "wooden furniture" as a material vector. Through these encodings, the system can convert the user's abstract preferences into specific numerical representations, providing a basis for subsequent matching calculations. At the same time, the system will perform spatial relationship encoding on the layout element set to obtain a spatial relationship vector. The purpose of spatial relationship encoding is to capture the spatial structural information in the furniture layout plan, such as the relative position, distance and connectivity between furniture. For example, the system will encode the distance between the sofa and the TV as a numerical value, and encode the relative position between the coffee table and the sofa as a vector. Through these encodings, the system can convert the spatial features of the layout plan into vector form, providing data support for subsequent correlation calculations. Based on the attention mechanism network and matching algorithm, the system will perform correlation calculations on the encoded semantic vector and spatial relationship vector to obtain a matching weight matrix. The attention mechanism network can dynamically focus on the key features in user preferences and layout plans, thereby improving the accuracy and rationality of matching. For example, the system will calculate the similarity between the semantic vector of the user's preference for modern minimalist style and the spatial relationship vector of furniture such as sofas and coffee tables in the living room, and generate a matching weight matrix. Each element in this matrix represents the degree of match between the user's preference and a piece of furniture or area in the layout plan. The higher the value, the better the match. Based on the generated matching weight matrix, the system will reorganize the layout of the screened furniture layout plan to obtain a layout reorganization plan.The purpose of layout reorganization is to adjust the position and matching of furniture according to the user's preferences to generate a design solution that better meets the user's needs. For example, if the matching weight matrix shows that the user prefers the combination of white sofa and gray carpet, the system will adjust the color of the sofa in the living room to white and the color of the carpet to gray. Through such reorganization, the system can generate a layout solution that meets the user's preferences and has a high degree of consistency. Finally, the system will check and adjust the style consistency of the layout reorganization solution to ensure that the final design solution is highly consistent in terms of color, material, and decorative style. The purpose of the style consistency check is to avoid the situation of style mixing in the design solution and ensure the coordination and aesthetics of the overall design. For example, the system will check whether the color, material, and style of the furniture in the adjusted layout solution are unified. If not, the system will make further adjustments until the best effect is achieved. After completing the style consistency check, the system will initially visualize the final matching design solution on the preset interactive interface so that the user can intuitively see the design effect. If the user has any comments or suggestions on the preliminary design solution, he can also make instant adjustments through the interactive interface. The system will update the design solution in real time according to the user's feedback until the user is satisfied. In summary, by matching the matching parameters and the selected furniture layout scheme, the whole process can not only accurately reflect the personalized needs of users, but also ensure the high consistency and practicality of the design scheme in style and function. For example, in the above-mentioned living room design case, the system not only generates a design scheme that meets the user's modern and simple style preferences through this series of steps, but also ensures the overall coordination and aesthetics of the design scheme through style consistency checks and adjustments. This process fully demonstrates the intelligent and technical advantages of personalized interior design methods, bringing users a more efficient and satisfactory home design experience.

[0042] In a specific embodiment, the target scheme is edited and rendered according to the interface interaction operation through a preset AI editing system to obtain a three-dimensional visualization effect diagram corresponding to the house plan, including: Performing real-time monitoring of the target user's interface interaction operations on the interaction interface to obtain the target user's operation instruction sequence for the matching design solution; wherein the operation instruction sequence includes recording the user's clicks, drags, and zooms; Using a multi-objective optimization algorithm, adjusting the parameters in the target solution based on the operation instruction sequence to obtain the adjusted design parameters and the adjusted design solution corresponding to the adjusted design parameters; wherein the parameters in the target solution include furniture layout parameters, color scheme parameters and material parameters; Performing spatial mapping on the adjustment design scheme through a preset spatial mapping algorithm to obtain a three-dimensional spatial layout model; Performing material and lighting rendering on the three-dimensional space layout model to obtain a three-dimensional space layout model with material texture and lighting effects; Through virtual reality technology, a three-dimensional visualization effect diagram is generated based on a three-dimensional space layout model with material texture and lighting effects.

