Design data interaction method and system for soft decoration engineering

Abstract

The invention discloses a design data interaction method and system for soft decoration engineering, and relates to the technical field of intelligent interior design and decoration, and the method comprises the steps: giving a design scheme through employing a preliminary design sketch, carrying out the adjustment according to the customer opinion, obtaining adjustment data, carrying out the optimization of the design scheme according to the adjustment data, and obtaining a customer feedback result, by optimizing the model, generating a plurality of optimization schemes and combining with the introduction of market trend data, not only are personalized demands of customers considered, but also the latest trend of the market is considered, so that the optimization schemes are more innovative, the customers can select the scheme which most conforms to own expectation and submit feedback, and a designer further adjusts the design scheme according to the feedback, so that the design efficiency is improved. And finally determining an optimal scheme. According to the intelligent optimization process, the design precision and efficiency are improved, the personalization and foresight of the design scheme are ensured, and the customer satisfaction and the project success rate are greatly improved.

Description

Technical Field

[0001] The present invention relates to the technical field of intelligent interior design and decoration, and in particular to a design data interaction method and system for soft decoration engineering. Background Art

[0002] As consumers' demands for personalization, intelligence and efficiency continue to increase, soft decoration projects have gradually become an important development direction in the field of home decoration. Traditional soft decoration design mainly relies on the designer's experience and the customer's subjective description. However, this traditional method has many limitations, especially in terms of customer demand analysis, design scheme interaction and construction quality monitoring, which makes it difficult to meet the modern consumers' requirements for high precision, real-time and interactivity.

[0003] The existing soft decoration project design scheme mainly relies on two-dimensional drawings or static three-dimensional models. It is difficult for customers to intuitively feel the actual effect of the design scheme, and it is even more difficult to participate in the design process and make adjustments in real time. Even with the introduction of AR technology, due to the lack of intelligent interactive means, customers still need to rely on the help of designers when adjusting the design scheme, and cannot achieve true self-service adjustment. Summary of the invention

[0004] In view of the above existing problems, the present invention is proposed.

[0005] Therefore, the present invention provides a design data interaction method for soft decoration engineering to solve the problem of unsmooth design data interaction in soft decoration engineering.

[0006] In order to solve the above technical problems, the present invention provides the following technical solutions:

[0007] In a first aspect, the present invention provides a design data interaction method for soft decoration engineering, which includes collecting personalized demand data and historical data of customers and generating a customer demand report;

[0008] Generate preliminary design sketches based on customer demand reports;

[0009] Based on the preliminary design sketch, give a design plan and adjust it according to customer opinions to obtain adjustment data;

[0010] Optimize the design plan based on the adjustment data and obtain customer feedback;

[0011] Based on customer feedback, confirm the final design plan and generate project documents;

[0012] Carry out construction according to project documents and get a complete project report.

[0013] As a preferred solution of the design data interaction method for soft decoration engineering described in the present invention, wherein: collecting the customer's personalized demand data and historical data, generating a customer demand report, including the following steps,

[0014] The customer's personalized demand data includes space size data, style preference data, budget limitation data and functional demand data;

[0015] The historical data includes preference record data, purchase behavior data, and communication record data left by customers in past interactions with designers;

[0016] The customer logs in to the design platform and fills in the project information form;

[0017] According to the project information form filled out by the customer, the BERT model is used to parse the information filled out by the customer to obtain the parsing result;

[0018] Based on the analysis results, a customer demand report is compiled.

[0019] As a preferred solution of the design data interaction method for soft decoration engineering of the present invention, wherein: generating a preliminary design sketch according to the customer demand report includes the following steps:

[0020] Select a generative adversarial network model;

[0021] Convert each customer requirement item in the customer requirement report into JSON format and combine them into a multi-dimensional tensor;

[0022] The multi-dimensional tensor is input into the generative adversarial network model to generate a preliminary design sketch.

[0023] As a preferred solution of the design data interaction method for soft decoration engineering described in the present invention, wherein: based on the preliminary design sketch, a design scheme is given and adjusted according to customer opinions to obtain adjustment data, including the following steps:

[0024] Select AR glasses as the AR device and guide customers to install the AR application;

[0025] Load the preliminary design sketch onto the AR glasses;

[0026] Start the AR glasses to detect the customer's current physical space and perform initial positioning and space mapping;

[0027] After the spatial mapping is completed, the captured spatial information is converted into a dense spatial grid, and the feature points in the spatial grid are extracted;

[0028] Based on the feature points, virtual objects are placed in the physical space and rendered;

[0029] Activate the real-time environment perception function of the AR glasses, use the camera sensor to detect the lighting conditions, material color and texture changes of the virtual objects, and obtain the detection results;

[0030] According to the detection results, the lighting conditions, material color and texture changes of virtual objects are dynamically adjusted using the lighting adaptation algorithm and color texture matching technology;

[0031] Customers view the initial effects of dynamically adjusted virtual objects;

[0032] Further adjust virtual objects according to customer needs and record user adjustment data;

[0033] The adjustment data includes data for changing the properties of virtual objects, customer interaction data, and environment perception data.

[0034] As a preferred solution of the design data interaction method for soft decoration engineering described in the present invention, wherein: optimizing the design solution according to the adjustment data to obtain the customer feedback result includes the following steps:

[0035] Extract key adjustment features based on adjustment data;

[0036] Convolutional neural network and recurrent neural network are selected as hybrid optimization models;

[0037] Input the key adjustment features into the hybrid optimization model to generate M optimization solutions;

[0038] Get the latest market trend data from external data sources;

[0039] Said external data sources include industry reports, social media;

[0040] Based on the market trend data, fine-tune the M optimization solutions and obtain the fine-tuning results;

[0041] Based on the fine-tuning results, a rendering engine is used to convert the M optimization solutions into high-quality and interactive two-dimensional images and generate virtual roaming videos.

