Soft furnishing arrangement data generation method, system and electronic device

Abstract

This application relates to the technical fields of artificial intelligence and interior design, and discloses a method, system, and electronic device for generating soft furnishing layout data. The method includes: acquiring original floor plan data of a target space from multiple data sources; performing structured processing on the original floor plan data to obtain original floor plan structured data conforming to a preset structured format; based on the original floor plan structured data, generating a preliminary soft furnishing layout plan using an AI large-scale model with soft furnishing constraint data in a rule knowledge base as constraints; performing geometric and rule consistency verification on the preliminary soft furnishing layout plan based on the soft furnishing constraint data in the rule knowledge base to obtain a target soft furnishing layout plan that meets the consistency verification results; and based on the target soft furnishing layout plan, retrieving soft furnishing layout data from a case tag library whose case tag matching degree meets preset conditions, and visually outputting the soft furnishing layout data. This method can improve the rationality of soft furnishing layout in the soft furnishing layout plan.

Description

Technical Field

[0001] This application relates to the fields of artificial intelligence, interior design and other technologies, and in particular to a method, system and electronic device for generating soft furnishing layout data. Background Technology

[0002] In traditional residential and commercial spaces, interior design is typically handled by designers using 2D CAD (Computer-Aided Design) or 3D modeling software to arrange furniture, lighting, and accessories piece by piece according to project standards and personal experience. This approach not only requires designers to perform a large amount of repetitive work for each apartment type, resulting in low efficiency, but the final arrangement is also often heavily influenced by personal experience and aesthetic preferences, leading to insufficient standardization. Furthermore, this method struggles to generate interior design schemes with a unified style and adhering to consistent rules for a large number of apartment types within a short period.

[0003] Therefore, to improve efficiency, some semi-automatic layout methods based on templates or preliminary rules have emerged in the industry. These methods involve pre-setting templates for typical apartment layouts or interior design. When a new apartment layout is imported, similar templates are matched and adapted based on factors such as area and room type. While these methods improve layout efficiency to some extent, the limited number of templates makes it difficult to cover complex and varied apartment combinations. If the layout or project style changes, a large number of new templates need to be created, resulting in insufficient scalability. Summary of the Invention

[0004] The embodiments of this application aim to at least partially address one of the technical problems in the related art. To this end, the embodiments of this application propose a method, system, and electronic device for generating interior design data.

[0005] This application provides a method for generating interior decoration layout data. The method includes: obtaining original floor plan data of a target space from multiple data sources; performing structured processing on the original floor plan data to obtain original floor plan structured data conforming to a preset structured format; generating a preliminary interior decoration layout plan based on the original floor plan structured data using an AI large model with interior decoration constraint data in a rule knowledge base as constraints; performing geometric and rule consistency verification on the preliminary interior decoration layout plan based on the interior decoration constraint data in the rule knowledge base to obtain a target interior decoration layout plan that meets the consistency verification results; and retrieving interior decoration layout data with case tag matching degrees meeting preset conditions from a case tag library based on the target interior decoration layout plan, and visually outputting the interior decoration layout data.

[0006] In some embodiments, based on the original structured apartment layout data, a preliminary soft furnishing arrangement scheme is generated using an AI big data model with soft furnishing constraint data in a rule knowledge base as constraints. This includes: inputting the original structured apartment layout data and soft furnishing constraint data into the AI ​​big data model; generating furniture bounding rectangle parameters, a furniture set, and first furniture layout data through the AI ​​big data model, wherein the furniture bounding rectangle parameters are used to characterize the placement position and orientation of the corresponding furniture; mapping the first furniture layout data to second furniture layout data that conforms to the standard dimensions of the corresponding furniture in the furniture model library based on the furniture bounding rectangle parameters and the furniture model library; and determining the preliminary soft furnishing arrangement scheme based on the furniture set and the second furniture layout data.

[0007] In some embodiments, the original floor plan structured data includes room data, wall data, door and window data, and fixed facility data. Based on a rule knowledge base, the preliminary soft furnishing layout scheme is geometrically and rule-wise consistent to obtain a target soft furnishing layout scheme that meets the consistency test results. This includes: performing geometric calculations on the room data, wall data, door and window data, and fixed facility data according to the furniture's circumscribed rectangle parameters to obtain geometric calculation results, wherein the geometric calculation results include the calculation results of furniture and room boundaries, the collision matrix between furniture, the calculation results of the minimum distance between furniture and doors and windows, and the calculation results of the passage path width; comparing and verifying the calculation results of furniture and room boundaries, the collision matrix between furniture, the calculation results of the minimum distance between doors and windows, and the calculation results of the passage path width with each soft furnishing constraint data in the rule knowledge base to obtain consistency test results, and outputting the calculation results that do not meet the consistency test results as structured error samples; adjusting the preliminary soft furnishing layout scheme according to the structured error samples to generate the target soft furnishing layout scheme.

[0008] In some embodiments, adjusting the preliminary soft furnishing layout scheme based on structured error samples to generate a target soft furnishing layout scheme includes: correcting the furniture bounding rectangle parameters corresponding to the structured error samples based on the structured error samples, and updating the second furniture layout data in the preliminary soft furnishing layout scheme based on the corrected furniture bounding rectangle parameters to generate the target soft furnishing layout scheme.

[0009] In some embodiments, the method further includes: iteratively updating the rule knowledge base based on accumulated structured error samples, wherein the iterative update process includes at least one of identifying rule defects, generating rule adjustment suggestions, adjusting rule conflicts, and adjusting rule redundancy.

[0010] In some embodiments, based on the target interior design scheme, interior design data with high matching degree of case tags are retrieved from the case tag library, and the interior design data is visualized and output. This includes: performing feature analysis on the target interior design scheme to obtain feature analysis results including furniture combination patterns and space types; based on the feature analysis results, retrieving interior design data with case tag matching degree that meets preset conditions from the case tag library, wherein the interior design data is constructed by extracting associated tags; and combining the 3D model resources or rendering templates associated with the retrieved interior design data with the target interior design scheme to generate and output the corresponding interior design visualization results.

[0011] In some embodiments, the process of inputting original floor plan structured data and soft furnishing constraint data into the AI ​​model, and generating furniture bounding rectangle parameters, furniture sets, and first furniture layout data through the AI ​​model, includes: inputting output format constraint instructions into the AI ​​model; the AI ​​model processing the original floor plan structured data and soft furnishing constraint data based on the output format constraint instructions to obtain furniture bounding rectangle parameters, furniture sets, and first furniture layout data that conform to the output format constraint instructions; wherein the output format constraint instructions include a soft furnishing arrangement result array and an arrangement description field; the soft furnishing arrangement result array includes at least one element, and each element includes furniture type and furniture coordinate information. In some embodiments, the rule knowledge base is iteratively updated based on accumulated structured error samples, including: obtaining a rule-based analysis request based on the structured error samples to analyze the reasons for non-compliance with consistency verification results; generating adjustment suggestions for the corresponding soft furnishing constraint data in the rule knowledge base according to the rule-based analysis request; and iteratively updating the rule knowledge base based on the adjustment suggestions, wherein the soft furnishing constraint data in the rule knowledge base includes rule number, applicable space type, triggering condition, geometric constraint condition, priority, and conflict handling strategy.