[0043] Specifically, in the personalized interior design process, editing and rendering the target solution according to the target user's interface interaction operation on the interactive interface through the preset AI editing system is an important step to enable users to participate in the design in real time and view the design effect immediately. This process not only requires real-time monitoring of the user's operation, but also generates a three-dimensional visualization effect map with material texture and lighting effects through multi-objective optimization algorithm and spatial mapping algorithm to ensure the accuracy and aesthetics of the design solution. First, the system will monitor the target user's interface interaction operation on the interactive interface in real time to obtain the target user's operation instruction sequence for matching the design solution. These operation instruction sequences include user actions such as clicking, dragging, and zooming on the interactive interface. For example, when a user clicks on a furniture icon on the interactive interface and drags it to a new position, or adjusts the size of the furniture through zoom gestures, the system will record these operations in real time. These operation instruction sequences not only reflect the user's immediate needs, but also provide a basis for subsequent design adjustments. Next, the system will use the multi-objective optimization algorithm to adjust the parameters in the target solution based on the operation instruction sequence to obtain the adjusted design parameters and the corresponding adjusted design solutions. The parameters in the target solution include furniture layout parameters, color scheme parameters, and material parameters. For example, if the user drags the sofa from one corner of the living room to another, the system will adjust the layout parameters of the sofa according to the user's operation instructions to ensure that the new position is in line with the user's intention and meets the rationality of the spatial layout. At the same time, if the user changes the color of the wall or chooses a different floor material, the system will adjust the color scheme parameters and material parameters accordingly to ensure the overall coordination of the design solution. Through the preset spatial mapping algorithm, the system will perform spatial mapping on the adjusted design solution to obtain a three-dimensional spatial layout model. The function of the spatial mapping algorithm is to convert the two-dimensional plane layout diagram into a three-dimensional spatial model to ensure that the position, size, and shape of each furniture in the three-dimensional space are accurate. For example, in the living room design, the system will map the adjusted sofa position, coffee table size and other information into the three-dimensional space to generate a complete three-dimensional spatial layout model. This model not only includes the specific location of the furniture, but also includes the structural information of the room's walls, doors, windows, etc., which provides a basis for subsequent rendering. Next, the system will perform material and lighting rendering on the three-dimensional spatial layout model to obtain a three-dimensional spatial layout model with material texture and lighting effects. The purpose of material rendering is to add realistic material textures to elements such as furniture and walls, such as the wood grain of wooden furniture and the fabric texture of fabric sofas. Lighting rendering simulates the effects of different light sources in the room, such as natural light entering from windows, light from chandeliers or wall lamps, etc. Through these renderings, the system can generate a highly realistic three-dimensional space layout model, allowing users to intuitively feel the actual effect of the design plan.Finally, the system will generate a three-dimensional visualization effect map based on the three-dimensional space layout model with material texture and lighting effects through virtual reality technology. Virtual reality technology can provide users with an immersive experience. Users can "walk" and "observe" in the virtual environment through VR devices to view every detail of the design plan in all directions. For example, users can walk around in the virtual living room, observe the material and color of the sofa up close, feel the softness of the light, and even simulate the natural lighting effects in different time periods to ensure that the design plan performs as expected in various environments. In summary, through the preset AI editing system, the target plan is edited and rendered according to the interface interaction operation of the target user on the interactive interface. The whole process can not only respond to the user's operation in real time, but also generate highly realistic three-dimensional visualization effect maps, which significantly improves the user's design participation and satisfaction. For example, in the aforementioned living room design case, users can adjust the furniture layout and color scheme in real time through simple clicks and drags, and preview the final effect in a three-dimensional environment through virtual reality technology. This process not only simplifies the design process, but also provides users with a more intuitive and flexible design experience, ensuring that the final design plan can fully meet the user's personalized needs and aesthetic preferences.