[0042] Show 2D images and virtual walkthrough videos to customers through AR applications;

[0043] The customer selects the required optimization solution and submits feedback to obtain customer feedback results.

[0044] As a preferred solution of the design data interaction method for soft decoration engineering of the present invention, the final design solution is confirmed and the project file is generated based on the customer feedback result, including the following steps:

[0045] The discriminator is used in combination with key adjustment features to calculate the quality score of the optimization solution required by the customer and perform normalization. The expression is:

[0046]

[0047] in, represents the quality score of the optimization solution required by the customer, j represents the index variable for traversing the key adjustment features, and J represents the total number of key adjustment features. Indicates the performance of the optimization solution required by the customer on the jth feature, represents the weight of the key adjustment feature, A normalization factor representing the quality score of the optimization solution required by the customer;

[0048] Set a threshold for the quality standard and determine whether the solution meets the customer's expectations based on whether the quality score of the optimization solution required by the customer is within the threshold.

[0049] Confirm the final design plan based on the judgment results and generate project documents;

[0050] The project documents include construction drawings, material lists, furniture and decoration lists, and the total budget for the project.

[0051] As a preferred solution of the design data interaction method for soft decoration engineering of the present invention, the following steps are included: construction is carried out according to the project file to obtain a complete project report:

[0052] Provide MagicLeap devices to construction teams;

[0053] Install IoT sensors at the construction site and set the data collection frequency for each sensor;

[0054] The IoT sensors include camera sensors, temperature and humidity sensors, air quality sensors, and vibration sensors;

[0055] Use IoT sensors to collect construction data, evaluate construction quality, identify risk points, and obtain monitoring results;

[0056] Based on the monitoring results and the immediate needs of the client, remote collaboration with the designer was carried out through the MagicLeap device to adjust the design plan again;

[0057] Deliver the adjusted design plan to the construction team for construction;

[0058] After the project construction is completed, a complete project report is generated;

[0059] The complete project report includes project overview, overall design plan, construction process, final effect and maintenance guidelines.

[0060] In a second aspect, the present invention provides a design data interaction system for soft decoration engineering, including a data collection module for collecting personalized demand data and historical data of customers and generating a customer demand report;

[0061] Sketch design module, generates preliminary design sketches based on customer demand reports;

[0062] The scheme adjustment module provides a design scheme based on the preliminary design sketch and adjusts it according to customer opinions;

[0063] Optimization module, which optimizes the adjusted design solution and obtains customer feedback results;

[0064] Solution confirmation module: based on customer feedback, confirms the final design solution and generates project documents;

[0065] Construction module, carry out construction according to project documents and get a complete project report.

[0066] In a third aspect, the present invention provides a computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: when the computer program is executed by the processor, any step of the design data interaction method for soft decoration engineering as described in the first aspect of the present invention is implemented.

[0067] In a fourth aspect, the present invention provides a computer-readable storage medium having a computer program stored thereon, wherein: when the computer program is executed by a processor, it implements any step of the design data interaction method for soft decoration engineering as described in the first aspect of the present invention.

[0068] The beneficial effects of the present invention are as follows: using a preliminary design sketch, a design scheme is given and adjusted according to customer opinions to obtain adjustment data, and the design scheme is optimized according to the adjustment data to obtain customer feedback results. Through the optimization model, multiple optimization schemes are generated to ensure that each scheme is scientific and reasonable. Combined with the introduction of market trend data, not only the personalized needs of customers are considered, but also the latest trends in the market are taken into account, making the optimization scheme more innovative. Customers can choose the scheme that best meets their expectations and submit feedback. Designers further adjust the design scheme based on the feedback and finally confirm the best scheme. This intelligent optimization process not only improves the accuracy and efficiency of the design, but also ensures the personalization and foresight of the design scheme, greatly improving customer satisfaction and project success rate. BRIEF DESCRIPTION OF THE DRAWINGS

[0069] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other accompanying drawings can be obtained based on these accompanying drawings without paying creative work.

[0070] Figure 1 This is a flow chart of the design data interaction method for soft decoration engineering in Example 1.

[0071] Figure 2 This is a decision diagram for determining whether the optimization solution in Example 1 meets customer expectations. DETAILED DESCRIPTION

[0072] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the specific implementation methods of the present invention are described in detail below in conjunction with the accompanying drawings.

[0073] In the following description, many specific details are set forth to facilitate a full understanding of the present invention, but the present invention may also be implemented in other ways different from those described herein, and those skilled in the art may make similar generalizations without violating the connotation of the present invention. Therefore, the present invention is not limited to the specific embodiments disclosed below.

[0074] Secondly, the term "one embodiment" or "embodiment" as used herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The term "in one embodiment" that appears in different places in this specification does not necessarily refer to the same embodiment, nor does it refer to a separate or selective embodiment that is mutually exclusive with other embodiments.

[0075] Example 1, reference Figure 1 and Figure 2 , which is the first embodiment of the present invention, and provides a design data interaction method for soft decoration engineering, comprising the following steps:

[0076] S1. Collect customers’ personalized demand data and historical data, and generate customer demand reports.