[0012] This application provides a soft furnishing arrangement system, comprising: an acquisition module for acquiring original floor plan data of a target space from multiple data sources; a processing module for performing structured processing on the original floor plan data to obtain original floor plan structured data conforming to a preset structured format; a generation module for generating a preliminary soft furnishing arrangement plan based on the original floor plan structured data, using an AI large model with soft furnishing constraint data in a rule knowledge base as constraints; a verification module for performing geometric and rule consistency verification on the preliminary soft furnishing arrangement plan based on the soft furnishing constraint data in the rule knowledge base to obtain a target soft furnishing arrangement plan that meets the consistency verification results; and an output module for retrieving soft furnishing arrangement data with case tag matching degrees meeting preset conditions from a case tag library based on the target soft furnishing arrangement plan, and visually outputting the soft furnishing arrangement data.

[0013] Embodiments of this application provide an electronic device, which includes: a memory, and one or more processors communicatively connected to the memory; the memory stores instructions executable by the one or more processors, which are executed by the one or more processors to cause the one or more processors to implement the steps of the method of any of the above embodiments.

[0014] Embodiments of this application provide a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the method of any of the above embodiments.

[0015] Embodiments of this application provide a computer program product, including a computer program that, when executed by a processor, implements the steps of the method according to any of the above embodiments.

[0016] The solution provided in this application obtains original apartment layout information from multi-source data and performs structured processing on the data to generate original apartment layout structured data conforming to a preset format, providing standardized input for subsequent analysis. Based on this, an AI-powered large-scale model generates a preliminary interior decoration layout plan according to the interior decoration constraint data in the rule knowledge base, ensuring that the layout process is subject to unified and manageable rule constraints. Furthermore, the solution provides a geometric and rule consistency verification module to verify the geometric and rule consistency of the preliminary plan, obtaining a target interior decoration layout plan that meets the constraint conditions, effectively improving the accuracy and usability of the layout results. Finally, based on the target interior decoration layout plan, interior decoration layout data that meets the preset conditions for matching is retrieved from the case tag library and visualized, realizing an integrated process of automatic matching and visualization output of geometrically reasonable automatic layout plans with existing renderings and 3D resources, improving the efficiency, standardization, and reusability of interior decoration layout. Attached Figure Description

[0017] Figure 1 A flowchart illustrating a method for generating interior design data according to an embodiment of this application; Figure 2 A flowchart illustrating a method for generating interior design data according to an embodiment of this application; Figure 3 A flowchart illustrating the process of generating a preliminary soft furnishing layout plan based on the original apartment structure data and soft furnishing constraint data is provided for the embodiments of this application. Figure 4 A flowchart illustrating the process of case tag matching and visualization output of target soft furnishing layout schemes provided in the embodiments of this application; Figure 5 A schematic diagram of a soft furnishing arrangement system provided in an embodiment of this application; Figure 6 A block diagram of an electronic device provided in an embodiment of this application. Detailed Implementation

[0018] The embodiments of this application are described in detail below. Examples of these embodiments are shown in the accompanying drawings, wherein the same or similar reference numerals denote the same or similar elements or elements having the same or similar functions throughout. The embodiments described below with reference to the accompanying drawings are exemplary and intended to explain this application, and should not be construed as limiting this application.

[0019] In traditional residential and commercial spaces, interior design is typically handled by designers using 2D CAD (Computer-Aided Design) or 3D modeling software to arrange furniture, lighting, and accessories piece by piece according to project standards and personal experience. This approach not only requires designers to perform a large amount of repetitive work for each apartment type, resulting in low efficiency, but the final arrangement is also often heavily influenced by personal experience and aesthetic preferences, leading to insufficient standardization. Furthermore, this method struggles to generate interior design schemes with a unified style and adhering to consistent rules for a large number of apartment types within a short period.

[0020] Therefore, to improve efficiency, some semi-automatic layout methods based on templates or preliminary rules have emerged in the industry. These methods involve pre-setting templates for typical apartment layouts or interior design. When a new apartment layout is imported, similar templates are matched and adapted based on factors such as area and room type. While these methods improve layout efficiency to some extent, the limited number of templates makes it difficult to cover complex and varied apartment combinations. If the layout or project style changes, a large number of new templates need to be created, resulting in insufficient scalability.

[0021] With the development of general-purpose AI models, exploratory solutions using AI to assist in interior design have emerged in the industry. For example, a general-purpose AI model can be input into a user's database by describing the room's area, function, and the location of doors and windows using natural language, along with interior design preferences and simple rule suggestions. The model then outputs design recommendations. These solutions attempt to reduce the workload of entirely manual design by leveraging AI's semantic understanding capabilities, and thus have some value in design assistance.

[0022] However, since the apartment layout and rule representations in the above schemes are still mainly unstructured text, the large models perform reasoning in the absence of accurate geometric information and systematic rule constraints, and the generated results often have significant deficiencies in terms of geometric rationality and feasibility.

[0023] Some examples typically incorporate general-purpose AI models into interior design decisions, but they still rely primarily on unstructured text for layout representation and rule management. For instance, firstly, information such as room names, areas, and approximate door and window locations is obtained through manual input, simple recognition, or semi-automatic tools. This information is then organized into natural language descriptions, such as: "This apartment includes a bedroom of approximately 12 square meters, with the door located slightly east of the south wall and a window in the middle of the north wall." Secondly, interior design rules are embedded as prompts, adding unstructured text rules to the layout description, such as: "Please place a double bed and two bedside tables in the bedroom. The headboard should be against the wall, and avoid having the bed directly facing doors or windows, while maintaining passageway space." Next, the system combines the above textual descriptions and rule prompts as input, calling a general-purpose AI model to directly generate a list of furniture and its planar location and orientation. In some implementations, the model may attempt to output numerical coordinates or directional descriptions. Finally, the system parses the model output, mapping furniture types and approximate locations to a 2D or 3D interface, generating candidate layout schemes for users to view and adjust.

[0024] This automated layout exploration path based on a general AI model helps reduce the workload of completely manual layout. However, it still relies heavily on textualized floor plan information, textual rules, and a one-time generation model. The floor plan information focuses on natural language summaries and lacks a unified structured geometric expression. Soft furnishing rules are scattered throughout prompts, making it difficult to form a systematic rule knowledge base. Furniture postures are mostly determined by AI directly outputting position and orientation, generally lacking automatic verification of geometric and rule consistency and error-sample-driven rule iteration mechanisms. Therefore, it has significant limitations in terms of structuring level, geometric constraint accuracy, and rule maintainability.

[0025] Specifically, the aforementioned exploration scheme for automated interior design layout based on a general AI model, while introducing AI into design decisions and reducing repetitive work for designers to some extent, still suffers from significant technical shortcomings in terms of engineering feasibility and stability because its key aspects still rely on unstructured text descriptions and lack precise geometric constraints and rule iteration mechanisms.

[0026] First, the apartment layout lacks a precise structured representation. Natural language forms are insufficient to convey the precise coordinates and topological relationships of geometric information such as room boundaries, walls, doors, and windows. Large models cannot accurately identify spatial boundaries, opening locations, and obstacle distribution, which can easily lead to geometric errors such as furniture crossing boundaries, being too close to doors, or occupying fixed facility areas.

[0027] Secondly, the rules cannot be controlled using a unified knowledge base. Interior design rules are scattered throughout the input text as natural language prompts, making centralized management and version control difficult. When rules need adjustment or expansion, the only solution is to modify the prompts, making it difficult to guarantee consistency across different tasks and time batches, and hindering stable reuse of rules across multiple projects. Thirdly, the way furniture postures are expressed is not conducive to stable AI understanding. Directly requiring large models to output the plane position coordinates and rotation angles of furniture causes the model to reason in a continuous angular space. Since large models have limited sensitivity to such geometrically continuous variables, this easily leads to incorrect orientation of beds, sofas, and other furniture, and improper wall alignment. Even with the same apartment layout and rule prompts, the differences in generated results remain significant.