[0044] The above describes the personalized interior design method based on the user's floor plan in the embodiment of the present invention. The following describes the personalized interior design system based on the user's floor plan in the embodiment of the present invention. Figure 2 In one embodiment of the present invention, a personalized interior design system based on a user's floor plan includes: The recommendation module 21 is used to analyze the behavior data of the target user through a preset recommendation algorithm to obtain the preference parameters of the target user; wherein the preference parameters include space function arrangement parameters and matching parameters; The acquisition module 22 is used to acquire the house plan uploaded by the target user, and use the plan image vectorization technology to convert the house plan into vectors to obtain vector space relationship feature data; A generation module 23, for inputting the vector space relationship feature data and the space function arrangement parameters into a preset graph neural network for space layout, and generating an initial furniture layout plan; A screening module 24 is used to screen the initial furniture layout scheme according to preset hard design rules to obtain a screened furniture layout scheme; A matching module 25 is used to match and design the matching parameters and the screened furniture layout scheme to obtain a matching design scheme, and to initially visualize the matching design scheme on a preset interactive interface; The editing module 26 is used to detect in real time whether the target user performs interface interaction operations on the interactive interface. If so, the matching design scheme targeted by the interface interaction operation is used as the target scheme, and the target scheme is edited and rendered according to the interface interaction operation through a preset AI editing system to obtain a three-dimensional visualization effect diagram corresponding to the house floor plan.

[0045] In this embodiment, for the specific implementation of each unit in the above system embodiment, please refer to the description in the above method embodiment, which will not be repeated here.

[0046] Reference Figure 3 The present invention also provides a computer device in an embodiment, wherein the internal structure of the computer device can be as follows: Figure 3 As shown. The computer device includes a processor, a memory, a display screen, an input device, a network interface and a database connected through a system bus. Among them, the processor designed by the computer is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program and a database. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The database of the computer device is used to store the corresponding data in this embodiment. The network interface of the computer device is used to communicate with an external terminal through a network connection. When the computer program is executed by the processor, the above method is implemented.

[0047] Those skilled in the art will understand that Figure 3 The structure shown in the figure is merely a block diagram of a portion of the structure related to the solution of the present invention, and does not constitute a limitation on the computer device to which the solution of the present invention is applied.

[0048] An embodiment of the present invention further provides a computer-readable storage medium on which a computer program is stored, and when the computer program is executed by a processor, the above method is implemented. It can be understood that the computer-readable storage medium in this embodiment can be a volatile readable storage medium or a non-volatile readable storage medium.

[0049] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to memory, storage, database or other media provided by the present invention and used in the embodiments may include non-volatile and / or volatile memory. Non-volatile memory may include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM) or flash memory. Volatile memory may include random access memory (RAM) or external cache memory. As an illustration and not limitation, RAM is available in a variety of forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double-speed data rate SDRAM (SSRSDRAM), enhanced SDRAM (ESDRAM), synchronous link (Synchlink) DRAM (SLDRAM), memory bus (Rambus) direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM.

[0050] It should be noted that, in this article, the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, device, article or method including a series of elements includes not only those elements, but also includes other elements not explicitly listed, or also includes elements inherent to such process, device, article or method. In the absence of further restrictions, an element defined by the sentence "includes a ..." does not exclude the presence of other identical elements in the process, device, article or method including the element.

[0051] The above description is only a preferred embodiment of the present invention, and does not limit the patent scope of the present invention. Any equivalent structure or equivalent process transformation made by using the contents of the present invention specification and drawings, or directly or indirectly applied in other related technical fields, are also included in the patent protection scope of the present invention.