[0077] The following steps are included:

[0078] The customer's personalized demand data includes space size data (specific dimensions of the house, such as the length, width, height of the room, the location and size of doors and windows, etc.), style preference data (customer's preferred decoration style, such as modern simplicity, European classical, Nordic style, etc.), budget constraint data (the total budget range of the project, and the budget allocation for different furniture, decorations, and materials) and functional demand data (functional requirements of each room, such as whether the living room needs to be set up with an audio-visual entertainment area, whether the bedroom needs a large-capacity wardrobe, whether the kitchen needs integrated appliances, etc.);

[0079] Historical data includes preference records left by customers in past interactions with designers (styles chosen by customers, furniture brands purchased, favorite colors, etc.), purchase behavior data (understanding customers' consumption habits for different products), and communication record data;

[0080] The customer logs in to the design platform and fills in the project information form;

[0081] In this step, customers can submit detailed personalized requirements by filling out online forms, avoiding the time cost and information omission problems of face-to-face communication in traditional ways.

[0082] Based on the project information form filled out by the customer, the BERT model is used to parse the information filled out by the customer to obtain the analysis results (explicit and implicit requirements);

[0083] The BERT model is a deep learning model based on natural language processing (NIP) that can understand the contextual semantics of text and extract explicit and implicit requirements;

[0084] The BERT model will directly extract explicit requirement information from the form, such as space size, budget constraints, etc.

[0085] In addition to explicit needs, the BERT model also analyzes the customer's description to discover some implicit needs. For example, if a customer mentions in the description that he hopes his home will have a cozy feeling, the BERT model will interpret it as a need for warm colors, soft lighting, comfortable furniture, and other elements.

[0086] According to the analysis results, it is compiled into a customer demand report;

[0087] The BERT model will categorize and summarize all customer needs and provide corresponding explanations;

[0088] Propose some personalized design suggestions based on the explicit and implicit needs of customers. For example, if the customer likes modern minimalist style but also wants a warm feeling at home, we will suggest adding some wooden elements and soft lighting design on the basis of minimalist style. At the same time, we will indicate possible design challenges or risks according to the customer's budget. For example, if the customer has a low budget but requires a high decoration standard, we will remind the customer that some alternatives may be needed.

[0089] Combine interpretation instructions and design recommendations into a detailed client requirements report.

[0090] S2. Generate preliminary design sketch based on customer demand report.

[0091] The following steps are included:

[0092] Select the Generative Adversarial Network model (GAN);

[0093] The reason for choosing GAN here is that it can generate diverse and high-quality design solutions based on complex multi-dimensional requirements in the design of soft decoration projects;

[0094] Convert each customer requirement item in the customer requirement report into JSON format and combine them into a multi-dimensional tensor;

[0095] A multidimensional tensor is a multidimensional array structure that can effectively represent complex data relationships. Each requirement item is mapped to different dimensions of the tensor according to its type and importance. For example, spatial dimensions can be mapped to the first dimension of the tensor, style preferences to the second dimension, budget constraints to the third dimension, and functional requirements to the fourth dimension. In this way, all the information in the customer requirement report can be integrated into a unified input format for use by the GAN model.

[0096] Input the multi-dimensional tensor into the generative adversarial network model to generate a preliminary design sketch;

[0097] After receiving the multidimensional tensor, the generator in the GAN model will generate a preliminary design sketch based on the functional requirements contained therein. The generated preliminary design sketch is not completed in one go, but is a process of gradual optimization through multiple iterations. In each iteration, the discriminator in the GAN model will evaluate the generated sketch to determine whether it meets the actual design standards. If the discriminator believes that the sketch is insufficient, the generator will adjust it based on the feedback and regenerate a new sketch.

[0098] S3. Based on the preliminary design sketch, give a design plan and make adjustments based on customer feedback.

[0099] Getting the adjustment data includes the following steps:

[0100] Select (HoloLens2) AR glasses as the AR device and guide customers to install the AR application;

[0101] The purpose of choosing HoloLens2 here is that it provides an advanced mixed reality experience, with a high-resolution holographic display, built-in environmental perception sensors, and natural human-computer interaction;

[0102] Load the preliminary design sketch onto the AR glasses;

[0103] Convert the preliminary design sketch from the original format (such as JPEG) to a 3D model format suitable for AR glasses display (such as FBX). After the format conversion is completed, the design sketch will be imported into the AR application in the form of a 3D model, and the application will optimize the model (such as adjusting the resolution and texture quality of the model);

[0104] Start the AR glasses to detect the customer's current physical space and perform initial positioning and space mapping;

[0105] When customers put on HoloLens2 and start the AR application, the built-in camera and sensors in the glasses automatically start scanning the physical environment around them (such as capturing the outline of the room, the position of furniture, and the shape of other obstacles), and collect environmental data to build a three-dimensional spatial map (determine the position of virtual objects in the physical space);

[0106] After the spatial mapping is completed, the captured spatial information is converted into a dense spatial grid, and the feature points in the spatial grid are extracted;

[0107] This spatial grid is made up of many small cubes and each cube represents a tiny area in space;

[0108] Extract a series of feature points from the spatial grid (usually located at prominent locations such as walls, floors, ceilings, and furniture edges). The selection of feature points provides stable anchor points for subsequent virtual object placement;

[0109] Based on the feature points, virtual objects are placed in the physical space and rendered;

[0110] According to the extracted feature points, select the appropriate location to place the virtual object. For example, the sofa will be placed in the center of the living room, and the TV cabinet will be placed against the wall. This process can be performed according to preset rules or user instructions.

[0111] Once the virtual object is correctly placed, HoloLens2 will render it to give it a realistic appearance in the AR glasses. The rendering process will take into account factors such as shadows and reflections to ensure that the virtual object blends perfectly with the physical environment.