[0028] Furthermore, the lack of systematic automatic verification of geometric and rule consistency means that traditional solutions often rely on manual checks of the layout's rationality through visual interfaces. There's a lack of a unified geometric calculation module to determine whether furniture crosses boundaries, causes collisions, meets passage width standards, or complies with door and window avoidance rules, making it difficult to establish unified compliance judgment standards. Finally, error samples are not systematically utilized. While generating numerous non-compliant layout cases during the batch generation of interior design schemes, these errors are typically not structured into analyzable error samples, nor are they linked to rule optimization mechanisms. This prevents the system from continuously improving its capabilities using data from its own operation.

[0029] Therefore, this application provides a method for generating interior design data. This method can represent the apartment layout in a unified structural format, enabling precise modeling of room boundaries, walls, doors, windows, and fixed facilities in a unified coordinate system using polygons, line segments, and regions. This provides a data foundation for geometric calculations and AI understanding. The method provided in this application constructs a machine-readable rule knowledge base, storing interior design rules in a structured manner, including rule numbers, applicable conditions, and geometric constraint parameters. It also provides corresponding standardized Chinese descriptions for AI, driving the interior design generation and verification process with the rule knowledge base.

[0030] Secondly, a layout representation centered on the bounding rectangle of furniture is introduced. AI generates the bounding rectangle parameters, and the system automatically calculates the specific location and orientation based on the furniture model library. This alleviates the AI's burden of processing continuous angle parameters and improves the geometric rationality and stability of the layout results. The method provided in this application also designs an automatic geometric and rule consistency verification mechanism. Through geometric formulas such as rectangle intersection judgment, polygon inclusion judgment, distance calculation between rectangle and door / window areas, and passageway width calculation, the generated scheme is systematically judged for compliance.

[0031] In addition, a rule iteration mechanism driven by error samples is constructed, which structures the results of rule violations into error samples, uses AI to summarize and analyze the error samples, generates rule adjustment suggestions and updates the rule knowledge base, so as to realize the automatic evolution of rules with usage.

[0032] Finally, the method provided in this application also links the interior design scheme with the case tag library and model library to achieve an integrated process of automatically matching and visually outputting geometrically reasonable automatic layout schemes with existing renderings and 3D resources.

[0033] Figure 1 This is a flowchart illustrating a method for generating interior design data, as provided in an embodiment of this application.

[0034] like Figure 1 As shown, the soft furnishing layout data generation method 100 provided in this application embodiment includes steps S110-S140.

[0035] Step S110: Obtain the original floor plan data of the target space from multiple data sources.

[0036] For example, the system can obtain original floor plan data from various data sources such as floor plan drawing tools, building information models, and 2D CAD files.

[0037] Step S120: Perform structuring processing on the original apartment layout data to obtain original apartment layout structured data that conforms to the preset structuring format.

[0038] For example, to facilitate unified processing, the system converts the raw data into a preset structured format, such as a JSON (JavaScript Object Notation) data structure, and establishes a unified Cartesian coordinate system. Furthermore, the system performs validity checks on the data, including verifying whether boundary polygons are closed, whether line segments are correctly connected, and whether there are any unreasonable intersections. It also normalizes the coordinate units and precision, providing a reliable foundation for subsequent geometric calculations and input into large AI models.

[0039] Step S130: Based on the original apartment layout structured data, a preliminary soft furnishing arrangement plan is generated using an AI big model with soft furnishing constraint data in the rule knowledge base as constraints.

[0040] For example, after completing the structured processing of the apartment layout data, the system can guide the AI ​​model to arrange the soft furnishings of the structured apartment layout under constraints by constructing a regularized AI question-and-answer description based on the original structured apartment layout data, and generate a preliminary soft furnishing arrangement plan.

[0041] Step S140: Based on the soft furnishing constraint data in the rule knowledge base, perform geometric and rule consistency verification on the preliminary soft furnishing layout scheme to obtain the target soft furnishing layout scheme that meets the consistency verification results.

[0042] For example, after the AI ​​model generates a preliminary layout plan, the system calls the geometry and rule verification module to automatically determine the preliminary soft furnishing layout, which can ensure that the generated target soft furnishing layout plan conforms to the soft furnishing constraint data in the rule knowledge base.

[0043] Step S150: Based on the target interior design scheme, retrieve interior design data from the case tag library that meets the preset conditions for case tag matching, and then visualize the interior design data.

[0044] For example, the system can perform feature analysis on existing interior design renderings and 3D model resources, automatically extracting tags such as space type, style type, color scheme, and furniture combination patterns to form a case tag library and a model tag library. After generating an interior design layout scheme that has passed geometric and rule verification, the system selects cases with high tag similarity from the case tag library based on the furniture type, combination relationship, and target style parameters in the scheme, and calls the corresponding 3D model resources or rendering templates to visualize the interior design layout data.

[0045] The solution provided in this application obtains original apartment layout information from multi-source data and performs structured processing on the data to generate original apartment layout structured data conforming to a preset format, providing standardized input for subsequent analysis. Based on this, an AI-powered large-scale model generates a preliminary interior decoration layout plan according to the interior decoration constraint data in the rule knowledge base, ensuring that the layout process is subject to unified and manageable rule constraints. Furthermore, the solution provides a geometric and rule consistency verification module to verify the geometric and rule consistency of the preliminary plan, obtaining a target interior decoration layout plan that meets the constraint conditions, effectively improving the accuracy and usability of the layout results. Finally, based on the target interior decoration layout plan, interior decoration layout data that meets the preset conditions for matching is retrieved from the case tag library and visualized, realizing an integrated process of automatic matching and visualization output of geometrically reasonable automatic layout plans with existing renderings and 3D resources, improving the efficiency, standardization, and reusability of interior decoration layout.

[0046] Figure 2 This is a flowchart illustrating a method for generating interior design data, as provided in an embodiment of this application.

[0047] like Figure 2As shown, after the process begins, the system performs structured transformation and standardization on the original apartment information. It then constructs and loads a rule knowledge base, integrating and initializing the soft furnishing constraints. Subsequently, within this rule-constrained framework, a large-scale model generates a soft furnishing scheme. This process allows the AI ​​model to output a preliminary soft furnishing layout based on the structured apartment and rule knowledge base. Next, an automatic geometric and rule consistency verification phase is initiated, checking and correcting the scheme to ensure it meets spatial geometric and rule requirements. Following this, iterative rule updates based on error samples are performed, utilizing the error samples discovered during verification to optimize and evolve the rule knowledge base. Finally, case tag matching and soft furnishing scheme visualization are output. By matching with historical case tags, compliant soft furnishing schemes are output and visualized.

[0048] Figure 3 This application provides a flowchart illustrating the process of generating a preliminary interior design scheme based on the original floor plan structured data and interior design constraint data.

[0049] like Figure 3 As shown, R&D designers can inject interior design constraint rules into the system through the rule base table, and simultaneously draw different apartment layouts using a layout tool. This tool can output the generated layout JSON to the AI ​​model and also allows programmers to modify attribute definitions to customize the layout structure. The AI ​​model receives input from the layout JSON and the rule base table, and combines the layout structure with the interior design rules to generate a preliminary interior design arrangement plan. Furthermore, R&D designers can iterate and optimize the rules based on the output preliminary interior design plan, thus forming a closed-loop optimization mechanism between the rule base and the AI ​​model, enabling continuous optimization of the interior design plan under rule constraints and apartment layout adaptation.

[0050] Now combined Figures 1-3 This application provides a detailed description of the proposed solution.