Claims

1. A personalized interior design method based on a user's floor plan, characterized in that: The following steps are involved: By using a preset recommendation algorithm, the behavior data of the target user is analyzed to obtain the target user's preference parameters; wherein the preference parameters include space function arrangement parameters and matching parameters; Obtaining a house plan uploaded by a target user, and performing vector conversion on the house plan using a plan image vectorization technology to obtain vector space relationship feature data; Inputting the vector space relationship feature data and the space function arrangement parameters into a preset graph neural network to perform space layout and generate an initial furniture layout plan; Screening the initial furniture layout schemes according to preset design hard rules to obtain a screened furniture layout scheme; Performing matching design on the matching parameters and the screened furniture layout scheme to obtain a matching design scheme, and initially visualizing the matching design scheme on a preset interactive interface; It is detected in real time whether the target user performs interface interaction operations on the interactive interface. If so, the matching design scheme targeted by the interface interaction operation is taken as the target scheme, and the target scheme is edited and rendered according to the interface interaction operation through a preset AI editing system to obtain a three-dimensional visualization effect diagram corresponding to the house floor plan.

2. The personalized interior design method based on the user's floor plan according to claim 1, characterized in that: The behavior data includes furniture browsing records and input text data. The behavior data of the target user is analyzed by a preset recommendation algorithm to obtain the target user's preference parameters, including: By using a preset recommendation algorithm, the furniture product browsing record and input text data of the target user are analyzed to obtain user-product interaction information; wherein the user-product interaction information is the browsing time, click times and collection status of different furniture products by the target user; Performing cluster analysis on the user-product interaction information to obtain a target user's preferred furniture style feature vector; Analyze the co-occurrence relationship of furniture products in the user-product interaction information to obtain a furniture matching pattern vector; Analyzing the furniture product description text in the user-product interaction information, and extracting the spatial function keywords in the furniture product description text to obtain a spatial function keyword vector; Constructing a target user preference matrix based on the preferred furniture style feature vector, the furniture matching pattern vector and the space function keyword vector; Inputting the target user preference matrix into a preset Bayesian inference algorithm to obtain the target user's preference probability distribution; The furniture selection behavior of the target user is predicted based on the preference probability distribution to obtain the preference parameters of the target user; wherein the preference parameters of the target user include space function arrangement parameters and matching parameters.

3. The personalized interior design method based on user floor plan according to claim 1, characterized in that: The method of converting the house plan into a vector using the plan image vectorization technology to obtain vector space relationship feature data includes: Identifying and marking each functional area in the house plan to obtain functional area marking data; Preprocessing the house plan using plan image vectorization technology to obtain a preprocessed image; Extracting contour lines in the preprocessed image to obtain a contour line image, and classifying and attribute-marking the contour lines in the contour line image based on the functional area annotation data to obtain a contour line image with functional area information; Converting the outline line image with the functional area information into a vector graphic to obtain vector graphic data; The vector graphics data is subjected to spatial relationship analysis to obtain vector spatial relationship feature data.

4. The personalized interior design method based on user floor plan according to claim 1, characterized in that: The step of inputting the vector space relationship feature data and the space function arrangement parameters into a preset graph neural network to optimize the space layout and generate an initial furniture layout plan includes: Performing spatial feature extraction on the vector spatial relationship feature data to obtain extracted spatial features; wherein the extracted spatial features include the area, shape, window positions and door opening direction information of the house plan; Classify and process the spatial function arrangement parameters to obtain functional area division information; wherein the functional area division information includes area requirements and relative position requirements of different functional areas; Based on the spatial feature vector and the functional area division information, construct a spatial layout relationship diagram; Based on a multi-scale graph neural network, feature propagation and fusion are performed on the spatial layout relationship graph to obtain a fused feature graph; Inputting the fused feature graph into a preset graph neural network to perform furniture layout exploration and space layout optimization to obtain a candidate furniture layout solution; The candidate furniture layout plans are evaluated and screened for rationality to obtain an initial furniture layout plan.