[0112] Activate the real-time environment perception function of AR glasses, use the depth sensor to detect the lighting conditions, material color and texture changes of virtual objects, and obtain the detection results;

[0113] Activate the camera sensor in the AR glasses (to capture light intensity, direction, and color temperature). Assuming the customer is using the AR glasses in a living room with floor-to-ceiling windows, the camera will capture the natural light coming in from the window as well as the brightness and color temperature of the indoor lights.

[0114] Use high dynamic range imaging (HDR) technology to capture images at different exposure levels. For example, the camera captures the bright areas formed by sunlight shining through the window and the darker parts of the room corners, thereby generating a detailed light map. The light map is analyzed by an image processing algorithm to extract key features of the light (such as using a shadow analysis algorithm to determine the position and intensity of the light source) and generate a lighting model (simulating the effect of sunlight shining in from the window and applying it to the rendering of virtual furniture to make its shadow and reflection effects consistent with the actual environment);

[0115] Use a camera to take photos or videos of the physical space to obtain the color distribution and texture characteristics of the current environment. For example, in a bedroom scene, the camera captures that the wall is off-white, the floor is marble-textured, and the bedside table is maroon.

[0116] Apply computer vision technology (color histogram) to extract the main colors and their distribution from the captured images, and use neural networks to extract texture features to identify the main texture patterns of the physical space (such as marble texture on the ground and velvet texture on the carpet);

[0117] In this step, by real-time detection of the lighting conditions, color, and texture of the environment, it is possible to ensure that the virtual objects are seamlessly integrated with the actual environment, thus enhancing the realism of the environment. Automated detection and adjustment reduces the workload of manual adjustment by the user, thus enhancing the user experience.

[0118] According to the detection results, the lighting conditions, material color and texture changes of virtual objects are dynamically adjusted using the lighting adaptation algorithm and color texture matching technology;

[0119] Apply the previously generated lighting model to the adjustment process of virtual objects (brightness, shadows, and reflection effects). When the lighting model shows that there is strong side light in the actual environment, the shadow of the virtual object should be adjusted accordingly (such as adjusting the shadow of the virtual coffee table to make it consistent with the shadow in the actual environment). Use physical rendering (PBR, accurately simulate the interaction between light and the surface of objects) technology to simulate real lighting effects, so that virtual objects appear natural under different lighting conditions. For example, the metal legs of the virtual coffee table show high reflectivity in strong light and appear dim in weak light, which simulates the characteristics of real metal;

[0120] Update the lighting effects of virtual objects in real time based on the user's movement (mobile AR glasses) and changes in the actual environment. For example, when a user walks from one end of the living room to the other, the shadows and reflections of the living room sofa will automatically adjust according to the new perspective and lighting conditions;

[0121] Match the extracted color and texture features with the material properties of the virtual object to find the closest color and texture combination (based on the color and texture features of the living room, choose a stone material for the virtual coffee table to coordinate with the marble texture of the floor). Based on the matching results, adjust the material properties of the virtual object to make its color and texture consistent with the surrounding environment;

[0122] This step provides users with an initial optimal adaptation state by dynamically adjusting the lighting conditions, color, and texture of virtual objects, while ensuring that the color and texture of virtual objects are consistent with the surrounding environment to avoid abrupt or unnatural effects;

[0123] Customers view the initial effects of dynamically adjusted virtual objects;

[0124] Provide a clear preview interface in the AR application for customers to see the effect of preliminary adjustments;

[0125] In this step, by showing the preliminary effect to the user, the user can more intuitively see how the design scheme performs in the actual environment, providing a good starting point to help the user achieve satisfactory results faster;

[0126] Further adjust virtual objects according to customer needs and record user adjustment data;

[0127] Adjustment data includes data on changing the properties of virtual objects, data on customer interaction methods, and environmental perception data.

[0128] The process of further adjustment can be explained from the following aspects:

[0129] Gesture recognition and drag operation: HoloLens2 supports gesture recognition. Customers can select and move virtual objects through simple hand movements (such as clicking, sliding, and pinching). For example, users can rotate virtual tables and chairs with their fingers to see how they look at different angles.

[0130] Voice commands and color changes: In addition to gestures, customers can also use voice commands to change the color or material of virtual objects. For example, customers can say "change the sofa to brown" or change the texture of the carpet, and HoloLens2 will immediately perform the corresponding operation according to the voice command;

[0131] Eye tracking: HoloLens 2 is also equipped with eye tracking, which allows customers to quickly select and modify a virtual object simply by looking at it and giving instructions. In addition, eye tracking can also be used to browse sample libraries of different materials, allowing customers to choose the most satisfactory material.

[0132] Adjustment data recording: During the entire adjustment process, HoloLens2 records every operation of the customer in real time, including changes in the attributes of virtual objects (such as position, color, material), customer interaction methods (such as gestures, voice, eye movements) and environmental perception data (such as light intensity, sound level).

[0133] S4. Optimize the design plan based on the adjustment data and obtain customer feedback.

[0134] The following steps are included:

[0135] Extract key adjustment features based on adjustment data;

[0136] Dimensionality reduction techniques (PCA principal component analysis) are used to identify key adjustment features (such as customer preferred color range, material type, and spatial layout pattern) from the adjustment data, and association rule mining algorithms are used to discover potential connections between different adjustment features, such as whether the choice of a certain material is always accompanied by a specific color preference;

[0137] Convolutional neural network (CNN) and recurrent neural network (RNN) are selected as hybrid optimization models;

[0138] CNN is good at processing images and visual data and can automatically learn local features in space, while RNN is suitable for processing sequence data. Therefore, the hybrid optimization model formed by the combination of the two can generate a more reasonable optimization plan while considering both static design elements (such as shape and color) and dynamic interactive behaviors (such as the user's operation sequence). It can not only process complex visual information, but also understand the user's behavior pattern, thus improving the quality of the optimization plan.