[0051] In some embodiments of this application, based on the original floor plan structured data, a preliminary soft furnishing arrangement scheme is generated using an AI large model with soft furnishing constraint data in a rule knowledge base as constraints. This includes: inputting the original floor plan structured data and soft furnishing constraint data into the AI ​​large model; generating furniture bounding rectangle parameters, a furniture set, and first furniture layout data through the AI ​​large model, wherein the furniture bounding rectangle parameters are used to characterize the placement position and orientation of the corresponding furniture; mapping the first furniture layout data to second furniture layout data that conforms to the standard dimensions of the corresponding furniture in the furniture model library based on the furniture bounding rectangle parameters and the furniture model library; and determining the preliminary soft furnishing arrangement scheme based on the furniture set and the second furniture layout data.

[0052] In other embodiments, the original floor plan structured data and soft furnishing constraint data are input into the AI ​​big model, and the furniture bounding rectangle parameters, furniture set, and first furniture layout data are generated by the AI ​​big model. This includes: inputting output format constraint instructions into the AI ​​big model; the AI ​​big model processes the original floor plan structured data and soft furnishing constraint data based on the output format constraint instructions to obtain furniture bounding rectangle parameters, furniture set, and first furniture layout data that conform to the output format constraint instructions. The output format constraint instructions include a soft furnishing arrangement result array and an arrangement description field. The soft furnishing arrangement result array includes at least one element, and each element includes furniture type and furniture coordinate information.

[0053] For example, the system constructs a unified input structure for the AI ​​large model, including the original floor plan structured data, the rule description section in the soft furnishing constraint data, and the output format constraint instructions. The original floor plan structured data is input data in which the structured data such as room boundaries, walls, doors and windows are compressed into a clearly formatted text or JSON substructure. For example, bedroom 1: boundary vertices [(x1,y1), (x2,y2), ...], door area D1=[xd1_min, xd1_max]×[yd1_min, yd1_max], window area W1=[xw1_min, xw1_max]×[yw1_min, yw1_max].

[0054] The rule description section in the soft furnishing constraint data automatically extracts the Chinese description of the corresponding rule from the rule knowledge base and lists them by number. For example, rule R1: The distance from any point on the outer rectangle of the bed to any point in the door area should not be less than D_door_bed_min; rule R2: The headboard boundary should be adjacent to a certain inner wall segment, and the average distance should be less than ε_wall_adhesive; rule R3: The effective width of the main passage area should not be less than W_min.

[0055] The output format constraint instructions include an array of interior design layout results and a layout description field. The interior design layout results array includes at least one element, and each element includes furniture type and furniture coordinate information. Specifically, the output format constraint instructions explicitly require the AI ​​large model to provide layout results in a structured format. For example, please output only JSON format results, where furnitures is an array, and each element contains type, bounding box, and necessary description fields. Here, type is the furniture type; bounding box is the coordinates of the bounding rectangle, including xmin, xmax, ymin, and ymax.

[0056] To improve the AI's ability to execute rules, the system adopts a rule-based question-and-answer structure when calling the AI. An example structure is as follows: A rule understanding class example, used to verify the model's understanding of rules: "Given the bedroom boundary, door area D, and bed circumscribed rectangle B, determine whether the bed violates the following rule: 'The distance from any point on the bed's circumscribed rectangle to any point on the door area should not be less than D_door_bed_min', and provide the judgment logic." A layout generation class example, used for actual layout: "Under the following bedroom boundary and door / window area conditions, please arrange a double bed and two bedside tables in the bedroom according to rules R1, R2, and R3. Output the type of each piece of furniture (e.g., bed_double, bedside_table) and the coordinates of its circumscribed rectangle, requiring compliance with the rules." During runtime, the system automatically fills the specific boundary coordinates, door and window areas, and rule thresholds into the above question and answer template to form a standardized call request. There is no need for manual input, thus ensuring the automation and consistency of the call process.

[0057] The AI ​​model, based on structured apartment layout data and soft furnishing constraints, generates corresponding bounding rectangle parameters for each piece of furniture. The bounding rectangle Ri can be used to accurately represent the placement and orientation of the corresponding furniture. This application uses the bounding rectangle as the core of the layout representation. For example, the bounding rectangle of furniture i can be represented as follows: Ri = [xi_min, xi_max] × [yi_min, yi_max] The AI ​​model simultaneously generates a furniture set {Ri}, which contains type information, size category, and bounding rectangle parameters of all furniture in the target space. The first furniture layout data is, for example, the raw layout information directly output by the AI ​​model. Then, based on the bounding rectangle parameters of the furniture generated by the AI ​​model and a pre-built furniture model library, the system maps the first furniture layout data to second furniture layout data that conforms to the standard dimensions of the corresponding furniture in the furniture model library.

[0058] Then, the system integrates the furniture set generated by the AI ​​model and the mapped second furniture layout data to form a preliminary soft furnishing arrangement plan.

[0059] This application uses a unified coordinate system to structurally represent information such as room boundaries, walls, doors and windows, and constructs a machine-readable rule knowledge base. It stores interior decoration rules in the form of rule numbers, applicable conditions, and geometric constraint parameters, enabling collaborative work between floor plan data, rule knowledge base, AI generation, and geometric verification. This solves the problem of reusability and maintenance of rules based on traditional text prompts.

[0060] In the embodiments provided in this application, the system can generate a preliminary soft furnishing layout plan based on the original structured apartment layout data and soft furnishing constraint data through an AI large-scale model. The generated preliminary soft furnishing layout plan can not only meet the constraints in the rule knowledge base, but also has high practicality and feasibility. In addition, this application also designs a rule-based question-and-answer structure adapted to the AI ​​large-scale model. By embedding geometric constraints and rule entries into the calling process in the form of standardized Chinese descriptions and example questions and answers, and automatically filling in specific parameters through the program, standardized input is formed, which greatly improves the AI's ability to understand and execute rules and reduces the reliance on manual intervention and subjective operation during operation.

[0061] In some embodiments of this application, the original floor plan structured data includes room data, wall data, door and window data, and fixed facility data; based on a rule knowledge base, the preliminary soft furnishing layout scheme is subjected to geometric and rule consistency verification to obtain a target soft furnishing layout scheme that meets the consistency verification results, including: Based on the bounding rectangle parameters of the furniture, geometric calculations are performed on the room data, wall data, door and window data, and fixed facility data to obtain the geometric calculation results. These results include the calculation results of the furniture and room boundaries, the collision matrix between furniture, the calculation results of the minimum distance between furniture and doors and windows, and the calculation results of the passage path width.

[0062] For example, room data may include a unique identifier for each room, its function type, and room boundaries. The unique identifier may be used to distinguish different rooms, the function type may indicate the room's usage (e.g., bedroom, living room), and the room boundaries may be represented as a sequence of polygon vertices {(x_k, y_k)}. Wall data may be represented by the starting coordinates (x_s, y_s) and ending coordinates (x_e, y_e) for each wall. Door and window data may include the center point coordinates, width, height, and opening direction vector for each door and window, and may be represented as a rectangular region D_j. Fixed facility data may include the geometric boundaries of columns, manholes, etc., and may be represented as rectangles or polygons.

[0063] After the AI ​​model generates a preliminary interior design plan, the system can call the geometry and rule verification module to automatically evaluate the plan.