5. The personalized interior design method based on user floor plan according to claim 1, characterized in that: The initial furniture layout scheme is screened by using preset hard design rules to obtain a screened furniture layout scheme, including: Performing spatial relationship analysis on the initial furniture layout plan using a spatial topology analysis algorithm to obtain a spatial topology relationship diagram between the furniture; wherein the spatial topology relationship diagram includes relative positions, distances, and connectivity between the furniture; Based on the ergonomic principles and the spatial topological relationship diagram, an ergonomic evaluation is performed on the initial furniture layout plan to obtain an ergonomic scoring matrix; By using a multi-objective optimization algorithm, the initial furniture layout scheme is optimized and calculated based on the ergonomic scoring matrix to obtain an ergonomic layout scheme set; wherein the ergonomic layout scheme set is a candidate layout scheme set that meets ergonomic requirements; Dividing the ergonomic layout solution set into functional areas to obtain a functional area semantic map; By using preset hard design rules, the functional area semantic graph is subjected to rule constraint check to obtain a set of ergonomic layout solutions after constraint check; The ergonomic layout scheme set after the constraint check is comprehensively evaluated and screened by a multi-criteria decision analysis method to obtain a screened furniture layout scheme.

6. The personalized interior design method based on user floor plan according to claim 1, characterized in that: The matching design is performed on the matching parameters and the screened furniture layout scheme to obtain a matching design scheme, including: Deconstructing the matching parameters and the selected furniture layout scheme respectively, and obtaining a deconstructed parameter set and a layout element set; Performing semantic encoding on the deconstructed parameter set to obtain an encoded semantic vector; Performing spatial relationship encoding on the layout element set to obtain a spatial relationship vector; Based on the attention mechanism network and matching algorithm, the correlation between the encoded semantic vector and the spatial relationship vector is calculated to obtain the matching weight matrix; Reorganize the layout of the screened furniture layout scheme based on the matching weight matrix to obtain a layout reorganization scheme; The layout reorganization scheme is checked and adjusted for style consistency to obtain a matching design scheme, and the matching design scheme is initially visualized in a preset interactive interface.

7. The personalized interior design method based on user floor plan according to claim 1, characterized in that: Through the preset AI editing system, the target scheme is edited and rendered according to the interface interaction operation to obtain a three-dimensional visualization effect diagram corresponding to the house plan, including: Performing real-time monitoring of the target user's interface interaction operations on the interaction interface to obtain the target user's operation instruction sequence for the matching design solution; wherein the operation instruction sequence includes recording the user's clicks, drags, and zooms; Using a multi-objective optimization algorithm, adjusting the parameters in the target solution based on the operation instruction sequence to obtain the adjusted design parameters and the adjusted design solution corresponding to the adjusted design parameters; wherein the parameters in the target solution include furniture layout parameters, color scheme parameters and material parameters; Performing spatial mapping on the adjustment design scheme through a preset spatial mapping algorithm to obtain a three-dimensional spatial layout model; Performing material and lighting rendering on the three-dimensional space layout model to obtain a three-dimensional space layout model with material texture and lighting effects; Through virtual reality technology, a three-dimensional visualization effect diagram is generated based on a three-dimensional space layout model with material texture and lighting effects.

8. A personalized interior design system based on a user's floor plan, characterized in that: include: The recommendation module is used to analyze the behavior data of the target user through a preset recommendation algorithm to obtain the preference parameters of the target user; wherein the preference parameters include space function arrangement parameters and matching parameters; An acquisition module is used to acquire a house plan uploaded by a target user, and to perform vector conversion on the house plan using a plan image vectorization technology to obtain vector space relationship feature data; Inputting the vector space relationship feature data and the space function arrangement parameters into a preset graph neural network to perform space layout and generate an initial furniture layout plan; A screening module, used to screen the initial furniture layout scheme according to preset hard design rules to obtain a screened furniture layout scheme; A matching module, used for matching and designing the matching parameters and the screened furniture layout scheme to obtain a matching design scheme, and initially visualizing the matching design scheme on a preset interactive interface; The editing module is used to detect in real time whether the target user performs interface interaction operations on the interactive interface. If so, the matching design scheme targeted by the interface interaction operation is taken as the target scheme, and the target scheme is edited and rendered according to the interface interaction operation through a preset AI editing system to obtain a three-dimensional visualization effect diagram corresponding to the house floor plan.

9. A computer device comprising a memory and a processor, wherein a computer program is stored in the memory, wherein: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 7 are implemented.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 7 are implemented.

Images