[0139] The key adjustment features are input into the hybrid optimization model to generate M (M=5) optimization solutions;

[0140] The extracted key adjustment features are converted into numerical form (to be accepted by the hybrid optimization model) and input into the hybrid optimization model. The model will generate 5 different optimization plans based on these features. Each plan contains specific modification suggestions, such as changing the position of a piece of furniture, changing the color of the wall, or adjusting the angle of the light.

[0141] In order to ensure that the generated optimization solutions are diverse and innovative enough, multiple output nodes can be set in the model so that the results generated each time are different;

[0142] Get the latest market trend data from external data sources;

[0143] External data sources include industry reports, social media;

[0144] Using web crawler technology, we regularly capture the latest market trend data (popular colors, materials, styles, etc. in current designs) from external data sources, and conduct in-depth analysis of market trend data through text mining algorithms to extract the most popular elements and future development directions.

[0145] Based on the market trend data, fine-tune the M optimization solutions and obtain the fine-tuning results;

[0146] Cosine similarity and Jaccard coefficient are used to quantitatively measure the fit between the five optimization schemes and the market trend data. For the schemes that do not match the market trend, targeted fine-tuning is carried out according to the analysis results. For example, if the navy blue color is popular in the market, the application ratio of the navy blue color can be increased in the scheme. If self-luminous materials (fluorescent coatings, LED optical fibers) are widely welcomed, they can be considered to be included in the design scheme.

[0147] Based on the fine-tuning results, the M optimization solutions are converted into high-quality and interactive two-dimensional images using a rendering engine (V-Ray), and a virtual roaming video is generated;

[0148] Prepare the materials needed for rendering. Users can choose preset materials or customize material properties such as color, reflectivity, transparency, roughness, etc. For specific materials (such as wood, glass, etc.), PBR materials can be used to ensure that they exhibit real physical properties under different lighting conditions;

[0149] According to the spatial layout, atmosphere requirements and rendering materials of the design, light sources (such as point light sources, regional light sources, and solar light sources) can be reasonably configured. Each light source can set different parameters, such as brightness, color, shadow intensity, etc., to simulate real lighting effects. For example, in the living room scene, warm-toned lights can be used to create a warm atmosphere, while in the kitchen scene, cool-toned lights can be used to highlight the sense of cleanliness.

[0150] Choose the appropriate camera angle according to customer needs. Camera angles include bird's-eye view (top view), horizontal view (front view), oblique view (perspective view), etc. Each angle can highlight different design elements and help customers understand the space layout and details from multiple angles. For example, a bird's-eye view can show the whole room, while a horizontal view can highlight the decorative effect of a specific area. After determining the angle, use the rule of thirds to further adjust the composition to ensure the balance and beauty of the image. In addition, you can also add some auxiliary elements (such as people, plants, ornaments, etc.) to enrich the content of the picture and add a sense of life;

[0151] Plan the camera's movement path (covering the main areas and highlights of the design). The path can be linear (such as walking from one room to another) or non-linear (such as rotating around a central point). Set appropriate animation parameters (such as speed and smoothness) for the camera's movement to ensure that the camera's movement is smooth and natural, avoiding abrupt jumps or freezes. In addition, you can also add animation effects (such as opening and closing, rotation) to the objects in the scene to increase the dynamics of the video;

[0152] Show 2D images and virtual walkthrough videos to customers through AR applications;

[0153] Integrate rendered 2D images and virtual tour videos into a complete presentation package. The application automatically adjusts the display mode according to the customer's device type (such as mobile phone, tablet). In addition to basic viewing functions, AR applications also support users to interact with 3D models, such as clicking to view details, dragging to change the perspective, etc. In addition, some interesting interactive links (such as choosing your favorite color) can be set to increase customer participation;

[0154] The customer selects the required optimization solution and submits feedback to obtain customer feedback results;

[0155] Customer feedback provides designers with direct suggestions for improvement, which helps to continuously optimize design solutions and service processes and achieve the goal of continuous improvement.

[0156] S5. Based on customer feedback, confirm the final design plan and generate project documents.

[0157] The following steps are included:

[0158] The discriminator is used in combination with key adjustment features to calculate the quality score of the optimization solution required by the customer and perform normalization. The expression is:

[0159]

[0160] in, represents the quality score of the optimization solution required by the customer, j represents the index variable for traversing the key adjustment features, and J represents the total number of key adjustment features. Indicates the performance of the optimization solution required by the customer on the jth feature. The value range is usually between 0 and 1, indicating the degree of satisfaction of the feature. Indicates the weight of the key adjustment feature. By introducing weights, some important features can be highlighted in the evaluation, while relatively minor features can be weakened. For example, if the customer attaches great importance to the spatial layout, the weight of the spatial layout feature may be higher. The normalization factor of the quality score of the optimization solution required by the customer, used to convert the quality score Normalize to a fixed range;

[0161] Set a threshold value of the quality standard (0.6 to 0.9), and judge whether the optimization solution required by the customer meets the customer's expectations based on whether the quality score of the optimization solution is within the threshold range;

[0162] When the quality score of the optimization solution required by the customer is greater than or equal to 0.9, it means that this optimization solution (excellent) fully meets the customer's expectations, and the customer is advised to confirm it as the final design solution;