[0064] For example, based on the parameters of the external rectangle of the furniture and the room data, the calculation result of the furniture and the room boundary is obtained. Specifically, the room boundary is represented as a polygon P, which is composed of a vertex sequence {(xk, yk)}. The external rectangle of furniture i is Ri. The system uses an algorithm for determining whether a point is inside a polygon, such as the ray method, to judge the four vertices of Ri. If all four vertices are inside the polygon P and there is no unreasonable crossing between the rectangle boundary and the polygon boundary, it is determined that the furniture does not cross the boundary; if there is a vertex outside the polygon P, it is determined that the furniture crosses the boundary, and the error type and relevant coordinates are recorded.

[0065] For example, based on the parameters of the external rectangles of the furniture, the collision matrix between the furniture is calculated. Specifically, for any two pieces of furniture i and j, their external rectangles are: Ri = [xi_min, xi_max] × [yi_min, yi_max], Rj = [xj_min, xj_max] × [yj_min, yj_max]. If all of the following conditions are met simultaneously: xi_max ≤ xj_min or xj_max ≤ xi_min; or yi_max ≤ yj_min or yj_max ≤ yi_min; then the two rectangles do not intersect on the plane. If none of the above non-intersection conditions are met, it is determined that there is an overlap or collision between the two pieces of furniture. The system detects all pairs of furniture and records the collision relationships for judging whether the constraints such as no intersection between the furniture are violated.

[0066] For example, based on the parameters of the external rectangle of the furniture and the door and window data, the calculation result of the minimum distance between the furniture and the door and window is obtained. Specifically, let the door and window area Dm be represented as a rectangle [xd_min, xd_max] × [yd_min, yd_max], and the external rectangle of the furniture be Ri = [xi_min, xi_max] × [yi_min, yi_max]. The system calculates the minimum distance d_min(Ri, Dm) between the two in the following way. If Ri and Dm intersect on the plane, then d_min = 0; otherwise, define: dx = max(0, max(xd_min xi_max, xi_min xd_max)), dy = max(0, max(yd_min yi_max, yi_min yd_max)), then the minimum distance d_min = √(dx² + dy²). The system compares d_min with the rule threshold D_door_window_min. If d_min < D_door_window_min, it is determined that the door and window avoidance rule is violated, and the rule number and relevant parameters are recorded.

[0067] For example, based on the parameters of the external rectangle of furniture, wall data, and fixture data, the calculation result of the passage path width is obtained. Specifically, the system can pre-define or automatically deduce the passage center line L, for example, a broken line connecting the door of a room and the main functional area of the room. At each point s on the center line, calculate the shortest distance from this point to the external rectangle of furniture or the wall in the normal direction, and obtain the distances d_left(s) and d_right(s) on both sides. The effective passage width can be expressed as: W(s) = d_left(s) + d_right(s).

[0068] The system calculates W(s) at the discrete sampling points on the center line. If there exists a point s such that W(s) < W_min, it is determined that the passage does not meet the rules, and an error sample violating the passage width rule is formed.

[0069] According to the above content, this application uses the external rectangle of furniture as the core layout expression. The boundary parameters of the external rectangle of furniture are generated by the AI large model, and then the system automatically solves the specific position and orientation in combination with the furniture model library, reducing the processing complexity of the continuous rotation angle by the AI and improving the stability of the layout plan in terms of the correctness of the orientation, the relationship with the wall, and geometric rationality.

[0070] Compare and verify the calculation results of the furniture and room boundary settlement, the furniture - to - furniture collision matrix, the minimum distance between doors and windows calculation result, and the passage path width calculation result with each piece of soft - decoration constraint data in the rule knowledge base to obtain the consistency verification result, and output the calculation results that do not meet the consistency verification result as structured error samples.

[0071] For example, compare and verify the calculation results of the furniture and room boundary settlement, the furniture - to - furniture collision matrix, the minimum distance between doors and windows calculation result, and the passage path width calculation result with each piece of soft - decoration constraint data in the rule knowledge base. For example, for the soft - decoration constraint data that the minimum distance between the bed and the door should be greater than or equal to D_door - bed_min, the system extracts the minimum distance d_min between the bed and the door from the calculation result of the minimum distance between furniture and doors and windows, and judges whether d_min ≥ D_door - bed_min holds. For the soft - decoration constraint data that there should be no collision between furniture, the system checks the furniture - to - furniture collision matrix to determine whether there is a collision; for the soft - decoration constraint data that the passage path width should be not less than W_min, the system extracts the minimum passage width W_min_actual from the calculation result of the passage path width and judges whether W_min_actual ≥ W_min holds. For the soft - decoration constraint data that furniture should not exceed the room boundary, the system checks whether there is an over - boundary situation from the calculation result of the furniture and room boundary.

[0072] The system aggregates the verification results of each soft furnishing constraint in the rule knowledge base to form a consistency check result. If all soft furnishing constraints are satisfied, the preliminary soft furnishing layout plan is deemed to have passed the consistency check; if there are any unsatisfied soft furnishing constraints, the plan is deemed to have failed the consistency check.

[0073] For calculation results that do not meet the consistency check, the system outputs them as structured error samples. Each error sample includes, for example, the apartment type ID, room ID, violation rule number, the bounding rectangle parameters of the relevant furniture, the specific geometric calculation results, and a data fragment from the original large model output.

[0074] In the embodiments provided in this application, geometric formulas such as point within polygon determination, rectangle intersection determination, minimum distance calculation between rectangle and door / window area, and passage path width calculation are systematically integrated and bound to threshold parameters in the rule knowledge base to form a unified automatic determination system for multiple rules such as furniture crossing boundaries, furniture collision, door / window avoidance, and passage width, thereby improving the accuracy and consistency of compliance verification of soft furnishing schemes.

[0075] In another embodiment, the preliminary soft furnishing layout plan is adjusted based on the structured error sample to generate the target soft furnishing layout plan, including: correcting the furniture bounding rectangle parameters corresponding to the structured error sample based on the structured error sample, and updating the second furniture layout data in the preliminary soft furnishing layout plan based on the corrected furniture bounding rectangle parameters to generate the target soft furnishing layout plan.

[0076] For example, after running for a period of time, the system will accumulate a large number of error samples. After obtaining the structured error samples, the system parses each structured error sample to identify error types not covered by the rules, error concentration ranges caused by rule thresholds being set too wide or too narrow, and situations where there are conflicts or redundancies between rules.

[0077] Based on the analysis results of structured error samples, the system corrects the corresponding furniture bounding rectangle parameters and updates the second furniture layout data in the preliminary soft furnishing arrangement plan based on the corrected furniture bounding rectangle parameters. The parameter correction process follows geometric constraints and design principles, ensuring that the target soft furnishing arrangement plan generated based on the updated second furniture layout data not only meets the rule requirements but also maintains the rationality and aesthetics of the layout.

[0078] In some embodiments, the method further includes: iteratively updating the rule knowledge base based on accumulated structured error samples, wherein the iterative update process includes at least one of identifying rule defects, generating rule adjustment suggestions, adjusting rule conflicts, and adjusting rule redundancy.

[0079] In some embodiments, the rule knowledge base is iteratively updated based on accumulated structured error samples, including: obtaining a rule-based analysis request based on the structured error samples to analyze the reasons for non-compliance with consistency verification results; generating adjustment suggestions for the corresponding soft furnishing constraint data in the rule knowledge base according to the rule-based analysis request; and iteratively updating the rule knowledge base based on the adjustment suggestions, wherein the soft furnishing constraint data in the rule knowledge base includes rule number, applicable space type, triggering condition, geometric constraint condition, priority, and conflict handling strategy.