[0163] When the quality score of the optimization solution required by the customer is greater than or equal to 0.9 and less than 0.7, it means that this optimization solution (good) basically meets the customer's expectations. According to the quality score results, a detailed fine-tuning task list (with examples) is compiled, including recommending a more durable alternative material (aerogel) if the customer has concerns about the durability of a certain floor material, and recommending embedding more lockers in the wall or furniture if the customer wants to increase storage space;

[0164] When the quality score of the optimization solution required by the customer is greater than or equal to 0.6 and less than 0.7, it means that this optimization solution (qualified) only meets part of the customer's expectations. Multiple optimization options are provided for the customer to choose from, and each option is accompanied by a detailed description:

[0165] Suppose the client is a retailer of a fashion brand and wants to design an attractive shopping environment for its new store. The initial solution generated is rated as qualified. The client has a certain degree of recognition of the overall design, but has proposed suggestions for improvement in some aspects. For example, the client believes that the design of the store's display area is not attractive enough and cannot highlight the characteristics of the products. At the same time, the client has some concerns about the location of the cashier area and feels that it is not conspicuous enough. According to the client's requirements, multiple optimization options are provided, including redesigning the layout of the display area and adjusting the location of the cashier area. The client selects one of the optimization solutions and views the adjusted renderings. At the same time, the quality score is recalculated. The final solution meets the good standard and the client confirms that the solution is the final design solution.

[0166] When the quality score of the optimization solution required by the customer is less than 0.6, it means that this optimization solution (unqualified) does not meet the customer's expectations. At this time, it is necessary to explain the design ideas, expected effects and cost changes of this solution to the customer, and the customer chooses the most appropriate optimization solution according to his own needs and budget:

[0167] Suppose the client is a chain restaurant operator who wants to design a warm and comfortable dining space for his restaurant. However, the initial scheme generated was rated as unqualified and the client was very dissatisfied with the overall design. The client believed that the distance between the tables was too small, which affected the dining experience of the customers and the lighting design of the restaurant was too glaring.

[0168] According to customer requirements, the following design ideas are proposed:

[0169] By increasing the distance between tables, customers’ dining comfort can be improved, and changing the soft lighting design can create a warm atmosphere, which is suitable for couples to date or family dinners (cost is 15,140 yuan);

[0170] Semi-open private rooms are set up to meet customers' need for privacy, and intelligent lighting control can provide a personalized dining experience suitable for different occasions (cost is 23,585 yuan);

[0171] Divide the restaurant into different themed areas, such as the garden area, the city night view area, etc. Each area has a unique decoration style and lighting design to attract different types of customers (cost 29,970 yuan);

[0172] Expected effect: Use renderings to show how the restaurant will look after adjusting the layout and lighting. Customers can see that the spacing between tables is wider and the lighting is softer, creating a warm dining atmosphere.

[0173] Provide interior renderings of the private room, customers can see the operation interface under intelligent lighting control, and customers can adjust the brightness and color of the lights according to their needs;

[0174] Through virtual tour videos showing the effects of different themed areas, customers can feel the unique style and atmosphere of each area;

[0175] Finally, the customer chose the semi-open private room + intelligent lighting control solution, believing that this would not only enhance the dining experience of customers, but also increase the attractiveness of the restaurant. A new design solution was generated based on the customer's choice, and the quality score was re-evaluated. It reached the excellent standard and the customer confirmed it as the final design solution.

[0176] Confirm the final design plan based on the judgment results and generate project documents;

[0177] Project documents include construction drawings (plan, elevation, section, node details), material list, furniture and decoration list, and total project budget (material cost, labor cost, transportation cost, installation cost);

[0178] The detailed budget and material information in this step makes the project's cost structure more transparent, allowing customers to clearly understand where every expense goes, enhancing their trust and satisfaction.

[0179] S6. Carry out construction according to project documents and obtain a complete project report.

[0180] The following steps are included:

[0181] Provide MagicLeap devices (augmented reality devices) to the construction team;

[0182] The construction team can use MagicLeap devices to view the 3D model of the design in real time at the construction site. For example, when laying floors or walls, construction workers can use the device to view virtual elevation lines to ensure construction accuracy.

[0183] Install IoT sensors at the construction site and set the data collection frequency for each sensor;

[0184] IoT sensors include camera sensors, temperature and humidity sensors, air quality sensors, and vibration sensors;

[0185] This step can detect construction problems in advance and make timely adjustments through real-time data collection, avoiding rework at a later stage and greatly reducing construction costs;

[0186] Use IoT sensors to collect construction data, evaluate construction quality, identify risk points, and obtain monitoring results;

[0187] The assessment of construction quality includes: for certain temperature-sensitive materials (such as paint and glue), set an appropriate construction temperature range. For example, the optimal construction temperature of paint is usually 15°C to 25°C. If the on-site temperature is lower than 15°C or higher than 25°C, remind the construction workers to suspend the painting operation to avoid uneven paint surface due to low temperature or too high temperature causing the paint surface to dry too quickly, which affects the construction quality;

[0188] For processes that require a dry environment (such as laying wooden floors and pasting wallpaper), set an appropriate humidity range. For example, when laying wooden floors, the air humidity should be controlled between 40% and 60%. If the humidity is too high, remind the construction workers to suspend laying to prevent the wooden floors from being damp and deformed. If the humidity is too low, it is recommended to take humidification measures to prevent the wooden floors from being over-dried and cracked.

[0189] When using boards or furniture containing formaldehyde, monitor the concentration of formaldehyde in the air. For example, the concentration of formaldehyde in indoor air should not exceed 0.1 mg / m 3 (National Standard), if the formaldehyde concentration is detected to be excessive, remind the construction workers to take measures (such as increasing ventilation time or using low-formaldehyde materials) to ensure that the air quality meets safety standards;

[0190] Risk point identification includes: When construction is carried out near sensitive areas such as residential areas, schools, and hospitals, a stricter vibration intensity range is set to protect the surrounding environment and the quality of life of residents. For example, when construction is carried out near residential areas, the vibration intensity threshold is set to 0.3m / s 2 If the vibration intensity exceeds 0.3m / s 2 , remind construction workers to suspend construction, and suggest adjusting construction time and using low-noise equipment to reduce the impact of vibration and avoid disrupting the lives of surrounding residents;

[0191] Based on the monitoring results and the immediate needs of the client, remote collaboration with the designer was carried out through the MagicLeap device to adjust the design plan again;

[0192] Provide a remote collaboration platform based on MagicLeap devices, through which customers, designers and construction teams can communicate and collaborate in real time;

[0193] Based on the monitoring results and the immediate needs of the customer, the designer can use the MagicLeap device to view the situation on the construction site in real time and adjust the design plan based on the actual situation. For example, if the color or texture of a certain material does not meet the customer's expectations, the designer can use the device to show the effects of different materials to help the customer make a better choice.

[0194] The difference between the re-adjustment of the design plan here and the optimization plan selected by the customer before is that one is when the customer puts forward new requirements after seeing the construction site during the construction process (for example, during the construction process, the customer finds that the window position of a room does not meet expectations, affecting the lighting effect. The designer checks the 3D model of the construction site in real time and proposes a new adjustment plan for the window position, and immediately communicates with the construction team to improve or perfect the newly proposed adjustment plan), and the other is when the customer puts forward preliminary opinions in the early stage of the design or the designer finds that certain design elements need to be optimized;

[0195] Deliver the adjusted design plan to the construction team for construction;

[0196] After the project construction is completed, a complete project report is generated;

[0197] The complete project report includes project overview, overall design plan (project background, customer needs, design ideas, construction period), construction process (construction photos, videos, progress reports), final effect (real photos, customer feedback) and maintenance guide (material maintenance, equipment maintenance, cleaning methods);

[0198] Through this step, through a complete project report, customers can fully understand the entire process of the project, including design, construction, final effect, etc., which enhances their trust and satisfaction.

[0199] This embodiment also provides a design data interaction system for soft decoration engineering, including: a data collection module, collecting customers' personalized demand data and historical data, and generating customer demand reports;

[0200] Sketch design module, generates preliminary design sketches based on customer demand reports;

[0201] The scheme adjustment module provides a design scheme based on the preliminary design sketch and adjusts it according to customer opinions;

[0202] Optimization module, which optimizes the adjusted design solution and obtains customer feedback results;

[0203] Solution confirmation module: based on customer feedback, confirms the final design solution and generates project documents;

[0204] Construction module, carry out construction according to project documents and get a complete project report.

[0205] This embodiment also provides a computer device, which is suitable for the design data interaction method for soft decoration engineering, including: a memory and a processor; the memory is used to store computer executable instructions, and the processor is used to execute computer executable instructions to implement the design data interaction method for soft decoration engineering proposed in the above embodiment.

[0206] The computer device may be a terminal, and the computer device includes a processor, a memory, a communication interface, a display screen and an input device connected via a system bus. The processor of the computer device 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 and a computer program. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The communication interface of the computer device is used to communicate with an external terminal in a wired or wireless manner, and the wireless manner can be achieved through WIFI, an operator network, NFC (near field communication) or other technologies. The display screen of the computer device may be a liquid crystal display screen or an electronic ink display screen, and the input device of the computer device may be a touch layer covering the display screen, or a key, trackball or touchpad provided on the housing of the computer device, or an external keyboard, touchpad or mouse, etc.

[0207] This embodiment also provides a storage medium on which a computer program is stored. When the program is executed by a processor, the design data interaction method for soft decoration engineering proposed in the above embodiment is implemented; the storage medium can be implemented by any type of volatile or non-volatile storage device or a combination of them, such as static random access memory (Static Random Access Memory, referred to as SRAM), electrically erasable programmable read-only memory (Electrically Erasable Programmable Read-Only Memory, referred to as EEPROM), erasable programmable read-only memory (Erasable Programmable Read Only Memory, referred to as EPROM), programmable read-only memory (Programmable Red-Only Memory, referred to as PROM), read-only memory (Read-Only Memory, referred to as ROM), magnetic storage, flash memory, magnetic disk or optical disk.

[0208] In summary, the present invention provides a specific design scheme through a preliminary design sketch and adjusts it according to customer opinions to obtain adjustment data, optimizes the design scheme according to the adjustment data, obtains customer feedback results, generates multiple optimization schemes through the optimization model, ensures that each scheme is scientific and reasonable, and combines the introduction of market trend data to not only consider the personalized needs of customers, but also take into account the latest market trends, making the optimization scheme more innovative. Customers can choose the scheme that best meets their expectations and submit feedback. Designers further adjust the design scheme based on the feedback and finally confirm the best scheme. This intelligent optimization process not only improves the accuracy and efficiency of the design, but also ensures the personalization and foresight of the design scheme, greatly improving customer satisfaction and project success rate.

[0209] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present invention, which should all be included in the scope of the claims of the present invention.