[0080] For example, the system can use error samples and existing rule descriptions as input to call an AI large model to generate rule adjustment suggestions. These suggestions might include reducing or increasing a certain distance threshold D_door_min; adding supplementary rules for specific door and window arrangements; or adjusting the priority of certain rules. The rule iteration and update module maps these natural language suggestions into structured rule table entries and automatically updates the rule knowledge base when preset conditions are met, such as the number of error samples of a certain type exceeding a threshold. In this way, the system can automatically correct rule defects during long-term operation, achieving self-evolution of its interior design capabilities.

[0081] In one embodiment, such as Figure 3 As shown, the decoration results and error samples output by the geometry and rule verification module can be centrally evaluated by R&D designers to identify missing or unreasonable rule entries in the rule base table. The optimized rules are then written back to the rule knowledge base through the rule iteration and update module, thus forming a closed-loop process from floor plan tools, rule base table, AI model to decoration results and then to rule optimization.

[0082] In one embodiment, the system can automatically construct rule-based error analysis question-and-answer requests based on error samples. For example, in the following bedroom layout scheme, the minimum distance d_min = 150mm between furniture A and the door area D has been calculated, which is less than the rule requirement D_door_bed_min = 300mm. Please analyze the possible reasons for violating rule R1 and provide more precise rule description suggestions. In this way, the AI ​​large model can output a language description of rule modifications for use in the next rule iteration update.

[0083] As can be seen, this application proposes a closed-loop mechanism from geometric and rule verification to error sample structuring and then to rule analysis and updating. It transforms non-compliant layout schemes into error samples containing rule numbers and geometric parameters. With the assistance of AI, it automatically summarizes rule defects, generates rule adjustment suggestions, and writes them back to the rule knowledge base in a structured manner. This enables the system to continuously optimize the rule system using its own operating data and achieve self-evolution of soft furnishing layout capabilities.

[0084] Figure 4 This is a flowchart illustrating the process of case tag matching and visual output of target interior design schemes provided in the embodiments of this application.

[0085] like Figure 4 As shown, R&D designers and users can interact with each other through the AI ​​module. R&D designers, for example, are responsible for building and maintaining the case library. The case library, as the core data storage node, transmits data to the AI ​​module and simultaneously receives mapping query data from the case tag library. The AI ​​module performs tag extraction processing on the case library data and then sends the results back to the case tag library. The case tag library not only stores tag data but also supports the case library through mapping queries and interacts with the AI ​​module through feature matching. Users can input their requirements into the AI ​​module using natural language, and the AI ​​module interacts with the case tag library through tag matching.

[0086] In some embodiments of this application, based on the target soft furnishing layout scheme, soft furnishing layout data with high case tag matching degree is retrieved from the case tag library, and the soft furnishing layout data is visualized and output. This includes: performing feature analysis on the target soft furnishing layout scheme to obtain feature analysis results including furniture combination patterns and space types; based on the feature analysis results, retrieving soft furnishing layout data with case tag matching degree meeting preset conditions from the case tag library, wherein the soft furnishing layout data is constructed by extracting associated tags; and combining the three-dimensional model resources or rendering templates associated with the retrieved soft furnishing layout data with the target soft furnishing layout scheme to generate and output the corresponding soft furnishing layout visualization results.

[0087] For example, the system performs feature analysis on existing interior design renderings and 3D model resources to obtain feature analysis results. These results include, for example, space type, style type, color tendency, and furniture combination pattern automatically extracted by the system. Then, based on the feature analysis results, the system retrieves tags such as space type, style type, color tendency, and furniture combination pattern from the case tag library to form a case tag library and a model tag library.

[0088] After generating a soft furnishing layout plan that has passed geometric and rule verification, the system selects cases with high tag similarity from the case tag library based on the furniture type, combination relationship and target style parameters in the plan, and calls the corresponding 3D model resources or rendering templates to achieve output methods such as quick preview and high-precision rendering. The quick preview can be used for design assistance and plan comparison; the high-precision rendering can be used for marketing presentation or formal report.

[0089] Through this embodiment, this application achieves a complete closed loop from inputting floor plan data and rules, to AI-generated solutions, geometric and rule verification, rule iterative updates, and finally, rendering output. Specifically, this application manages existing renderings and 3D model resources by tagging them, mapping furniture combinations in the soft furnishing generation solution to tag spaces. This enables case matching and visualization output based on tag similarity, achieving automatic linkage between geometrically reasonable solutions and high-quality visualization resources without requiring specialized training of image models.

[0090] In one embodiment, such as Figure 4 As shown, the construction and use of the case library and case tag library can adopt the following workflow: R&D designers enter the soft furnishing renderings and 3D renderings of completed projects into the case library. The image data in the case library is sent as image input to the AI ​​tag extraction module. The AI ​​tag extraction module performs style recognition, space type recognition, color matching recognition, and furniture combination pattern recognition for each case image, automatically generates the corresponding tag set, and writes the tag set and case identifier into the case tag library.

[0091] When a user expresses requirements such as target space type, style preference, and functional needs in natural language, the natural language matching AI module receives the user input, maps the natural language to the same tag space as the case tag library, and forms target tags or tag weight vectors. Then, based on the tag matching results, it retrieves several case identifiers with high tag similarity from the case tag library, and obtains corresponding renderings or 3D model resources from the case library through mapping queries, returning them to the user or the visualization output module. Through this embodiment, this application can realize natural language-based case retrieval and the linkage between the case library and the case tag library.

[0092] Figure 5 This is a schematic diagram of a soft furnishing arrangement system provided in an embodiment of this application.

[0093] like Figure 5 As shown, an embodiment of this application provides a soft furnishing arrangement system 500, comprising: The acquisition module 510 is used to obtain the original floor plan data of the target space from multiple data sources.

[0094] The processing module 520 is used to perform structured processing on the original apartment layout data to obtain original apartment layout structured data that conforms to the preset structured format.

[0095] Module 530 generates a preliminary interior design scheme based on the original apartment layout structured data and using an AI model with interior design constraint data from the rule knowledge base as constraints.

[0096] The verification module 540 is used to perform geometric and rule consistency verification on the preliminary soft furnishing layout scheme based on the soft furnishing constraint data in the rule knowledge base, and obtain the target soft furnishing layout scheme that meets the consistency verification results.

[0097] The output module 550 is used to retrieve soft furnishing layout data that meets the preset conditions of case tag matching degree from the case tag library based on the target soft furnishing layout scheme, and to visualize the soft furnishing layout data.

[0098] In some embodiments, the generation module 530 is further configured to: input original floor plan structured data and soft furnishing constraint data into the AI ​​large model, and generate furniture circumscribed rectangle parameters, furniture set, and first furniture layout data through the AI ​​large model, wherein the furniture circumscribed rectangle parameters are used to characterize the placement position and orientation of the corresponding furniture; based on the furniture circumscribed rectangle parameters and the furniture model library, map the first furniture layout data to second furniture layout data that conforms to the standard dimensions of the corresponding furniture in the furniture model library; and determine a preliminary soft furnishing arrangement scheme based on the furniture set and the second furniture layout data.

[0099] In some embodiments, the original floor plan structured data includes room data, wall data, door and window data, and fixed facility data; the verification module 540 is further configured to: perform geometric calculations on the room data, wall data, door and window data, and fixed facility data according to the furniture circumscribed rectangle parameters to obtain geometric calculation results, wherein the geometric calculation results include the calculation results of furniture and room boundaries, the collision matrix between furniture, the calculation results of minimum distance between furniture and doors and windows, and the calculation results of passage path width; compare and verify the calculation results of furniture and room boundaries, the collision matrix between furniture, the calculation results of minimum distance between doors and windows, and the calculation results of passage path width with each soft furnishing constraint data in the rule knowledge base to obtain consistency verification results, and output the calculation results that do not meet the consistency verification results as structured error samples; the generation module 530 is further configured to: adjust the preliminary soft furnishing layout scheme according to the structured error samples to generate the target soft furnishing layout scheme.