Claims

1. A design data interaction method for soft decoration engineering, characterized by: include, Collect customers' personalized demand data and historical data and generate customer demand reports; Generate preliminary design sketches based on customer demand reports; Based on the preliminary design sketch, give a design plan and adjust it according to customer opinions to obtain adjustment data; Optimize the design plan based on the adjustment data and obtain customer feedback; Based on customer feedback, confirm the final design plan and generate project documents; Carry out construction according to project documents and get a complete project report.

2. The design data interaction method for soft decoration engineering according to claim 1, characterized in that: Collect customers' personalized demand data and historical data, and generate customer demand reports. The following steps are included: The customer's personalized demand data includes space size data, style preference data, budget limitation data and functional demand data; The historical data includes preference record data, purchase behavior data, and communication record data left by customers in past interactions with designers; The customer logs in to the design platform and fills in the project information form; According to the project information form filled out by the customer, the BERT model is used to parse the information filled out by the customer to obtain the parsing result; According to the analysis results, it is compiled into a customer demand report.

3. The design data interaction method for soft decoration engineering according to claim 2, characterized in that: According to the customer demand report, generate a preliminary design sketch, including the following steps, Select a generative adversarial network model; Convert each customer requirement item in the customer requirement report into JSON format and combine them into a multi-dimensional tensor; The multi-dimensional tensor is input into the generative adversarial network model to generate a preliminary design sketch.

4. The design data interaction method for soft decoration engineering according to claim 3, characterized in that: Based on the preliminary design sketch, a design plan is given and adjusted according to customer opinions to obtain adjustment data, including the following steps: Select AR glasses as the AR device and guide customers to install the AR application; Load the preliminary design sketch onto the AR glasses; Start the AR glasses to detect the customer's current physical space and perform initial positioning and spatial mapping; After the spatial mapping is completed, the captured spatial information is converted into a dense spatial grid, and the feature points in the spatial grid are extracted; Based on the feature points, virtual objects are placed in the physical space and rendered; Activate the real-time environment perception function of the AR glasses, use the camera sensor to detect the lighting conditions, material color and texture changes of the virtual objects, and obtain the detection results; According to the detection results, the lighting conditions, material color and texture changes of virtual objects are dynamically adjusted using the lighting adaptation algorithm and color texture matching technology; Customers view the initial effects of dynamically adjusted virtual objects; Further adjust virtual objects according to customer needs and record user adjustment data; The adjustment data includes data for changing the properties of virtual objects, customer interaction data, and environment perception data.

5. The design data interaction method for soft decoration engineering according to claim 4, characterized in that: Optimize the design scheme according to the adjustment data and obtain customer feedback results, including the following steps: Extract key adjustment features based on adjustment data; Convolutional neural network and recurrent neural network are selected as hybrid optimization models; Input the key adjustment features into the hybrid optimization model to generate M optimization solutions; Get the latest market trend data from external data sources; Said external data sources include industry reports, social media; Based on the market trend data, fine-tune the M optimization solutions and obtain the fine-tuning results; Based on the fine-tuning results, a rendering engine is used to convert the M optimization solutions into high-quality and interactive two-dimensional images and generate virtual roaming videos. Show 2D images and virtual walkthrough videos to customers through AR applications; The customer selects the required optimization solution and submits feedback to obtain customer feedback results.

6. The design data interaction method for soft decoration engineering according to claim 5, characterized in that: Based on customer feedback, confirm the final design plan and generate project documents. The following steps are included: The discriminator is used in combination with key adjustment features to calculate the quality score of the optimization solution required by the customer and perform normalization. The expression is: in, represents the quality score of the optimization solution required by the customer, j represents the index variable for traversing the key adjustment features, and J represents the total number of key adjustment features. Indicates the performance of the optimization solution required by the customer on the jth feature, represents the weight of the key adjustment feature, A normalization factor representing the quality score of the optimization solution required by the customer; Set a threshold for the quality standard and determine whether the solution meets the customer's expectations based on whether the quality score of the optimization solution required by the customer is within the threshold. Confirm the final design plan based on the judgment results and generate project documents; The project documents include construction drawings, material lists, furniture and decoration lists, and the total budget for the project.

7. The design data interaction method for soft decoration engineering according to claim 6, characterized in that: Carry out construction according to the project documents and get a complete project report, including the following steps, Provide MagicLeap devices to construction teams; Install IoT sensors at the construction site and set the data collection frequency for each sensor; The IoT sensors include camera sensors, temperature and humidity sensors, air quality sensors, and vibration sensors; Use IoT sensors to collect construction data, evaluate construction quality, identify risk points, and obtain monitoring results; Based on the monitoring results and the immediate needs of the client, remote collaboration with the designer was carried out through the MagicLeap device to adjust the design plan again; Deliver the adjusted design plan to the construction team for construction; After the project construction is completed, a complete project report is generated; The complete project report includes project overview, overall design plan, construction process, final effect and maintenance guidelines.

8. A design data interaction system for soft decoration engineering, based on the design data interaction method for soft decoration engineering according to any one of claims 1 to 7, characterized in that: include, Data collection module, which collects customers’ personalized demand data and historical data and generates customer demand reports; Sketch design module, generates preliminary design sketches based on customer demand reports; The scheme adjustment module provides a design scheme based on the preliminary design sketch and adjusts it according to customer opinions; Optimization module, which optimizes the adjusted design solution and obtains customer feedback results; Solution confirmation module: based on customer feedback, confirms the final design solution and generates project documents; Construction module, carry out construction according to project documents and get a complete project report.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the design data interaction method for soft decoration engineering described in 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 design data interaction method for soft decoration engineering described in any one of claims 1 to 7 are implemented.

Images