[0100] In some embodiments, the verification module 540 is further configured to: correct the furniture bounding rectangle parameters corresponding to the structured error samples based on the structured error samples, and update the second furniture layout data in the preliminary soft furnishing layout scheme based on the corrected furniture bounding rectangle parameters, so as to generate the target soft furnishing layout scheme.

[0101] In some embodiments, the soft furnishing system 500 further includes an update module, which is used to: iteratively update the rule knowledge base based on accumulated structured error samples, wherein the iterative update process includes at least one of identifying rule defects, generating rule adjustment suggestions, adjusting rule conflicts, and adjusting rule redundancy.

[0102] In some embodiments, the output module 550 is further configured to: perform feature analysis on the target soft furnishing arrangement scheme to obtain feature analysis results including furniture combination patterns and space types; based on the feature analysis results, retrieve soft furnishing arrangement data from the case tag library that meets preset conditions for case tag matching, wherein the soft furnishing arrangement data is constructed by extracting associated tags; combine the three-dimensional model resources or rendering templates associated with the retrieved soft furnishing arrangement data with the target soft furnishing arrangement scheme to generate and output the corresponding soft furnishing arrangement visualization results.

[0103] In some embodiments, the generation module 530 is further configured to: input output format constraint instructions to the AI ​​large model, wherein the AI ​​large model processes the original floor plan structured data and soft furnishing constraint data based on the output format constraint instructions to obtain furniture bounding rectangle parameters, furniture sets, and first furniture layout data that conform to the output format constraint instructions, wherein the output format constraint instructions include a soft furnishing layout result array and a layout description field, the soft furnishing layout result array includes at least one element, each element including furniture type and furniture coordinate information. In some embodiments, the update module is further configured to: obtain a rule-based analysis request based on structured error samples to analyze the reasons for non-compliance verification results; generate adjustment suggestions for the corresponding soft furnishing constraint data in the rule knowledge base according to the rule-based analysis request; and perform iterative update processing on the rule knowledge base based on the adjustment suggestions, wherein the soft furnishing constraint data in the rule knowledge base includes rule number, applicable space type, triggering condition, geometric constraint condition, priority, and conflict handling strategy.

[0104] It is understandable that for a detailed description of the soft furnishing system 500, please refer to the description of the methods applied to the soft furnishing system above.

[0105] This application provides a computer-readable storage medium storing a computer program thereon, which, when executed by a processor, implements the steps of the method in any of the above embodiments.

[0106] This application provides a computer program product that includes instructions that, when executed by a processor of a computer device, enable the computer device to perform the steps of the method described in any of the above embodiments.

[0107] Figure 6 A block diagram of an electronic device provided in an embodiment of this application.

[0108] This application provides an electronic device, including a memory and a processor. The memory stores a computer program, and the processor executes the computer program to implement the method in any of the above embodiments.

[0109] like Figure 6As shown, for ease of understanding, an embodiment of this application illustrates a specific electronic device 600.

[0110] Electronic device 600 is intended to represent various forms of digital computers, such as laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. Electronic device 600 may also represent various forms of mobile devices, such as personal digital processors, cellular phones, smartphones, wearable devices, and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely illustrative and are not intended to limit the implementation of the present disclosure described and / or claimed herein.

[0111] like Figure 6 As shown, device 600 includes a computing unit 601, which can perform various appropriate actions and processes based on a computer program stored in read-only memory (ROM) 602 or a computer program loaded from storage unit 608 into random access memory (RAM) 603. RAM 603 may also store various programs and data required for the operation of electronic device 600. The computing unit 601, ROM 602, and RAM 603 are interconnected via bus 604. Input / output (I / O) interface 605 is also connected to bus 604.

[0112] Multiple components in electronic device 600 are connected to I / O interface 605. These components include: input unit 606, such as a keyboard or mouse; output unit 607, such as various types of displays or speakers; storage unit 608, such as a disk or optical disk; and communication unit 609, such as a network interface card (NIC), modem, or wireless transceiver. Communication unit 609 allows electronic device 600 to exchange information / data with other devices through computer networks such as the Internet and / or various telecommunications networks.

[0113] The computing unit 601 can be a variety of general-purpose and / or special-purpose processing components with processing and computing capabilities. Some examples of the computing unit 601 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various special-purpose artificial intelligence (AI) computing chips, various computing units running machine learning model algorithms, a digital signal processor (DSP), and any suitable processor, controller, microcontroller, etc. The computing unit 601 performs the various methods described above. For example, in some embodiments, any one or more of the methods described above can be implemented as a computer software program tangibly contained in a machine-readable medium, such as storage unit 608. In some embodiments, part or all of the computer program can be loaded and / or installed on the electronic device 600 via ROM 602 and / or communication unit 609. When the computer program is loaded into RAM 603 and executed by the computing unit 601, one or more steps of any one or more of the methods described above can be performed. Alternatively, in other embodiments, the computing unit 601 can be configured to perform any one or more of the methods described above by any other suitable means (e.g., by means of firmware).

[0114] It should be noted that the logic and / or steps represented in the flowchart or otherwise described herein, for example, can be considered as a sequenced list of executable instructions for implementing logical functions, and can be specifically implemented in any computer-readable medium for use by, or in conjunction with, an instruction execution system, apparatus, or device (such as a computer-based system, a processor-included system, or other system that can fetch and execute instructions from, an instruction execution system, apparatus, or device). For the purposes of this application, "computer-readable medium" can be any means that can contain, store, communicate, propagate, or transmit programs for use by, or in conjunction with, an instruction execution system, apparatus, or device. More specific examples (a non-exhaustive list) of computer-readable media include: electrical connections (electronic devices) having one or more wires, portable computer disk drives (magnetic devices), random access memory (RAM), read-only memory (ROM), erasable and editable read-only memory (EPROM or flash memory), fiber optic devices, and portable optical disc read-only memory (CDROM). Furthermore, computer-readable media can even be paper or other suitable media on which programs can be printed, because programs can be obtained electronically, for example, by optically scanning the paper or other media, followed by editing, interpreting, or otherwise processing as necessary, and then stored in computer memory.

[0115] It should be understood that various parts of this application can be implemented using hardware, software, firmware, or a combination thereof. In the above embodiments, multiple steps or methods can be implemented using software or firmware stored in memory and executed by a suitable instruction execution system. For example, if implemented in hardware, as in another embodiment, it can be implemented using any one or a combination of the following techniques known in the art: discrete logic circuits having logic gates for implementing logical functions on data signals, application-specific integrated circuits (ASICs) having suitable combinational logic gates, programmable gate arrays (PGAs), field-programmable gate arrays (FPGAs), etc.

[0116] In the description of this application, the references to terms such as "one embodiment," "some embodiments," "example," "specific example," or "some examples," etc., indicate that a specific feature, structure, material, or characteristic described in connection with that embodiment or example is included in at least one embodiment or example of this application. In this application, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples.

[0117] In the description of this application, it should be understood that the terms "center", "longitudinal", "lateral", "length", "width", "thickness", "upper", "lower", "front", "rear", "left", "right", "vertical", "horizontal", "top", "bottom", "inner", "outer", "clockwise", "counterclockwise", "axial", "radial", "circumferential", etc., indicating the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings, are only for the convenience of describing this application and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation, and therefore should not be construed as a limitation of this application.

[0118] Furthermore, the terms "first," "second," etc., used in the embodiments of this application are for descriptive purposes only and should not be construed as indicating or implying relative importance, or implicitly specifying the number of technical features indicated in this embodiment. Therefore, features defined with terms such as "first" and "second" in the embodiments of this application can explicitly or implicitly indicate that the embodiment includes at least one of those features. In the description of this application, the word "multiple" means at least two or more, such as two, three, four, etc., unless otherwise explicitly and specifically defined in the embodiments.

[0119] In this application, unless otherwise explicitly specified or limited in the embodiments, the terms "installation," "connection," "joining," and "fixing" appearing in the embodiments should be interpreted broadly. For example, a connection can be a fixed connection, a detachable connection, or an integral part; it can also be a mechanical connection, an electrical connection, etc. Of course, it can also be a direct connection, or an indirect connection through an intermediate medium, or it can be the internal communication between two components, or the interaction between two components. Those skilled in the art can understand the specific meaning of the above terms in this application based on the specific implementation.

[0120] In this application, unless otherwise expressly specified and limited, "above" or "below" the second feature can mean that the first feature is in direct contact with the second feature, or that the first feature is in indirect contact with the second feature through an intermediate medium. Furthermore, "above," "on top of," and "over" the second feature can mean that the first feature is directly above or diagonally above the second feature, or simply that the first feature is at a higher horizontal level than the second feature. "Below," "below," and "under" the second feature can mean that the first feature is directly below or diagonally below the second feature, or simply that the first feature is at a lower horizontal level than the second feature.

Claims

1. A method for generating soft furnishing layout data, characterized in that, The method includes: Obtain the original floor plan data of the target space from multiple data sources; The original apartment layout data is processed into a structured format to obtain original apartment layout structured data that conforms to a preset structured format. Based on the original structured data of the apartment layout, a preliminary soft furnishing arrangement plan is generated through an AI big model with soft furnishing constraint data in the rule knowledge base as the constraint condition. Based on the soft furnishing constraint data in the rule knowledge base, the preliminary soft furnishing layout scheme is subjected to geometric and rule consistency verification to obtain the target soft furnishing layout scheme that meets the consistency verification results. Based on the target interior design scheme, interior design data that meet the preset conditions for case tag matching are retrieved from the case tag library, and the interior design data is visualized and output.

2. The method according to claim 1, characterized in that, Based on the original structured apartment layout data, an AI model is used with soft furnishing constraints from a rule-based knowledge base as the constraint conditions to generate a preliminary soft furnishing arrangement plan, including: The original floor plan structured data and the soft furnishing constraint data are input into the AI ​​model, and the AI ​​model generates furniture bounding rectangle parameters, furniture set and first furniture layout data. The furniture bounding rectangle parameters are used to characterize the placement position and orientation of the corresponding furniture. Based on the furniture bounding rectangle parameters and the furniture model library, the first furniture layout data is mapped to second furniture layout data that conforms to the standard dimensions of the corresponding furniture in the furniture model library; Based on the furniture set and the second furniture layout data, the preliminary soft furnishing arrangement plan is determined.

3. The method according to claim 2, characterized in that, The original structured apartment layout data includes room data, wall data, door and window data, and fixed facility data; based on the rule knowledge base, the preliminary interior decoration layout plan is subjected to geometric and rule consistency verification to obtain a target interior decoration layout plan that meets the consistency verification results, including: Based on the bounding rectangle parameters of the furniture, geometric calculations are performed on the room data, the wall data, the door and window data, and the fixed facility data to obtain geometric calculation results. The geometric calculation results include the calculation results of the furniture and room boundaries, the collision matrix between furniture, the calculation results of the minimum distance between furniture and doors and windows, and the calculation results of the passage path width. The results of the furniture and room boundary settlement, the collision matrix between furniture, the minimum distance calculation result of doors and windows, and the width calculation result of the passage path are compared and verified with each soft furnishing constraint data in the rule knowledge base to obtain the consistency verification result. The calculation results that do not meet the consistency verification result are output as structured error samples. Based on the structured error samples, the preliminary interior design scheme is adjusted to generate the target interior design scheme.

4. The method according to claim 3, characterized in that, The step of adjusting the preliminary interior design scheme based on the structured error samples to generate the target interior design scheme includes: Based on the structured error sample, the bounding rectangle parameters of the furniture corresponding to the structured error sample are corrected, and the second furniture layout data in the preliminary soft furnishing layout scheme is updated based on the corrected bounding rectangle parameters to generate the target soft furnishing layout scheme.

5. The method according to claim 3, characterized in that, The method further includes: The rule knowledge base is iteratively updated based on the accumulated structured error samples, wherein the iterative update process includes at least one of identifying rule defects, generating rule adjustment suggestions, adjusting rule conflicts, and adjusting rule redundancy.

6. The method according to claim 1, characterized in that, The step of retrieving interior design data with high case tag matching from the case tag library based on the target interior design scheme, and visualizing the interior design data, includes: The target interior design scheme is subjected to feature analysis to obtain feature analysis results including furniture combination patterns and space types; Based on the feature analysis results, soft furnishing layout data that meet the preset conditions for case tag matching is retrieved from the case tag library, wherein the soft furnishing layout data is constructed by extracting related tags; The retrieved interior design data is associated with a 3D model resource or rendering template, which is then combined with the target interior design scheme to generate and output the corresponding interior design visualization result.

7. The method according to claim 2, characterized in that, The process of inputting the original floor plan structured data and the soft furnishing constraint data into the AI ​​model, and generating furniture bounding rectangle parameters, furniture sets, and first furniture layout data through the AI ​​model, includes: The AI ​​model is fed with output format constraint instructions. Based on these instructions, the AI ​​model processes the original floor plan structured data and the interior decoration constraint data to obtain the bounding rectangle parameters of the furniture, the furniture set, and the layout data of the first furniture, all conforming to the output format constraint instructions. The output format constraint instruction includes an array of soft furnishing layout results and an layout description field. The array of soft furnishing layout results includes at least one element, and each element includes furniture type and furniture coordinate information.

8. The method according to claim 5, characterized in that, The iterative update process of the rule knowledge base based on the accumulated structured error samples includes: Based on the structured error samples, a rule-based analysis request is obtained to analyze the reasons why the consistency verification results are not satisfied; Based on the rule-based analysis request, adjustment suggestions are generated for the corresponding soft furnishing constraint data in the rule knowledge base; Based on the aforementioned adjustment suggestions, the rule knowledge base is iteratively updated. The rule knowledge base includes soft furnishing constraint data such as rule number, applicable space type, triggering condition, geometric constraint condition, priority, and conflict handling strategy.

9. A soft furnishing arrangement system, characterized in that, The system includes: The acquisition module is used to obtain the original floor plan data of the target space from multiple data sources; The processing module is used to perform structured processing on the original apartment layout data to obtain original apartment layout structured data that conforms to a preset structured format. The generation module, based on the original apartment structure data, uses an AI big model with soft furnishing constraint data in the rule knowledge base as constraints to generate a preliminary soft furnishing layout plan; The verification module is used to perform geometric and rule consistency verification on the preliminary soft furnishing layout scheme based on the soft furnishing constraint data in the rule knowledge base, so as to obtain the target soft furnishing layout scheme that meets the consistency verification results. The output module is used to retrieve soft furnishing layout data that meets the preset conditions for case tag matching from the case tag library based on the target soft furnishing layout scheme, and to visualize and output the soft furnishing layout data.

10. An electronic device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that, When the processor executes the computer program, it implements the steps of the method according to any one of claims 1-8.

Images